diff --git a/Documentation/DocBook/genericirq.tmpl b/Documentation/DocBook/genericirq.tmpl
index 1448b33..fb10fd0 100644
--- a/Documentation/DocBook/genericirq.tmpl
+++ b/Documentation/DocBook/genericirq.tmpl
@@ -28,7 +28,7 @@
   </authorgroup>
 
   <copyright>
-   <year>2005-2006</year>
+   <year>2005-2010</year>
    <holder>Thomas Gleixner</holder>
   </copyright>
   <copyright>
@@ -100,6 +100,10 @@
 	  <listitem><para>Edge type</para></listitem>
 	  <listitem><para>Simple type</para></listitem>
 	</itemizedlist>
+	During the implementation we identified another type:
+	<itemizedlist>
+	  <listitem><para>Fast EOI type</para></listitem>
+	</itemizedlist>
 	In the SMP world of the __do_IRQ() super-handler another type
 	was identified:
 	<itemizedlist>
@@ -153,6 +157,7 @@
 	is still available. This leads to a kind of duality for the time
 	being. Over time the new model should be used in more and more
 	architectures, as it enables smaller and cleaner IRQ subsystems.
+	It's deprecated for three years now and about to be removed.
 	</para>
   </chapter>
   <chapter id="bugs">
@@ -217,6 +222,7 @@
 	  <itemizedlist>
 	  <listitem><para>handle_level_irq</para></listitem>
 	  <listitem><para>handle_edge_irq</para></listitem>
+	  <listitem><para>handle_fasteoi_irq</para></listitem>
 	  <listitem><para>handle_simple_irq</para></listitem>
 	  <listitem><para>handle_percpu_irq</para></listitem>
 	  </itemizedlist>
@@ -233,33 +239,33 @@
 		are used by the default flow implementations.
 		The following helper functions are implemented (simplified excerpt):
 		<programlisting>
-default_enable(irq)
+default_enable(struct irq_data *data)
 {
-	desc->chip->unmask(irq);
+	desc->chip->irq_unmask(data);
 }
 
-default_disable(irq)
+default_disable(struct irq_data *data)
 {
-	if (!delay_disable(irq))
-		desc->chip->mask(irq);
+	if (!delay_disable(data))
+		desc->chip->irq_mask(data);
 }
 
-default_ack(irq)
+default_ack(struct irq_data *data)
 {
-	chip->ack(irq);
+	chip->irq_ack(data);
 }
 
-default_mask_ack(irq)
+default_mask_ack(struct irq_data *data)
 {
-	if (chip->mask_ack) {
-		chip->mask_ack(irq);
+	if (chip->irq_mask_ack) {
+		chip->irq_mask_ack(data);
 	} else {
-		chip->mask(irq);
-		chip->ack(irq);
+		chip->irq_mask(data);
+		chip->irq_ack(data);
 	}
 }
 
-noop(irq)
+noop(struct irq_data *data))
 {
 }
 
@@ -278,12 +284,27 @@
 		<para>
 		The following control flow is implemented (simplified excerpt):
 		<programlisting>
-desc->chip->start();
+desc->chip->irq_mask();
 handle_IRQ_event(desc->action);
-desc->chip->end();
+desc->chip->irq_unmask();
 		</programlisting>
 		</para>
-   	    </sect3>
+	    </sect3>
+	    <sect3 id="Default_FASTEOI_IRQ_flow_handler">
+		<title>Default Fast EOI IRQ flow handler</title>
+		<para>
+		handle_fasteoi_irq provides a generic implementation
+		for interrupts, which only need an EOI at the end of
+		the handler
+		</para>
+		<para>
+		The following control flow is implemented (simplified excerpt):
+		<programlisting>
+handle_IRQ_event(desc->action);
+desc->chip->irq_eoi();
+		</programlisting>
+		</para>
+	    </sect3>
 	    <sect3 id="Default_Edge_IRQ_flow_handler">
 	 	<title>Default Edge IRQ flow handler</title>
 		<para>
@@ -294,20 +315,19 @@
 		The following control flow is implemented (simplified excerpt):
 		<programlisting>
 if (desc->status &amp; running) {
-	desc->chip->hold();
+	desc->chip->irq_mask();
 	desc->status |= pending | masked;
 	return;
 }
-desc->chip->start();
+desc->chip->irq_ack();
 desc->status |= running;
 do {
 	if (desc->status &amp; masked)
-		desc->chip->enable();
+		desc->chip->irq_unmask();
 	desc->status &amp;= ~pending;
 	handle_IRQ_event(desc->action);
 } while (status &amp; pending);
 desc->status &amp;= ~running;
-desc->chip->end();
 		</programlisting>
 		</para>
    	    </sect3>
@@ -342,9 +362,9 @@
 		<para>
 		The following control flow is implemented (simplified excerpt):
 		<programlisting>
-desc->chip->start();
 handle_IRQ_event(desc->action);
-desc->chip->end();
+if (desc->chip->irq_eoi)
+        desc->chip->irq_eoi();
 		</programlisting>
 		</para>
    	    </sect3>
@@ -375,8 +395,7 @@
 	mechanism. (It's necessary to enable CONFIG_HARDIRQS_SW_RESEND when
 	you want to use the delayed interrupt disable feature and your
 	hardware is not capable of retriggering	an interrupt.)
-	The delayed interrupt disable can be runtime enabled, per interrupt,
-	by setting the IRQ_DELAYED_DISABLE flag in the irq_desc status field.
+	The delayed interrupt disable is not configurable.
 	</para>
 	</sect2>
     </sect1>
@@ -387,13 +406,13 @@
 	contains all the direct chip relevant functions, which
 	can be utilized by the irq flow implementations.
 	  <itemizedlist>
-	  <listitem><para>ack()</para></listitem>
-	  <listitem><para>mask_ack() - Optional, recommended for performance</para></listitem>
-	  <listitem><para>mask()</para></listitem>
-	  <listitem><para>unmask()</para></listitem>
-	  <listitem><para>retrigger() - Optional</para></listitem>
-	  <listitem><para>set_type() - Optional</para></listitem>
-	  <listitem><para>set_wake() - Optional</para></listitem>
+	  <listitem><para>irq_ack()</para></listitem>
+	  <listitem><para>irq_mask_ack() - Optional, recommended for performance</para></listitem>
+	  <listitem><para>irq_mask()</para></listitem>
+	  <listitem><para>irq_unmask()</para></listitem>
+	  <listitem><para>irq_retrigger() - Optional</para></listitem>
+	  <listitem><para>irq_set_type() - Optional</para></listitem>
+	  <listitem><para>irq_set_wake() - Optional</para></listitem>
 	  </itemizedlist>
 	These primitives are strictly intended to mean what they say: ack means
 	ACK, masking means masking of an IRQ line, etc. It is up to the flow
@@ -458,6 +477,7 @@
      <para>
      This chapter contains the autogenerated documentation of the internal functions.
      </para>
+!Ikernel/irq/irqdesc.c
 !Ikernel/irq/handle.c
 !Ikernel/irq/chip.c
   </chapter>
diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl
index a0d479d..f66f4df 100644
--- a/Documentation/DocBook/kernel-locking.tmpl
+++ b/Documentation/DocBook/kernel-locking.tmpl
@@ -1645,7 +1645,9 @@
       all the readers who were traversing the list when we deleted the
       element are finished.  We use <function>call_rcu()</function> to
       register a callback which will actually destroy the object once
-      the readers are finished.
+      all pre-existing readers are finished.  Alternatively,
+      <function>synchronize_rcu()</function> may be used to block until
+      all pre-existing are finished.
     </para>
     <para>
       But how does Read Copy Update know when the readers are
@@ -1714,7 +1716,7 @@
 -        object_put(obj);
 +        list_del_rcu(&amp;obj-&gt;list);
          cache_num--;
-+        call_rcu(&amp;obj-&gt;rcu, cache_delete_rcu, obj);
++        call_rcu(&amp;obj-&gt;rcu, cache_delete_rcu);
  }
 
  /* Must be holding cache_lock */
@@ -1725,14 +1727,6 @@
          if (++cache_num > MAX_CACHE_SIZE) {
                  struct object *i, *outcast = NULL;
                  list_for_each_entry(i, &amp;cache, list) {
-@@ -85,6 +94,7 @@
-         obj-&gt;popularity = 0;
-         atomic_set(&amp;obj-&gt;refcnt, 1); /* The cache holds a reference */
-         spin_lock_init(&amp;obj-&gt;lock);
-+        INIT_RCU_HEAD(&amp;obj-&gt;rcu);
-
-         spin_lock_irqsave(&amp;cache_lock, flags);
-         __cache_add(obj);
 @@ -104,12 +114,11 @@
  struct object *cache_find(int id)
  {
diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt
index 790d1a8..0c134f8 100644
--- a/Documentation/RCU/checklist.txt
+++ b/Documentation/RCU/checklist.txt
@@ -218,13 +218,22 @@
 	include:
 
 	a.	Keeping a count of the number of data-structure elements
-		used by the RCU-protected data structure, including those
-		waiting for a grace period to elapse.  Enforce a limit
-		on this number, stalling updates as needed to allow
-		previously deferred frees to complete.
+		used by the RCU-protected data structure, including
+		those waiting for a grace period to elapse.  Enforce a
+		limit on this number, stalling updates as needed to allow
+		previously deferred frees to complete.	Alternatively,
+		limit only the number awaiting deferred free rather than
+		the total number of elements.
 
-		Alternatively, limit only the number awaiting deferred
-		free rather than the total number of elements.
+		One way to stall the updates is to acquire the update-side
+		mutex.	(Don't try this with a spinlock -- other CPUs
+		spinning on the lock could prevent the grace period
+		from ever ending.)  Another way to stall the updates
+		is for the updates to use a wrapper function around
+		the memory allocator, so that this wrapper function
+		simulates OOM when there is too much memory awaiting an
+		RCU grace period.  There are of course many other
+		variations on this theme.
 
 	b.	Limiting update rate.  For example, if updates occur only
 		once per hour, then no explicit rate limiting is required,
@@ -365,3 +374,26 @@
 	and the compiler to freely reorder code into and out of RCU
 	read-side critical sections.  It is the responsibility of the
 	RCU update-side primitives to deal with this.
+
+17.	Use CONFIG_PROVE_RCU, CONFIG_DEBUG_OBJECTS_RCU_HEAD, and
+	the __rcu sparse checks to validate your RCU code.  These
+	can help find problems as follows:
+
+	CONFIG_PROVE_RCU: check that accesses to RCU-protected data
+		structures are carried out under the proper RCU
+		read-side critical section, while holding the right
+		combination of locks, or whatever other conditions
+		are appropriate.
+
+	CONFIG_DEBUG_OBJECTS_RCU_HEAD: check that you don't pass the
+		same object to call_rcu() (or friends) before an RCU
+		grace period has elapsed since the last time that you
+		passed that same object to call_rcu() (or friends).
+
+	__rcu sparse checks: tag the pointer to the RCU-protected data
+		structure with __rcu, and sparse will warn you if you
+		access that pointer without the services of one of the
+		variants of rcu_dereference().
+
+	These debugging aids can help you find problems that are
+	otherwise extremely difficult to spot.
diff --git a/Documentation/RCU/stallwarn.txt b/Documentation/RCU/stallwarn.txt
index 44c6dcc..862c08e 100644
--- a/Documentation/RCU/stallwarn.txt
+++ b/Documentation/RCU/stallwarn.txt
@@ -80,6 +80,24 @@
 o	For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel
 	without invoking schedule().
 
+o	A CPU-bound real-time task in a CONFIG_PREEMPT kernel, which might
+	happen to preempt a low-priority task in the middle of an RCU
+	read-side critical section.   This is especially damaging if
+	that low-priority task is not permitted to run on any other CPU,
+	in which case the next RCU grace period can never complete, which
+	will eventually cause the system to run out of memory and hang.
+	While the system is in the process of running itself out of
+	memory, you might see stall-warning messages.
+
+o	A CPU-bound real-time task in a CONFIG_PREEMPT_RT kernel that
+	is running at a higher priority than the RCU softirq threads.
+	This will prevent RCU callbacks from ever being invoked,
+	and in a CONFIG_TREE_PREEMPT_RCU kernel will further prevent
+	RCU grace periods from ever completing.  Either way, the
+	system will eventually run out of memory and hang.  In the
+	CONFIG_TREE_PREEMPT_RCU case, you might see stall-warning
+	messages.
+
 o	A bug in the RCU implementation.
 
 o	A hardware failure.  This is quite unlikely, but has occurred
diff --git a/Documentation/RCU/trace.txt b/Documentation/RCU/trace.txt
index efd8cc9..a851118 100644
--- a/Documentation/RCU/trace.txt
+++ b/Documentation/RCU/trace.txt
@@ -125,6 +125,17 @@
 	of RCU callbacks is ready to invoke, then the remainder will
 	be deferred.
 
+o	"ci" is the number of RCU callbacks that have been invoked for
+	this CPU.  Note that ci+ql is the number of callbacks that have
+	been registered in absence of CPU-hotplug activity.
+
+o	"co" is the number of RCU callbacks that have been orphaned due to
+	this CPU going offline.
+
+o	"ca" is the number of RCU callbacks that have been adopted due to
+	other CPUs going offline.  Note that ci+co-ca+ql is the number of
+	RCU callbacks registered on this CPU.
+
 There is also an rcu/rcudata.csv file with the same information in
 comma-separated-variable spreadsheet format.
 
@@ -180,7 +191,7 @@
 
 o	"jfq" is the number of jiffies remaining for this grace period
 	before force_quiescent_state() is invoked to help push things
-	along.  Note that CPUs in dyntick-idle mode thoughout the grace
+	along.  Note that CPUs in dyntick-idle mode throughout the grace
 	period will not report on their own, but rather must be check by
 	some other CPU via force_quiescent_state().
 
diff --git a/Documentation/cputopology.txt b/Documentation/cputopology.txt
index f1c5c4b..902d315 100644
--- a/Documentation/cputopology.txt
+++ b/Documentation/cputopology.txt
@@ -14,25 +14,39 @@
 	identifier (rather than the kernel's).  The actual value is
 	architecture and platform dependent.
 
-3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
+3) /sys/devices/system/cpu/cpuX/topology/book_id:
+
+	the book ID of cpuX. Typically it is the hardware platform's
+	identifier (rather than the kernel's).	The actual value is
+	architecture and platform dependent.
+
+4) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
 
 	internel kernel map of cpuX's hardware threads within the same
 	core as cpuX
 
-4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
+5) /sys/devices/system/cpu/cpuX/topology/core_siblings:
 
 	internal kernel map of cpuX's hardware threads within the same
 	physical_package_id.
 
+6) /sys/devices/system/cpu/cpuX/topology/book_siblings:
+
+	internal kernel map of cpuX's hardware threads within the same
+	book_id.
+
 To implement it in an architecture-neutral way, a new source file,
-drivers/base/topology.c, is to export the 4 attributes.
+drivers/base/topology.c, is to export the 4 or 6 attributes. The two book
+related sysfs files will only be created if CONFIG_SCHED_BOOK is selected.
 
 For an architecture to support this feature, it must define some of
 these macros in include/asm-XXX/topology.h:
 #define topology_physical_package_id(cpu)
 #define topology_core_id(cpu)
+#define topology_book_id(cpu)
 #define topology_thread_cpumask(cpu)
 #define topology_core_cpumask(cpu)
+#define topology_book_cpumask(cpu)
 
 The type of **_id is int.
 The type of siblings is (const) struct cpumask *.
@@ -45,6 +59,9 @@
 3) thread_siblings: just the given CPU
 4) core_siblings: just the given CPU
 
+For architectures that don't support books (CONFIG_SCHED_BOOK) there are no
+default definitions for topology_book_id() and topology_book_cpumask().
+
 Additionally, CPU topology information is provided under
 /sys/devices/system/cpu and includes these files.  The internal
 source for the output is in brackets ("[]").
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 842aa9d..5e2bc4a 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -386,34 +386,6 @@
 
 ----------------------------
 
-What:	Support for VMware's guest paravirtuliazation technique [VMI] will be
-	dropped.
-When:	2.6.37 or earlier.
-Why:	With the recent innovations in CPU hardware acceleration technologies
-	from Intel and AMD, VMware ran a few experiments to compare these
-	techniques to guest paravirtualization technique on VMware's platform.
-	These hardware assisted virtualization techniques have outperformed the
-	performance benefits provided by VMI in most of the workloads. VMware
-	expects that these hardware features will be ubiquitous in a couple of
-	years, as a result, VMware has started a phased retirement of this
-	feature from the hypervisor. We will be removing this feature from the
-	Kernel too. Right now we are targeting 2.6.37 but can retire earlier if
-	technical reasons (read opportunity to remove major chunk of pvops)
-	arise.
-
-	Please note that VMI has always been an optimization and non-VMI kernels
-	still work fine on VMware's platform.
-	Latest versions of VMware's product which support VMI are,
-	Workstation 7.0 and VSphere 4.0 on ESX side, future maintainence
-	releases for these products will continue supporting VMI.
-
-	For more details about VMI retirement take a look at this,
-	http://blogs.vmware.com/guestosguide/2009/09/vmi-retirement.html
-
-Who:	Alok N Kataria <akataria@vmware.com>
-
-----------------------------
-
 What:	Support for lcd_switch and display_get in asus-laptop driver
 When:	March 2010
 Why:	These two features use non-standard interfaces. There are the
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 8dd7248..3a0009e 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -455,7 +455,7 @@
 			[ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
 				pxa_timer,timer3,32k_counter,timer0_1
 			[AVR32] avr32
-			[X86-32] pit,hpet,tsc,vmi-timer;
+			[X86-32] pit,hpet,tsc;
 				scx200_hrt on Geode; cyclone on IBM x440
 			[MIPS] MIPS
 			[PARISC] cr16
@@ -2153,6 +2153,11 @@
 			Reserves a hole at the top of the kernel virtual
 			address space.
 
+	reservelow=	[X86]
+			Format: nn[K]
+			Set the amount of memory to reserve for BIOS at
+			the bottom of the address space.
+
 	reset_devices	[KNL] Force drivers to reset the underlying device
 			during initialization.
 
@@ -2435,6 +2440,10 @@
 			disables clocksource verification at runtime.
 			Used to enable high-resolution timer mode on older
 			hardware, and in virtualized environment.
+			[x86] noirqtime: Do not use TSC to do irq accounting.
+			Used to run time disable IRQ_TIME_ACCOUNTING on any
+			platforms where RDTSC is slow and this accounting
+			can add overhead.
 
 	turbografx.map[2|3]=	[HW,JOY]
 			TurboGraFX parallel port interface
diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt
index 1762b81..741fe66 100644
--- a/Documentation/kprobes.txt
+++ b/Documentation/kprobes.txt
@@ -542,9 +542,11 @@
 registration and unregistration.
 
 Probe handlers are run with preemption disabled.  Depending on the
-architecture, handlers may also run with interrupts disabled.  In any
-case, your handler should not yield the CPU (e.g., by attempting to
-acquire a semaphore).
+architecture and optimization state, handlers may also run with
+interrupts disabled (e.g., kretprobe handlers and optimized kprobe
+handlers run without interrupt disabled on x86/x86-64).  In any case,
+your handler should not yield the CPU (e.g., by attempting to acquire
+a semaphore).
 
 Since a return probe is implemented by replacing the return
 address with the trampoline's address, stack backtraces and calls
diff --git a/Documentation/networking/e1000.txt b/Documentation/networking/e1000.txt
index 2df7186..d9271e7 100644
--- a/Documentation/networking/e1000.txt
+++ b/Documentation/networking/e1000.txt
@@ -1,82 +1,35 @@
 Linux* Base Driver for the Intel(R) PRO/1000 Family of Adapters
 ===============================================================
 
-September 26, 2006
-
+Intel Gigabit Linux driver.
+Copyright(c) 1999 - 2010 Intel Corporation.
 
 Contents
 ========
 
-- In This Release
 - Identifying Your Adapter
-- Building and Installation
 - Command Line Parameters
 - Speed and Duplex Configuration
 - Additional Configurations
-- Known Issues
 - Support
 
-
-In This Release
-===============
-
-This file describes the Linux* Base Driver for the Intel(R) PRO/1000 Family
-of Adapters.  This driver includes support for Itanium(R)2-based systems.
-
-For questions related to hardware requirements, refer to the documentation
-supplied with your Intel PRO/1000 adapter. All hardware requirements listed
-apply to use with Linux.
-
-The following features are now available in supported kernels:
- - Native VLANs
- - Channel Bonding (teaming)
- - SNMP
-
-Channel Bonding documentation can be found in the Linux kernel source:
-/Documentation/networking/bonding.txt
-
-The driver information previously displayed in the /proc filesystem is not
-supported in this release.  Alternatively, you can use ethtool (version 1.6
-or later), lspci, and ifconfig to obtain the same information.
-
-Instructions on updating ethtool can be found in the section "Additional
-Configurations" later in this document.
-
-NOTE: The Intel(R) 82562v 10/100 Network Connection only provides 10/100
-support.
-
-
 Identifying Your Adapter
 ========================
 
 For more information on how to identify your adapter, go to the Adapter &
 Driver ID Guide at:
 
-    http://support.intel.com/support/network/adapter/pro100/21397.htm
+    http://support.intel.com/support/go/network/adapter/idguide.htm
 
 For the latest Intel network drivers for Linux, refer to the following
 website.  In the search field, enter your adapter name or type, or use the
 networking link on the left to search for your adapter:
 
-    http://downloadfinder.intel.com/scripts-df/support_intel.asp
-
+    http://support.intel.com/support/go/network/adapter/home.htm
 
 Command Line Parameters
 =======================
 
-If the driver is built as a module, the  following optional parameters
-are used by entering them on the command line with the modprobe command
-using this syntax:
-
-     modprobe e1000 [<option>=<VAL1>,<VAL2>,...]
-
-For example, with two PRO/1000 PCI adapters, entering:
-
-     modprobe e1000 TxDescriptors=80,128
-
-loads the e1000 driver with 80 TX descriptors for the first adapter and
-128 TX descriptors for the second adapter.
-
 The default value for each parameter is generally the recommended setting,
 unless otherwise noted.
 
@@ -89,10 +42,6 @@
         parameters, see the application note at:
         http://www.intel.com/design/network/applnots/ap450.htm
 
-        A descriptor describes a data buffer and attributes related to
-        the data buffer.  This information is accessed by the hardware.
-
-
 AutoNeg
 -------
 (Supported only on adapters with copper connections)
@@ -106,7 +55,6 @@
 NOTE:  Refer to the Speed and Duplex section of this readme for more
        information on the AutoNeg parameter.
 
-
 Duplex
 ------
 (Supported only on adapters with copper connections)
@@ -119,7 +67,6 @@
 link partner is forced (either full or half), Duplex defaults to half-
 duplex.
 
-
 FlowControl
 -----------
 Valid Range:   0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx)
@@ -128,16 +75,16 @@
 This parameter controls the automatic generation(Tx) and response(Rx)
 to Ethernet PAUSE frames.
 
-
 InterruptThrottleRate
 ---------------------
 (not supported on Intel(R) 82542, 82543 or 82544-based adapters)
-Valid Range:   0,1,3,100-100000 (0=off, 1=dynamic, 3=dynamic conservative)
+Valid Range:   0,1,3,4,100-100000 (0=off, 1=dynamic, 3=dynamic conservative,
+                                   4=simplified balancing)
 Default Value: 3
 
 The driver can limit the amount of interrupts per second that the adapter
-will generate for incoming packets. It does this by writing a value to the 
-adapter that is based on the maximum amount of interrupts that the adapter 
+will generate for incoming packets. It does this by writing a value to the
+adapter that is based on the maximum amount of interrupts that the adapter
 will generate per second.
 
 Setting InterruptThrottleRate to a value greater or equal to 100
@@ -146,37 +93,43 @@
 load on the system and can lower CPU utilization under heavy load,
 but will increase latency as packets are not processed as quickly.
 
-The default behaviour of the driver previously assumed a static 
-InterruptThrottleRate value of 8000, providing a good fallback value for 
-all traffic types,but lacking in small packet performance and latency. 
-The hardware can handle many more small packets per second however, and 
+The default behaviour of the driver previously assumed a static
+InterruptThrottleRate value of 8000, providing a good fallback value for
+all traffic types,but lacking in small packet performance and latency.
+The hardware can handle many more small packets per second however, and
 for this reason an adaptive interrupt moderation algorithm was implemented.
 
 Since 7.3.x, the driver has two adaptive modes (setting 1 or 3) in which
-it dynamically adjusts the InterruptThrottleRate value based on the traffic 
+it dynamically adjusts the InterruptThrottleRate value based on the traffic
 that it receives. After determining the type of incoming traffic in the last
-timeframe, it will adjust the InterruptThrottleRate to an appropriate value 
+timeframe, it will adjust the InterruptThrottleRate to an appropriate value
 for that traffic.
 
 The algorithm classifies the incoming traffic every interval into
-classes.  Once the class is determined, the InterruptThrottleRate value is 
-adjusted to suit that traffic type the best. There are three classes defined: 
+classes.  Once the class is determined, the InterruptThrottleRate value is
+adjusted to suit that traffic type the best. There are three classes defined:
 "Bulk traffic", for large amounts of packets of normal size; "Low latency",
 for small amounts of traffic and/or a significant percentage of small
-packets; and "Lowest latency", for almost completely small packets or 
+packets; and "Lowest latency", for almost completely small packets or
 minimal traffic.
 
-In dynamic conservative mode, the InterruptThrottleRate value is set to 4000 
-for traffic that falls in class "Bulk traffic". If traffic falls in the "Low 
-latency" or "Lowest latency" class, the InterruptThrottleRate is increased 
+In dynamic conservative mode, the InterruptThrottleRate value is set to 4000
+for traffic that falls in class "Bulk traffic". If traffic falls in the "Low
+latency" or "Lowest latency" class, the InterruptThrottleRate is increased
 stepwise to 20000. This default mode is suitable for most applications.
 
 For situations where low latency is vital such as cluster or
 grid computing, the algorithm can reduce latency even more when
 InterruptThrottleRate is set to mode 1. In this mode, which operates
-the same as mode 3, the InterruptThrottleRate will be increased stepwise to 
+the same as mode 3, the InterruptThrottleRate will be increased stepwise to
 70000 for traffic in class "Lowest latency".
 
+In simplified mode the interrupt rate is based on the ratio of Tx and
+Rx traffic.  If the bytes per second rate is approximately equal, the
+interrupt rate will drop as low as 2000 interrupts per second.  If the
+traffic is mostly transmit or mostly receive, the interrupt rate could
+be as high as 8000.
+
 Setting InterruptThrottleRate to 0 turns off any interrupt moderation
 and may improve small packet latency, but is generally not suitable
 for bulk throughput traffic.
@@ -212,8 +165,6 @@
        be platform-specific.  If CPU utilization is not a concern, use
        RX_POLLING (NAPI) and default driver settings.
 
-
-
 RxDescriptors
 -------------
 Valid Range:   80-256 for 82542 and 82543-based adapters
@@ -225,15 +176,14 @@
 incoming packets, at the expense of increased system memory utilization.
 
 Each descriptor is 16 bytes.  A receive buffer is also allocated for each
-descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending 
+descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending
 on the MTU setting. The maximum MTU size is 16110.
 
-NOTE:  MTU designates the frame size.  It only needs to be set for Jumbo 
-       Frames.  Depending on the available system resources, the request 
-       for a higher number of receive descriptors may be denied.  In this 
+NOTE:  MTU designates the frame size.  It only needs to be set for Jumbo
+       Frames.  Depending on the available system resources, the request
+       for a higher number of receive descriptors may be denied.  In this
        case, use a lower number.
 
-
 RxIntDelay
 ----------
 Valid Range:   0-65535 (0=off)
@@ -254,7 +204,6 @@
           restoring the network connection.  To eliminate the potential
           for the hang ensure that RxIntDelay is set to 0.
 
-
 RxAbsIntDelay
 -------------
 (This parameter is supported only on 82540, 82545 and later adapters.)
@@ -268,7 +217,6 @@
 along with RxIntDelay, may improve traffic throughput in specific network
 conditions.
 
-
 Speed
 -----
 (This parameter is supported only on adapters with copper connections.)
@@ -280,7 +228,6 @@
 partner is set to auto-negotiate, the board will auto-detect the correct
 speed.  Duplex should also be set when Speed is set to either 10 or 100.
 
-
 TxDescriptors
 -------------
 Valid Range:   80-256 for 82542 and 82543-based adapters
@@ -295,6 +242,36 @@
        higher number of transmit descriptors may be denied.  In this case,
        use a lower number.
 
+TxDescriptorStep
+----------------
+Valid Range:    1 (use every Tx Descriptor)
+		4 (use every 4th Tx Descriptor)
+
+Default Value:  1 (use every Tx Descriptor)
+
+On certain non-Intel architectures, it has been observed that intense TX
+traffic bursts of short packets may result in an improper descriptor
+writeback. If this occurs, the driver will report a "TX Timeout" and reset
+the adapter, after which the transmit flow will restart, though data may
+have stalled for as much as 10 seconds before it resumes.
+
+The improper writeback does not occur on the first descriptor in a system
+memory cache-line, which is typically 32 bytes, or 4 descriptors long.
+
+Setting TxDescriptorStep to a value of 4 will ensure that all TX descriptors
+are aligned to the start of a system memory cache line, and so this problem
+will not occur.
+
+NOTES: Setting TxDescriptorStep to 4 effectively reduces the number of
+       TxDescriptors available for transmits to 1/4 of the normal allocation.
+       This has a possible negative performance impact, which may be
+       compensated for by allocating more descriptors using the TxDescriptors
+       module parameter.
+
+       There are other conditions which may result in "TX Timeout", which will
+       not be resolved by the use of the TxDescriptorStep parameter. As the
+       issue addressed by this parameter has never been observed on Intel
+       Architecture platforms, it should not be used on Intel platforms.
 
 TxIntDelay
 ----------
@@ -307,7 +284,6 @@
 system is reporting dropped transmits, this value may be set too high
 causing the driver to run out of available transmit descriptors.
 
-
 TxAbsIntDelay
 -------------
 (This parameter is supported only on 82540, 82545 and later adapters.)
@@ -330,6 +306,35 @@
 A value of '1' indicates that the driver should enable IP checksum
 offload for received packets (both UDP and TCP) to the adapter hardware.
 
+Copybreak
+---------
+Valid Range:   0-xxxxxxx (0=off)
+Default Value: 256
+Usage: insmod e1000.ko copybreak=128
+
+Driver copies all packets below or equaling this size to a fresh Rx
+buffer before handing it up the stack.
+
+This parameter is different than other parameters, in that it is a
+single (not 1,1,1 etc.) parameter applied to all driver instances and
+it is also available during runtime at
+/sys/module/e1000/parameters/copybreak
+
+SmartPowerDownEnable
+--------------------
+Valid Range: 0-1
+Default Value:  0 (disabled)
+
+Allows PHY to turn off in lower power states. The user can turn off
+this parameter in supported chipsets.
+
+KumeranLockLoss
+---------------
+Valid Range: 0-1
+Default Value: 1 (enabled)
+
+This workaround skips resetting the PHY at shutdown for the initial
+silicon releases of ICH8 systems.
 
 Speed and Duplex Configuration
 ==============================
@@ -385,40 +390,9 @@
 parameter should not be used.  Instead, use the Speed and Duplex parameters
 previously mentioned to force the adapter to the same speed and duplex.
 
-
 Additional Configurations
 =========================
 
-  Configuring the Driver on Different Distributions
-  -------------------------------------------------
-  Configuring a network driver to load properly when the system is started
-  is distribution dependent.  Typically, the configuration process involves
-  adding an alias line to /etc/modules.conf or /etc/modprobe.conf as well
-  as editing other system startup scripts and/or configuration files.  Many
-  popular Linux distributions ship with tools to make these changes for you.
-  To learn the proper way to configure a network device for your system,
-  refer to your distribution documentation.  If during this process you are
-  asked for the driver or module name, the name for the Linux Base Driver
-  for the Intel(R) PRO/1000 Family of Adapters is e1000.
-
-  As an example, if you install the e1000 driver for two PRO/1000 adapters
-  (eth0 and eth1) and set the speed and duplex to 10full and 100half, add
-  the following to modules.conf or or modprobe.conf:
-
-       alias eth0 e1000
-       alias eth1 e1000
-       options e1000 Speed=10,100 Duplex=2,1
-
-  Viewing Link Messages
-  ---------------------
-  Link messages will not be displayed to the console if the distribution is
-  restricting system messages.  In order to see network driver link messages
-  on your console, set dmesg to eight by entering the following:
-
-       dmesg -n 8
-
-  NOTE: This setting is not saved across reboots.
-
   Jumbo Frames
   ------------
   Jumbo Frames support is enabled by changing the MTU to a value larger than
@@ -437,9 +411,11 @@
    setting in a different location.
 
   Notes:
-
-  - To enable Jumbo Frames, increase the MTU size on the interface beyond
-    1500.
+  Degradation in throughput performance may be observed in some Jumbo frames
+  environments. If this is observed, increasing the application's socket buffer
+  size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help.
+  See the specific application manual and /usr/src/linux*/Documentation/
+  networking/ip-sysctl.txt for more details.
 
   - The maximum MTU setting for Jumbo Frames is 16110.  This value coincides
     with the maximum Jumbo Frames size of 16128.
@@ -447,40 +423,11 @@
   - Using Jumbo Frames at 10 or 100 Mbps may result in poor performance or
     loss of link.
 
-  - Some Intel gigabit adapters that support Jumbo Frames have a frame size
-    limit of 9238 bytes, with a corresponding MTU size limit of 9216 bytes.
-    The adapters with this limitation are based on the Intel(R) 82571EB,
-    82572EI, 82573L and 80003ES2LAN controller.  These correspond to the
-    following product names:
-     Intel(R) PRO/1000 PT Server Adapter
-     Intel(R) PRO/1000 PT Desktop Adapter
-     Intel(R) PRO/1000 PT Network Connection
-     Intel(R) PRO/1000 PT Dual Port Server Adapter
-     Intel(R) PRO/1000 PT Dual Port Network Connection
-     Intel(R) PRO/1000 PF Server Adapter
-     Intel(R) PRO/1000 PF Network Connection
-     Intel(R) PRO/1000 PF Dual Port Server Adapter
-     Intel(R) PRO/1000 PB Server Connection
-     Intel(R) PRO/1000 PL Network Connection
-     Intel(R) PRO/1000 EB Network Connection with I/O Acceleration
-     Intel(R) PRO/1000 EB Backplane Connection with I/O Acceleration
-     Intel(R) PRO/1000 PT Quad Port Server Adapter
-
   - Adapters based on the Intel(R) 82542 and 82573V/E controller do not
     support Jumbo Frames. These correspond to the following product names:
      Intel(R) PRO/1000 Gigabit Server Adapter
      Intel(R) PRO/1000 PM Network Connection
 
-  - The following adapters do not support Jumbo Frames:
-     Intel(R) 82562V 10/100 Network Connection
-     Intel(R) 82566DM Gigabit Network Connection
-     Intel(R) 82566DC Gigabit Network Connection
-     Intel(R) 82566MM Gigabit Network Connection
-     Intel(R) 82566MC Gigabit Network Connection
-     Intel(R) 82562GT 10/100 Network Connection
-     Intel(R) 82562G 10/100 Network Connection
-
-
   Ethtool
   -------
   The driver utilizes the ethtool interface for driver configuration and
@@ -490,142 +437,14 @@
   The latest release of ethtool can be found from
   http://sourceforge.net/projects/gkernel.
 
-  NOTE: Ethtool 1.6 only supports a limited set of ethtool options.  Support
-  for a more complete ethtool feature set can be enabled by upgrading
-  ethtool to ethtool-1.8.1.
-
   Enabling Wake on LAN* (WoL)
   ---------------------------
-  WoL is configured through the Ethtool* utility.  Ethtool is included with
-  all versions of Red Hat after Red Hat 7.2.  For other Linux distributions,
-  download and install Ethtool from the following website:
-  http://sourceforge.net/projects/gkernel.
-
-  For instructions on enabling WoL with Ethtool, refer to the website listed
-  above.
+  WoL is configured through the Ethtool* utility.
 
   WoL will be enabled on the system during the next shut down or reboot.
   For this driver version, in order to enable WoL, the e1000 driver must be
   loaded when shutting down or rebooting the system.
 
-  Wake On LAN is only supported on port A for the following devices:
-  Intel(R) PRO/1000 PT Dual Port Network Connection
-  Intel(R) PRO/1000 PT Dual Port Server Connection
-  Intel(R) PRO/1000 PT Dual Port Server Adapter
-  Intel(R) PRO/1000 PF Dual Port Server Adapter
-  Intel(R) PRO/1000 PT Quad Port Server Adapter
-
-  NAPI
-  ----
-  NAPI (Rx polling mode) is enabled in the e1000 driver.
-
-  See www.cyberus.ca/~hadi/usenix-paper.tgz for more information on NAPI.
-
-
-Known Issues
-============
-
-Dropped Receive Packets on Half-duplex 10/100 Networks
-------------------------------------------------------
-If you have an Intel PCI Express adapter running at 10mbps or 100mbps, half-
-duplex, you may observe occasional dropped receive packets.  There are no
-workarounds for this problem in this network configuration.  The network must
-be updated to operate in full-duplex, and/or 1000mbps only.
-
-Jumbo Frames System Requirement
--------------------------------
-Memory allocation failures have been observed on Linux systems with 64 MB
-of RAM or less that are running Jumbo Frames.  If you are using Jumbo
-Frames, your system may require more than the advertised minimum
-requirement of 64 MB of system memory.
-
-Performance Degradation with Jumbo Frames
------------------------------------------
-Degradation in throughput performance may be observed in some Jumbo frames
-environments.  If this is observed, increasing the application's socket
-buffer size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values
-may help.  See the specific application manual and
-/usr/src/linux*/Documentation/
-networking/ip-sysctl.txt for more details.
-
-Jumbo Frames on Foundry BigIron 8000 switch
--------------------------------------------
-There is a known issue using Jumbo frames when connected to a Foundry
-BigIron 8000 switch.  This is a 3rd party limitation.  If you experience
-loss of packets, lower the MTU size.
-
-Allocating Rx Buffers when Using Jumbo Frames 
----------------------------------------------
-Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if 
-the available memory is heavily fragmented. This issue may be seen with PCI-X 
-adapters or with packet split disabled. This can be reduced or eliminated 
-by changing the amount of available memory for receive buffer allocation, by
-increasing /proc/sys/vm/min_free_kbytes. 
-
-Multiple Interfaces on Same Ethernet Broadcast Network
-------------------------------------------------------
-Due to the default ARP behavior on Linux, it is not possible to have
-one system on two IP networks in the same Ethernet broadcast domain
-(non-partitioned switch) behave as expected.  All Ethernet interfaces
-will respond to IP traffic for any IP address assigned to the system.
-This results in unbalanced receive traffic.
-
-If you have multiple interfaces in a server, either turn on ARP
-filtering by entering:
-
-    echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter
-(this only works if your kernel's version is higher than 2.4.5),
-
-NOTE: This setting is not saved across reboots.  The configuration
-change can be made permanent by adding the line:
-    net.ipv4.conf.all.arp_filter = 1
-to the file /etc/sysctl.conf
-
-      or,
-
-install the interfaces in separate broadcast domains (either in
-different switches or in a switch partitioned to VLANs).
-
-82541/82547 can't link or are slow to link with some link partners
------------------------------------------------------------------
-There is a known compatibility issue with 82541/82547 and some
-low-end switches where the link will not be established, or will
-be slow to establish.  In particular, these switches are known to
-be incompatible with 82541/82547:
-
-    Planex FXG-08TE
-    I-O Data ETG-SH8
-
-To workaround this issue, the driver can be compiled with an override
-of the PHY's master/slave setting.  Forcing master or forcing slave
-mode will improve time-to-link.
-
-    # make CFLAGS_EXTRA=-DE1000_MASTER_SLAVE=<n>
-
-Where <n> is:
-
-    0 = Hardware default
-    1 = Master mode
-    2 = Slave mode
-    3 = Auto master/slave
-
-Disable rx flow control with ethtool
-------------------------------------
-In order to disable receive flow control using ethtool, you must turn
-off auto-negotiation on the same command line.
-
-For example:
-
-   ethtool -A eth? autoneg off rx off
-
-Unplugging network cable while ethtool -p is running
-----------------------------------------------------
-In kernel versions 2.5.50 and later (including 2.6 kernel), unplugging
-the network cable while ethtool -p is running will cause the system to
-become unresponsive to keyboard commands, except for control-alt-delete.
-Restarting the system appears to be the only remedy.
-
-
 Support
 =======
 
diff --git a/Documentation/networking/e1000e.txt b/Documentation/networking/e1000e.txt
new file mode 100644
index 0000000..6aa048b
--- /dev/null
+++ b/Documentation/networking/e1000e.txt
@@ -0,0 +1,302 @@
+Linux* Driver for Intel(R) Network Connection
+===============================================================
+
+Intel Gigabit Linux driver.
+Copyright(c) 1999 - 2010 Intel Corporation.
+
+Contents
+========
+
+- Identifying Your Adapter
+- Command Line Parameters
+- Additional Configurations
+- Support
+
+Identifying Your Adapter
+========================
+
+The e1000e driver supports all PCI Express Intel(R) Gigabit Network
+Connections, except those that are 82575, 82576 and 82580-based*.
+
+* NOTE: The Intel(R) PRO/1000 P Dual Port Server Adapter is supported by
+  the e1000 driver, not the e1000e driver due to the 82546 part being used
+  behind a PCI Express bridge.
+
+For more information on how to identify your adapter, go to the Adapter &
+Driver ID Guide at:
+
+    http://support.intel.com/support/go/network/adapter/idguide.htm
+
+For the latest Intel network drivers for Linux, refer to the following
+website.  In the search field, enter your adapter name or type, or use the
+networking link on the left to search for your adapter:
+
+    http://support.intel.com/support/go/network/adapter/home.htm
+
+Command Line Parameters
+=======================
+
+The default value for each parameter is generally the recommended setting,
+unless otherwise noted.
+
+NOTES:  For more information about the InterruptThrottleRate,
+        RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay
+        parameters, see the application note at:
+        http://www.intel.com/design/network/applnots/ap450.htm
+
+InterruptThrottleRate
+---------------------
+Valid Range:   0,1,3,4,100-100000 (0=off, 1=dynamic, 3=dynamic conservative,
+                                   4=simplified balancing)
+Default Value: 3
+
+The driver can limit the amount of interrupts per second that the adapter
+will generate for incoming packets. It does this by writing a value to the
+adapter that is based on the maximum amount of interrupts that the adapter
+will generate per second.
+
+Setting InterruptThrottleRate to a value greater or equal to 100
+will program the adapter to send out a maximum of that many interrupts
+per second, even if more packets have come in. This reduces interrupt
+load on the system and can lower CPU utilization under heavy load,
+but will increase latency as packets are not processed as quickly.
+
+The driver has two adaptive modes (setting 1 or 3) in which
+it dynamically adjusts the InterruptThrottleRate value based on the traffic
+that it receives. After determining the type of incoming traffic in the last
+timeframe, it will adjust the InterruptThrottleRate to an appropriate value
+for that traffic.
+
+The algorithm classifies the incoming traffic every interval into
+classes.  Once the class is determined, the InterruptThrottleRate value is
+adjusted to suit that traffic type the best. There are three classes defined:
+"Bulk traffic", for large amounts of packets of normal size; "Low latency",
+for small amounts of traffic and/or a significant percentage of small
+packets; and "Lowest latency", for almost completely small packets or
+minimal traffic.
+
+In dynamic conservative mode, the InterruptThrottleRate value is set to 4000
+for traffic that falls in class "Bulk traffic". If traffic falls in the "Low
+latency" or "Lowest latency" class, the InterruptThrottleRate is increased
+stepwise to 20000. This default mode is suitable for most applications.
+
+For situations where low latency is vital such as cluster or
+grid computing, the algorithm can reduce latency even more when
+InterruptThrottleRate is set to mode 1. In this mode, which operates
+the same as mode 3, the InterruptThrottleRate will be increased stepwise to
+70000 for traffic in class "Lowest latency".
+
+In simplified mode the interrupt rate is based on the ratio of Tx and
+Rx traffic.  If the bytes per second rate is approximately equal the
+interrupt rate will drop as low as 2000 interrupts per second.  If the
+traffic is mostly transmit or mostly receive, the interrupt rate could
+be as high as 8000.
+
+Setting InterruptThrottleRate to 0 turns off any interrupt moderation
+and may improve small packet latency, but is generally not suitable
+for bulk throughput traffic.
+
+NOTE:  InterruptThrottleRate takes precedence over the TxAbsIntDelay and
+       RxAbsIntDelay parameters.  In other words, minimizing the receive
+       and/or transmit absolute delays does not force the controller to
+       generate more interrupts than what the Interrupt Throttle Rate
+       allows.
+
+NOTE:  When e1000e is loaded with default settings and multiple adapters
+       are in use simultaneously, the CPU utilization may increase non-
+       linearly.  In order to limit the CPU utilization without impacting
+       the overall throughput, we recommend that you load the driver as
+       follows:
+
+           modprobe e1000e InterruptThrottleRate=3000,3000,3000
+
+       This sets the InterruptThrottleRate to 3000 interrupts/sec for
+       the first, second, and third instances of the driver.  The range
+       of 2000 to 3000 interrupts per second works on a majority of
+       systems and is a good starting point, but the optimal value will
+       be platform-specific.  If CPU utilization is not a concern, use
+       RX_POLLING (NAPI) and default driver settings.
+
+RxIntDelay
+----------
+Valid Range:   0-65535 (0=off)
+Default Value: 0
+
+This value delays the generation of receive interrupts in units of 1.024
+microseconds.  Receive interrupt reduction can improve CPU efficiency if
+properly tuned for specific network traffic.  Increasing this value adds
+extra latency to frame reception and can end up decreasing the throughput
+of TCP traffic.  If the system is reporting dropped receives, this value
+may be set too high, causing the driver to run out of available receive
+descriptors.
+
+CAUTION:  When setting RxIntDelay to a value other than 0, adapters may
+          hang (stop transmitting) under certain network conditions.  If
+          this occurs a NETDEV WATCHDOG message is logged in the system
+          event log.  In addition, the controller is automatically reset,
+          restoring the network connection.  To eliminate the potential
+          for the hang ensure that RxIntDelay is set to 0.
+
+RxAbsIntDelay
+-------------
+Valid Range:   0-65535 (0=off)
+Default Value: 8
+
+This value, in units of 1.024 microseconds, limits the delay in which a
+receive interrupt is generated.  Useful only if RxIntDelay is non-zero,
+this value ensures that an interrupt is generated after the initial
+packet is received within the set amount of time.  Proper tuning,
+along with RxIntDelay, may improve traffic throughput in specific network
+conditions.
+
+TxIntDelay
+----------
+Valid Range:   0-65535 (0=off)
+Default Value: 8
+
+This value delays the generation of transmit interrupts in units of
+1.024 microseconds.  Transmit interrupt reduction can improve CPU
+efficiency if properly tuned for specific network traffic.  If the
+system is reporting dropped transmits, this value may be set too high
+causing the driver to run out of available transmit descriptors.
+
+TxAbsIntDelay
+-------------
+Valid Range:   0-65535 (0=off)
+Default Value: 32
+
+This value, in units of 1.024 microseconds, limits the delay in which a
+transmit interrupt is generated.  Useful only if TxIntDelay is non-zero,
+this value ensures that an interrupt is generated after the initial
+packet is sent on the wire within the set amount of time.  Proper tuning,
+along with TxIntDelay, may improve traffic throughput in specific
+network conditions.
+
+Copybreak
+---------
+Valid Range:   0-xxxxxxx (0=off)
+Default Value: 256
+
+Driver copies all packets below or equaling this size to a fresh Rx
+buffer before handing it up the stack.
+
+This parameter is different than other parameters, in that it is a
+single (not 1,1,1 etc.) parameter applied to all driver instances and
+it is also available during runtime at
+/sys/module/e1000e/parameters/copybreak
+
+SmartPowerDownEnable
+--------------------
+Valid Range: 0-1
+Default Value:  0 (disabled)
+
+Allows PHY to turn off in lower power states. The user can set this parameter
+in supported chipsets.
+
+KumeranLockLoss
+---------------
+Valid Range: 0-1
+Default Value: 1 (enabled)
+
+This workaround skips resetting the PHY at shutdown for the initial
+silicon releases of ICH8 systems.
+
+IntMode
+-------
+Valid Range: 0-2 (0=legacy, 1=MSI, 2=MSI-X)
+Default Value: 2
+
+Allows changing the interrupt mode at module load time, without requiring a
+recompile. If the driver load fails to enable a specific interrupt mode, the
+driver will try other interrupt modes, from least to most compatible.  The
+interrupt order is MSI-X, MSI, Legacy.  If specifying MSI (IntMode=1)
+interrupts, only MSI and Legacy will be attempted.
+
+CrcStripping
+------------
+Valid Range: 0-1
+Default Value: 1 (enabled)
+
+Strip the CRC from received packets before sending up the network stack.  If
+you have a machine with a BMC enabled but cannot receive IPMI traffic after
+loading or enabling the driver, try disabling this feature.
+
+WriteProtectNVM
+---------------
+Valid Range: 0-1
+Default Value: 1 (enabled)
+
+Set the hardware to ignore all write/erase cycles to the GbE region in the
+ICHx NVM (non-volatile memory).  This feature can be disabled by the
+WriteProtectNVM module parameter (enabled by default) only after a hardware
+reset, but the machine must be power cycled before trying to enable writes.
+
+Note: the kernel boot option iomem=relaxed may need to be set if the kernel
+config option CONFIG_STRICT_DEVMEM=y, if the root user wants to write the
+NVM from user space via ethtool.
+
+Additional Configurations
+=========================
+
+  Jumbo Frames
+  ------------
+  Jumbo Frames support is enabled by changing the MTU to a value larger than
+  the default of 1500.  Use the ifconfig command to increase the MTU size.
+  For example:
+
+       ifconfig eth<x> mtu 9000 up
+
+  This setting is not saved across reboots.
+
+  Notes:
+
+  - The maximum MTU setting for Jumbo Frames is 9216.  This value coincides
+    with the maximum Jumbo Frames size of 9234 bytes.
+
+  - Using Jumbo Frames at 10 or 100 Mbps is not supported and may result in
+    poor performance or loss of link.
+
+  - Some adapters limit Jumbo Frames sized packets to a maximum of
+    4096 bytes and some adapters do not support Jumbo Frames.
+
+
+  Ethtool
+  -------
+  The driver utilizes the ethtool interface for driver configuration and
+  diagnostics, as well as displaying statistical information.  We
+  strongly recommend downloading the latest version of Ethtool at:
+
+  http://sourceforge.net/projects/gkernel.
+
+  Speed and Duplex
+  ----------------
+  Speed and Duplex are configured through the Ethtool* utility. For
+  instructions,  refer to the Ethtool man page.
+
+  Enabling Wake on LAN* (WoL)
+  ---------------------------
+  WoL is configured through the Ethtool* utility. For instructions on
+  enabling WoL with Ethtool, refer to the Ethtool man page.
+
+  WoL will be enabled on the system during the next shut down or reboot.
+  For this driver version, in order to enable WoL, the e1000e driver must be
+  loaded when shutting down or rebooting the system.
+
+  In most cases Wake On LAN is only supported on port A for multiple port
+  adapters. To verify if a port supports Wake on LAN run ethtool eth<X>.
+
+
+Support
+=======
+
+For general information, go to the Intel support website at:
+
+    www.intel.com/support/
+
+or the Intel Wired Networking project hosted by Sourceforge at:
+
+    http://sourceforge.net/projects/e1000
+
+If an issue is identified with the released source code on the supported
+kernel with a supported adapter, email the specific information related
+to the issue to e1000-devel@lists.sf.net
diff --git a/Documentation/networking/ixgbevf.txt b/Documentation/networking/ixgbevf.txt
old mode 100755
new mode 100644
index 19015de..21dd5d1
--- a/Documentation/networking/ixgbevf.txt
+++ b/Documentation/networking/ixgbevf.txt
@@ -1,19 +1,16 @@
 Linux* Base Driver for Intel(R) Network Connection
 ==================================================
 
-November 24, 2009
+Intel Gigabit Linux driver.
+Copyright(c) 1999 - 2010 Intel Corporation.
 
 Contents
 ========
 
-- In This Release
 - Identifying Your Adapter
 - Known Issues/Troubleshooting
 - Support
 
-In This Release
-===============
-
 This file describes the ixgbevf Linux* Base Driver for Intel Network
 Connection.
 
@@ -33,7 +30,7 @@
 For more information on how to identify your adapter, go to the Adapter &
 Driver ID Guide at:
 
-    http://support.intel.com/support/network/sb/CS-008441.htm
+    http://support.intel.com/support/go/network/adapter/idguide.htm
 
 Known Issues/Troubleshooting
 ============================
@@ -57,34 +54,3 @@
 If an issue is identified with the released source code on the supported
 kernel with a supported adapter, email the specific information related
 to the issue to e1000-devel@lists.sf.net
-
-License
-=======
-
-Intel 10 Gigabit Linux driver.
-Copyright(c) 1999 - 2009 Intel Corporation.
-
-This program is free software; you can redistribute it and/or modify it
-under the terms and conditions of the GNU General Public License,
-version 2, as published by the Free Software Foundation.
-
-This program is distributed in the hope it will be useful, but WITHOUT
-ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-more details.
-
-You should have received a copy of the GNU General Public License along with
-this program; if not, write to the Free Software Foundation, Inc.,
-51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
-The full GNU General Public License is included in this distribution in
-the file called "COPYING".
-
-Trademarks
-==========
-
-Intel, Itanium, and Pentium are trademarks or registered trademarks of
-Intel Corporation or its subsidiaries in the United States and other
-countries.
-
-* Other names and brands may be claimed as the property of others.
diff --git a/Documentation/pcmcia/driver-changes.txt b/Documentation/pcmcia/driver-changes.txt
index 26c0f9c..dd04361 100644
--- a/Documentation/pcmcia/driver-changes.txt
+++ b/Documentation/pcmcia/driver-changes.txt
@@ -1,4 +1,29 @@
 This file details changes in 2.6 which affect PCMCIA card driver authors:
+* pcmcia_loop_config() and autoconfiguration (as of 2.6.36)
+   If struct pcmcia_device *p_dev->config_flags is set accordingly,
+   pcmcia_loop_config() now sets up certain configuration values
+   automatically, though the driver may still override the settings
+   in the callback function. The following autoconfiguration options
+   are provided at the moment:
+	CONF_AUTO_CHECK_VCC : check for matching Vcc
+	CONF_AUTO_SET_VPP   : set Vpp
+	CONF_AUTO_AUDIO     : auto-enable audio line, if required
+	CONF_AUTO_SET_IO    : set ioport resources (->resource[0,1])
+	CONF_AUTO_SET_IOMEM : set first iomem resource (->resource[2])
+
+* pcmcia_request_configuration -> pcmcia_enable_device (as of 2.6.36)
+   pcmcia_request_configuration() got renamed to pcmcia_enable_device(),
+   as it mirrors pcmcia_disable_device(). Configuration settings are now
+   stored in struct pcmcia_device, e.g. in the fields config_flags,
+   config_index, config_base, vpp.
+
+* pcmcia_request_window changes (as of 2.6.36)
+   Instead of win_req_t, drivers are now requested to fill out
+   struct pcmcia_device *p_dev->resource[2,3,4,5] for up to four ioport
+   ranges. After a call to pcmcia_request_window(), the regions found there
+   are reserved and may be used immediately -- until pcmcia_release_window()
+   is called.
+
 * pcmcia_request_io changes (as of 2.6.36)
    Instead of io_req_t, drivers are now requested to fill out
    struct pcmcia_device *p_dev->resource[0,1] for up to two ioport
diff --git a/Documentation/powerpc/dts-bindings/fsl/spi.txt b/Documentation/powerpc/dts-bindings/fsl/spi.txt
index 80510c0..777abd7 100644
--- a/Documentation/powerpc/dts-bindings/fsl/spi.txt
+++ b/Documentation/powerpc/dts-bindings/fsl/spi.txt
@@ -1,7 +1,9 @@
 * SPI (Serial Peripheral Interface)
 
 Required properties:
-- cell-index : SPI controller index.
+- cell-index : QE SPI subblock index.
+		0: QE subblock SPI1
+		1: QE subblock SPI2
 - compatible : should be "fsl,spi".
 - mode : the SPI operation mode, it can be "cpu" or "cpu-qe".
 - reg : Offset and length of the register set for the device
@@ -29,3 +31,23 @@
 		gpios = <&gpio 18 1	// device reg=<0>
 			 &gpio 19 1>;	// device reg=<1>
 	};
+
+
+* eSPI (Enhanced Serial Peripheral Interface)
+
+Required properties:
+- compatible : should be "fsl,mpc8536-espi".
+- reg : Offset and length of the register set for the device.
+- interrupts : should contain eSPI interrupt, the device has one interrupt.
+- fsl,espi-num-chipselects : the number of the chipselect signals.
+
+Example:
+	spi@110000 {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		compatible = "fsl,mpc8536-espi";
+		reg = <0x110000 0x1000>;
+		interrupts = <53 0x2>;
+		interrupt-parent = <&mpic>;
+		fsl,espi-num-chipselects = <4>;
+	};
diff --git a/Documentation/vm/page-types.c b/Documentation/vm/page-types.c
index ccd951f..cc96ee2 100644
--- a/Documentation/vm/page-types.c
+++ b/Documentation/vm/page-types.c
@@ -478,7 +478,7 @@
 	}
 
 	if (opt_unpoison && !hwpoison_forget_fd) {
-		sprintf(buf, "%s/renew-pfn", hwpoison_debug_fs);
+		sprintf(buf, "%s/unpoison-pfn", hwpoison_debug_fs);
 		hwpoison_forget_fd = checked_open(buf, O_WRONLY);
 	}
 }
diff --git a/MAINTAINERS b/MAINTAINERS
index f46d8e6..6f5b5b2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -969,6 +969,16 @@
 S:	Maintained
 F:	arch/arm/mach-s5p*/
 
+ARM/SAMSUNG S5P SERIES FIMC SUPPORT
+M:	Kyungmin Park <kyungmin.park@samsung.com>
+M:	Sylwester Nawrocki <s.nawrocki@samsung.com>
+L:	linux-arm-kernel@lists.infradead.org
+L:	linux-media@vger.kernel.org
+S:	Maintained
+F:	arch/arm/plat-s5p/dev-fimc*
+F:	arch/arm/plat-samsung/include/plat/*fimc*
+F:	drivers/media/video/s5p-fimc/
+
 ARM/SHMOBILE ARM ARCHITECTURE
 M:	Paul Mundt <lethal@linux-sh.org>
 M:	Magnus Damm <magnus.damm@gmail.com>
@@ -1517,6 +1527,8 @@
 S:	Supported
 F:	Documentation/filesystems/ceph.txt
 F:	fs/ceph
+F:	net/ceph
+F:	include/linux/ceph
 
 CERTIFIED WIRELESS USB (WUSB) SUBSYSTEM:
 M:	David Vrabel <david.vrabel@csr.com>
@@ -2535,7 +2547,7 @@
 F:	drivers/scsi/gdt*
 
 GENERIC GPIO I2C DRIVER
-M:	Haavard Skinnemoen <hskinnemoen@atmel.com>
+M:	Haavard Skinnemoen <hskinnemoen@gmail.com>
 S:	Supported
 F:	drivers/i2c/busses/i2c-gpio.c
 F:	include/linux/i2c-gpio.h
@@ -3063,16 +3075,27 @@
 S:	Maintained
 F:	drivers/net/ixp2000/
 
-INTEL ETHERNET DRIVERS (e100/e1000/e1000e/igb/igbvf/ixgb/ixgbe)
+INTEL ETHERNET DRIVERS (e100/e1000/e1000e/igb/igbvf/ixgb/ixgbe/ixgbevf)
 M:	Jeff Kirsher <jeffrey.t.kirsher@intel.com>
 M:	Jesse Brandeburg <jesse.brandeburg@intel.com>
 M:	Bruce Allan <bruce.w.allan@intel.com>
-M:	Alex Duyck <alexander.h.duyck@intel.com>
+M:	Carolyn Wyborny <carolyn.wyborny@intel.com>
+M:	Don Skidmore <donald.c.skidmore@intel.com>
+M:	Greg Rose <gregory.v.rose@intel.com>
 M:	PJ Waskiewicz <peter.p.waskiewicz.jr@intel.com>
+M:	Alex Duyck <alexander.h.duyck@intel.com>
 M:	John Ronciak <john.ronciak@intel.com>
 L:	e1000-devel@lists.sourceforge.net
 W:	http://e1000.sourceforge.net/
 S:	Supported
+F:	Documentation/networking/e100.txt
+F:	Documentation/networking/e1000.txt
+F:	Documentation/networking/e1000e.txt
+F:	Documentation/networking/igb.txt
+F:	Documentation/networking/igbvf.txt
+F:	Documentation/networking/ixgb.txt
+F:	Documentation/networking/ixgbe.txt
+F:	Documentation/networking/ixgbevf.txt
 F:	drivers/net/e100.c
 F:	drivers/net/e1000/
 F:	drivers/net/e1000e/
@@ -3080,6 +3103,7 @@
 F:	drivers/net/igbvf/
 F:	drivers/net/ixgb/
 F:	drivers/net/ixgbe/
+F:	drivers/net/ixgbevf/
 
 INTEL PRO/WIRELESS 2100 NETWORK CONNECTION SUPPORT
 L:	linux-wireless@vger.kernel.org
@@ -3140,7 +3164,7 @@
 
 IOC3 SERIAL DRIVER
 M:	Pat Gefre <pfg@sgi.com>
-L:	linux-mips@linux-mips.org
+L:	linux-serial@vger.kernel.org
 S:	Maintained
 F:	drivers/serial/ioc3_serial.c
 
@@ -3217,6 +3241,12 @@
 F:	include/net/irda/
 F:	net/irda/
 
+IRQ SUBSYSTEM
+M:	Thomas Gleixner <tglx@linutronix.de>
+S:	Maintained
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git irq/core
+F:	kernel/irq/
+
 ISAPNP
 M:	Jaroslav Kysela <perex@perex.cz>
 S:	Maintained
@@ -4783,6 +4813,15 @@
 F:	include/linux/qnx4_fs.h
 F:	include/linux/qnxtypes.h
 
+RADOS BLOCK DEVICE (RBD)
+F:	include/linux/qnxtypes.h
+M:	Yehuda Sadeh <yehuda@hq.newdream.net>
+M:	Sage Weil <sage@newdream.net>
+M:	ceph-devel@vger.kernel.org
+S:	Supported
+F:	drivers/block/rbd.c
+F:	drivers/block/rbd_types.h
+
 RADEON FRAMEBUFFER DISPLAY DRIVER
 M:	Benjamin Herrenschmidt <benh@kernel.crashing.org>
 L:	linux-fbdev@vger.kernel.org
@@ -5008,6 +5047,12 @@
 F:	drivers/media/video/*7146*
 F:	include/media/*7146*
 
+SAMSUNG AUDIO (ASoC) DRIVERS
+M:	Jassi Brar <jassi.brar@samsung.com>
+L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
+S:	Supported
+F:	sound/soc/s3c24xx
+
 TLG2300 VIDEO4LINUX-2 DRIVER
 M:	Huang Shijie <shijie8@gmail.com>
 M:	Kang Yong <kangyong@telegent.com>
@@ -6450,8 +6495,10 @@
 WOLFSON MICROELECTRONICS DRIVERS
 M:	Mark Brown <broonie@opensource.wolfsonmicro.com>
 M:	Ian Lartey <ian@opensource.wolfsonmicro.com>
+M:	Dimitris Papastamos <dp@opensource.wolfsonmicro.com>
+T:	git git://opensource.wolfsonmicro.com/linux-2.6-asoc
 T:	git git://opensource.wolfsonmicro.com/linux-2.6-audioplus
-W:	http://opensource.wolfsonmicro.com/node/8
+W:	http://opensource.wolfsonmicro.com/content/linux-drivers-wolfson-devices
 S:	Supported
 F:	Documentation/hwmon/wm83??
 F:	drivers/leds/leds-wm83*.c
diff --git a/Makefile b/Makefile
index 77b5c6e..d3c1071 100644
--- a/Makefile
+++ b/Makefile
@@ -1,8 +1,8 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 36
-EXTRAVERSION = -rc7
-NAME = Sheep on Meth
+EXTRAVERSION =
+NAME = Flesh-Eating Bats with Fangs
 
 # *DOCUMENTATION*
 # To see a list of typical targets execute "make help"
@@ -568,6 +568,12 @@
 
 ifdef CONFIG_FUNCTION_TRACER
 KBUILD_CFLAGS	+= -pg
+ifdef CONFIG_DYNAMIC_FTRACE
+	ifdef CONFIG_HAVE_C_RECORDMCOUNT
+		BUILD_C_RECORDMCOUNT := y
+		export BUILD_C_RECORDMCOUNT
+	endif
+endif
 endif
 
 # We trigger additional mismatches with less inlining
@@ -591,6 +597,11 @@
 # conserve stack if available
 KBUILD_CFLAGS   += $(call cc-option,-fconserve-stack)
 
+# check for 'asm goto'
+ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(CC)), y)
+	KBUILD_CFLAGS += -DCC_HAVE_ASM_GOTO
+endif
+
 # Add user supplied CPPFLAGS, AFLAGS and CFLAGS as the last assignments
 # But warn user when we do so
 warn-assign = \
diff --git a/arch/Kconfig b/arch/Kconfig
index fe48fc7..53d7f61 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -158,4 +158,7 @@
 	  subsystem.  Also has support for calculating CPU cycle events
 	  to determine how many clock cycles in a given period.
 
+config HAVE_ARCH_JUMP_LABEL
+	bool
+
 source "kernel/gcov/Kconfig"
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index b9647bb..d04ccd7 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -9,6 +9,7 @@
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_SYSCALL_WRAPPERS
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_DMA_ATTRS
 	help
diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h
index 4157cd3..fe792ca 100644
--- a/arch/alpha/include/asm/perf_event.h
+++ b/arch/alpha/include/asm/perf_event.h
@@ -1,11 +1,6 @@
 #ifndef __ASM_ALPHA_PERF_EVENT_H
 #define __ASM_ALPHA_PERF_EVENT_H
 
-/* Alpha only supports software events through this interface. */
-extern void set_perf_event_pending(void);
-
-#define PERF_EVENT_INDEX_OFFSET 0
-
 #ifdef CONFIG_PERF_EVENTS
 extern void init_hw_perf_events(void);
 #else
diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 85d8e4f..1cc4968 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -307,7 +307,7 @@
 			     new_raw_count) != prev_raw_count)
 		goto again;
 
-	delta = (new_raw_count  - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
+	delta = (new_raw_count - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
 
 	/* It is possible on very rare occasions that the PMC has overflowed
 	 * but the interrupt is yet to come.  Detect and fix this situation.
@@ -402,14 +402,13 @@
 		struct hw_perf_event *hwc = &pe->hw;
 		int idx = hwc->idx;
 
-		if (cpuc->current_idx[j] != PMC_NO_INDEX) {
-			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
-			continue;
+		if (cpuc->current_idx[j] == PMC_NO_INDEX) {
+			alpha_perf_event_set_period(pe, hwc, idx);
+			cpuc->current_idx[j] = idx;
 		}
 
-		alpha_perf_event_set_period(pe, hwc, idx);
-		cpuc->current_idx[j] = idx;
-		cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
+		if (!(hwc->state & PERF_HES_STOPPED))
+			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
 	}
 	cpuc->config = cpuc->event[0]->hw.config_base;
 }
@@ -420,12 +419,13 @@
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static int alpha_pmu_enable(struct perf_event *event)
+static int alpha_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
 	int n0;
 	int ret;
-	unsigned long flags;
+	unsigned long irq_flags;
 
 	/*
 	 * The Sparc code has the IRQ disable first followed by the perf
@@ -435,8 +435,8 @@
 	 * nevertheless we disable the PMCs first to enable a potential
 	 * final PMI to occur before we disable interrupts.
 	 */
-	perf_disable();
-	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+	local_irq_save(irq_flags);
 
 	/* Default to error to be returned */
 	ret = -EAGAIN;
@@ -455,8 +455,12 @@
 		}
 	}
 
-	local_irq_restore(flags);
-	perf_enable();
+	hwc->state = PERF_HES_UPTODATE;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_STOPPED;
+
+	local_irq_restore(irq_flags);
+	perf_pmu_enable(event->pmu);
 
 	return ret;
 }
@@ -467,15 +471,15 @@
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static void alpha_pmu_disable(struct perf_event *event)
+static void alpha_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
-	unsigned long flags;
+	unsigned long irq_flags;
 	int j;
 
-	perf_disable();
-	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+	local_irq_save(irq_flags);
 
 	for (j = 0; j < cpuc->n_events; j++) {
 		if (event == cpuc->event[j]) {
@@ -501,8 +505,8 @@
 		}
 	}
 
-	local_irq_restore(flags);
-	perf_enable();
+	local_irq_restore(irq_flags);
+	perf_pmu_enable(event->pmu);
 }
 
 
@@ -514,13 +518,44 @@
 }
 
 
-static void alpha_pmu_unthrottle(struct perf_event *event)
+static void alpha_pmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		cpuc->idx_mask &= ~(1UL<<hwc->idx);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		alpha_perf_event_update(event, hwc, hwc->idx, 0);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_DISABLE, (1UL<<hwc->idx));
+}
+
+
+static void alpha_pmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+		alpha_perf_event_set_period(event, hwc, hwc->idx);
+	}
+
+	hwc->state = 0;
+
 	cpuc->idx_mask |= 1UL<<hwc->idx;
-	wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
 }
 
 
@@ -642,39 +677,36 @@
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= alpha_pmu_enable,
-	.disable	= alpha_pmu_disable,
-	.read		= alpha_pmu_read,
-	.unthrottle	= alpha_pmu_unthrottle,
-};
-
-
 /*
  * Main entry point to initialise a HW performance event.
  */
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int alpha_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!alpha_pmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	/* Do the real initialisation work. */
 	err = __hw_perf_event_init(event);
 
-	if (err)
-		return ERR_PTR(err);
-
-	return &pmu;
+	return err;
 }
 
-
-
 /*
  * Main entry point - enable HW performance counters.
  */
-void hw_perf_enable(void)
+static void alpha_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -700,7 +732,7 @@
  * Main entry point - disable HW performance counters.
  */
 
-void hw_perf_disable(void)
+static void alpha_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -713,6 +745,17 @@
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= alpha_pmu_enable,
+	.pmu_disable	= alpha_pmu_disable,
+	.event_init	= alpha_pmu_event_init,
+	.add		= alpha_pmu_add,
+	.del		= alpha_pmu_del,
+	.start		= alpha_pmu_start,
+	.stop		= alpha_pmu_stop,
+	.read		= alpha_pmu_read,
+};
+
 
 /*
  * Main entry point - don't know when this is called but it
@@ -766,7 +809,7 @@
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 
 	/* la_ptr is the counter that overflowed. */
-	if (unlikely(la_ptr >= perf_max_events)) {
+	if (unlikely(la_ptr >= alpha_pmu->num_pmcs)) {
 		/* This should never occur! */
 		irq_err_count++;
 		pr_warning("PMI: silly index %ld\n", la_ptr);
@@ -807,7 +850,7 @@
 			/* Interrupts coming too quickly; "throttle" the
 			 * counter, i.e., disable it for a little while.
 			 */
-			cpuc->idx_mask &= ~(1UL<<idx);
+			alpha_pmu_stop(event, 0);
 		}
 	}
 	wrperfmon(PERFMON_CMD_ENABLE, cpuc->idx_mask);
@@ -837,6 +880,7 @@
 
 	/* And set up PMU specification */
 	alpha_pmu = &ev67_pmu;
-	perf_max_events = alpha_pmu->num_pmcs;
+
+	perf_pmu_register(&pmu);
 }
 
diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c
index 396af17..0f1d849 100644
--- a/arch/alpha/kernel/time.c
+++ b/arch/alpha/kernel/time.c
@@ -41,7 +41,7 @@
 #include <linux/init.h>
 #include <linux/bcd.h>
 #include <linux/profile.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -83,25 +83,25 @@
 
 unsigned long est_cycle_freq;
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()  __get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()      __get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()     __get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()  __get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()      __get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()     __get_cpu_var(irq_work_pending) = 0
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
-	set_perf_event_pending_flag();
+	set_irq_work_pending_flag();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()      0
-#define clear_perf_event_pending()
+#define test_irq_work_pending()      0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 
 static inline __u32 rpcc(void)
@@ -191,9 +191,9 @@
 
 	write_sequnlock(&xtime_lock);
 
-	if (test_perf_event_pending()) {
-		clear_perf_event_pending();
-		perf_event_do_pending();
+	if (test_irq_work_pending()) {
+		clear_irq_work_pending();
+		irq_work_run();
 	}
 
 #ifndef CONFIG_SMP
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 88c97bc..9103904 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -23,6 +23,7 @@
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZO
 	select HAVE_KERNEL_LZMA
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_REGS_AND_STACK_ACCESS_API
@@ -1101,6 +1102,20 @@
 	  invalidated are not, resulting in an incoherency in the system page
 	  tables. The workaround changes the TLB flushing routines to invalidate
 	  entries regardless of the ASID.
+
+config ARM_ERRATA_743622
+	bool "ARM errata: Faulty hazard checking in the Store Buffer may lead to data corruption"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 743622 Cortex-A9
+	  (r2p0..r2p2) erratum. Under very rare conditions, a faulty
+	  optimisation in the Cortex-A9 Store Buffer may lead to data
+	  corruption. This workaround sets a specific bit in the diagnostic
+	  register of the Cortex-A9 which disables the Store Buffer
+	  optimisation, preventing the defect from occurring. This has no
+	  visible impact on the overall performance or power consumption of the
+	  processor.
+
 endmenu
 
 source "arch/arm/common/Kconfig"
diff --git a/arch/arm/include/asm/hw_irq.h b/arch/arm/include/asm/hw_irq.h
index 90831f6..5586b7c 100644
--- a/arch/arm/include/asm/hw_irq.h
+++ b/arch/arm/include/asm/hw_irq.h
@@ -24,4 +24,6 @@
 #define IRQF_PROBE	(1 << 1)
 #define IRQF_NOAUTOEN	(1 << 2)
 
+#define ARCH_IRQ_INIT_FLAGS	(IRQ_NOREQUEST | IRQ_NOPROBE)
+
 #endif
diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h
index b5799a3..c4aa4e8 100644
--- a/arch/arm/include/asm/perf_event.h
+++ b/arch/arm/include/asm/perf_event.h
@@ -12,18 +12,6 @@
 #ifndef __ARM_PERF_EVENT_H__
 #define __ARM_PERF_EVENT_H__
 
-/*
- * NOP: on *most* (read: all supported) ARM platforms, the performance
- * counter interrupts are regular interrupts and not an NMI. This
- * means that when we receive the interrupt we can call
- * perf_event_do_pending() that handles all of the work with
- * interrupts disabled.
- */
-static inline void
-set_perf_event_pending(void)
-{
-}
-
 /* ARM performance counters start from 1 (in the cp15 accesses) so use the
  * same indexes here for consistency. */
 #define PERF_EVENT_INDEX_OFFSET 1
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index c0d5c3b..36ad3be 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -154,14 +154,6 @@
 
 void __init init_IRQ(void)
 {
-	struct irq_desc *desc;
-	int irq;
-
-	for (irq = 0; irq < nr_irqs; irq++) {
-		desc = irq_to_desc_alloc_node(irq, 0);
-		desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE;
-	}
-
 	init_arch_irq();
 }
 
@@ -169,7 +161,7 @@
 int __init arch_probe_nr_irqs(void)
 {
 	nr_irqs = arch_nr_irqs ? arch_nr_irqs : NR_IRQS;
-	return 0;
+	return nr_irqs;
 }
 #endif
 
diff --git a/arch/arm/kernel/kprobes-decode.c b/arch/arm/kernel/kprobes-decode.c
index 8bccbfa..2c1f005 100644
--- a/arch/arm/kernel/kprobes-decode.c
+++ b/arch/arm/kernel/kprobes-decode.c
@@ -1162,11 +1162,12 @@
 {
 	/*
 	 * MSR   : cccc 0011 0x10 xxxx xxxx xxxx xxxx xxxx
-	 * Undef : cccc 0011 0x00 xxxx xxxx xxxx xxxx xxxx
+	 * Undef : cccc 0011 0100 xxxx xxxx xxxx xxxx xxxx
 	 * ALU op with S bit and Rd == 15 :
 	 *	   cccc 001x xxx1 xxxx 1111 xxxx xxxx xxxx
 	 */
-	if ((insn & 0x0f900000) == 0x03200000 ||	/* MSR & Undef */
+	if ((insn & 0x0fb00000) == 0x03200000 ||	/* MSR */
+	    (insn & 0x0ff00000) == 0x03400000 ||	/* Undef */
 	    (insn & 0x0e10f000) == 0x0210f000)		/* ALU s-bit, R15  */
 		return INSN_REJECTED;
 
@@ -1177,7 +1178,7 @@
 	 * *S (bit 20) updates condition codes
 	 * ADC/SBC/RSC reads the C flag
 	 */
-	insn &= 0xfff00fff;	/* Rn = r0, Rd = r0 */
+	insn &= 0xffff0fff;	/* Rd = r0 */
 	asi->insn[0] = insn;
 	asi->insn_handler = (insn & (1 << 20)) ?  /* S-bit */
 			emulate_alu_imm_rwflags : emulate_alu_imm_rflags;
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index ecbb028..49643b1 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -123,6 +123,12 @@
 }
 EXPORT_SYMBOL_GPL(armpmu_get_max_events);
 
+int perf_num_counters(void)
+{
+	return armpmu_get_max_events();
+}
+EXPORT_SYMBOL_GPL(perf_num_counters);
+
 #define HW_OP_UNSUPPORTED		0xFFFF
 
 #define C(_x) \
@@ -221,27 +227,6 @@
 }
 
 static void
-armpmu_disable(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-	int idx = hwc->idx;
-
-	WARN_ON(idx < 0);
-
-	clear_bit(idx, cpuc->active_mask);
-	armpmu->disable(hwc, idx);
-
-	barrier();
-
-	armpmu_event_update(event, hwc, idx);
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
-
-	perf_event_update_userpage(event);
-}
-
-static void
 armpmu_read(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -254,13 +239,44 @@
 }
 
 static void
-armpmu_unthrottle(struct perf_event *event)
+armpmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 
+	if (!armpmu)
+		return;
+
+	/*
+	 * ARM pmu always has to update the counter, so ignore
+	 * PERF_EF_UPDATE, see comments in armpmu_start().
+	 */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		armpmu->disable(hwc, hwc->idx);
+		barrier(); /* why? */
+		armpmu_event_update(event, hwc, hwc->idx);
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
+}
+
+static void
+armpmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!armpmu)
+		return;
+
+	/*
+	 * ARM pmu always has to reprogram the period, so ignore
+	 * PERF_EF_RELOAD, see the comment below.
+	 */
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
 	/*
 	 * Set the period again. Some counters can't be stopped, so when we
-	 * were throttled we simply disabled the IRQ source and the counter
+	 * were stopped we simply disabled the IRQ source and the counter
 	 * may have been left counting. If we don't do this step then we may
 	 * get an interrupt too soon or *way* too late if the overflow has
 	 * happened since disabling.
@@ -269,14 +285,33 @@
 	armpmu->enable(hwc, hwc->idx);
 }
 
+static void
+armpmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	WARN_ON(idx < 0);
+
+	clear_bit(idx, cpuc->active_mask);
+	armpmu_stop(event, PERF_EF_UPDATE);
+	cpuc->events[idx] = NULL;
+	clear_bit(idx, cpuc->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
 static int
-armpmu_enable(struct perf_event *event)
+armpmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx;
 	int err = 0;
 
+	perf_pmu_disable(event->pmu);
+
 	/* If we don't have a space for the counter then finish early. */
 	idx = armpmu->get_event_idx(cpuc, hwc);
 	if (idx < 0) {
@@ -293,25 +328,19 @@
 	cpuc->events[idx] = event;
 	set_bit(idx, cpuc->active_mask);
 
-	/* Set the period for the event. */
-	armpmu_event_set_period(event, hwc, idx);
-
-	/* Enable the event. */
-	armpmu->enable(hwc, idx);
+	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	if (flags & PERF_EF_START)
+		armpmu_start(event, PERF_EF_RELOAD);
 
 	/* Propagate our changes to the userspace mapping. */
 	perf_event_update_userpage(event);
 
 out:
+	perf_pmu_enable(event->pmu);
 	return err;
 }
 
-static struct pmu pmu = {
-	.enable	    = armpmu_enable,
-	.disable    = armpmu_disable,
-	.unthrottle = armpmu_unthrottle,
-	.read	    = armpmu_read,
-};
+static struct pmu pmu;
 
 static int
 validate_event(struct cpu_hw_events *cpuc,
@@ -491,20 +520,29 @@
 	return err;
 }
 
-const struct pmu *
-hw_perf_event_init(struct perf_event *event)
+static int armpmu_event_init(struct perf_event *event)
 {
 	int err = 0;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!armpmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	event->destroy = hw_perf_event_destroy;
 
 	if (!atomic_inc_not_zero(&active_events)) {
-		if (atomic_read(&active_events) > perf_max_events) {
+		if (atomic_read(&active_events) > armpmu->num_events) {
 			atomic_dec(&active_events);
-			return ERR_PTR(-ENOSPC);
+			return -ENOSPC;
 		}
 
 		mutex_lock(&pmu_reserve_mutex);
@@ -518,17 +556,16 @@
 	}
 
 	if (err)
-		return ERR_PTR(err);
+		return err;
 
 	err = __hw_perf_event_init(event);
 	if (err)
 		hw_perf_event_destroy(event);
 
-	return err ? ERR_PTR(err) : &pmu;
+	return err;
 }
 
-void
-hw_perf_enable(void)
+static void armpmu_enable(struct pmu *pmu)
 {
 	/* Enable all of the perf events on hardware. */
 	int idx;
@@ -549,13 +586,23 @@
 	armpmu->start();
 }
 
-void
-hw_perf_disable(void)
+static void armpmu_disable(struct pmu *pmu)
 {
 	if (armpmu)
 		armpmu->stop();
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= armpmu_enable,
+	.pmu_disable	= armpmu_disable,
+	.event_init	= armpmu_event_init,
+	.add		= armpmu_add,
+	.del		= armpmu_del,
+	.start		= armpmu_start,
+	.stop		= armpmu_stop,
+	.read		= armpmu_read,
+};
+
 /*
  * ARMv6 Performance counter handling code.
  *
@@ -1045,7 +1092,7 @@
 	 * platforms that can have the PMU interrupts raised as an NMI, this
 	 * will not work.
 	 */
-	perf_event_do_pending();
+	irq_work_run();
 
 	return IRQ_HANDLED;
 }
@@ -2021,7 +2068,7 @@
 	 * platforms that can have the PMU interrupts raised as an NMI, this
 	 * will not work.
 	 */
-	perf_event_do_pending();
+	irq_work_run();
 
 	return IRQ_HANDLED;
 }
@@ -2389,7 +2436,7 @@
 			armpmu->disable(hwc, idx);
 	}
 
-	perf_event_do_pending();
+	irq_work_run();
 
 	/*
 	 * Re-enable the PMU.
@@ -2716,7 +2763,7 @@
 			armpmu->disable(hwc, idx);
 	}
 
-	perf_event_do_pending();
+	irq_work_run();
 
 	/*
 	 * Re-enable the PMU.
@@ -2933,14 +2980,12 @@
 			armpmu = &armv6pmu;
 			memcpy(armpmu_perf_cache_map, armv6_perf_cache_map,
 					sizeof(armv6_perf_cache_map));
-			perf_max_events	= armv6pmu.num_events;
 			break;
 		case 0xB020:	/* ARM11mpcore */
 			armpmu = &armv6mpcore_pmu;
 			memcpy(armpmu_perf_cache_map,
 			       armv6mpcore_perf_cache_map,
 			       sizeof(armv6mpcore_perf_cache_map));
-			perf_max_events = armv6mpcore_pmu.num_events;
 			break;
 		case 0xC080:	/* Cortex-A8 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA8;
@@ -2952,7 +2997,6 @@
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		case 0xC090:	/* Cortex-A9 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA9;
@@ -2964,7 +3008,6 @@
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		}
 	/* Intel CPUs [xscale]. */
@@ -2975,13 +3018,11 @@
 			armpmu = &xscale1pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale1pmu.num_events;
 			break;
 		case 2:
 			armpmu = &xscale2pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale2pmu.num_events;
 			break;
 		}
 	}
@@ -2991,9 +3032,10 @@
 				arm_pmu_names[armpmu->id], armpmu->num_events);
 	} else {
 		pr_info("no hardware support available\n");
-		perf_max_events = -1;
 	}
 
+	perf_pmu_register(&pmu);
+
 	return 0;
 }
 arch_initcall(init_hw_perf_events);
@@ -3001,13 +3043,6 @@
 /*
  * Callchain handling code.
  */
-static inline void
-callchain_store(struct perf_callchain_entry *entry,
-		u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 /*
  * The registers we're interested in are at the end of the variable
@@ -3039,7 +3074,7 @@
 	if (__copy_from_user_inatomic(&buftail, tail, sizeof(buftail)))
 		return NULL;
 
-	callchain_store(entry, buftail.lr);
+	perf_callchain_store(entry, buftail.lr);
 
 	/*
 	 * Frame pointers should strictly progress back up the stack
@@ -3051,16 +3086,11 @@
 	return buftail.fp - 1;
 }
 
-static void
-perf_callchain_user(struct pt_regs *regs,
-		    struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct frame_tail *tail;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
 
 	tail = (struct frame_tail *)regs->ARM_fp - 1;
 
@@ -3078,56 +3108,18 @@
 		void *data)
 {
 	struct perf_callchain_entry *entry = data;
-	callchain_store(entry, fr->pc);
+	perf_callchain_store(entry, fr->pc);
 	return 0;
 }
 
-static void
-perf_callchain_kernel(struct pt_regs *regs,
-		      struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stackframe fr;
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
 	fr.fp = regs->ARM_fp;
 	fr.sp = regs->ARM_sp;
 	fr.lr = regs->ARM_lr;
 	fr.pc = regs->ARM_pc;
 	walk_stackframe(&fr, callchain_trace, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs,
-		  struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (!current || !current->pid)
-		return;
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-
-struct perf_callchain_entry *
-perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-	perf_do_callchain(regs, entry);
-	return entry;
-}
diff --git a/arch/arm/mach-at91/include/mach/system.h b/arch/arm/mach-at91/include/mach/system.h
index c80e090..ee8db15 100644
--- a/arch/arm/mach-at91/include/mach/system.h
+++ b/arch/arm/mach-at91/include/mach/system.h
@@ -28,17 +28,16 @@
 
 static inline void arch_idle(void)
 {
-#ifndef CONFIG_DEBUG_KERNEL
 	/*
 	 * Disable the processor clock.  The processor will be automatically
 	 * re-enabled by an interrupt or by a reset.
 	 */
 	at91_sys_write(AT91_PMC_SCDR, AT91_PMC_PCK);
-#else
+#ifndef CONFIG_CPU_ARM920T
 	/*
 	 * Set the processor (CP15) into 'Wait for Interrupt' mode.
-	 * Unlike disabling the processor clock via the PMC (above)
-	 *  this allows the processor to be woken via JTAG.
+	 * Post-RM9200 processors need this in conjunction with the above
+	 * to save power when idle.
 	 */
 	cpu_do_idle();
 #endif
diff --git a/arch/arm/mach-bcmring/dma.c b/arch/arm/mach-bcmring/dma.c
index 29c0a91..77eb35c 100644
--- a/arch/arm/mach-bcmring/dma.c
+++ b/arch/arm/mach-bcmring/dma.c
@@ -691,7 +691,7 @@
 
 	memset(&gDMA, 0, sizeof(gDMA));
 
-	init_MUTEX_LOCKED(&gDMA.lock);
+	sema_init(&gDMA.lock, 0);
 	init_waitqueue_head(&gDMA.freeChannelQ);
 
 	/* Initialize the Hardware */
@@ -1574,7 +1574,7 @@
 {
 	memset(memMap, 0, sizeof(*memMap));
 
-	init_MUTEX(&memMap->lock);
+	sema_init(&memMap->lock, 1);
 
 	return 0;
 }
diff --git a/arch/arm/mach-bcmring/irq.c b/arch/arm/mach-bcmring/irq.c
index dc1c493..e315263 100644
--- a/arch/arm/mach-bcmring/irq.c
+++ b/arch/arm/mach-bcmring/irq.c
@@ -67,21 +67,21 @@
 }
 
 static struct irq_chip bcmring_irq0_chip = {
-	.typename = "ARM-INTC0",
+	.name = "ARM-INTC0",
 	.ack = bcmring_mask_irq0,
 	.mask = bcmring_mask_irq0,	/* mask a specific interrupt, blocking its delivery. */
 	.unmask = bcmring_unmask_irq0,	/* unmaks an interrupt */
 };
 
 static struct irq_chip bcmring_irq1_chip = {
-	.typename = "ARM-INTC1",
+	.name = "ARM-INTC1",
 	.ack = bcmring_mask_irq1,
 	.mask = bcmring_mask_irq1,
 	.unmask = bcmring_unmask_irq1,
 };
 
 static struct irq_chip bcmring_irq2_chip = {
-	.typename = "ARM-SINTC",
+	.name = "ARM-SINTC",
 	.ack = bcmring_mask_irq2,
 	.mask = bcmring_mask_irq2,
 	.unmask = bcmring_unmask_irq2,
diff --git a/arch/arm/mach-ep93xx/dma-m2p.c b/arch/arm/mach-ep93xx/dma-m2p.c
index 8904ca4..a696d35 100644
--- a/arch/arm/mach-ep93xx/dma-m2p.c
+++ b/arch/arm/mach-ep93xx/dma-m2p.c
@@ -276,7 +276,7 @@
 	v &= ~(M2P_CONTROL_STALL_IRQ_EN | M2P_CONTROL_NFB_IRQ_EN);
 	m2p_set_control(ch, v);
 
-	while (m2p_channel_state(ch) == STATE_ON)
+	while (m2p_channel_state(ch) >= STATE_ON)
 		cpu_relax();
 
 	m2p_set_control(ch, 0x0);
diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig
index c5c0369..2f7e272 100644
--- a/arch/arm/mach-imx/Kconfig
+++ b/arch/arm/mach-imx/Kconfig
@@ -122,6 +122,7 @@
 	select IMX_HAVE_PLATFORM_IMX_I2C
 	select IMX_HAVE_PLATFORM_IMX_UART
 	select IMX_HAVE_PLATFORM_MXC_NAND
+	select MXC_ULPI if USB_ULPI
 	help
 	  Include support for Eukrea CPUIMX27 platform. This includes
 	  specific configurations for the module and its peripherals.
diff --git a/arch/arm/mach-imx/mach-cpuimx27.c b/arch/arm/mach-imx/mach-cpuimx27.c
index 339150a..6830afd 100644
--- a/arch/arm/mach-imx/mach-cpuimx27.c
+++ b/arch/arm/mach-imx/mach-cpuimx27.c
@@ -259,7 +259,7 @@
 	i2c_register_board_info(0, eukrea_cpuimx27_i2c_devices,
 				ARRAY_SIZE(eukrea_cpuimx27_i2c_devices));
 
-	imx27_add_i2c_imx1(&cpuimx27_i2c1_data);
+	imx27_add_i2c_imx0(&cpuimx27_i2c1_data);
 
 	platform_add_devices(platform_devices, ARRAY_SIZE(platform_devices));
 
diff --git a/arch/arm/mach-iop13xx/msi.c b/arch/arm/mach-iop13xx/msi.c
index f34b0ed..7149fcc 100644
--- a/arch/arm/mach-iop13xx/msi.c
+++ b/arch/arm/mach-iop13xx/msi.c
@@ -164,10 +164,10 @@
 static struct irq_chip iop13xx_msi_chip = {
 	.name = "PCI-MSI",
 	.ack = iop13xx_msi_nop,
-	.enable = unmask_msi_irq,
-	.disable = mask_msi_irq,
-	.mask = mask_msi_irq,
-	.unmask = unmask_msi_irq,
+	.irq_enable = unmask_msi_irq,
+	.irq_disable = mask_msi_irq,
+	.irq_mask = mask_msi_irq,
+	.irq_unmask = unmask_msi_irq,
 };
 
 int arch_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
diff --git a/arch/arm/mach-lpc32xx/phy3250.c b/arch/arm/mach-lpc32xx/phy3250.c
index bc9a42d..0c936cf 100644
--- a/arch/arm/mach-lpc32xx/phy3250.c
+++ b/arch/arm/mach-lpc32xx/phy3250.c
@@ -172,18 +172,12 @@
 }
 
 static struct pl022_config_chip spi0_chip_info = {
-	.lbm			= LOOPBACK_DISABLED,
 	.com_mode		= INTERRUPT_TRANSFER,
 	.iface			= SSP_INTERFACE_MOTOROLA_SPI,
 	.hierarchy		= SSP_MASTER,
 	.slave_tx_disable	= 0,
-	.endian_tx		= SSP_TX_LSB,
-	.endian_rx		= SSP_RX_LSB,
-	.data_size		= SSP_DATA_BITS_8,
 	.rx_lev_trig		= SSP_RX_4_OR_MORE_ELEM,
 	.tx_lev_trig		= SSP_TX_4_OR_MORE_EMPTY_LOC,
-	.clk_phase		= SSP_CLK_FIRST_EDGE,
-	.clk_pol		= SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len		= SSP_BITS_8,
 	.wait_state		= SSP_MWIRE_WAIT_ZERO,
 	.duplex			= SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -239,6 +233,7 @@
 			.max_speed_hz = 5000000,
 			.bus_num = 0,
 			.chip_select = 0,
+			.mode = SPI_MODE_0,
 			.platform_data = &eeprom,
 			.controller_data = &spi0_chip_info,
 		},
diff --git a/arch/arm/mach-s5p6440/cpu.c b/arch/arm/mach-s5p6440/cpu.c
index 526f33a..ec592e8 100644
--- a/arch/arm/mach-s5p6440/cpu.c
+++ b/arch/arm/mach-s5p6440/cpu.c
@@ -19,6 +19,7 @@
 #include <linux/sysdev.h>
 #include <linux/serial_core.h>
 #include <linux/platform_device.h>
+#include <linux/sched.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
diff --git a/arch/arm/mach-s5p6442/cpu.c b/arch/arm/mach-s5p6442/cpu.c
index a48fb55..70ac681 100644
--- a/arch/arm/mach-s5p6442/cpu.c
+++ b/arch/arm/mach-s5p6442/cpu.c
@@ -19,6 +19,7 @@
 #include <linux/sysdev.h>
 #include <linux/serial_core.h>
 #include <linux/platform_device.h>
+#include <linux/sched.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
diff --git a/arch/arm/mach-s5pc100/cpu.c b/arch/arm/mach-s5pc100/cpu.c
index 251c92a..cd1afbc 100644
--- a/arch/arm/mach-s5pc100/cpu.c
+++ b/arch/arm/mach-s5pc100/cpu.c
@@ -21,6 +21,7 @@
 #include <linux/sysdev.h>
 #include <linux/serial_core.h>
 #include <linux/platform_device.h>
+#include <linux/sched.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
diff --git a/arch/arm/mach-s5pv210/clock.c b/arch/arm/mach-s5pv210/clock.c
index cfecd70..d562670 100644
--- a/arch/arm/mach-s5pv210/clock.c
+++ b/arch/arm/mach-s5pv210/clock.c
@@ -173,11 +173,6 @@
 	return s5p_gatectrl(S5P_CLKGATE_IP3, clk, enable);
 }
 
-static int s5pv210_clk_ip4_ctrl(struct clk *clk, int enable)
-{
-	return s5p_gatectrl(S5P_CLKGATE_IP4, clk, enable);
-}
-
 static int s5pv210_clk_mask0_ctrl(struct clk *clk, int enable)
 {
 	return s5p_gatectrl(S5P_CLK_SRC_MASK0, clk, enable);
diff --git a/arch/arm/mach-s5pv210/cpu.c b/arch/arm/mach-s5pv210/cpu.c
index 77f456c..245b82b 100644
--- a/arch/arm/mach-s5pv210/cpu.c
+++ b/arch/arm/mach-s5pv210/cpu.c
@@ -19,6 +19,7 @@
 #include <linux/io.h>
 #include <linux/sysdev.h>
 #include <linux/platform_device.h>
+#include <linux/sched.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
diff --git a/arch/arm/mach-u300/dummyspichip.c b/arch/arm/mach-u300/dummyspichip.c
index 5f55012..03f7936 100644
--- a/arch/arm/mach-u300/dummyspichip.c
+++ b/arch/arm/mach-u300/dummyspichip.c
@@ -46,7 +46,6 @@
 	 * struct, this is just used here to alter the behaviour of the chip
 	 * in order to perform tests.
 	 */
-	struct pl022_config_chip *chip_info = spi->controller_data;
 	int status;
 	u8 txbuf[14] = {0xDE, 0xAD, 0xBE, 0xEF, 0x2B, 0xAD,
 			0xCA, 0xFE, 0xBA, 0xBE, 0xB1, 0x05,
@@ -72,7 +71,7 @@
 	 * Force chip to 8 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_8;
+	spi->bits_per_word = 8;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
@@ -159,7 +158,7 @@
 	 * Force chip to 16 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_16;
+	spi->bits_per_word = 16;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
diff --git a/arch/arm/mach-u300/spi.c b/arch/arm/mach-u300/spi.c
index f0e887b..edb2c0d 100644
--- a/arch/arm/mach-u300/spi.c
+++ b/arch/arm/mach-u300/spi.c
@@ -30,8 +30,6 @@
 }
 
 struct pl022_config_chip dummy_chip_info = {
-	/* Nominally this is LOOPBACK_DISABLED, but this is our dummy chip! */
-	.lbm = LOOPBACK_ENABLED,
 	/*
 	 * available POLLING_TRANSFER and INTERRUPT_TRANSFER,
 	 * DMA_TRANSFER does not work
@@ -42,14 +40,8 @@
 	.hierarchy = SSP_MASTER,
 	/* 0 = drive TX even as slave, 1 = do not drive TX as slave */
 	.slave_tx_disable = 0,
-	/* LSB first */
-	.endian_tx = SSP_TX_LSB,
-	.endian_rx = SSP_RX_LSB,
-	.data_size = SSP_DATA_BITS_8, /* used to be 12 in some default */
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len = SSP_BITS_12,
 	.wait_state = SSP_MWIRE_WAIT_ZERO,
 	.duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -75,7 +67,7 @@
 		.bus_num        = 0, /* Only one bus on this chip */
 		.chip_select    = 0,
 		/* Means SPI_CS_HIGH, change if e.g low CS */
-		.mode           = 0,
+		.mode           = SPI_MODE_1 | SPI_LSB_FIRST | SPI_LOOP,
 	},
 #endif
 };
diff --git a/arch/arm/mach-ux500/board-mop500.c b/arch/arm/mach-ux500/board-mop500.c
index 0e8fd13..219ae0c 100644
--- a/arch/arm/mach-ux500/board-mop500.c
+++ b/arch/arm/mach-ux500/board-mop500.c
@@ -55,19 +55,13 @@
 }
 
 struct pl022_config_chip ab4500_chip_info = {
-	.lbm = LOOPBACK_DISABLED,
 	.com_mode = INTERRUPT_TRANSFER,
 	.iface = SSP_INTERFACE_MOTOROLA_SPI,
 	/* we can act as master only */
 	.hierarchy = SSP_MASTER,
 	.slave_tx_disable = 0,
-	.endian_rx = SSP_RX_MSB,
-	.endian_tx = SSP_TX_MSB,
-	.data_size = SSP_DATA_BITS_24,
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_HIGH,
 	.cs_control = ab4500_spi_cs_control,
 };
 
@@ -83,7 +77,7 @@
 		.max_speed_hz = 12000000,
 		.bus_num = 0,
 		.chip_select = 0,
-		.mode = SPI_MODE_0,
+		.mode = SPI_MODE_3,
 		.irq = IRQ_DB8500_AB8500,
 	},
 };
diff --git a/arch/arm/mach-vexpress/ct-ca9x4.c b/arch/arm/mach-vexpress/ct-ca9x4.c
index efb1270..71fb173 100644
--- a/arch/arm/mach-vexpress/ct-ca9x4.c
+++ b/arch/arm/mach-vexpress/ct-ca9x4.c
@@ -68,7 +68,7 @@
 }
 
 #if 0
-static void ct_ca9x4_timer_init(void)
+static void __init ct_ca9x4_timer_init(void)
 {
 	writel(0, MMIO_P2V(CT_CA9X4_TIMER0) + TIMER_CTRL);
 	writel(0, MMIO_P2V(CT_CA9X4_TIMER1) + TIMER_CTRL);
@@ -222,7 +222,7 @@
 	.resource	= pmu_resources,
 };
 
-static void ct_ca9x4_init(void)
+static void __init ct_ca9x4_init(void)
 {
 	int i;
 
diff --git a/arch/arm/mach-vexpress/v2m.c b/arch/arm/mach-vexpress/v2m.c
index 817f0ad..7eaa232 100644
--- a/arch/arm/mach-vexpress/v2m.c
+++ b/arch/arm/mach-vexpress/v2m.c
@@ -48,7 +48,7 @@
 }
 
 
-static void v2m_timer_init(void)
+static void __init v2m_timer_init(void)
 {
 	writel(0, MMIO_P2V(V2M_TIMER0) + TIMER_CTRL);
 	writel(0, MMIO_P2V(V2M_TIMER1) + TIMER_CTRL);
diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index ab50627..17e7b0b 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -204,8 +204,12 @@
 	/*
 	 * Don't allow RAM to be mapped - this causes problems with ARMv6+
 	 */
-	if (WARN_ON(pfn_valid(pfn)))
-		return NULL;
+	if (pfn_valid(pfn)) {
+		printk(KERN_WARNING "BUG: Your driver calls ioremap() on system memory.  This leads\n"
+		       KERN_WARNING "to architecturally unpredictable behaviour on ARMv6+, and ioremap()\n"
+		       KERN_WARNING "will fail in the next kernel release.  Please fix your driver.\n");
+		WARN_ON(1);
+	}
 
 	type = get_mem_type(mtype);
 	if (!type)
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index 6a3a2d0..e8ed9dc 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -248,7 +248,7 @@
 	},
 	[MT_MEMORY] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_USER | L_PTE_EXEC,
+				L_PTE_WRITE | L_PTE_EXEC,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
 		.domain    = DOMAIN_KERNEL,
@@ -259,7 +259,7 @@
 	},
 	[MT_MEMORY_NONCACHED] = {
 		.prot_pte  = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
-				L_PTE_USER | L_PTE_EXEC | L_PTE_MT_BUFFERABLE,
+				L_PTE_WRITE | L_PTE_EXEC | L_PTE_MT_BUFFERABLE,
 		.prot_l1   = PMD_TYPE_TABLE,
 		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE,
 		.domain    = DOMAIN_KERNEL,
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index 7563ff0..197f21b 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -253,6 +253,14 @@
 	orreq	r10, r10, #1 << 22		@ set bit #22
 	mcreq	p15, 0, r10, c15, c0, 1		@ write diagnostic register
 #endif
+#ifdef CONFIG_ARM_ERRATA_743622
+	teq	r6, #0x20			@ present in r2p0
+	teqne	r6, #0x21			@ present in r2p1
+	teqne	r6, #0x22			@ present in r2p2
+	mrceq	p15, 0, r10, c15, c0, 1		@ read diagnostic register
+	orreq	r10, r10, #1 << 6		@ set bit #6
+	mcreq	p15, 0, r10, c15, c0, 1		@ write diagnostic register
+#endif
 
 3:	mov	r10, #0
 #ifdef HARVARD_CACHE
@@ -365,7 +373,7 @@
 	b	__v7_ca9mp_setup
 	.long	cpu_arch_name
 	.long	cpu_elf_name
-	.long	HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
+	.long	HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP|HWCAP_TLS
 	.long	cpu_v7_name
 	.long	v7_processor_functions
 	.long	v7wbi_tlb_fns
diff --git a/arch/arm/oprofile/Makefile b/arch/arm/oprofile/Makefile
index e666eaf..b2215c6 100644
--- a/arch/arm/oprofile/Makefile
+++ b/arch/arm/oprofile/Makefile
@@ -6,4 +6,8 @@
 		oprofilefs.o oprofile_stats.o \
 		timer_int.o )
 
+ifeq ($(CONFIG_HW_PERF_EVENTS),y)
+DRIVER_OBJS += $(addprefix ../../../drivers/oprofile/, oprofile_perf.o)
+endif
+
 oprofile-y				:= $(DRIVER_OBJS) common.o
diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index 72e09eb..8aa9744 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -25,139 +25,10 @@
 #include <asm/ptrace.h>
 
 #ifdef CONFIG_HW_PERF_EVENTS
-/*
- * Per performance monitor configuration as set via oprofilefs.
- */
-struct op_counter_config {
-	unsigned long count;
-	unsigned long enabled;
-	unsigned long event;
-	unsigned long unit_mask;
-	unsigned long kernel;
-	unsigned long user;
-	struct perf_event_attr attr;
-};
-
-static int op_arm_enabled;
-static DEFINE_MUTEX(op_arm_mutex);
-
-static struct op_counter_config *counter_config;
-static struct perf_event **perf_events[nr_cpumask_bits];
-static int perf_num_counters;
-
-/*
- * Overflow callback for oprofile.
- */
-static void op_overflow_handler(struct perf_event *event, int unused,
-			struct perf_sample_data *data, struct pt_regs *regs)
+char *op_name_from_perf_id(void)
 {
-	int id;
-	u32 cpu = smp_processor_id();
+	enum arm_perf_pmu_ids id = armpmu_get_pmu_id();
 
-	for (id = 0; id < perf_num_counters; ++id)
-		if (perf_events[cpu][id] == event)
-			break;
-
-	if (id != perf_num_counters)
-		oprofile_add_sample(regs, id);
-	else
-		pr_warning("oprofile: ignoring spurious overflow "
-				"on cpu %u\n", cpu);
-}
-
-/*
- * Called by op_arm_setup to create perf attributes to mirror the oprofile
- * settings in counter_config. Attributes are created as `pinned' events and
- * so are permanently scheduled on the PMU.
- */
-static void op_perf_setup(void)
-{
-	int i;
-	u32 size = sizeof(struct perf_event_attr);
-	struct perf_event_attr *attr;
-
-	for (i = 0; i < perf_num_counters; ++i) {
-		attr = &counter_config[i].attr;
-		memset(attr, 0, size);
-		attr->type		= PERF_TYPE_RAW;
-		attr->size		= size;
-		attr->config		= counter_config[i].event;
-		attr->sample_period	= counter_config[i].count;
-		attr->pinned		= 1;
-	}
-}
-
-static int op_create_counter(int cpu, int event)
-{
-	int ret = 0;
-	struct perf_event *pevent;
-
-	if (!counter_config[event].enabled || (perf_events[cpu][event] != NULL))
-		return ret;
-
-	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
-						  cpu, -1,
-						  op_overflow_handler);
-
-	if (IS_ERR(pevent)) {
-		ret = PTR_ERR(pevent);
-	} else if (pevent->state != PERF_EVENT_STATE_ACTIVE) {
-		perf_event_release_kernel(pevent);
-		pr_warning("oprofile: failed to enable event %d "
-				"on CPU %d\n", event, cpu);
-		ret = -EBUSY;
-	} else {
-		perf_events[cpu][event] = pevent;
-	}
-
-	return ret;
-}
-
-static void op_destroy_counter(int cpu, int event)
-{
-	struct perf_event *pevent = perf_events[cpu][event];
-
-	if (pevent) {
-		perf_event_release_kernel(pevent);
-		perf_events[cpu][event] = NULL;
-	}
-}
-
-/*
- * Called by op_arm_start to create active perf events based on the
- * perviously configured attributes.
- */
-static int op_perf_start(void)
-{
-	int cpu, event, ret = 0;
-
-	for_each_online_cpu(cpu) {
-		for (event = 0; event < perf_num_counters; ++event) {
-			ret = op_create_counter(cpu, event);
-			if (ret)
-				goto out;
-		}
-	}
-
-out:
-	return ret;
-}
-
-/*
- * Called by op_arm_stop at the end of a profiling run.
- */
-static void op_perf_stop(void)
-{
-	int cpu, event;
-
-	for_each_online_cpu(cpu)
-		for (event = 0; event < perf_num_counters; ++event)
-			op_destroy_counter(cpu, event);
-}
-
-
-static char *op_name_from_perf_id(enum arm_perf_pmu_ids id)
-{
 	switch (id) {
 	case ARM_PERF_PMU_ID_XSCALE1:
 		return "arm/xscale1";
@@ -176,116 +47,6 @@
 	}
 }
 
-static int op_arm_create_files(struct super_block *sb, struct dentry *root)
-{
-	unsigned int i;
-
-	for (i = 0; i < perf_num_counters; i++) {
-		struct dentry *dir;
-		char buf[4];
-
-		snprintf(buf, sizeof buf, "%d", i);
-		dir = oprofilefs_mkdir(sb, root, buf);
-		oprofilefs_create_ulong(sb, dir, "enabled", &counter_config[i].enabled);
-		oprofilefs_create_ulong(sb, dir, "event", &counter_config[i].event);
-		oprofilefs_create_ulong(sb, dir, "count", &counter_config[i].count);
-		oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask);
-		oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel);
-		oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user);
-	}
-
-	return 0;
-}
-
-static int op_arm_setup(void)
-{
-	spin_lock(&oprofilefs_lock);
-	op_perf_setup();
-	spin_unlock(&oprofilefs_lock);
-	return 0;
-}
-
-static int op_arm_start(void)
-{
-	int ret = -EBUSY;
-
-	mutex_lock(&op_arm_mutex);
-	if (!op_arm_enabled) {
-		ret = 0;
-		op_perf_start();
-		op_arm_enabled = 1;
-	}
-	mutex_unlock(&op_arm_mutex);
-	return ret;
-}
-
-static void op_arm_stop(void)
-{
-	mutex_lock(&op_arm_mutex);
-	if (op_arm_enabled)
-		op_perf_stop();
-	op_arm_enabled = 0;
-	mutex_unlock(&op_arm_mutex);
-}
-
-#ifdef CONFIG_PM
-static int op_arm_suspend(struct platform_device *dev, pm_message_t state)
-{
-	mutex_lock(&op_arm_mutex);
-	if (op_arm_enabled)
-		op_perf_stop();
-	mutex_unlock(&op_arm_mutex);
-	return 0;
-}
-
-static int op_arm_resume(struct platform_device *dev)
-{
-	mutex_lock(&op_arm_mutex);
-	if (op_arm_enabled && op_perf_start())
-		op_arm_enabled = 0;
-	mutex_unlock(&op_arm_mutex);
-	return 0;
-}
-
-static struct platform_driver oprofile_driver = {
-	.driver		= {
-		.name		= "arm-oprofile",
-	},
-	.resume		= op_arm_resume,
-	.suspend	= op_arm_suspend,
-};
-
-static struct platform_device *oprofile_pdev;
-
-static int __init init_driverfs(void)
-{
-	int ret;
-
-	ret = platform_driver_register(&oprofile_driver);
-	if (ret)
-		goto out;
-
-	oprofile_pdev =	platform_device_register_simple(
-				oprofile_driver.driver.name, 0, NULL, 0);
-	if (IS_ERR(oprofile_pdev)) {
-		ret = PTR_ERR(oprofile_pdev);
-		platform_driver_unregister(&oprofile_driver);
-	}
-
-out:
-	return ret;
-}
-
-static void  exit_driverfs(void)
-{
-	platform_device_unregister(oprofile_pdev);
-	platform_driver_unregister(&oprofile_driver);
-}
-#else
-static int __init init_driverfs(void) { return 0; }
-#define exit_driverfs() do { } while (0)
-#endif /* CONFIG_PM */
-
 static int report_trace(struct stackframe *frame, void *d)
 {
 	unsigned int *depth = d;
@@ -350,74 +111,14 @@
 
 int __init oprofile_arch_init(struct oprofile_operations *ops)
 {
-	int cpu, ret = 0;
-
-	perf_num_counters = armpmu_get_max_events();
-
-	counter_config = kcalloc(perf_num_counters,
-			sizeof(struct op_counter_config), GFP_KERNEL);
-
-	if (!counter_config) {
-		pr_info("oprofile: failed to allocate %d "
-				"counters\n", perf_num_counters);
-		return -ENOMEM;
-	}
-
-	ret = init_driverfs();
-	if (ret) {
-		kfree(counter_config);
-		counter_config = NULL;
-		return ret;
-	}
-
-	for_each_possible_cpu(cpu) {
-		perf_events[cpu] = kcalloc(perf_num_counters,
-				sizeof(struct perf_event *), GFP_KERNEL);
-		if (!perf_events[cpu]) {
-			pr_info("oprofile: failed to allocate %d perf events "
-					"for cpu %d\n", perf_num_counters, cpu);
-			while (--cpu >= 0)
-				kfree(perf_events[cpu]);
-			return -ENOMEM;
-		}
-	}
-
 	ops->backtrace		= arm_backtrace;
-	ops->create_files	= op_arm_create_files;
-	ops->setup		= op_arm_setup;
-	ops->start		= op_arm_start;
-	ops->stop		= op_arm_stop;
-	ops->shutdown		= op_arm_stop;
-	ops->cpu_type		= op_name_from_perf_id(armpmu_get_pmu_id());
 
-	if (!ops->cpu_type)
-		ret = -ENODEV;
-	else
-		pr_info("oprofile: using %s\n", ops->cpu_type);
-
-	return ret;
+	return oprofile_perf_init(ops);
 }
 
-void oprofile_arch_exit(void)
+void __exit oprofile_arch_exit(void)
 {
-	int cpu, id;
-	struct perf_event *event;
-
-	if (*perf_events) {
-		for_each_possible_cpu(cpu) {
-			for (id = 0; id < perf_num_counters; ++id) {
-				event = perf_events[cpu][id];
-				if (event != NULL)
-					perf_event_release_kernel(event);
-			}
-			kfree(perf_events[cpu]);
-		}
-	}
-
-	if (counter_config) {
-		kfree(counter_config);
-		exit_driverfs();
-	}
+	oprofile_perf_exit();
 }
 #else
 int __init oprofile_arch_init(struct oprofile_operations *ops)
@@ -425,5 +126,5 @@
 	pr_info("oprofile: hardware counters not available\n");
 	return -ENODEV;
 }
-void oprofile_arch_exit(void) {}
+void __exit oprofile_arch_exit(void) {}
 #endif /* CONFIG_HW_PERF_EVENTS */
diff --git a/arch/arm/plat-omap/iommu.c b/arch/arm/plat-omap/iommu.c
index a202a2c..6cd151b 100644
--- a/arch/arm/plat-omap/iommu.c
+++ b/arch/arm/plat-omap/iommu.c
@@ -320,6 +320,7 @@
 		if ((start <= da) && (da < start + bytes)) {
 			dev_dbg(obj->dev, "%s: %08x<=%08x(%x)\n",
 				__func__, start, da, bytes);
+			iotlb_load_cr(obj, &cr);
 			iommu_write_reg(obj, 1, MMU_FLUSH_ENTRY);
 		}
 	}
diff --git a/arch/arm/plat-samsung/adc.c b/arch/arm/plat-samsung/adc.c
index 04d9521..e8f2be2 100644
--- a/arch/arm/plat-samsung/adc.c
+++ b/arch/arm/plat-samsung/adc.c
@@ -435,7 +435,6 @@
 static int s3c_adc_resume(struct platform_device *pdev)
 {
 	struct adc_device *adc = platform_get_drvdata(pdev);
-	unsigned long flags;
 
 	clk_enable(adc->clk);
 	enable_irq(adc->irq);
diff --git a/arch/arm/plat-samsung/clock.c b/arch/arm/plat-samsung/clock.c
index 90a2051..e8d20b0 100644
--- a/arch/arm/plat-samsung/clock.c
+++ b/arch/arm/plat-samsung/clock.c
@@ -48,6 +48,9 @@
 #include <plat/clock.h>
 #include <plat/cpu.h>
 
+#include <linux/serial_core.h>
+#include <plat/regs-serial.h> /* for s3c24xx_uart_devs */
+
 /* clock information */
 
 static LIST_HEAD(clocks);
@@ -65,6 +68,28 @@
 	return 0;
 }
 
+static int dev_is_s3c_uart(struct device *dev)
+{
+	struct platform_device **pdev = s3c24xx_uart_devs;
+	int i;
+	for (i = 0; i < ARRAY_SIZE(s3c24xx_uart_devs); i++, pdev++)
+		if (*pdev && dev == &(*pdev)->dev)
+			return 1;
+	return 0;
+}
+
+/*
+ * Serial drivers call get_clock() very early, before platform bus
+ * has been set up, this requires a special check to let them get
+ * a proper clock
+ */
+
+static int dev_is_platform_device(struct device *dev)
+{
+	return dev->bus == &platform_bus_type ||
+	       (dev->bus == NULL && dev_is_s3c_uart(dev));
+}
+
 /* Clock API calls */
 
 struct clk *clk_get(struct device *dev, const char *id)
@@ -73,7 +98,7 @@
 	struct clk *clk = ERR_PTR(-ENOENT);
 	int idno;
 
-	if (dev == NULL || dev->bus != &platform_bus_type)
+	if (dev == NULL || !dev_is_platform_device(dev))
 		idno = -1;
 	else
 		idno = to_platform_device(dev)->id;
diff --git a/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h b/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
index e5aba8f..b226f740 100644
--- a/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
+++ b/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
@@ -32,6 +32,8 @@
  * struct s3c64xx_spi_info - SPI Controller defining structure
  * @src_clk_nr: Clock source index for the CLK_CFG[SPI_CLKSEL] field.
  * @src_clk_name: Platform name of the corresponding clock.
+ * @clk_from_cmu: If the SPI clock/prescalar control block is present
+ *     by the platform's clock-management-unit and not in SPI controller.
  * @num_cs: Number of CS this controller emulates.
  * @cfg_gpio: Configure pins for this SPI controller.
  * @fifo_lvl_mask: All tx fifo_lvl fields start at offset-6
@@ -41,6 +43,7 @@
 struct s3c64xx_spi_info {
 	int src_clk_nr;
 	char *src_clk_name;
+	bool clk_from_cmu;
 
 	int num_cs;
 
diff --git a/arch/blackfin/include/asm/bfin5xx_spi.h b/arch/blackfin/include/asm/bfin5xx_spi.h
index ed4f8c6..4223cf0 100644
--- a/arch/blackfin/include/asm/bfin5xx_spi.h
+++ b/arch/blackfin/include/asm/bfin5xx_spi.h
@@ -11,26 +11,17 @@
 
 #define MIN_SPI_BAUD_VAL	2
 
-#define SPI_READ              0
-#define SPI_WRITE             1
-
-#define SPI_CTRL_OFF            0x0
-#define SPI_FLAG_OFF            0x4
-#define SPI_STAT_OFF            0x8
-#define SPI_TXBUFF_OFF          0xc
-#define SPI_RXBUFF_OFF          0x10
-#define SPI_BAUD_OFF            0x14
-#define SPI_SHAW_OFF            0x18
-
-
 #define BIT_CTL_ENABLE      0x4000
 #define BIT_CTL_OPENDRAIN   0x2000
 #define BIT_CTL_MASTER      0x1000
-#define BIT_CTL_POLAR       0x0800
-#define BIT_CTL_PHASE       0x0400
-#define BIT_CTL_BITORDER    0x0200
+#define BIT_CTL_CPOL        0x0800
+#define BIT_CTL_CPHA        0x0400
+#define BIT_CTL_LSBF        0x0200
 #define BIT_CTL_WORDSIZE    0x0100
-#define BIT_CTL_MISOENABLE  0x0020
+#define BIT_CTL_EMISO       0x0020
+#define BIT_CTL_PSSE        0x0010
+#define BIT_CTL_GM          0x0008
+#define BIT_CTL_SZ          0x0004
 #define BIT_CTL_RXMOD       0x0000
 #define BIT_CTL_TXMOD       0x0001
 #define BIT_CTL_TIMOD_DMA_TX 0x0003
@@ -50,61 +41,7 @@
 #define BIT_STU_SENDOVER    0x0001
 #define BIT_STU_RECVFULL    0x0020
 
-#define CFG_SPI_ENABLE      1
-#define CFG_SPI_DISABLE     0
-
-#define CFG_SPI_OUTENABLE   1
-#define CFG_SPI_OUTDISABLE  0
-
-#define CFG_SPI_ACTLOW      1
-#define CFG_SPI_ACTHIGH     0
-
-#define CFG_SPI_PHASESTART  1
-#define CFG_SPI_PHASEMID    0
-
-#define CFG_SPI_MASTER      1
-#define CFG_SPI_SLAVE       0
-
-#define CFG_SPI_SENELAST    0
-#define CFG_SPI_SENDZERO    1
-
-#define CFG_SPI_RCVFLUSH    1
-#define CFG_SPI_RCVDISCARD  0
-
-#define CFG_SPI_LSBFIRST    1
-#define CFG_SPI_MSBFIRST    0
-
-#define CFG_SPI_WORDSIZE16  1
-#define CFG_SPI_WORDSIZE8   0
-
-#define CFG_SPI_MISOENABLE   1
-#define CFG_SPI_MISODISABLE  0
-
-#define CFG_SPI_READ      0x00
-#define CFG_SPI_WRITE     0x01
-#define CFG_SPI_DMAREAD   0x02
-#define CFG_SPI_DMAWRITE  0x03
-
-#define CFG_SPI_CSCLEARALL  0
-#define CFG_SPI_CHIPSEL1    1
-#define CFG_SPI_CHIPSEL2    2
-#define CFG_SPI_CHIPSEL3    3
-#define CFG_SPI_CHIPSEL4    4
-#define CFG_SPI_CHIPSEL5    5
-#define CFG_SPI_CHIPSEL6    6
-#define CFG_SPI_CHIPSEL7    7
-
-#define CFG_SPI_CS1VALUE    1
-#define CFG_SPI_CS2VALUE    2
-#define CFG_SPI_CS3VALUE    3
-#define CFG_SPI_CS4VALUE    4
-#define CFG_SPI_CS5VALUE    5
-#define CFG_SPI_CS6VALUE    6
-#define CFG_SPI_CS7VALUE    7
-
-#define CMD_SPI_SET_BAUDRATE  2
-#define CMD_SPI_GET_SYSTEMCLOCK   25
-#define CMD_SPI_SET_WRITECONTINUOUS     26
+#define MAX_CTRL_CS          8  /* cs in spi controller */
 
 /* device.platform_data for SSP controller devices */
 struct bfin5xx_spi_master {
@@ -120,9 +57,7 @@
 	u16 ctl_reg;
 	u8 enable_dma;
 	u8 bits_per_word;
-	u8 cs_change_per_word;
 	u16 cs_chg_udelay; /* Some devices require 16-bit delays */
-	u32 cs_gpio;
 	/* Value to send if no TX value is supplied, usually 0x0 or 0xFFFF */
 	u16 idle_tx_val;
 	u8 pio_interrupt; /* Enable spi data irq */
diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 16399bd..0f2417d 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -7,6 +7,7 @@
 	default y
 	select HAVE_IDE
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 
 config ZONE_DMA
diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile
index f470975..4ff2fb1 100644
--- a/arch/frv/lib/Makefile
+++ b/arch/frv/lib/Makefile
@@ -5,4 +5,4 @@
 lib-y := \
 	__ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \
 	checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \
-	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_event.o
+	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o
diff --git a/arch/frv/lib/perf_event.c b/arch/frv/lib/perf_event.c
deleted file mode 100644
index 9ac5acf..0000000
--- a/arch/frv/lib/perf_event.c
+++ /dev/null
@@ -1,19 +0,0 @@
-/* Performance event handling
- *
- * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#include <linux/perf_event.h>
-
-/*
- * mark the performance event as pending
- */
-void set_perf_event_pending(void)
-{
-}
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index ba22849..7c82fa1 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -53,6 +53,9 @@
 	bool
 	default y
 
+config ARCH_DMA_ADDR_T_64BIT
+	def_bool y
+
 config NEED_DMA_MAP_STATE
 	def_bool y
 
@@ -62,6 +65,9 @@
 config SWIOTLB
        bool
 
+config STACKTRACE_SUPPORT
+	def_bool y
+
 config GENERIC_LOCKBREAK
 	def_bool n
 
@@ -683,8 +689,10 @@
 # Use the generic interrupt handling code in kernel/irq/:
 #
 config GENERIC_HARDIRQS
-	bool
-	default y
+	def_bool y
+
+config GENERIC_HARDIRQS_NO__DO_IRQ
+	def_bool y
 
 config GENERIC_IRQ_PROBE
 	bool
diff --git a/arch/ia64/include/asm/compat.h b/arch/ia64/include/asm/compat.h
deleted file mode 100644
index 9301a28..0000000
--- a/arch/ia64/include/asm/compat.h
+++ /dev/null
@@ -1,208 +0,0 @@
-#ifndef _ASM_IA64_COMPAT_H
-#define _ASM_IA64_COMPAT_H
-/*
- * Architecture specific compatibility types
- */
-#include <linux/types.h>
-
-#define COMPAT_USER_HZ		100
-#define COMPAT_UTS_MACHINE	"i686\0\0\0"
-
-typedef u32		compat_size_t;
-typedef s32		compat_ssize_t;
-typedef s32		compat_time_t;
-typedef s32		compat_clock_t;
-typedef s32		compat_key_t;
-typedef s32		compat_pid_t;
-typedef u16		__compat_uid_t;
-typedef u16		__compat_gid_t;
-typedef u32		__compat_uid32_t;
-typedef u32		__compat_gid32_t;
-typedef u16		compat_mode_t;
-typedef u32		compat_ino_t;
-typedef u16		compat_dev_t;
-typedef s32		compat_off_t;
-typedef s64		compat_loff_t;
-typedef u16		compat_nlink_t;
-typedef u16		compat_ipc_pid_t;
-typedef s32		compat_daddr_t;
-typedef u32		compat_caddr_t;
-typedef __kernel_fsid_t	compat_fsid_t;
-typedef s32		compat_timer_t;
-
-typedef s32		compat_int_t;
-typedef s32		compat_long_t;
-typedef s64 __attribute__((aligned(4))) compat_s64;
-typedef u32		compat_uint_t;
-typedef u32		compat_ulong_t;
-typedef u64 __attribute__((aligned(4))) compat_u64;
-
-struct compat_timespec {
-	compat_time_t	tv_sec;
-	s32		tv_nsec;
-};
-
-struct compat_timeval {
-	compat_time_t	tv_sec;
-	s32		tv_usec;
-};
-
-struct compat_stat {
-	compat_dev_t	st_dev;
-	u16		__pad1;
-	compat_ino_t	st_ino;
-	compat_mode_t	st_mode;
-	compat_nlink_t	st_nlink;
-	__compat_uid_t	st_uid;
-	__compat_gid_t	st_gid;
-	compat_dev_t	st_rdev;
-	u16		__pad2;
-	u32		st_size;
-	u32		st_blksize;
-	u32		st_blocks;
-	u32		st_atime;
-	u32		st_atime_nsec;
-	u32		st_mtime;
-	u32		st_mtime_nsec;
-	u32		st_ctime;
-	u32		st_ctime_nsec;
-	u32		__unused4;
-	u32		__unused5;
-};
-
-struct compat_flock {
-	short		l_type;
-	short		l_whence;
-	compat_off_t	l_start;
-	compat_off_t	l_len;
-	compat_pid_t	l_pid;
-};
-
-#define F_GETLK64	12
-#define F_SETLK64	13
-#define F_SETLKW64	14
-
-/*
- * IA32 uses 4 byte alignment for 64 bit quantities,
- * so we need to pack this structure.
- */
-struct compat_flock64 {
-	short		l_type;
-	short		l_whence;
-	compat_loff_t	l_start;
-	compat_loff_t	l_len;
-	compat_pid_t	l_pid;
-} __attribute__((packed));
-
-struct compat_statfs {
-	int		f_type;
-	int		f_bsize;
-	int		f_blocks;
-	int		f_bfree;
-	int		f_bavail;
-	int		f_files;
-	int		f_ffree;
-	compat_fsid_t	f_fsid;
-	int		f_namelen;	/* SunOS ignores this field. */
-	int		f_frsize;
-	int		f_spare[5];
-};
-
-#define COMPAT_RLIM_OLD_INFINITY       0x7fffffff
-#define COMPAT_RLIM_INFINITY           0xffffffff
-
-typedef u32		compat_old_sigset_t;	/* at least 32 bits */
-
-#define _COMPAT_NSIG		64
-#define _COMPAT_NSIG_BPW	32
-
-typedef u32		compat_sigset_word;
-
-#define COMPAT_OFF_T_MAX	0x7fffffff
-#define COMPAT_LOFF_T_MAX	0x7fffffffffffffffL
-
-struct compat_ipc64_perm {
-	compat_key_t key;
-	__compat_uid32_t uid;
-	__compat_gid32_t gid;
-	__compat_uid32_t cuid;
-	__compat_gid32_t cgid;
-	unsigned short mode;
-	unsigned short __pad1;
-	unsigned short seq;
-	unsigned short __pad2;
-	compat_ulong_t unused1;
-	compat_ulong_t unused2;
-};
-
-struct compat_semid64_ds {
-	struct compat_ipc64_perm sem_perm;
-	compat_time_t  sem_otime;
-	compat_ulong_t __unused1;
-	compat_time_t  sem_ctime;
-	compat_ulong_t __unused2;
-	compat_ulong_t sem_nsems;
-	compat_ulong_t __unused3;
-	compat_ulong_t __unused4;
-};
-
-struct compat_msqid64_ds {
-	struct compat_ipc64_perm msg_perm;
-	compat_time_t  msg_stime;
-	compat_ulong_t __unused1;
-	compat_time_t  msg_rtime;
-	compat_ulong_t __unused2;
-	compat_time_t  msg_ctime;
-	compat_ulong_t __unused3;
-	compat_ulong_t msg_cbytes;
-	compat_ulong_t msg_qnum;
-	compat_ulong_t msg_qbytes;
-	compat_pid_t   msg_lspid;
-	compat_pid_t   msg_lrpid;
-	compat_ulong_t __unused4;
-	compat_ulong_t __unused5;
-};
-
-struct compat_shmid64_ds {
-	struct compat_ipc64_perm shm_perm;
-	compat_size_t  shm_segsz;
-	compat_time_t  shm_atime;
-	compat_ulong_t __unused1;
-	compat_time_t  shm_dtime;
-	compat_ulong_t __unused2;
-	compat_time_t  shm_ctime;
-	compat_ulong_t __unused3;
-	compat_pid_t   shm_cpid;
-	compat_pid_t   shm_lpid;
-	compat_ulong_t shm_nattch;
-	compat_ulong_t __unused4;
-	compat_ulong_t __unused5;
-};
-
-/*
- * A pointer passed in from user mode. This should not be used for syscall parameters,
- * just declare them as pointers because the syscall entry code will have appropriately
- * converted them already.
- */
-typedef	u32		compat_uptr_t;
-
-static inline void __user *
-compat_ptr (compat_uptr_t uptr)
-{
-	return (void __user *) (unsigned long) uptr;
-}
-
-static inline compat_uptr_t
-ptr_to_compat(void __user *uptr)
-{
-	return (u32)(unsigned long)uptr;
-}
-
-static __inline__ void __user *
-arch_compat_alloc_user_space (long len)
-{
-	struct pt_regs *regs = task_pt_regs(current);
-	return (void __user *) (((regs->r12 & 0xffffffff) & -16) - len);
-}
-
-#endif /* _ASM_IA64_COMPAT_H */
diff --git a/arch/ia64/include/asm/hardirq.h b/arch/ia64/include/asm/hardirq.h
index d514cd9..8fb7d33 100644
--- a/arch/ia64/include/asm/hardirq.h
+++ b/arch/ia64/include/asm/hardirq.h
@@ -6,12 +6,6 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
-
-#include <linux/threads.h>
-#include <linux/irq.h>
-
-#include <asm/processor.h>
-
 /*
  * No irq_cpustat_t for IA-64.  The data is held in the per-CPU data structure.
  */
@@ -20,6 +14,11 @@
 
 #define local_softirq_pending()		(local_cpu_data->softirq_pending)
 
+#include <linux/threads.h>
+#include <linux/irq.h>
+
+#include <asm/processor.h>
+
 extern void __iomem *ipi_base_addr;
 
 void ack_bad_irq(unsigned int irq);
diff --git a/arch/ia64/include/asm/iommu_table.h b/arch/ia64/include/asm/iommu_table.h
new file mode 100644
index 0000000..92c8d36
--- /dev/null
+++ b/arch/ia64/include/asm/iommu_table.h
@@ -0,0 +1,6 @@
+#ifndef _ASM_IA64_IOMMU_TABLE_H
+#define _ASM_IA64_IOMMU_TABLE_H
+
+#define IOMMU_INIT_POST(_detect)
+
+#endif /* _ASM_IA64_IOMMU_TABLE_H */
diff --git a/arch/ia64/include/asm/system.h b/arch/ia64/include/asm/system.h
index 2feb7f6..6cca307 100644
--- a/arch/ia64/include/asm/system.h
+++ b/arch/ia64/include/asm/system.h
@@ -196,10 +196,6 @@
 
 void default_idle(void);
 
-#ifdef CONFIG_VIRT_CPU_ACCOUNTING
-extern void account_system_vtime(struct task_struct *);
-#endif
-
 #endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index db10b1e..395c2f2 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -34,6 +34,7 @@
 obj-$(CONFIG_PCI_MSI)		+= msi_ia64.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 obj-$(CONFIG_IA64_MC_ERR_INJECT)+= err_inject.o
+obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 
 obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirtentry.o \
 				   paravirt_patch.o
diff --git a/arch/ia64/kernel/cyclone.c b/arch/ia64/kernel/cyclone.c
index 71e3586..d52f1f7 100644
--- a/arch/ia64/kernel/cyclone.c
+++ b/arch/ia64/kernel/cyclone.c
@@ -59,13 +59,13 @@
 		return -ENODEV;
 	}
 	base = readq(reg);
+	iounmap(reg);
 	if(!base){
 		printk(KERN_ERR "Summit chipset: Could not find valid CBAR"
 				" value.\n");
 		use_cyclone = 0;
 		return -ENODEV;
 	}
-	iounmap(reg);
 
 	/* setup PMCC */
 	offset = (base + CYCLONE_PMCC_OFFSET);
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 7ded766..22c3840 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -108,10 +108,6 @@
 #define DBG(fmt...)
 #endif
 
-#define NR_PREALLOCATE_RTE_ENTRIES \
-	(PAGE_SIZE / sizeof(struct iosapic_rte_info))
-#define RTE_PREALLOCATED	(1)
-
 static DEFINE_SPINLOCK(iosapic_lock);
 
 /*
@@ -136,7 +132,6 @@
 	struct list_head rte_list;	/* RTEs sharing the same vector */
 	char		rte_index;	/* IOSAPIC RTE index */
 	int		refcnt;		/* reference counter */
-	unsigned int	flags;		/* flags */
 	struct iosapic	*iosapic;
 } ____cacheline_aligned;
 
@@ -155,9 +150,6 @@
 
 static unsigned char pcat_compat __devinitdata;	/* 8259 compatibility flag */
 
-static int iosapic_kmalloc_ok;
-static LIST_HEAD(free_rte_list);
-
 static inline void
 iosapic_write(struct iosapic *iosapic, unsigned int reg, u32 val)
 {
@@ -394,7 +386,7 @@
 }
 
 static void
-iosapic_end_level_irq (unsigned int irq)
+iosapic_unmask_level_irq (unsigned int irq)
 {
 	ia64_vector vec = irq_to_vector(irq);
 	struct iosapic_rte_info *rte;
@@ -404,7 +396,8 @@
 	if (unlikely(irq_desc[irq].status & IRQ_MOVE_PENDING)) {
 		do_unmask_irq = 1;
 		mask_irq(irq);
-	}
+	} else
+		unmask_irq(irq);
 
 	list_for_each_entry(rte, &iosapic_intr_info[irq].rtes, rte_list)
 		iosapic_eoi(rte->iosapic->addr, vec);
@@ -427,9 +420,8 @@
 	.enable =	iosapic_enable_level_irq,
 	.disable =	iosapic_disable_level_irq,
 	.ack =		iosapic_ack_level_irq,
-	.end =		iosapic_end_level_irq,
 	.mask =		mask_irq,
-	.unmask =	unmask_irq,
+	.unmask =	iosapic_unmask_level_irq,
 	.set_affinity =	iosapic_set_affinity
 };
 
@@ -552,37 +544,6 @@
 	}
 }
 
-static struct iosapic_rte_info * __init_refok iosapic_alloc_rte (void)
-{
-	int i;
-	struct iosapic_rte_info *rte;
-	int preallocated = 0;
-
-	if (!iosapic_kmalloc_ok && list_empty(&free_rte_list)) {
-		rte = alloc_bootmem(sizeof(struct iosapic_rte_info) *
-				    NR_PREALLOCATE_RTE_ENTRIES);
-		for (i = 0; i < NR_PREALLOCATE_RTE_ENTRIES; i++, rte++)
-			list_add(&rte->rte_list, &free_rte_list);
-	}
-
-	if (!list_empty(&free_rte_list)) {
-		rte = list_entry(free_rte_list.next, struct iosapic_rte_info,
-				 rte_list);
-		list_del(&rte->rte_list);
-		preallocated++;
-	} else {
-		rte = kmalloc(sizeof(struct iosapic_rte_info), GFP_ATOMIC);
-		if (!rte)
-			return NULL;
-	}
-
-	memset(rte, 0, sizeof(struct iosapic_rte_info));
-	if (preallocated)
-		rte->flags |= RTE_PREALLOCATED;
-
-	return rte;
-}
-
 static inline int irq_is_shared (int irq)
 {
 	return (iosapic_intr_info[irq].count > 1);
@@ -615,7 +576,7 @@
 
 	rte = find_rte(irq, gsi);
 	if (!rte) {
-		rte = iosapic_alloc_rte();
+		rte = kzalloc(sizeof (*rte), GFP_ATOMIC);
 		if (!rte) {
 			printk(KERN_WARNING "%s: cannot allocate memory\n",
 			       __func__);
@@ -658,6 +619,10 @@
 			       idesc->chip->name, irq_type->name);
 		idesc->chip = irq_type;
 	}
+	if (trigger == IOSAPIC_EDGE)
+		__set_irq_handler_unlocked(irq, handle_edge_irq);
+	else
+		__set_irq_handler_unlocked(irq, handle_level_irq);
 	return 0;
 }
 
@@ -1161,10 +1126,3 @@
 	return;
 }
 #endif
-
-static int __init iosapic_enable_kmalloc (void)
-{
-	iosapic_kmalloc_ok = 1;
-	return 0;
-}
-core_initcall (iosapic_enable_kmalloc);
diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c
index f14c35f..9a26015 100644
--- a/arch/ia64/kernel/irq_ia64.c
+++ b/arch/ia64/kernel/irq_ia64.c
@@ -30,6 +30,7 @@
 #include <linux/bitops.h>
 #include <linux/irq.h>
 #include <linux/ratelimit.h>
+#include <linux/acpi.h>
 
 #include <asm/delay.h>
 #include <asm/intrinsics.h>
@@ -635,6 +636,7 @@
 	desc->chip = &irq_type_ia64_lsapic;
 	if (action)
 		setup_irq(irq, action);
+	set_irq_handler(irq, handle_percpu_irq);
 }
 
 void __init
@@ -650,6 +652,9 @@
 void __init
 init_IRQ (void)
 {
+#ifdef CONFIG_ACPI
+	acpi_boot_init();
+#endif
 	ia64_register_ipi();
 	register_percpu_irq(IA64_SPURIOUS_INT_VECTOR, NULL);
 #ifdef CONFIG_SMP
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index a0220dc..1753f6a 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -2055,25 +2055,6 @@
 
 	IA64_MCA_DEBUG("%s: registered OS INIT handler with SAL\n", __func__);
 
-	/*
-	 *  Configure the CMCI/P vector and handler. Interrupts for CMC are
-	 *  per-processor, so AP CMC interrupts are setup in smp_callin() (smpboot.c).
-	 */
-	register_percpu_irq(IA64_CMC_VECTOR, &cmci_irqaction);
-	register_percpu_irq(IA64_CMCP_VECTOR, &cmcp_irqaction);
-	ia64_mca_cmc_vector_setup();       /* Setup vector on BSP */
-
-	/* Setup the MCA rendezvous interrupt vector */
-	register_percpu_irq(IA64_MCA_RENDEZ_VECTOR, &mca_rdzv_irqaction);
-
-	/* Setup the MCA wakeup interrupt vector */
-	register_percpu_irq(IA64_MCA_WAKEUP_VECTOR, &mca_wkup_irqaction);
-
-#ifdef CONFIG_ACPI
-	/* Setup the CPEI/P handler */
-	register_percpu_irq(IA64_CPEP_VECTOR, &mca_cpep_irqaction);
-#endif
-
 	/* Initialize the areas set aside by the OS to buffer the
 	 * platform/processor error states for MCA/INIT/CMC
 	 * handling.
@@ -2103,6 +2084,25 @@
 	if (!mca_init)
 		return 0;
 
+	/*
+	 *  Configure the CMCI/P vector and handler. Interrupts for CMC are
+	 *  per-processor, so AP CMC interrupts are setup in smp_callin() (smpboot.c).
+	 */
+	register_percpu_irq(IA64_CMC_VECTOR, &cmci_irqaction);
+	register_percpu_irq(IA64_CMCP_VECTOR, &cmcp_irqaction);
+	ia64_mca_cmc_vector_setup();       /* Setup vector on BSP */
+
+	/* Setup the MCA rendezvous interrupt vector */
+	register_percpu_irq(IA64_MCA_RENDEZ_VECTOR, &mca_rdzv_irqaction);
+
+	/* Setup the MCA wakeup interrupt vector */
+	register_percpu_irq(IA64_MCA_WAKEUP_VECTOR, &mca_wkup_irqaction);
+
+#ifdef CONFIG_ACPI
+	/* Setup the CPEI/P handler */
+	register_percpu_irq(IA64_CPEP_VECTOR, &mca_cpep_irqaction);
+#endif
+
 	register_hotcpu_notifier(&mca_cpu_notifier);
 
 	/* Setup the CMCI/P vector and handler */
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index 4a746ea..00b19a4 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -104,8 +104,8 @@
  */
 static struct irq_chip ia64_msi_chip = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= ia64_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity	= ia64_set_msi_irq_affinity,
@@ -160,8 +160,8 @@
 
 static struct irq_chip dmar_msi_type = {
 	.name = "DMAR_MSI",
-	.unmask = dmar_msi_unmask,
-	.mask = dmar_msi_mask,
+	.irq_unmask = dmar_msi_unmask,
+	.irq_mask = dmar_msi_mask,
 	.ack = ia64_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity = dmar_msi_set_affinity,
diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c
index fdf6f9d..77597e5 100644
--- a/arch/ia64/kernel/palinfo.c
+++ b/arch/ia64/kernel/palinfo.c
@@ -434,7 +434,7 @@
 	unsigned long phys_stacked;
 	pal_hints_u_t hints;
 	unsigned long iregs, dregs;
-	char *info_type[]={
+	static const char * const info_type[] = {
 		"Implemented AR(s)",
 		"AR(s) with read side-effects",
 		"Implemented CR(s)",
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index cce050e..6b1852f 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -1573,7 +1573,7 @@
 		return -EINVAL;
 	}
 
-	ctx = (pfm_context_t *)filp->private_data;
+	ctx = filp->private_data;
 	if (ctx == NULL) {
 		printk(KERN_ERR "perfmon: pfm_read: NULL ctx [%d]\n", task_pid_nr(current));
 		return -EINVAL;
@@ -1673,7 +1673,7 @@
 		return 0;
 	}
 
-	ctx = (pfm_context_t *)filp->private_data;
+	ctx = filp->private_data;
 	if (ctx == NULL) {
 		printk(KERN_ERR "perfmon: pfm_poll: NULL ctx [%d]\n", task_pid_nr(current));
 		return 0;
@@ -1733,7 +1733,7 @@
 		return -EBADF;
 	}
 
-	ctx = (pfm_context_t *)filp->private_data;
+	ctx = filp->private_data;
 	if (ctx == NULL) {
 		printk(KERN_ERR "perfmon: pfm_fasync NULL ctx [%d]\n", task_pid_nr(current));
 		return -EBADF;
@@ -1841,7 +1841,7 @@
 		return -EBADF;
 	}
 
-	ctx = (pfm_context_t *)filp->private_data;
+	ctx = filp->private_data;
 	if (ctx == NULL) {
 		printk(KERN_ERR "perfmon: pfm_flush: NULL ctx [%d]\n", task_pid_nr(current));
 		return -EBADF;
@@ -1984,7 +1984,7 @@
 		return -EBADF;
 	}
 	
-	ctx = (pfm_context_t *)filp->private_data;
+	ctx = filp->private_data;
 	if (ctx == NULL) {
 		printk(KERN_ERR "perfmon: pfm_close: NULL ctx [%d]\n", task_pid_nr(current));
 		return -EBADF;
@@ -4907,7 +4907,7 @@
 		goto error_args;
 	}
 
-	ctx = (pfm_context_t *)file->private_data;
+	ctx = file->private_data;
 	if (unlikely(ctx == NULL)) {
 		DPRINT(("no context for fd %d\n", fd));
 		goto error_args;
diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c
index aa8b5fa..45d7543 100644
--- a/arch/ia64/kernel/salinfo.c
+++ b/arch/ia64/kernel/salinfo.c
@@ -642,7 +642,7 @@
 	for (i = 0; i < ARRAY_SIZE(salinfo_log_name); i++) {
 		data = salinfo_data + i;
 		data->type = i;
-		init_MUTEX(&data->mutex);
+		sema_init(&data->mutex, 1);
 		dir = proc_mkdir(salinfo_log_name[i], salinfo_dir);
 		if (!dir)
 			continue;
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 8fb958a..911cf97 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -594,10 +594,6 @@
 	cpu_init();	/* initialize the bootstrap CPU */
 	mmu_context_init();	/* initialize context_id bitmap */
 
-#ifdef CONFIG_ACPI
-	acpi_boot_init();
-#endif
-
 	paravirt_banner();
 	paravirt_arch_setup_console(cmdline_p);
 
diff --git a/arch/ia64/kernel/stacktrace.c b/arch/ia64/kernel/stacktrace.c
new file mode 100644
index 0000000..5af2783
--- /dev/null
+++ b/arch/ia64/kernel/stacktrace.c
@@ -0,0 +1,39 @@
+/*
+ * arch/ia64/kernel/stacktrace.c
+ *
+ * Stack trace management functions
+ *
+ */
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/module.h>
+
+static void
+ia64_do_save_stack(struct unw_frame_info *info, void *arg)
+{
+	struct stack_trace *trace = arg;
+	unsigned long ip;
+	int skip = trace->skip;
+
+	trace->nr_entries = 0;
+	do {
+		unw_get_ip(info, &ip);
+		if (ip == 0)
+			break;
+		if (skip == 0) {
+			trace->entries[trace->nr_entries++] = ip;
+			if (trace->nr_entries == trace->max_entries)
+				break;
+		} else
+			skip--;
+	} while (unw_unwind(info) >= 0);
+}
+
+/*
+ * Save stack-backtrace addresses into a stack_trace buffer.
+ */
+void save_stack_trace(struct stack_trace *trace)
+{
+	unw_init_running(ia64_do_save_stack, trace);
+}
+EXPORT_SYMBOL(save_stack_trace);
diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c
index b6c0e63..fed6afa 100644
--- a/arch/ia64/kernel/unwind.c
+++ b/arch/ia64/kernel/unwind.c
@@ -1204,10 +1204,10 @@
 static inline unw_hash_index_t
 hash (unsigned long ip)
 {
-#	define hashmagic	0x9e3779b97f4a7c16UL	/* based on (sqrt(5)/2-1)*2^64 */
+	/* magic number = ((sqrt(5)-1)/2)*2^64 */
+	static const unsigned long hashmagic = 0x9e3779b97f4a7c16UL;
 
-	return (ip >> 4)*hashmagic >> (64 - UNW_LOG_HASH_SIZE);
-#undef hashmagic
+	return (ip >> 4) * hashmagic >> (64 - UNW_LOG_HASH_SIZE);
 }
 
 static inline long
@@ -1531,7 +1531,7 @@
 	struct unw_labeled_state *ls, *next;
 	unsigned long ip = info->ip;
 	struct unw_state_record sr;
-	struct unw_table *table;
+	struct unw_table *table, *prev;
 	struct unw_reg_info *r;
 	struct unw_insn insn;
 	u8 *dp, *desc_end;
@@ -1560,11 +1560,26 @@
 
 	STAT(parse_start = ia64_get_itc());
 
+	prev = NULL;
 	for (table = unw.tables; table; table = table->next) {
 		if (ip >= table->start && ip < table->end) {
+			/*
+			 * Leave the kernel unwind table at the very front,
+			 * lest moving it breaks some assumption elsewhere.
+			 * Otherwise, move the matching table to the second
+			 * position in the list so that traversals can benefit
+			 * from commonality in backtrace paths.
+			 */
+			if (prev && prev != unw.tables) {
+				/* unw is safe - we're already spinlocked */
+				prev->next = table->next;
+				table->next = unw.tables->next;
+				unw.tables->next = table;
+			}
 			e = lookup(table, ip - table->segment_base);
 			break;
 		}
+		prev = table;
 	}
 	if (!e) {
 		/* no info, return default unwinder (leaf proc, no mem stack, no saved regs)  */
diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c
index 0c72dd4..a5e500f 100644
--- a/arch/ia64/sn/kernel/msi_sn.c
+++ b/arch/ia64/sn/kernel/msi_sn.c
@@ -228,8 +228,8 @@
 
 static struct irq_chip sn_msi_chip = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= sn_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity	= sn_set_msi_irq_affinity,
diff --git a/arch/ia64/xen/xen_pv_ops.c b/arch/ia64/xen/xen_pv_ops.c
index 8adc6a1..3e8d350 100644
--- a/arch/ia64/xen/xen_pv_ops.c
+++ b/arch/ia64/xen/xen_pv_ops.c
@@ -1136,7 +1136,6 @@
 static void __init
 xen_patch_branch(unsigned long tag, unsigned long type)
 {
-	const unsigned long nelem =
-		sizeof(xen_branch_target) / sizeof(xen_branch_target[0]);
-	__paravirt_patch_apply_branch(tag, type, xen_branch_target, nelem);
+	__paravirt_patch_apply_branch(tag, type, xen_branch_target,
+					ARRAY_SIZE(xen_branch_target));
 }
diff --git a/arch/m32r/include/asm/elf.h b/arch/m32r/include/asm/elf.h
index 2f85412..b8da7d0 100644
--- a/arch/m32r/include/asm/elf.h
+++ b/arch/m32r/include/asm/elf.h
@@ -82,9 +82,9 @@
  * These are used to set parameters in the core dumps.
  */
 #define ELF_CLASS	ELFCLASS32
-#if defined(__LITTLE_ENDIAN)
+#if defined(__LITTLE_ENDIAN__)
 #define ELF_DATA	ELFDATA2LSB
-#elif defined(__BIG_ENDIAN)
+#elif defined(__BIG_ENDIAN__)
 #define ELF_DATA	ELFDATA2MSB
 #else
 #error no endian defined
diff --git a/arch/m32r/kernel/.gitignore b/arch/m32r/kernel/.gitignore
new file mode 100644
index 0000000..c5f676c
--- /dev/null
+++ b/arch/m32r/kernel/.gitignore
@@ -0,0 +1 @@
+vmlinux.lds
diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c
index 3c71f77..7db26f1 100644
--- a/arch/m32r/kernel/irq.c
+++ b/arch/m32r/kernel/irq.c
@@ -51,7 +51,7 @@
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c
index 7bbe386..a08697f 100644
--- a/arch/m32r/kernel/signal.c
+++ b/arch/m32r/kernel/signal.c
@@ -28,6 +28,8 @@
 
 #define DEBUG_SIG 0
 
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
 asmlinkage int
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		unsigned long r2, unsigned long r3, unsigned long r4,
@@ -254,7 +256,7 @@
 static int prev_insn(struct pt_regs *regs)
 {
 	u16 inst;
-	if (get_user(&inst, (u16 __user *)(regs->bpc - 2)))
+	if (get_user(inst, (u16 __user *)(regs->bpc - 2)))
 		return -EFAULT;
 	if ((inst & 0xfff0) == 0x10f0)	/* trap ? */
 		regs->bpc -= 2;
diff --git a/arch/m32r/platforms/m32104ut/setup.c b/arch/m32r/platforms/m32104ut/setup.c
index 922fdfd..402a59d 100644
--- a/arch/m32r/platforms/m32104ut/setup.c
+++ b/arch/m32r/platforms/m32104ut/setup.c
@@ -65,7 +65,7 @@
 
 static struct irq_chip m32104ut_irq_type =
 {
-	.typename = "M32104UT-IRQ",
+	.name = "M32104UT-IRQ",
 	.startup = startup_m32104ut_irq,
 	.shutdown = shutdown_m32104ut_irq,
 	.enable = enable_m32104ut_irq,
diff --git a/arch/m32r/platforms/m32700ut/setup.c b/arch/m32r/platforms/m32700ut/setup.c
index 9c1bc74..80b1a02 100644
--- a/arch/m32r/platforms/m32700ut/setup.c
+++ b/arch/m32r/platforms/m32700ut/setup.c
@@ -71,7 +71,7 @@
 
 static struct irq_chip m32700ut_irq_type =
 {
-	.typename = "M32700UT-IRQ",
+	.name = "M32700UT-IRQ",
 	.startup = startup_m32700ut_irq,
 	.shutdown = shutdown_m32700ut_irq,
 	.enable = enable_m32700ut_irq,
@@ -148,7 +148,7 @@
 
 static struct irq_chip m32700ut_pld_irq_type =
 {
-	.typename = "M32700UT-PLD-IRQ",
+	.name = "M32700UT-PLD-IRQ",
 	.startup = startup_m32700ut_pld_irq,
 	.shutdown = shutdown_m32700ut_pld_irq,
 	.enable = enable_m32700ut_pld_irq,
@@ -217,7 +217,7 @@
 
 static struct irq_chip m32700ut_lanpld_irq_type =
 {
-	.typename = "M32700UT-PLD-LAN-IRQ",
+	.name = "M32700UT-PLD-LAN-IRQ",
 	.startup = startup_m32700ut_lanpld_irq,
 	.shutdown = shutdown_m32700ut_lanpld_irq,
 	.enable = enable_m32700ut_lanpld_irq,
@@ -286,7 +286,7 @@
 
 static struct irq_chip m32700ut_lcdpld_irq_type =
 {
-	.typename = "M32700UT-PLD-LCD-IRQ",
+	.name = "M32700UT-PLD-LCD-IRQ",
 	.startup = startup_m32700ut_lcdpld_irq,
 	.shutdown = shutdown_m32700ut_lcdpld_irq,
 	.enable = enable_m32700ut_lcdpld_irq,
diff --git a/arch/m32r/platforms/mappi/setup.c b/arch/m32r/platforms/mappi/setup.c
index fb4b177..ea00c84 100644
--- a/arch/m32r/platforms/mappi/setup.c
+++ b/arch/m32r/platforms/mappi/setup.c
@@ -65,7 +65,7 @@
 
 static struct irq_chip mappi_irq_type =
 {
-	.typename = "MAPPI-IRQ",
+	.name = "MAPPI-IRQ",
 	.startup = startup_mappi_irq,
 	.shutdown = shutdown_mappi_irq,
 	.enable = enable_mappi_irq,
diff --git a/arch/m32r/platforms/mappi2/setup.c b/arch/m32r/platforms/mappi2/setup.c
index 6a65eda..c049376 100644
--- a/arch/m32r/platforms/mappi2/setup.c
+++ b/arch/m32r/platforms/mappi2/setup.c
@@ -72,7 +72,7 @@
 
 static struct irq_chip mappi2_irq_type =
 {
-	.typename = "MAPPI2-IRQ",
+	.name = "MAPPI2-IRQ",
 	.startup = startup_mappi2_irq,
 	.shutdown = shutdown_mappi2_irq,
 	.enable = enable_mappi2_irq,
diff --git a/arch/m32r/platforms/mappi3/setup.c b/arch/m32r/platforms/mappi3/setup.c
index 9c337ae..882de25 100644
--- a/arch/m32r/platforms/mappi3/setup.c
+++ b/arch/m32r/platforms/mappi3/setup.c
@@ -72,7 +72,7 @@
 
 static struct irq_chip mappi3_irq_type =
 {
-	.typename = "MAPPI3-IRQ",
+	.name = "MAPPI3-IRQ",
 	.startup = startup_mappi3_irq,
 	.shutdown = shutdown_mappi3_irq,
 	.enable = enable_mappi3_irq,
diff --git a/arch/m32r/platforms/oaks32r/setup.c b/arch/m32r/platforms/oaks32r/setup.c
index ed86574..d11d93b 100644
--- a/arch/m32r/platforms/oaks32r/setup.c
+++ b/arch/m32r/platforms/oaks32r/setup.c
@@ -63,7 +63,7 @@
 
 static struct irq_chip oaks32r_irq_type =
 {
-	.typename = "OAKS32R-IRQ",
+	.name = "OAKS32R-IRQ",
 	.startup = startup_oaks32r_irq,
 	.shutdown = shutdown_oaks32r_irq,
 	.enable = enable_oaks32r_irq,
diff --git a/arch/m32r/platforms/opsput/setup.c b/arch/m32r/platforms/opsput/setup.c
index 80d6806..5f3402a 100644
--- a/arch/m32r/platforms/opsput/setup.c
+++ b/arch/m32r/platforms/opsput/setup.c
@@ -72,7 +72,7 @@
 
 static struct irq_chip opsput_irq_type =
 {
-	.typename = "OPSPUT-IRQ",
+	.name = "OPSPUT-IRQ",
 	.startup = startup_opsput_irq,
 	.shutdown = shutdown_opsput_irq,
 	.enable = enable_opsput_irq,
@@ -149,7 +149,7 @@
 
 static struct irq_chip opsput_pld_irq_type =
 {
-	.typename = "OPSPUT-PLD-IRQ",
+	.name = "OPSPUT-PLD-IRQ",
 	.startup = startup_opsput_pld_irq,
 	.shutdown = shutdown_opsput_pld_irq,
 	.enable = enable_opsput_pld_irq,
@@ -218,7 +218,7 @@
 
 static struct irq_chip opsput_lanpld_irq_type =
 {
-	.typename = "OPSPUT-PLD-LAN-IRQ",
+	.name = "OPSPUT-PLD-LAN-IRQ",
 	.startup = startup_opsput_lanpld_irq,
 	.shutdown = shutdown_opsput_lanpld_irq,
 	.enable = enable_opsput_lanpld_irq,
diff --git a/arch/m32r/platforms/usrv/setup.c b/arch/m32r/platforms/usrv/setup.c
index 7573026..1beac7a 100644
--- a/arch/m32r/platforms/usrv/setup.c
+++ b/arch/m32r/platforms/usrv/setup.c
@@ -63,7 +63,7 @@
 
 static struct irq_chip mappi_irq_type =
 {
-	.typename = "M32700-IRQ",
+	.name = "M32700-IRQ",
 	.startup = startup_mappi_irq,
 	.shutdown = shutdown_mappi_irq,
 	.enable = enable_mappi_irq,
@@ -136,7 +136,7 @@
 
 static struct irq_chip m32700ut_pld_irq_type =
 {
-	.typename = "USRV-PLD-IRQ",
+	.name = "USRV-PLD-IRQ",
 	.startup = startup_m32700ut_pld_irq,
 	.shutdown = shutdown_m32700ut_pld_irq,
 	.enable = enable_m32700ut_pld_irq,
diff --git a/arch/mips/Kbuild b/arch/mips/Kbuild
index e322d65..7dd65cf 100644
--- a/arch/mips/Kbuild
+++ b/arch/mips/Kbuild
@@ -7,6 +7,10 @@
 include arch/mips/Kbuild.platforms
 obj-y := $(platform-y)
 
+# make clean traverses $(obj-) without having included .config, so
+# everything ends up here
+obj- := $(platform-)
+
 # mips object files
 # The object files are linked as core-y files would be linked
 
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 5526faa..4c9f402 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -881,11 +881,15 @@
 config GENERIC_ISA_DMA
 	bool
 	select ZONE_DMA if GENERIC_ISA_DMA_SUPPORT_BROKEN=n
+	select ISA_DMA_API
 
 config GENERIC_ISA_DMA_SUPPORT_BROKEN
 	bool
 	select GENERIC_ISA_DMA
 
+config ISA_DMA_API
+	bool
+
 config GENERIC_GPIO
 	bool
 
diff --git a/arch/mips/boot/compressed/Makefile b/arch/mips/boot/compressed/Makefile
index 5fd7f7a..5042d51 100644
--- a/arch/mips/boot/compressed/Makefile
+++ b/arch/mips/boot/compressed/Makefile
@@ -105,4 +105,4 @@
 vmlinuz.srec: vmlinuz
 	$(call cmd,objcopy)
 
-clean-files := $(objtree)/vmlinuz.*
+clean-files := $(objtree)/vmlinuz $(objtree)/vmlinuz.{32,ecoff,bin,srec}
diff --git a/arch/mips/dec/Platform b/arch/mips/dec/Platform
index 3adbcbd..cf55a6f 100644
--- a/arch/mips/dec/Platform
+++ b/arch/mips/dec/Platform
@@ -1,7 +1,7 @@
 #
 # DECstation family
 #
-platform-$(CONFIG_MACH_DECSTATION)	= dec/
+platform-$(CONFIG_MACH_DECSTATION)	+= dec/
 cflags-$(CONFIG_MACH_DECSTATION)	+= \
 			-I$(srctree)/arch/mips/include/asm/mach-dec
 libs-$(CONFIG_MACH_DECSTATION)		+= arch/mips/dec/prom/
diff --git a/arch/mips/include/asm/fcntl.h b/arch/mips/include/asm/fcntl.h
index e482fe9..75edded 100644
--- a/arch/mips/include/asm/fcntl.h
+++ b/arch/mips/include/asm/fcntl.h
@@ -56,6 +56,7 @@
  */
 
 #ifdef CONFIG_32BIT
+#include <linux/types.h>
 
 struct flock {
 	short	l_type;
diff --git a/arch/mips/include/asm/siginfo.h b/arch/mips/include/asm/siginfo.h
index 96e28f1..1ca64b4 100644
--- a/arch/mips/include/asm/siginfo.h
+++ b/arch/mips/include/asm/siginfo.h
@@ -88,6 +88,7 @@
 #ifdef __ARCH_SI_TRAPNO
 			int _trapno;	/* TRAP # which caused the signal */
 #endif
+			short _addr_lsb;
 		} _sigfault;
 
 		/* SIGPOLL, SIGXFSZ (To do ...)  */
diff --git a/arch/mips/jz4740/Platform b/arch/mips/jz4740/Platform
index 6a97230..ba91be9 100644
--- a/arch/mips/jz4740/Platform
+++ b/arch/mips/jz4740/Platform
@@ -1,3 +1,3 @@
-core-$(CONFIG_MACH_JZ4740)	+= arch/mips/jz4740/
+platform-$(CONFIG_MACH_JZ4740)	+= jz4740/
 cflags-$(CONFIG_MACH_JZ4740)	+= -I$(srctree)/arch/mips/include/asm/mach-jz4740
 load-$(CONFIG_MACH_JZ4740)	+= 0xffffffff80010000
diff --git a/arch/mips/kernel/branch.c b/arch/mips/kernel/branch.c
index 0176ed0..32103cc 100644
--- a/arch/mips/kernel/branch.c
+++ b/arch/mips/kernel/branch.c
@@ -40,7 +40,6 @@
 		return -EFAULT;
 	}
 
-	regs->regs[0] = 0;
 	switch (insn.i_format.opcode) {
 	/*
 	 * jr and jalr are in r_format format.
diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c
index 2340f11..9a526ba 100644
--- a/arch/mips/kernel/mips-mt-fpaff.c
+++ b/arch/mips/kernel/mips-mt-fpaff.c
@@ -103,7 +103,7 @@
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
 
-	retval = security_task_setscheduler(p, 0, NULL);
+	retval = security_task_setscheduler(p)
 	if (retval)
 		goto out_unlock;
 
diff --git a/arch/mips/kernel/ptrace.c b/arch/mips/kernel/ptrace.c
index c51b95f..c877733 100644
--- a/arch/mips/kernel/ptrace.c
+++ b/arch/mips/kernel/ptrace.c
@@ -536,7 +536,7 @@
 {
 	/* do the secure computing check first */
 	if (!entryexit)
-		secure_computing(regs->regs[0]);
+		secure_computing(regs->regs[2]);
 
 	if (unlikely(current->audit_context) && entryexit)
 		audit_syscall_exit(AUDITSC_RESULT(regs->regs[2]),
@@ -565,7 +565,7 @@
 
 out:
 	if (unlikely(current->audit_context) && !entryexit)
-		audit_syscall_entry(audit_arch(), regs->regs[0],
+		audit_syscall_entry(audit_arch(), regs->regs[2],
 				    regs->regs[4], regs->regs[5],
 				    regs->regs[6], regs->regs[7]);
 }
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index 584415e..fbaabad 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -63,9 +63,9 @@
 	sw	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	lw	t1, PT_R2(sp)		# syscall number
 	negu	v0			# error
-	sw	v0, PT_R0(sp)		# set flag for syscall
-					# restarting
+	sw	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sw	v0, PT_R2(sp)		# result
 
 o32_syscall_exit:
@@ -104,9 +104,9 @@
 	sw	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	lw	t1, PT_R2(sp)		# syscall number
 	negu	v0			# error
-	sw	v0, PT_R0(sp)		# set flag for syscall
-					# restarting
+	sw	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sw	v0, PT_R2(sp)		# result
 
 	j	syscall_exit
@@ -169,8 +169,7 @@
 	 * We probably should handle this case a bit more drastic.
 	 */
 bad_stack:
-	negu	v0				# error
-	sw	v0, PT_R0(sp)
+	li	v0, EFAULT
 	sw	v0, PT_R2(sp)
 	li	t0, 1				# set error flag
 	sw	t0, PT_R7(sp)
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index 5573f8e..3f41792 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -66,9 +66,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# set flag for syscall
-					# restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 n64_syscall_exit:
@@ -109,8 +109,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# set flag for syscall restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 	j	syscall_exit
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 1e38ec9..f08ece6 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -65,8 +65,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# set flag for syscall restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 	local_irq_disable		# make sure need_resched and
@@ -106,8 +107,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# set flag for syscall restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 	j	syscall_exit
@@ -320,10 +322,10 @@
 	PTR	sys_cacheflush
 	PTR	sys_cachectl
 	PTR	sys_sysmips
-	PTR	sys_io_setup			/* 6200 */
+	PTR	compat_sys_io_setup			/* 6200 */
 	PTR	sys_io_destroy
-	PTR	sys_io_getevents
-	PTR	sys_io_submit
+	PTR	compat_sys_io_getevents
+	PTR	compat_sys_io_submit
 	PTR	sys_io_cancel
 	PTR	sys_exit_group			/* 6205 */
 	PTR	sys_lookup_dcookie
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 171979f..78d768a 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -93,8 +93,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# flag for syscall restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 o32_syscall_exit:
@@ -142,8 +143,9 @@
 	sd	t0, PT_R7(sp)		# set error flag
 	beqz	t0, 1f
 
+	ld	t1, PT_R2(sp)		# syscall number
 	dnegu	v0			# error
-	sd	v0, PT_R0(sp)		# set flag for syscall restarting
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
 1:	sd	v0, PT_R2(sp)		# result
 
 	j	syscall_exit
@@ -154,8 +156,7 @@
 	 * The stackpointer for a call with more than 4 arguments is bad.
 	 */
 bad_stack:
-	dnegu	v0			# error
-	sd	v0, PT_R0(sp)
+	li	v0, EFAULT
 	sd	v0, PT_R2(sp)
 	li	t0, 1			# set error flag
 	sd	t0, PT_R7(sp)
@@ -444,10 +445,10 @@
 	PTR	compat_sys_futex
 	PTR	compat_sys_sched_setaffinity
 	PTR	compat_sys_sched_getaffinity	/* 4240 */
-	PTR	sys_io_setup
+	PTR	compat_sys_io_setup
 	PTR	sys_io_destroy
-	PTR	sys_io_getevents
-	PTR	sys_io_submit
+	PTR	compat_sys_io_getevents
+	PTR	compat_sys_io_submit
 	PTR	sys_io_cancel			/* 4245 */
 	PTR	sys_exit_group
 	PTR	sys32_lookup_dcookie
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index 2099d5a..5922342 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -390,7 +390,6 @@
 {
 	struct rt_sigframe __user *frame;
 	sigset_t set;
-	stack_t st;
 	int sig;
 
 	frame = (struct rt_sigframe __user *) regs.regs[29];
@@ -411,11 +410,9 @@
 	else if (sig)
 		force_sig(sig, current);
 
-	if (__copy_from_user(&st, &frame->rs_uc.uc_stack, sizeof(st)))
-		goto badframe;
 	/* It is more difficult to avoid calling this function than to
 	   call it and ignore errors.  */
-	do_sigaltstack((stack_t __user *)&st, NULL, regs.regs[29]);
+	do_sigaltstack(&frame->rs_uc.uc_stack, NULL, regs.regs[29]);
 
 	/*
 	 * Don't let your children do this ...
@@ -550,23 +547,26 @@
 	struct mips_abi *abi = current->thread.abi;
 	void *vdso = current->mm->context.vdso;
 
-	switch(regs->regs[0]) {
-	case ERESTART_RESTARTBLOCK:
-	case ERESTARTNOHAND:
-		regs->regs[2] = EINTR;
-		break;
-	case ERESTARTSYS:
-		if (!(ka->sa.sa_flags & SA_RESTART)) {
+	if (regs->regs[0]) {
+		switch(regs->regs[2]) {
+		case ERESTART_RESTARTBLOCK:
+		case ERESTARTNOHAND:
 			regs->regs[2] = EINTR;
 			break;
+		case ERESTARTSYS:
+			if (!(ka->sa.sa_flags & SA_RESTART)) {
+				regs->regs[2] = EINTR;
+				break;
+			}
+		/* fallthrough */
+		case ERESTARTNOINTR:
+			regs->regs[7] = regs->regs[26];
+			regs->regs[2] = regs->regs[0];
+			regs->cp0_epc -= 4;
 		}
-	/* fallthrough */
-	case ERESTARTNOINTR:		/* Userland will reload $v0.  */
-		regs->regs[7] = regs->regs[26];
-		regs->cp0_epc -= 8;
-	}
 
-	regs->regs[0] = 0;		/* Don't deal with this again.  */
+		regs->regs[0] = 0;		/* Don't deal with this again.  */
+	}
 
 	if (sig_uses_siginfo(ka))
 		ret = abi->setup_rt_frame(vdso + abi->rt_signal_return_offset,
@@ -575,6 +575,9 @@
 		ret = abi->setup_frame(vdso + abi->signal_return_offset,
 				       ka, regs, sig, oldset);
 
+	if (ret)
+		return ret;
+
 	spin_lock_irq(&current->sighand->siglock);
 	sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
 	if (!(ka->sa.sa_flags & SA_NODEFER))
@@ -622,17 +625,13 @@
 		return;
 	}
 
-	/*
-	 * Who's code doesn't conform to the restartable syscall convention
-	 * dies here!!!  The li instruction, a single machine instruction,
-	 * must directly be followed by the syscall instruction.
-	 */
 	if (regs->regs[0]) {
 		if (regs->regs[2] == ERESTARTNOHAND ||
 		    regs->regs[2] == ERESTARTSYS ||
 		    regs->regs[2] == ERESTARTNOINTR) {
+			regs->regs[2] = regs->regs[0];
 			regs->regs[7] = regs->regs[26];
-			regs->cp0_epc -= 8;
+			regs->cp0_epc -= 4;
 		}
 		if (regs->regs[2] == ERESTART_RESTARTBLOCK) {
 			regs->regs[2] = current->thread.abi->restart;
diff --git a/arch/mips/kernel/signal_n32.c b/arch/mips/kernel/signal_n32.c
index 2c5df81..ee24d81 100644
--- a/arch/mips/kernel/signal_n32.c
+++ b/arch/mips/kernel/signal_n32.c
@@ -109,6 +109,7 @@
 asmlinkage void sysn32_rt_sigreturn(nabi_no_regargs struct pt_regs regs)
 {
 	struct rt_sigframe_n32 __user *frame;
+	mm_segment_t old_fs;
 	sigset_t set;
 	stack_t st;
 	s32 sp;
@@ -143,7 +144,11 @@
 
 	/* It is more difficult to avoid calling this function than to
 	   call it and ignore errors.  */
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
 	do_sigaltstack((stack_t __user *)&st, NULL, regs.regs[29]);
+	set_fs(old_fs);
+
 
 	/*
 	 * Don't let your children do this ...
diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c
index 69b039c..33d5a5c 100644
--- a/arch/mips/kernel/unaligned.c
+++ b/arch/mips/kernel/unaligned.c
@@ -109,8 +109,6 @@
 	unsigned long value;
 	unsigned int res;
 
-	regs->regs[0] = 0;
-
 	/*
 	 * This load never faults.
 	 */
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 907417d..79a04a9 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -16,6 +16,7 @@
 	select RTC_DRV_GENERIC
 	select INIT_ALL_POSSIBLE
 	select BUG
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select GENERIC_ATOMIC64 if !64BIT
 	help
diff --git a/arch/parisc/include/asm/perf_event.h b/arch/parisc/include/asm/perf_event.h
index cc146427..1e0fd8b 100644
--- a/arch/parisc/include/asm/perf_event.h
+++ b/arch/parisc/include/asm/perf_event.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_PARISC_PERF_EVENT_H
 #define __ASM_PARISC_PERF_EVENT_H
 
-/* parisc only supports software events through this interface. */
-static inline void set_perf_event_pending(void) { }
+/* Empty, just to avoid compiling error */
 
 #endif /* __ASM_PARISC_PERF_EVENT_H */
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 631e5a0..4b1e521 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -138,6 +138,7 @@
 	select HAVE_OPROFILE
 	select HAVE_SYSCALL_WRAPPERS if PPC64
 	select GENERIC_ATOMIC64 if PPC32
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64
diff --git a/arch/powerpc/boot/dts/mpc8536ds.dts b/arch/powerpc/boot/dts/mpc8536ds.dts
index 815cebb..a75c10e 100644
--- a/arch/powerpc/boot/dts/mpc8536ds.dts
+++ b/arch/powerpc/boot/dts/mpc8536ds.dts
@@ -108,6 +108,58 @@
 			};
 		};
 
+		spi@7000 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc8536-espi";
+			reg = <0x7000 0x1000>;
+			interrupts = <59 0x2>;
+			interrupt-parent = <&mpic>;
+			fsl,espi-num-chipselects = <4>;
+
+			flash@0 {
+				#address-cells = <1>;
+				#size-cells = <1>;
+				compatible = "spansion,s25sl12801";
+				reg = <0>;
+				spi-max-frequency = <40000000>;
+				partition@u-boot {
+					label = "u-boot";
+					reg = <0x00000000 0x00100000>;
+					read-only;
+				};
+				partition@kernel {
+					label = "kernel";
+					reg = <0x00100000 0x00500000>;
+					read-only;
+				};
+				partition@dtb {
+					label = "dtb";
+					reg = <0x00600000 0x00100000>;
+					read-only;
+				};
+				partition@fs {
+					label = "file system";
+					reg = <0x00700000 0x00900000>;
+				};
+			};
+			flash@1 {
+				compatible = "spansion,s25sl12801";
+				reg = <1>;
+				spi-max-frequency = <40000000>;
+			};
+			flash@2 {
+				compatible = "spansion,s25sl12801";
+				reg = <2>;
+				spi-max-frequency = <40000000>;
+			};
+			flash@3 {
+				compatible = "spansion,s25sl12801";
+				reg = <3>;
+				spi-max-frequency = <40000000>;
+			};
+		};
+
 		dma@21300 {
 			#address-cells = <1>;
 			#size-cells = <1>;
diff --git a/arch/powerpc/boot/dts/p4080ds.dts b/arch/powerpc/boot/dts/p4080ds.dts
index 2f0de24..5b7fc29 100644
--- a/arch/powerpc/boot/dts/p4080ds.dts
+++ b/arch/powerpc/boot/dts/p4080ds.dts
@@ -236,22 +236,19 @@
 		};
 
 		spi@110000 {
-			cell-index = <0>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			compatible = "fsl,espi";
+			compatible = "fsl,p4080-espi", "fsl,mpc8536-espi";
 			reg = <0x110000 0x1000>;
 			interrupts = <53 0x2>;
 			interrupt-parent = <&mpic>;
-			espi,num-ss-bits = <4>;
-			mode = "cpu";
+			fsl,espi-num-chipselects = <4>;
 
-			fsl_m25p80@0 {
+			flash@0 {
 				#address-cells = <1>;
 				#size-cells = <1>;
-				compatible = "fsl,espi-flash";
+				compatible = "spansion,s25sl12801";
 				reg = <0>;
-				linux,modalias = "fsl_m25p80";
 				spi-max-frequency = <40000000>; /* input clock */
 				partition@u-boot {
 					label = "u-boot";
diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h
index 1ff6662..9b287fd 100644
--- a/arch/powerpc/include/asm/paca.h
+++ b/arch/powerpc/include/asm/paca.h
@@ -129,7 +129,7 @@
 	u8 soft_enabled;		/* irq soft-enable flag */
 	u8 hard_enabled;		/* set if irqs are enabled in MSR */
 	u8 io_sync;			/* writel() needs spin_unlock sync */
-	u8 perf_event_pending;		/* PM interrupt while soft-disabled */
+	u8 irq_work_pending;		/* IRQ_WORK interrupt while soft-disable */
 
 	/* Stuff for accurate time accounting */
 	u64 user_time;			/* accumulated usermode TB ticks */
diff --git a/arch/powerpc/include/asm/system.h b/arch/powerpc/include/asm/system.h
index 6c294ac..9c3d160 100644
--- a/arch/powerpc/include/asm/system.h
+++ b/arch/powerpc/include/asm/system.h
@@ -542,10 +542,6 @@
 
 #define PTRRELOC(x)	((typeof(x)) add_reloc_offset((unsigned long)(x)))
 
-#ifdef CONFIG_VIRT_CPU_ACCOUNTING
-extern void account_system_vtime(struct task_struct *);
-#endif
-
 extern struct dentry *powerpc_debugfs_root;
 
 #endif /* __KERNEL__ */
diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c
index 95ad9da..d05ae42 100644
--- a/arch/powerpc/kernel/perf_callchain.c
+++ b/arch/powerpc/kernel/perf_callchain.c
@@ -23,18 +23,6 @@
 #include "ppc32.h"
 #endif
 
-/*
- * Store another value in a callchain_entry.
- */
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	unsigned int nr = entry->nr;
-
-	if (nr < PERF_MAX_STACK_DEPTH) {
-		entry->ip[nr] = ip;
-		entry->nr = nr + 1;
-	}
-}
 
 /*
  * Is sp valid as the address of the next kernel stack frame after prev_sp?
@@ -58,8 +46,8 @@
 	return 0;
 }
 
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -69,8 +57,7 @@
 
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->nip);
+	perf_callchain_store(entry, regs->nip);
 
 	if (!validate_sp(sp, current, STACK_FRAME_OVERHEAD))
 		return;
@@ -89,7 +76,7 @@
 			next_ip = regs->nip;
 			lr = regs->link;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_KERNEL);
+			perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 
 		} else {
 			if (level == 0)
@@ -111,7 +98,7 @@
 			++level;
 		}
 
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		if (!valid_next_sp(next_sp, sp))
 			return;
 		sp = next_sp;
@@ -233,8 +220,8 @@
 		puc == (unsigned long) &sf->uc;
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -246,8 +233,7 @@
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, next_ip);
 
 	for (;;) {
 		fp = (unsigned long __user *) sp;
@@ -276,14 +262,14 @@
 			    read_user_stack_64(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
@@ -315,8 +301,8 @@
 	return __get_user_inatomic(*ret, ptr);
 }
 
-static inline void perf_callchain_user_64(struct pt_regs *regs,
-					  struct perf_callchain_entry *entry)
+static inline void perf_callchain_user_64(struct perf_callchain_entry *entry,
+					  struct pt_regs *regs)
 {
 }
 
@@ -435,8 +421,8 @@
 	return mctx->mc_gregs;
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned int sp, next_sp;
 	unsigned int next_ip;
@@ -447,8 +433,7 @@
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, next_ip);
 
 	while (entry->nr < PERF_MAX_STACK_DEPTH) {
 		fp = (unsigned int __user *) (unsigned long) sp;
@@ -470,45 +455,24 @@
 			    read_user_stack_32(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
 }
 
-/*
- * Since we can't get PMU interrupts inside a PMU interrupt handler,
- * we don't need separate irq and nmi entries here.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, cpu_perf_callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(cpu_perf_callchain);
-
-	entry->nr = 0;
-
-	if (!user_mode(regs)) {
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-
-	if (regs) {
-		if (current_is_64bit())
-			perf_callchain_user_64(regs, entry);
-		else
-			perf_callchain_user_32(regs, entry);
-	}
-
-	return entry;
+	if (current_is_64bit())
+		perf_callchain_user_64(entry, regs);
+	else
+		perf_callchain_user_32(entry, regs);
 }
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index d301a30..3129c85 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -402,6 +402,9 @@
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	if (!event->hw.idx)
 		return;
 	/*
@@ -517,7 +520,7 @@
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void power_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -565,7 +568,7 @@
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void power_pmu_enable(struct pmu *pmu)
 {
 	struct perf_event *event;
 	struct cpu_hw_events *cpuhw;
@@ -672,6 +675,8 @@
 		}
 		local64_set(&event->hw.prev_count, val);
 		event->hw.idx = idx;
+		if (event->hw.state & PERF_HES_STOPPED)
+			val = 0;
 		write_pmc(idx, val);
 		perf_event_update_userpage(event);
 	}
@@ -727,7 +732,7 @@
  * re-enable the PMU in order to get hw_perf_enable to do the
  * actual work of reconfiguring the PMU.
  */
-static int power_pmu_enable(struct perf_event *event)
+static int power_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -735,7 +740,7 @@
 	int ret = -EAGAIN;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	/*
 	 * Add the event to the list (if there is room)
@@ -749,6 +754,9 @@
 	cpuhw->events[n0] = event->hw.config;
 	cpuhw->flags[n0] = event->hw.event_base;
 
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -769,7 +777,7 @@
 
 	ret = 0;
  out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
@@ -777,14 +785,14 @@
 /*
  * Remove a event from the PMU.
  */
-static void power_pmu_disable(struct perf_event *event)
+static void power_pmu_del(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	long i;
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	power_pmu_read(event);
 
@@ -821,34 +829,60 @@
 		cpuhw->mmcr[0] &= ~(MMCR0_PMXE | MMCR0_FCECE);
 	}
 
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
 /*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
+ * POWER-PMU does not support disabling individual counters, hence
+ * program their cycle counter to their max value and ignore the interrupts.
  */
-static void power_pmu_unthrottle(struct perf_event *event)
+
+static void power_pmu_start(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
+	unsigned long flags;
+	s64 left;
+
+	if (!event->hw.idx || !event->hw.sample_period)
+		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void power_pmu_stop(struct perf_event *event, int ef_flags)
+{
 	unsigned long flags;
 
 	if (!event->hw.idx || !event->hw.sample_period)
 		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
+
 	power_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -857,10 +891,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-void power_pmu_start_txn(const struct pmu *pmu)
+void power_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 	cpuhw->n_txn_start = cpuhw->n_events;
 }
@@ -870,11 +905,12 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-void power_pmu_cancel_txn(const struct pmu *pmu)
+void power_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -882,7 +918,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-int power_pmu_commit_txn(const struct pmu *pmu)
+int power_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	long i, n;
@@ -901,19 +937,10 @@
 		cpuhw->event[i]->hw.config = cpuhw->events[i];
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-struct pmu power_pmu = {
-	.enable		= power_pmu_enable,
-	.disable	= power_pmu_disable,
-	.read		= power_pmu_read,
-	.unthrottle	= power_pmu_unthrottle,
-	.start_txn	= power_pmu_start_txn,
-	.cancel_txn	= power_pmu_cancel_txn,
-	.commit_txn	= power_pmu_commit_txn,
-};
-
 /*
  * Return 1 if we might be able to put event on a limited PMC,
  * or 0 if not.
@@ -1014,7 +1041,7 @@
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int power_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	unsigned long flags;
@@ -1026,25 +1053,27 @@
 	struct cpu_hw_events *cpuhw;
 
 	if (!ppmu)
-		return ERR_PTR(-ENXIO);
+		return -ENOENT;
+
 	switch (event->attr.type) {
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 	case PERF_TYPE_RAW:
 		ev = event->attr.config;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
+
 	event->hw.config_base = ev;
 	event->hw.idx = 0;
 
@@ -1063,7 +1092,7 @@
 	 * XXX we should check if the task is an idle task.
 	 */
 	flags = 0;
-	if (event->ctx->task)
+	if (event->attach_state & PERF_ATTACH_TASK)
 		flags |= PPMU_ONLY_COUNT_RUN;
 
 	/*
@@ -1081,7 +1110,7 @@
 			 */
 			ev = normal_pmc_alternative(ev, flags);
 			if (!ev)
-				return ERR_PTR(-EINVAL);
+				return -EINVAL;
 		}
 	}
 
@@ -1095,19 +1124,19 @@
 		n = collect_events(event->group_leader, ppmu->n_counter - 1,
 				   ctrs, events, cflags);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 	events[n] = ev;
 	ctrs[n] = event;
 	cflags[n] = flags;
 	if (check_excludes(ctrs, cflags, n, 1))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 	err = power_check_constraints(cpuhw, events, cflags, n + 1);
 	put_cpu_var(cpu_hw_events);
 	if (err)
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	event->hw.config = events[n];
 	event->hw.event_base = cflags[n];
@@ -1132,11 +1161,23 @@
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &power_pmu;
+	return err;
 }
 
+struct pmu power_pmu = {
+	.pmu_enable	= power_pmu_enable,
+	.pmu_disable	= power_pmu_disable,
+	.event_init	= power_pmu_event_init,
+	.add		= power_pmu_add,
+	.del		= power_pmu_del,
+	.start		= power_pmu_start,
+	.stop		= power_pmu_stop,
+	.read		= power_pmu_read,
+	.start_txn	= power_pmu_start_txn,
+	.cancel_txn	= power_pmu_cancel_txn,
+	.commit_txn	= power_pmu_commit_txn,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -1149,6 +1190,11 @@
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -1171,6 +1217,11 @@
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -1183,23 +1234,9 @@
 		if (event->attr.sample_type & PERF_SAMPLE_ADDR)
 			perf_get_data_addr(regs, &data.addr);
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			power_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 /*
@@ -1342,6 +1379,7 @@
 		freeze_events_kernel = MMCR0_FCHV;
 #endif /* CONFIG_PPC64 */
 
+	perf_pmu_register(&power_pmu);
 	perf_cpu_notifier(power_pmu_notifier);
 
 	return 0;
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index 1ba4547..7ecca59 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -156,6 +156,9 @@
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	/*
 	 * Performance monitor interrupts come even when interrupts
 	 * are soft-disabled, as long as interrupts are hard-enabled.
@@ -177,7 +180,7 @@
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void fsl_emb_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -216,7 +219,7 @@
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void fsl_emb_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -262,8 +265,8 @@
 	return n;
 }
 
-/* perf must be disabled, context locked on entry */
-static int fsl_emb_pmu_enable(struct perf_event *event)
+/* context locked on entry */
+static int fsl_emb_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int ret = -EAGAIN;
@@ -271,6 +274,7 @@
 	u64 val;
 	int i;
 
+	perf_pmu_disable(event->pmu);
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED)
@@ -301,6 +305,12 @@
 			val = 0x80000000L - left;
 	}
 	local64_set(&event->hw.prev_count, val);
+
+	if (!(flags & PERF_EF_START)) {
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+		val = 0;
+	}
+
 	write_pmc(i, val);
 	perf_event_update_userpage(event);
 
@@ -310,15 +320,17 @@
 	ret = 0;
  out:
 	put_cpu_var(cpu_hw_events);
+	perf_pmu_enable(event->pmu);
 	return ret;
 }
 
-/* perf must be disabled, context locked on entry */
-static void fsl_emb_pmu_disable(struct perf_event *event)
+/* context locked on entry */
+static void fsl_emb_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int i = event->hw.idx;
 
+	perf_pmu_disable(event->pmu);
 	if (i < 0)
 		goto out;
 
@@ -346,45 +358,58 @@
 	cpuhw->n_events--;
 
  out:
+	perf_pmu_enable(event->pmu);
 	put_cpu_var(cpu_hw_events);
 }
 
-/*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
- *
- * Context is locked on entry, but perf is not disabled.
- */
-static void fsl_emb_pmu_unthrottle(struct perf_event *event)
+static void fsl_emb_pmu_start(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
+	unsigned long flags;
+	s64 left;
+
+	if (event->hw.idx < 0 || !event->hw.sample_period)
+		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void fsl_emb_pmu_stop(struct perf_event *event, int ef_flags)
+{
 	unsigned long flags;
 
 	if (event->hw.idx < 0 || !event->hw.sample_period)
 		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
+
 	fsl_emb_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
-static struct pmu fsl_emb_pmu = {
-	.enable		= fsl_emb_pmu_enable,
-	.disable	= fsl_emb_pmu_disable,
-	.read		= fsl_emb_pmu_read,
-	.unthrottle	= fsl_emb_pmu_unthrottle,
-};
-
 /*
  * Release the PMU if this is the last perf_event.
  */
@@ -428,7 +453,7 @@
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int fsl_emb_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	struct perf_event *events[MAX_HWEVENTS];
@@ -441,14 +466,14 @@
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 
 	case PERF_TYPE_RAW:
@@ -456,12 +481,12 @@
 		break;
 
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
 
 	event->hw.config = ppmu->xlate_event(ev);
 	if (!(event->hw.config & FSL_EMB_EVENT_VALID))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	/*
 	 * If this is in a group, check if it can go on with all the
@@ -473,7 +498,7 @@
 		n = collect_events(event->group_leader,
 		                   ppmu->n_counter - 1, events);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) {
@@ -484,7 +509,7 @@
 		}
 
 		if (num_restricted >= ppmu->n_restricted)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	event->hw.idx = -1;
@@ -497,7 +522,7 @@
 	if (event->attr.exclude_kernel)
 		event->hw.config_base |= PMLCA_FCS;
 	if (event->attr.exclude_idle)
-		return ERR_PTR(-ENOTSUPP);
+		return -ENOTSUPP;
 
 	event->hw.last_period = event->hw.sample_period;
 	local64_set(&event->hw.period_left, event->hw.last_period);
@@ -523,11 +548,20 @@
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &fsl_emb_pmu;
+	return err;
 }
 
+static struct pmu fsl_emb_pmu = {
+	.pmu_enable	= fsl_emb_pmu_enable,
+	.pmu_disable	= fsl_emb_pmu_disable,
+	.event_init	= fsl_emb_pmu_event_init,
+	.add		= fsl_emb_pmu_add,
+	.del		= fsl_emb_pmu_del,
+	.start		= fsl_emb_pmu_start,
+	.stop		= fsl_emb_pmu_stop,
+	.read		= fsl_emb_pmu_read,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -540,6 +574,11 @@
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -562,6 +601,11 @@
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -571,23 +615,9 @@
 		perf_sample_data_init(&data, 0);
 		data.period = event->hw.last_period;
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			fsl_emb_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 static void perf_event_interrupt(struct pt_regs *regs)
@@ -651,5 +681,7 @@
 	pr_info("%s performance monitor hardware support registered\n",
 		pmu->name);
 
+	perf_pmu_register(&fsl_emb_pmu);
+
 	return 0;
 }
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 8533b3b..54888eb 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -53,7 +53,7 @@
 #include <linux/posix-timers.h>
 #include <linux/irq.h>
 #include <linux/delay.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <asm/trace.h>
 
 #include <asm/io.h>
@@ -493,60 +493,60 @@
 }
 #endif /* CONFIG_PPC_ISERIES */
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
 /*
  * 64-bit uses a byte in the PACA, 32-bit uses a per-cpu variable...
  */
 #ifdef CONFIG_PPC64
-static inline unsigned long test_perf_event_pending(void)
+static inline unsigned long test_irq_work_pending(void)
 {
 	unsigned long x;
 
 	asm volatile("lbz %0,%1(13)"
 		: "=r" (x)
-		: "i" (offsetof(struct paca_struct, perf_event_pending)));
+		: "i" (offsetof(struct paca_struct, irq_work_pending)));
 	return x;
 }
 
-static inline void set_perf_event_pending_flag(void)
+static inline void set_irq_work_pending_flag(void)
 {
 	asm volatile("stb %0,%1(13)" : :
 		"r" (1),
-		"i" (offsetof(struct paca_struct, perf_event_pending)));
+		"i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
-static inline void clear_perf_event_pending(void)
+static inline void clear_irq_work_pending(void)
 {
 	asm volatile("stb %0,%1(13)" : :
 		"r" (0),
-		"i" (offsetof(struct paca_struct, perf_event_pending)));
+		"i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
 #else /* 32-bit */
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()	__get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()	__get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()	__get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()	__get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()		__get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()	__get_cpu_var(irq_work_pending) = 0
 
 #endif /* 32 vs 64 bit */
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
 	preempt_disable();
-	set_perf_event_pending_flag();
+	set_irq_work_pending_flag();
 	set_dec(1);
 	preempt_enable();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()	0
-#define clear_perf_event_pending()
+#define test_irq_work_pending()	0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 /*
  * For iSeries shared processors, we have to let the hypervisor
@@ -587,9 +587,9 @@
 
 	calculate_steal_time();
 
-	if (test_perf_event_pending()) {
-		clear_perf_event_pending();
-		perf_event_do_pending();
+	if (test_irq_work_pending()) {
+		clear_irq_work_pending();
+		irq_work_run();
 	}
 
 #ifdef CONFIG_PPC_ISERIES
diff --git a/arch/powerpc/platforms/cell/axon_msi.c b/arch/powerpc/platforms/cell/axon_msi.c
index 9708553..e3e379c 100644
--- a/arch/powerpc/platforms/cell/axon_msi.c
+++ b/arch/powerpc/platforms/cell/axon_msi.c
@@ -310,9 +310,9 @@
 }
 
 static struct irq_chip msic_irq_chip = {
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
-	.shutdown	= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_shutdown	= mask_msi_irq,
 	.name		= "AXON-MSI",
 };
 
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index 93834b0..67e2c4b 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -243,7 +243,7 @@
 	 * at that level, so we do it here by hand.
 	 */
 	if (irq_to_desc(virq)->msi_desc)
-		unmask_msi_irq(virq);
+		unmask_msi_irq(irq_get_irq_data(virq));
 
 	/* unmask it */
 	xics_unmask_irq(virq);
diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c
index 87991d3..bdbd896 100644
--- a/arch/powerpc/sysdev/fsl_msi.c
+++ b/arch/powerpc/sysdev/fsl_msi.c
@@ -51,8 +51,8 @@
 }
 
 static struct irq_chip fsl_msi_chip = {
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= fsl_msi_end_irq,
 	.name		= "FSL-MSI",
 };
diff --git a/arch/powerpc/sysdev/mpic_pasemi_msi.c b/arch/powerpc/sysdev/mpic_pasemi_msi.c
index 3b6a9a4..320ad5a 100644
--- a/arch/powerpc/sysdev/mpic_pasemi_msi.c
+++ b/arch/powerpc/sysdev/mpic_pasemi_msi.c
@@ -39,24 +39,24 @@
 static struct mpic *msi_mpic;
 
 
-static void mpic_pasemi_msi_mask_irq(unsigned int irq)
+static void mpic_pasemi_msi_mask_irq(struct irq_data *data)
 {
-	pr_debug("mpic_pasemi_msi_mask_irq %d\n", irq);
-	mask_msi_irq(irq);
-	mpic_mask_irq(irq);
+	pr_debug("mpic_pasemi_msi_mask_irq %d\n", data->irq);
+	mask_msi_irq(data);
+	mpic_mask_irq(data->irq);
 }
 
-static void mpic_pasemi_msi_unmask_irq(unsigned int irq)
+static void mpic_pasemi_msi_unmask_irq(struct irq_data *data)
 {
-	pr_debug("mpic_pasemi_msi_unmask_irq %d\n", irq);
-	mpic_unmask_irq(irq);
-	unmask_msi_irq(irq);
+	pr_debug("mpic_pasemi_msi_unmask_irq %d\n", data->irq);
+	mpic_unmask_irq(data->irq);
+	unmask_msi_irq(data);
 }
 
 static struct irq_chip mpic_pasemi_msi_chip = {
-	.shutdown	= mpic_pasemi_msi_mask_irq,
-	.mask		= mpic_pasemi_msi_mask_irq,
-	.unmask		= mpic_pasemi_msi_unmask_irq,
+	.irq_shutdown	= mpic_pasemi_msi_mask_irq,
+	.irq_mask	= mpic_pasemi_msi_mask_irq,
+	.irq_unmask	= mpic_pasemi_msi_unmask_irq,
 	.eoi		= mpic_end_irq,
 	.set_type	= mpic_set_irq_type,
 	.set_affinity	= mpic_set_affinity,
diff --git a/arch/powerpc/sysdev/mpic_u3msi.c b/arch/powerpc/sysdev/mpic_u3msi.c
index bcbfe79..a2b028b 100644
--- a/arch/powerpc/sysdev/mpic_u3msi.c
+++ b/arch/powerpc/sysdev/mpic_u3msi.c
@@ -23,22 +23,22 @@
 /* A bit ugly, can we get this from the pci_dev somehow? */
 static struct mpic *msi_mpic;
 
-static void mpic_u3msi_mask_irq(unsigned int irq)
+static void mpic_u3msi_mask_irq(struct irq_data *data)
 {
-	mask_msi_irq(irq);
-	mpic_mask_irq(irq);
+	mask_msi_irq(data);
+	mpic_mask_irq(data->irq);
 }
 
-static void mpic_u3msi_unmask_irq(unsigned int irq)
+static void mpic_u3msi_unmask_irq(struct irq_data *data)
 {
-	mpic_unmask_irq(irq);
-	unmask_msi_irq(irq);
+	mpic_unmask_irq(data->irq);
+	unmask_msi_irq(data);
 }
 
 static struct irq_chip mpic_u3msi_chip = {
-	.shutdown	= mpic_u3msi_mask_irq,
-	.mask		= mpic_u3msi_mask_irq,
-	.unmask		= mpic_u3msi_unmask_irq,
+	.irq_shutdown	= mpic_u3msi_mask_irq,
+	.irq_mask	= mpic_u3msi_mask_irq,
+	.irq_unmask	= mpic_u3msi_unmask_irq,
 	.eoi		= mpic_end_irq,
 	.set_type	= mpic_set_irq_type,
 	.set_affinity	= mpic_set_affinity,
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index f0777a4..75976a1 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -95,6 +95,7 @@
 	select HAVE_KVM if 64BIT
 	select HAVE_ARCH_TRACEHOOK
 	select INIT_ALL_POSSIBLE
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
@@ -198,6 +199,13 @@
 	  can be controlled through /sys/devices/system/cpu/cpu#.
 	  Say N if you want to disable CPU hotplug.
 
+config SCHED_BOOK
+	bool "Book scheduler support"
+	depends on SMP
+	help
+	  Book scheduler support improves the CPU scheduler's decision making
+	  when dealing with machines that have several books.
+
 config MATHEMU
 	bool "IEEE FPU emulation"
 	depends on MARCH_G5
diff --git a/arch/s390/include/asm/hardirq.h b/arch/s390/include/asm/hardirq.h
index 498bc38..881d945 100644
--- a/arch/s390/include/asm/hardirq.h
+++ b/arch/s390/include/asm/hardirq.h
@@ -12,10 +12,6 @@
 #ifndef __ASM_HARDIRQ_H
 #define __ASM_HARDIRQ_H
 
-#include <linux/threads.h>
-#include <linux/sched.h>
-#include <linux/cache.h>
-#include <linux/interrupt.h>
 #include <asm/lowcore.h>
 
 #define local_softirq_pending() (S390_lowcore.softirq_pending)
diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h
index 3840cbe..a75f168 100644
--- a/arch/s390/include/asm/perf_event.h
+++ b/arch/s390/include/asm/perf_event.h
@@ -4,7 +4,6 @@
  * Copyright 2009 Martin Schwidefsky, IBM Corporation.
  */
 
-static inline void set_perf_event_pending(void) {}
-static inline void clear_perf_event_pending(void) {}
+/* Empty, just to avoid compiling error */
 
 #define PERF_EVENT_INDEX_OFFSET 0
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index 8e8a50e..1f2ebc4 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -97,7 +97,6 @@
 
 extern void account_vtime(struct task_struct *, struct task_struct *);
 extern void account_tick_vtime(struct task_struct *);
-extern void account_system_vtime(struct task_struct *);
 
 #ifdef CONFIG_PFAULT
 extern void pfault_irq_init(void);
diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h
index 831bd03..051107a 100644
--- a/arch/s390/include/asm/topology.h
+++ b/arch/s390/include/asm/topology.h
@@ -3,15 +3,32 @@
 
 #include <linux/cpumask.h>
 
-#define mc_capable()	(1)
-
-const struct cpumask *cpu_coregroup_mask(unsigned int cpu);
-
 extern unsigned char cpu_core_id[NR_CPUS];
 extern cpumask_t cpu_core_map[NR_CPUS];
 
+static inline const struct cpumask *cpu_coregroup_mask(unsigned int cpu)
+{
+	return &cpu_core_map[cpu];
+}
+
 #define topology_core_id(cpu)		(cpu_core_id[cpu])
 #define topology_core_cpumask(cpu)	(&cpu_core_map[cpu])
+#define mc_capable()			(1)
+
+#ifdef CONFIG_SCHED_BOOK
+
+extern unsigned char cpu_book_id[NR_CPUS];
+extern cpumask_t cpu_book_map[NR_CPUS];
+
+static inline const struct cpumask *cpu_book_mask(unsigned int cpu)
+{
+	return &cpu_book_map[cpu];
+}
+
+#define topology_book_id(cpu)		(cpu_book_id[cpu])
+#define topology_book_cpumask(cpu)	(&cpu_book_map[cpu])
+
+#endif /* CONFIG_SCHED_BOOK */
 
 int topology_set_cpu_management(int fc);
 void topology_schedule_update(void);
@@ -30,6 +47,8 @@
 };
 #endif
 
+#define SD_BOOK_INIT	SD_CPU_INIT
+
 #include <asm-generic/topology.h>
 
 #endif /* _ASM_S390_TOPOLOGY_H */
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index bcef007..13559c9 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -57,8 +57,8 @@
 	union tl_entry tle[0];
 };
 
-struct core_info {
-	struct core_info *next;
+struct mask_info {
+	struct mask_info *next;
 	unsigned char id;
 	cpumask_t mask;
 };
@@ -66,7 +66,6 @@
 static int topology_enabled;
 static void topology_work_fn(struct work_struct *work);
 static struct tl_info *tl_info;
-static struct core_info core_info;
 static int machine_has_topology;
 static struct timer_list topology_timer;
 static void set_topology_timer(void);
@@ -74,38 +73,37 @@
 /* topology_lock protects the core linked list */
 static DEFINE_SPINLOCK(topology_lock);
 
+static struct mask_info core_info;
 cpumask_t cpu_core_map[NR_CPUS];
 unsigned char cpu_core_id[NR_CPUS];
 
-static cpumask_t cpu_coregroup_map(unsigned int cpu)
+#ifdef CONFIG_SCHED_BOOK
+static struct mask_info book_info;
+cpumask_t cpu_book_map[NR_CPUS];
+unsigned char cpu_book_id[NR_CPUS];
+#endif
+
+static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu)
 {
-	struct core_info *core = &core_info;
-	unsigned long flags;
 	cpumask_t mask;
 
 	cpus_clear(mask);
 	if (!topology_enabled || !machine_has_topology)
 		return cpu_possible_map;
-	spin_lock_irqsave(&topology_lock, flags);
-	while (core) {
-		if (cpu_isset(cpu, core->mask)) {
-			mask = core->mask;
+	while (info) {
+		if (cpu_isset(cpu, info->mask)) {
+			mask = info->mask;
 			break;
 		}
-		core = core->next;
+		info = info->next;
 	}
-	spin_unlock_irqrestore(&topology_lock, flags);
 	if (cpus_empty(mask))
 		mask = cpumask_of_cpu(cpu);
 	return mask;
 }
 
-const struct cpumask *cpu_coregroup_mask(unsigned int cpu)
-{
-	return &cpu_core_map[cpu];
-}
-
-static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core)
+static void add_cpus_to_mask(struct tl_cpu *tl_cpu, struct mask_info *book,
+			     struct mask_info *core)
 {
 	unsigned int cpu;
 
@@ -117,23 +115,35 @@
 
 		rcpu = CPU_BITS - 1 - cpu + tl_cpu->origin;
 		for_each_present_cpu(lcpu) {
-			if (cpu_logical_map(lcpu) == rcpu) {
-				cpu_set(lcpu, core->mask);
-				cpu_core_id[lcpu] = core->id;
-				smp_cpu_polarization[lcpu] = tl_cpu->pp;
-			}
+			if (cpu_logical_map(lcpu) != rcpu)
+				continue;
+#ifdef CONFIG_SCHED_BOOK
+			cpu_set(lcpu, book->mask);
+			cpu_book_id[lcpu] = book->id;
+#endif
+			cpu_set(lcpu, core->mask);
+			cpu_core_id[lcpu] = core->id;
+			smp_cpu_polarization[lcpu] = tl_cpu->pp;
 		}
 	}
 }
 
-static void clear_cores(void)
+static void clear_masks(void)
 {
-	struct core_info *core = &core_info;
+	struct mask_info *info;
 
-	while (core) {
-		cpus_clear(core->mask);
-		core = core->next;
+	info = &core_info;
+	while (info) {
+		cpus_clear(info->mask);
+		info = info->next;
 	}
+#ifdef CONFIG_SCHED_BOOK
+	info = &book_info;
+	while (info) {
+		cpus_clear(info->mask);
+		info = info->next;
+	}
+#endif
 }
 
 static union tl_entry *next_tle(union tl_entry *tle)
@@ -146,29 +156,36 @@
 
 static void tl_to_cores(struct tl_info *info)
 {
+#ifdef CONFIG_SCHED_BOOK
+	struct mask_info *book = &book_info;
+#else
+	struct mask_info *book = NULL;
+#endif
+	struct mask_info *core = &core_info;
 	union tl_entry *tle, *end;
-	struct core_info *core = &core_info;
+
 
 	spin_lock_irq(&topology_lock);
-	clear_cores();
+	clear_masks();
 	tle = info->tle;
 	end = (union tl_entry *)((unsigned long)info + info->length);
 	while (tle < end) {
 		switch (tle->nl) {
-		case 5:
-		case 4:
-		case 3:
+#ifdef CONFIG_SCHED_BOOK
 		case 2:
+			book = book->next;
+			book->id = tle->container.id;
 			break;
+#endif
 		case 1:
 			core = core->next;
 			core->id = tle->container.id;
 			break;
 		case 0:
-			add_cpus_to_core(&tle->cpu, core);
+			add_cpus_to_mask(&tle->cpu, book, core);
 			break;
 		default:
-			clear_cores();
+			clear_masks();
 			machine_has_topology = 0;
 			goto out;
 		}
@@ -221,10 +238,29 @@
 
 static void update_cpu_core_map(void)
 {
+	unsigned long flags;
 	int cpu;
 
-	for_each_possible_cpu(cpu)
-		cpu_core_map[cpu] = cpu_coregroup_map(cpu);
+	spin_lock_irqsave(&topology_lock, flags);
+	for_each_possible_cpu(cpu) {
+		cpu_core_map[cpu] = cpu_group_map(&core_info, cpu);
+#ifdef CONFIG_SCHED_BOOK
+		cpu_book_map[cpu] = cpu_group_map(&book_info, cpu);
+#endif
+	}
+	spin_unlock_irqrestore(&topology_lock, flags);
+}
+
+static void store_topology(struct tl_info *info)
+{
+#ifdef CONFIG_SCHED_BOOK
+	int rc;
+
+	rc = stsi(info, 15, 1, 3);
+	if (rc != -ENOSYS)
+		return;
+#endif
+	stsi(info, 15, 1, 2);
 }
 
 int arch_update_cpu_topology(void)
@@ -238,7 +274,7 @@
 		topology_update_polarization_simple();
 		return 0;
 	}
-	stsi(info, 15, 1, 2);
+	store_topology(info);
 	tl_to_cores(info);
 	update_cpu_core_map();
 	for_each_online_cpu(cpu) {
@@ -299,12 +335,24 @@
 }
 __initcall(init_topology_update);
 
+static void alloc_masks(struct tl_info *info, struct mask_info *mask, int offset)
+{
+	int i, nr_masks;
+
+	nr_masks = info->mag[NR_MAG - offset];
+	for (i = 0; i < info->mnest - offset; i++)
+		nr_masks *= info->mag[NR_MAG - offset - 1 - i];
+	nr_masks = max(nr_masks, 1);
+	for (i = 0; i < nr_masks; i++) {
+		mask->next = alloc_bootmem(sizeof(struct mask_info));
+		mask = mask->next;
+	}
+}
+
 void __init s390_init_cpu_topology(void)
 {
 	unsigned long long facility_bits;
 	struct tl_info *info;
-	struct core_info *core;
-	int nr_cores;
 	int i;
 
 	if (stfle(&facility_bits, 1) <= 0)
@@ -315,25 +363,13 @@
 
 	tl_info = alloc_bootmem_pages(PAGE_SIZE);
 	info = tl_info;
-	stsi(info, 15, 1, 2);
-
-	nr_cores = info->mag[NR_MAG - 2];
-	for (i = 0; i < info->mnest - 2; i++)
-		nr_cores *= info->mag[NR_MAG - 3 - i];
-
+	store_topology(info);
 	pr_info("The CPU configuration topology of the machine is:");
 	for (i = 0; i < NR_MAG; i++)
 		printk(" %d", info->mag[i]);
 	printk(" / %d\n", info->mnest);
-
-	core = &core_info;
-	for (i = 0; i < nr_cores; i++) {
-		core->next = alloc_bootmem(sizeof(struct core_info));
-		core = core->next;
-		if (!core)
-			goto error;
-	}
-	return;
-error:
-	machine_has_topology = 0;
+	alloc_masks(info, &core_info, 2);
+#ifdef CONFIG_SCHED_BOOK
+	alloc_masks(info, &book_info, 3);
+#endif
 }
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 33990fa..35b6879 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -16,6 +16,7 @@
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_ATTRS
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_KERNEL_GZIP
@@ -249,6 +250,11 @@
 	select PM
 	select PM_RUNTIME
 
+config CPU_HAS_PMU
+       depends on CPU_SH4 || CPU_SH4A
+       default y
+       bool
+
 if SUPERH32
 
 choice
@@ -738,6 +744,14 @@
 	  LLSC, this should be more efficient than the other alternative of
 	  disabling interrupts around the atomic sequence.
 
+config HW_PERF_EVENTS
+	bool "Enable hardware performance counter support for perf events"
+	depends on PERF_EVENTS && CPU_HAS_PMU
+	default y
+	help
+	  Enable hardware performance counter support for perf events. If
+	  disabled, perf events will use software events only.
+
 source "drivers/sh/Kconfig"
 
 endmenu
diff --git a/arch/sh/include/asm/perf_event.h b/arch/sh/include/asm/perf_event.h
index 3d0c9f3..14308be 100644
--- a/arch/sh/include/asm/perf_event.h
+++ b/arch/sh/include/asm/perf_event.h
@@ -26,11 +26,4 @@
 extern int reserve_pmc_hardware(void);
 extern void release_pmc_hardware(void);
 
-static inline void set_perf_event_pending(void)
-{
-	/* Nothing to see here, move along. */
-}
-
-#define PERF_EVENT_INDEX_OFFSET	0
-
 #endif /* __ASM_SH_PERF_EVENT_H */
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 257de1f..ae5bac3 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -290,7 +290,7 @@
 int __init arch_probe_nr_irqs(void)
 {
 	nr_irqs = sh_mv.mv_nr_irqs;
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 #endif
 
diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c
index a9dd3ab..d5ca1ef 100644
--- a/arch/sh/kernel/perf_callchain.c
+++ b/arch/sh/kernel/perf_callchain.c
@@ -14,11 +14,6 @@
 #include <asm/unwinder.h>
 #include <asm/ptrace.h>
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 static void callchain_warning(void *data, char *msg)
 {
@@ -39,7 +34,7 @@
 	struct perf_callchain_entry *entry = data;
 
 	if (reliable)
-		callchain_store(entry, addr);
+		perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops callchain_ops = {
@@ -49,47 +44,10 @@
 	.address	= callchain_address,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->pc);
+	perf_callchain_store(entry, regs->pc);
 
 	unwind_stack(NULL, regs, NULL, &callchain_ops, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	/*
-	 * Only the kernel side is implemented for now.
-	 */
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-}
-
-/*
- * No need for separate IRQ and NMI entries.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
-
-	return entry;
-}
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 7a3dc35..5a4b334 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -59,6 +59,24 @@
 	return !!sh_pmu;
 }
 
+const char *perf_pmu_name(void)
+{
+	if (!sh_pmu)
+		return NULL;
+
+	return sh_pmu->name;
+}
+EXPORT_SYMBOL_GPL(perf_pmu_name);
+
+int perf_num_counters(void)
+{
+	if (!sh_pmu)
+		return 0;
+
+	return sh_pmu->num_events;
+}
+EXPORT_SYMBOL_GPL(perf_num_counters);
+
 /*
  * Release the PMU if this is the last perf_event.
  */
@@ -206,50 +224,80 @@
 	local64_add(delta, &event->count);
 }
 
-static void sh_pmu_disable(struct perf_event *event)
+static void sh_pmu_stop(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
 
-	clear_bit(idx, cpuc->active_mask);
-	sh_pmu->disable(hwc, idx);
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sh_pmu->disable(hwc, idx);
+		cpuc->events[idx] = NULL;
+		event->hw.state |= PERF_HES_STOPPED;
+	}
 
-	barrier();
+	if ((flags & PERF_EF_UPDATE) && !(event->hw.state & PERF_HES_UPTODATE)) {
+		sh_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
 
-	sh_perf_event_update(event, &event->hw, idx);
+static void sh_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
 
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	cpuc->events[idx] = event;
+	event->hw.state = 0;
+	sh_pmu->enable(hwc, idx);
+}
+
+static void sh_pmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	sh_pmu_stop(event, PERF_EF_UPDATE);
+	__clear_bit(event->hw.idx, cpuc->used_mask);
 
 	perf_event_update_userpage(event);
 }
 
-static int sh_pmu_enable(struct perf_event *event)
+static int sh_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
+	int ret = -EAGAIN;
 
-	if (test_and_set_bit(idx, cpuc->used_mask)) {
+	perf_pmu_disable(event->pmu);
+
+	if (__test_and_set_bit(idx, cpuc->used_mask)) {
 		idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events);
 		if (idx == sh_pmu->num_events)
-			return -EAGAIN;
+			goto out;
 
-		set_bit(idx, cpuc->used_mask);
+		__set_bit(idx, cpuc->used_mask);
 		hwc->idx = idx;
 	}
 
 	sh_pmu->disable(hwc, idx);
 
-	cpuc->events[idx] = event;
-	set_bit(idx, cpuc->active_mask);
-
-	sh_pmu->enable(hwc, idx);
+	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (flags & PERF_EF_START)
+		sh_pmu_start(event, PERF_EF_RELOAD);
 
 	perf_event_update_userpage(event);
-
-	return 0;
+	ret = 0;
+out:
+	perf_pmu_enable(event->pmu);
+	return ret;
 }
 
 static void sh_pmu_read(struct perf_event *event)
@@ -257,24 +305,56 @@
 	sh_perf_event_update(event, &event->hw, event->hw.idx);
 }
 
-static const struct pmu pmu = {
-	.enable		= sh_pmu_enable,
-	.disable	= sh_pmu_disable,
-	.read		= sh_pmu_read,
-};
-
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int sh_pmu_event_init(struct perf_event *event)
 {
-	int err = __hw_perf_event_init(event);
+	int err;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HW_CACHE:
+	case PERF_TYPE_HARDWARE:
+		err = __hw_perf_event_init(event);
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (unlikely(err)) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static void sh_pmu_enable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->enable_all();
+}
+
+static void sh_pmu_disable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->disable_all();
+}
+
+static struct pmu pmu = {
+	.pmu_enable	= sh_pmu_enable,
+	.pmu_disable	= sh_pmu_disable,
+	.event_init	= sh_pmu_event_init,
+	.add		= sh_pmu_add,
+	.del		= sh_pmu_del,
+	.start		= sh_pmu_start,
+	.stop		= sh_pmu_stop,
+	.read		= sh_pmu_read,
+};
+
 static void sh_pmu_setup(int cpu)
 {
 	struct cpu_hw_events *cpuhw = &per_cpu(cpu_hw_events, cpu);
@@ -299,32 +379,17 @@
 	return NOTIFY_OK;
 }
 
-void hw_perf_enable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->enable_all();
-}
-
-void hw_perf_disable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->disable_all();
-}
-
-int __cpuinit register_sh_pmu(struct sh_pmu *pmu)
+int __cpuinit register_sh_pmu(struct sh_pmu *_pmu)
 {
 	if (sh_pmu)
 		return -EBUSY;
-	sh_pmu = pmu;
+	sh_pmu = _pmu;
 
-	pr_info("Performance Events: %s support registered\n", pmu->name);
+	pr_info("Performance Events: %s support registered\n", _pmu->name);
 
-	WARN_ON(pmu->num_events > MAX_HWEVENTS);
+	WARN_ON(_pmu->num_events > MAX_HWEVENTS);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(sh_pmu_notifier);
 	return 0;
 }
diff --git a/arch/sh/oprofile/Makefile b/arch/sh/oprofile/Makefile
index 4886c5c..e85aae7 100644
--- a/arch/sh/oprofile/Makefile
+++ b/arch/sh/oprofile/Makefile
@@ -6,4 +6,8 @@
 		oprofilefs.o oprofile_stats.o \
 		timer_int.o )
 
+ifeq ($(CONFIG_HW_PERF_EVENTS),y)
+DRIVER_OBJS += $(addprefix ../../../drivers/oprofile/, oprofile_perf.o)
+endif
+
 oprofile-y	:= $(DRIVER_OBJS) common.o backtrace.o
diff --git a/arch/sh/oprofile/common.c b/arch/sh/oprofile/common.c
index ac60493..e10d893 100644
--- a/arch/sh/oprofile/common.c
+++ b/arch/sh/oprofile/common.c
@@ -17,114 +17,45 @@
 #include <linux/init.h>
 #include <linux/errno.h>
 #include <linux/smp.h>
+#include <linux/perf_event.h>
 #include <asm/processor.h>
-#include "op_impl.h"
 
-static struct op_sh_model *model;
-
-static struct op_counter_config ctr[20];
-
+#ifdef CONFIG_HW_PERF_EVENTS
 extern void sh_backtrace(struct pt_regs * const regs, unsigned int depth);
 
-static int op_sh_setup(void)
+char *op_name_from_perf_id(void)
 {
-	/* Pre-compute the values to stuff in the hardware registers.  */
-	model->reg_setup(ctr);
+	const char *pmu;
+	char buf[20];
+	int size;
 
-	/* Configure the registers on all cpus.  */
-	on_each_cpu(model->cpu_setup, NULL, 1);
+	pmu = perf_pmu_name();
+	if (!pmu)
+		return NULL;
 
-        return 0;
-}
+	size = snprintf(buf, sizeof(buf), "sh/%s", pmu);
+	if (size > -1 && size < sizeof(buf))
+		return buf;
 
-static int op_sh_create_files(struct super_block *sb, struct dentry *root)
-{
-	int i, ret = 0;
-
-	for (i = 0; i < model->num_counters; i++) {
-		struct dentry *dir;
-		char buf[4];
-
-		snprintf(buf, sizeof(buf), "%d", i);
-		dir = oprofilefs_mkdir(sb, root, buf);
-
-		ret |= oprofilefs_create_ulong(sb, dir, "enabled", &ctr[i].enabled);
-		ret |= oprofilefs_create_ulong(sb, dir, "event", &ctr[i].event);
-		ret |= oprofilefs_create_ulong(sb, dir, "kernel", &ctr[i].kernel);
-		ret |= oprofilefs_create_ulong(sb, dir, "user", &ctr[i].user);
-
-		if (model->create_files)
-			ret |= model->create_files(sb, dir);
-		else
-			ret |= oprofilefs_create_ulong(sb, dir, "count", &ctr[i].count);
-
-		/* Dummy entries */
-		ret |= oprofilefs_create_ulong(sb, dir, "unit_mask", &ctr[i].unit_mask);
-	}
-
-	return ret;
-}
-
-static int op_sh_start(void)
-{
-	/* Enable performance monitoring for all counters.  */
-	on_each_cpu(model->cpu_start, NULL, 1);
-
-	return 0;
-}
-
-static void op_sh_stop(void)
-{
-	/* Disable performance monitoring for all counters.  */
-	on_each_cpu(model->cpu_stop, NULL, 1);
+	return NULL;
 }
 
 int __init oprofile_arch_init(struct oprofile_operations *ops)
 {
-	struct op_sh_model *lmodel = NULL;
-	int ret;
-
-	/*
-	 * Always assign the backtrace op. If the counter initialization
-	 * fails, we fall back to the timer which will still make use of
-	 * this.
-	 */
 	ops->backtrace = sh_backtrace;
 
-	/*
-	 * XXX
-	 *
-	 * All of the SH7750/SH-4A counters have been converted to perf,
-	 * this infrastructure hook is left for other users until they've
-	 * had a chance to convert over, at which point all of this
-	 * will be deleted.
-	 */
-
-	if (!lmodel)
-		return -ENODEV;
-	if (!(current_cpu_data.flags & CPU_HAS_PERF_COUNTER))
-		return -ENODEV;
-
-	ret = lmodel->init();
-	if (unlikely(ret != 0))
-		return ret;
-
-	model = lmodel;
-
-	ops->setup		= op_sh_setup;
-	ops->create_files	= op_sh_create_files;
-	ops->start		= op_sh_start;
-	ops->stop		= op_sh_stop;
-	ops->cpu_type		= lmodel->cpu_type;
-
-	printk(KERN_INFO "oprofile: using %s performance monitoring.\n",
-	       lmodel->cpu_type);
-
-	return 0;
+	return oprofile_perf_init(ops);
 }
 
-void oprofile_arch_exit(void)
+void __exit oprofile_arch_exit(void)
 {
-	if (model && model->exit)
-		model->exit();
+	oprofile_perf_exit();
 }
+#else
+int __init oprofile_arch_init(struct oprofile_operations *ops)
+{
+	pr_info("oprofile: hardware counters not available\n");
+	return -ENODEV;
+}
+void __exit oprofile_arch_exit(void) {}
+#endif /* CONFIG_HW_PERF_EVENTS */
diff --git a/arch/sh/oprofile/op_impl.h b/arch/sh/oprofile/op_impl.h
deleted file mode 100644
index 1244479..0000000
--- a/arch/sh/oprofile/op_impl.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __OP_IMPL_H
-#define __OP_IMPL_H
-
-/* Per-counter configuration as set via oprofilefs.  */
-struct op_counter_config {
-	unsigned long enabled;
-	unsigned long event;
-
-	unsigned long count;
-
-	/* Dummy values for userspace tool compliance */
-	unsigned long kernel;
-	unsigned long user;
-	unsigned long unit_mask;
-};
-
-/* Per-architecture configury and hooks.  */
-struct op_sh_model {
-	void (*reg_setup)(struct op_counter_config *);
-	int (*create_files)(struct super_block *sb, struct dentry *dir);
-	void (*cpu_setup)(void *dummy);
-	int (*init)(void);
-	void (*exit)(void);
-	void (*cpu_start)(void *args);
-	void (*cpu_stop)(void *args);
-	char *cpu_type;
-	unsigned char num_counters;
-};
-
-/* arch/sh/oprofile/common.c */
-extern void sh_backtrace(struct pt_regs * const regs, unsigned int depth);
-
-#endif /* __OP_IMPL_H */
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 491e9d6..3e9d314 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -26,10 +26,12 @@
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select RTC_CLASS
 	select RTC_DRV_M48T59
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
+	select HAVE_ARCH_JUMP_LABEL
 
 config SPARC32
 	def_bool !64BIT
@@ -53,6 +55,7 @@
 	select RTC_DRV_BQ4802
 	select RTC_DRV_SUN4V
 	select RTC_DRV_STARFIRE
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 
diff --git a/arch/sparc/include/asm/jump_label.h b/arch/sparc/include/asm/jump_label.h
new file mode 100644
index 0000000..62e66d7
--- /dev/null
+++ b/arch/sparc/include/asm/jump_label.h
@@ -0,0 +1,32 @@
+#ifndef _ASM_SPARC_JUMP_LABEL_H
+#define _ASM_SPARC_JUMP_LABEL_H
+
+#ifdef __KERNEL__
+
+#include <linux/types.h>
+#include <asm/system.h>
+
+#define JUMP_LABEL_NOP_SIZE 4
+
+#define JUMP_LABEL(key, label)					\
+	do {							\
+		asm goto("1:\n\t"				\
+			 "nop\n\t"				\
+			 "nop\n\t"				\
+			 ".pushsection __jump_table,  \"a\"\n\t"\
+			 ".word 1b, %l[" #label "], %c0\n\t"	\
+			 ".popsection \n\t"			\
+			 : :  "i" (key) :  : label);\
+	} while (0)
+
+#endif /* __KERNEL__ */
+
+typedef u32 jump_label_t;
+
+struct jump_entry {
+	jump_label_t code;
+	jump_label_t target;
+	jump_label_t key;
+};
+
+#endif
diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h
index 727af70..6e8bfa1 100644
--- a/arch/sparc/include/asm/perf_event.h
+++ b/arch/sparc/include/asm/perf_event.h
@@ -1,10 +1,6 @@
 #ifndef __ASM_SPARC_PERF_EVENT_H
 #define __ASM_SPARC_PERF_EVENT_H
 
-extern void set_perf_event_pending(void);
-
-#define	PERF_EVENT_INDEX_OFFSET	0
-
 #ifdef CONFIG_PERF_EVENTS
 #include <asm/ptrace.h>
 
diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile
index 0c2dc1f..599398f 100644
--- a/arch/sparc/kernel/Makefile
+++ b/arch/sparc/kernel/Makefile
@@ -119,3 +119,5 @@
 
 pc--$(CONFIG_PERF_EVENTS) := perf_event.o
 obj-$(CONFIG_SPARC64)	+= $(pc--y)
+
+obj-$(CONFIG_SPARC64)	+= jump_label.o
diff --git a/arch/sparc/kernel/jump_label.c b/arch/sparc/kernel/jump_label.c
new file mode 100644
index 0000000..ea2dafc
--- /dev/null
+++ b/arch/sparc/kernel/jump_label.c
@@ -0,0 +1,47 @@
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+
+#include <linux/jump_label.h>
+#include <linux/memory.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+void arch_jump_label_transform(struct jump_entry *entry,
+			       enum jump_label_type type)
+{
+	u32 val;
+	u32 *insn = (u32 *) (unsigned long) entry->code;
+
+	if (type == JUMP_LABEL_ENABLE) {
+		s32 off = (s32)entry->target - (s32)entry->code;
+
+#ifdef CONFIG_SPARC64
+		/* ba,pt %xcc, . + (off << 2) */
+		val = 0x10680000 | ((u32) off >> 2);
+#else
+		/* ba . + (off << 2) */
+		val = 0x10800000 | ((u32) off >> 2);
+#endif
+	} else {
+		val = 0x01000000;
+	}
+
+	get_online_cpus();
+	mutex_lock(&text_mutex);
+	*insn = val;
+	flushi(insn);
+	mutex_unlock(&text_mutex);
+	put_online_cpus();
+}
+
+void arch_jump_label_text_poke_early(jump_label_t addr)
+{
+	u32 *insn_p = (u32 *) (unsigned long) addr;
+
+	*insn_p = 0x01000000;
+	flushi(insn_p);
+}
+
+#endif
diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c
index f848aad..ee3c7dd 100644
--- a/arch/sparc/kernel/module.c
+++ b/arch/sparc/kernel/module.c
@@ -18,6 +18,9 @@
 #include <asm/spitfire.h>
 
 #ifdef CONFIG_SPARC64
+
+#include <linux/jump_label.h>
+
 static void *module_map(unsigned long size)
 {
 	struct vm_struct *area;
@@ -227,6 +230,9 @@
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
 {
+	/* make jump label nops */
+	jump_label_apply_nops(me);
+
 	/* Cheetah's I-cache is fully coherent.  */
 	if (tlb_type == spitfire) {
 		unsigned long va;
diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c
index 548b8ca..b210416 100644
--- a/arch/sparc/kernel/pci_msi.c
+++ b/arch/sparc/kernel/pci_msi.c
@@ -114,10 +114,10 @@
 
 static struct irq_chip msi_irq = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
-	.enable		= unmask_msi_irq,
-	.disable	= mask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_enable	= unmask_msi_irq,
+	.irq_disable	= mask_msi_irq,
 	/* XXX affinity XXX */
 };
 
diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c
index c4a6a50..b87873c 100644
--- a/arch/sparc/kernel/pcr.c
+++ b/arch/sparc/kernel/pcr.c
@@ -7,7 +7,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <linux/ftrace.h>
 
 #include <asm/pil.h>
@@ -43,14 +43,14 @@
 
 	old_regs = set_irq_regs(regs);
 	irq_enter();
-#ifdef CONFIG_PERF_EVENTS
-	perf_event_do_pending();
+#ifdef CONFIG_IRQ_WORK
+	irq_work_run();
 #endif
 	irq_exit();
 	set_irq_regs(old_regs);
 }
 
-void set_perf_event_pending(void)
+void arch_irq_work_raise(void)
 {
 	set_softint(1 << PIL_DEFERRED_PCR_WORK);
 }
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 6318e62..0d6deb5 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -658,13 +658,16 @@
 
 		enc = perf_event_get_enc(cpuc->events[i]);
 		pcr &= ~mask_for_index(idx);
-		pcr |= event_encoding(enc, idx);
+		if (hwc->state & PERF_HES_STOPPED)
+			pcr |= nop_for_index(idx);
+		else
+			pcr |= event_encoding(enc, idx);
 	}
 out:
 	return pcr;
 }
 
-void hw_perf_enable(void)
+static void sparc_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 pcr;
@@ -691,7 +694,7 @@
 	pcr_ops->write(cpuc->pcr);
 }
 
-void hw_perf_disable(void)
+static void sparc_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 val;
@@ -710,48 +713,6 @@
 	pcr_ops->write(cpuc->pcr);
 }
 
-static void sparc_pmu_disable(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-	unsigned long flags;
-	int i;
-
-	local_irq_save(flags);
-	perf_disable();
-
-	for (i = 0; i < cpuc->n_events; i++) {
-		if (event == cpuc->event[i]) {
-			int idx = cpuc->current_idx[i];
-
-			/* Shift remaining entries down into
-			 * the existing slot.
-			 */
-			while (++i < cpuc->n_events) {
-				cpuc->event[i - 1] = cpuc->event[i];
-				cpuc->events[i - 1] = cpuc->events[i];
-				cpuc->current_idx[i - 1] =
-					cpuc->current_idx[i];
-			}
-
-			/* Absorb the final count and turn off the
-			 * event.
-			 */
-			sparc_pmu_disable_event(cpuc, hwc, idx);
-			barrier();
-			sparc_perf_event_update(event, hwc, idx);
-
-			perf_event_update_userpage(event);
-
-			cpuc->n_events--;
-			break;
-		}
-	}
-
-	perf_enable();
-	local_irq_restore(flags);
-}
-
 static int active_event_index(struct cpu_hw_events *cpuc,
 			      struct perf_event *event)
 {
@@ -765,6 +726,74 @@
 	return cpuc->current_idx[i];
 }
 
+static void sparc_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		sparc_perf_event_set_period(event, &event->hw, idx);
+	}
+
+	event->hw.state = 0;
+
+	sparc_pmu_enable_event(cpuc, &event->hw, idx);
+}
+
+static void sparc_pmu_stop(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sparc_pmu_disable_event(cpuc, &event->hw, idx);
+		event->hw.state |= PERF_HES_STOPPED;
+	}
+
+	if (!(event->hw.state & PERF_HES_UPTODATE) && (flags & PERF_EF_UPDATE)) {
+		sparc_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
+
+static void sparc_pmu_del(struct perf_event *event, int _flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	unsigned long flags;
+	int i;
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	for (i = 0; i < cpuc->n_events; i++) {
+		if (event == cpuc->event[i]) {
+			/* Absorb the final count and turn off the
+			 * event.
+			 */
+			sparc_pmu_stop(event, PERF_EF_UPDATE);
+
+			/* Shift remaining entries down into
+			 * the existing slot.
+			 */
+			while (++i < cpuc->n_events) {
+				cpuc->event[i - 1] = cpuc->event[i];
+				cpuc->events[i - 1] = cpuc->events[i];
+				cpuc->current_idx[i - 1] =
+					cpuc->current_idx[i];
+			}
+
+			perf_event_update_userpage(event);
+
+			cpuc->n_events--;
+			break;
+		}
+	}
+
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
 static void sparc_pmu_read(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -774,15 +803,6 @@
 	sparc_perf_event_update(event, hwc, idx);
 }
 
-static void sparc_pmu_unthrottle(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	int idx = active_event_index(cpuc, event);
-	struct hw_perf_event *hwc = &event->hw;
-
-	sparc_pmu_enable_event(cpuc, hwc, idx);
-}
-
 static atomic_t active_events = ATOMIC_INIT(0);
 static DEFINE_MUTEX(pmc_grab_mutex);
 
@@ -877,7 +897,7 @@
 	if (!n_ev)
 		return 0;
 
-	if (n_ev > perf_max_events)
+	if (n_ev > MAX_HWEVENTS)
 		return -1;
 
 	msk0 = perf_event_get_msk(events[0]);
@@ -984,23 +1004,27 @@
 	return n;
 }
 
-static int sparc_pmu_enable(struct perf_event *event)
+static int sparc_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n0, ret = -EAGAIN;
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	n0 = cpuc->n_events;
-	if (n0 >= perf_max_events)
+	if (n0 >= MAX_HWEVENTS)
 		goto out;
 
 	cpuc->event[n0] = event;
 	cpuc->events[n0] = event->hw.event_base;
 	cpuc->current_idx[n0] = PIC_NO_INDEX;
 
+	event->hw.state = PERF_HES_UPTODATE;
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state |= PERF_HES_STOPPED;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -1020,12 +1044,12 @@
 
 	ret = 0;
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
 
-static int __hw_perf_event_init(struct perf_event *event)
+static int sparc_pmu_event_init(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
 	struct perf_event *evts[MAX_HWEVENTS];
@@ -1038,22 +1062,33 @@
 	if (atomic_read(&nmi_active) < 0)
 		return -ENODEV;
 
-	pmap = NULL;
-	if (attr->type == PERF_TYPE_HARDWARE) {
+	switch (attr->type) {
+	case PERF_TYPE_HARDWARE:
 		if (attr->config >= sparc_pmu->max_events)
 			return -EINVAL;
 		pmap = sparc_pmu->event_map(attr->config);
-	} else if (attr->type == PERF_TYPE_HW_CACHE) {
+		break;
+
+	case PERF_TYPE_HW_CACHE:
 		pmap = sparc_map_cache_event(attr->config);
 		if (IS_ERR(pmap))
 			return PTR_ERR(pmap);
-	} else if (attr->type != PERF_TYPE_RAW)
-		return -EOPNOTSUPP;
+		break;
+
+	case PERF_TYPE_RAW:
+		pmap = NULL;
+		break;
+
+	default:
+		return -ENOENT;
+
+	}
 
 	if (pmap) {
 		hwc->event_base = perf_event_encode(pmap);
 	} else {
-		/* User gives us "(encoding << 16) | pic_mask" for
+		/*
+		 * User gives us "(encoding << 16) | pic_mask" for
 		 * PERF_TYPE_RAW events.
 		 */
 		hwc->event_base = attr->config;
@@ -1071,7 +1106,7 @@
 	n = 0;
 	if (event->group_leader != event) {
 		n = collect_events(event->group_leader,
-				   perf_max_events - 1,
+				   MAX_HWEVENTS - 1,
 				   evts, events, current_idx_dmy);
 		if (n < 0)
 			return -EINVAL;
@@ -1107,10 +1142,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void sparc_pmu_start_txn(const struct pmu *pmu)
+static void sparc_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 }
 
@@ -1119,11 +1155,12 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void sparc_pmu_cancel_txn(const struct pmu *pmu)
+static void sparc_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1131,7 +1168,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int sparc_pmu_commit_txn(const struct pmu *pmu)
+static int sparc_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n;
@@ -1147,28 +1184,24 @@
 		return -EAGAIN;
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= sparc_pmu_enable,
-	.disable	= sparc_pmu_disable,
+static struct pmu pmu = {
+	.pmu_enable	= sparc_pmu_enable,
+	.pmu_disable	= sparc_pmu_disable,
+	.event_init	= sparc_pmu_event_init,
+	.add		= sparc_pmu_add,
+	.del		= sparc_pmu_del,
+	.start		= sparc_pmu_start,
+	.stop		= sparc_pmu_stop,
 	.read		= sparc_pmu_read,
-	.unthrottle	= sparc_pmu_unthrottle,
 	.start_txn	= sparc_pmu_start_txn,
 	.cancel_txn	= sparc_pmu_cancel_txn,
 	.commit_txn	= sparc_pmu_commit_txn,
 };
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	int err = __hw_perf_event_init(event);
-
-	if (err)
-		return ERR_PTR(err);
-	return &pmu;
-}
-
 void perf_event_print_debug(void)
 {
 	unsigned long flags;
@@ -1244,7 +1277,7 @@
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			sparc_pmu_disable_event(cpuc, hwc, idx);
+			sparc_pmu_stop(event, 0);
 	}
 
 	return NOTIFY_STOP;
@@ -1285,28 +1318,21 @@
 
 	pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type);
 
-	/* All sparc64 PMUs currently have 2 events.  */
-	perf_max_events = 2;
-
+	perf_pmu_register(&pmu);
 	register_die_notifier(&perf_event_nmi_notifier);
 }
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
-
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void perf_callchain_kernel(struct perf_callchain_entry *entry,
+			   struct pt_regs *regs)
 {
 	unsigned long ksp, fp;
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 	int graph = 0;
 #endif
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->tpc);
+	stack_trace_flush();
+
+	perf_callchain_store(entry, regs->tpc);
 
 	ksp = regs->u_regs[UREG_I6];
 	fp = ksp + STACK_BIAS;
@@ -1330,13 +1356,13 @@
 			pc = sf->callers_pc;
 			fp = (unsigned long)sf->fp + STACK_BIAS;
 		}
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 		if ((pc + 8UL) == (unsigned long) &return_to_handler) {
 			int index = current->curr_ret_stack;
 			if (current->ret_stack && index >= graph) {
 				pc = current->ret_stack[index - graph].ret;
-				callchain_store(entry, pc);
+				perf_callchain_store(entry, pc);
 				graph++;
 			}
 		}
@@ -1344,13 +1370,12 @@
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] + STACK_BIAS;
 	do {
@@ -1363,17 +1388,16 @@
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp + STACK_BIAS;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] & 0xffffffffUL;
 	do {
@@ -1386,34 +1410,16 @@
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-/* Like powerpc we can't get PMU interrupts within the PMU handler,
- * so no need for separate NMI and IRQ chains as on x86.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-	if (!user_mode(regs)) {
-		stack_trace_flush();
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-	if (regs) {
-		flushw_user();
-		if (test_thread_flag(TIF_32BIT))
-			perf_callchain_user_32(regs, entry);
-		else
-			perf_callchain_user_64(regs, entry);
-	}
-	return entry;
+	flushw_user();
+	if (test_thread_flag(TIF_32BIT))
+		perf_callchain_user_32(entry, regs);
+	else
+		perf_callchain_user_64(entry, regs);
 }
diff --git a/arch/tile/kernel/irq.c b/arch/tile/kernel/irq.c
index 596c600..9a27d56 100644
--- a/arch/tile/kernel/irq.c
+++ b/arch/tile/kernel/irq.c
@@ -208,7 +208,7 @@
 }
 
 static struct irq_chip tile_irq_chip = {
-	.typename = "tile_irq_chip",
+	.name = "tile_irq_chip",
 	.ack = tile_irq_chip_ack,
 	.eoi = tile_irq_chip_eoi,
 	.mask = tile_irq_chip_mask,
@@ -288,7 +288,7 @@
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action = action->next; action; action = action->next)
diff --git a/arch/um/drivers/hostaudio_kern.c b/arch/um/drivers/hostaudio_kern.c
index 0c46e39..63c740a 100644
--- a/arch/um/drivers/hostaudio_kern.c
+++ b/arch/um/drivers/hostaudio_kern.c
@@ -40,6 +40,11 @@
 "    This is used to specify the host mixer device to the hostaudio driver.\n"\
 "    The default is \"" HOSTAUDIO_DEV_MIXER "\".\n\n"
 
+module_param(dsp, charp, 0644);
+MODULE_PARM_DESC(dsp, DSP_HELP);
+module_param(mixer, charp, 0644);
+MODULE_PARM_DESC(mixer, MIXER_HELP);
+
 #ifndef MODULE
 static int set_dsp(char *name, int *add)
 {
@@ -56,15 +61,6 @@
 }
 
 __uml_setup("mixer=", set_mixer, "mixer=<mixer device>\n" MIXER_HELP);
-
-#else /*MODULE*/
-
-module_param(dsp, charp, 0644);
-MODULE_PARM_DESC(dsp, DSP_HELP);
-
-module_param(mixer, charp, 0644);
-MODULE_PARM_DESC(mixer, MIXER_HELP);
-
 #endif
 
 /* /dev/dsp file operations */
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 1bcd208..9734994 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -163,6 +163,7 @@
 	struct scatterlist sg[MAX_SG];
 	struct request *request;
 	int start_sg, end_sg;
+	sector_t rq_pos;
 };
 
 #define DEFAULT_COW { \
@@ -187,6 +188,7 @@
 	.request =		NULL, \
 	.start_sg =		0, \
 	.end_sg =		0, \
+	.rq_pos =		0, \
 }
 
 /* Protected by ubd_lock */
@@ -1228,7 +1230,6 @@
 {
 	struct io_thread_req *io_req;
 	struct request *req;
-	sector_t sector;
 	int n;
 
 	while(1){
@@ -1239,12 +1240,12 @@
 				return;
 
 			dev->request = req;
+			dev->rq_pos = blk_rq_pos(req);
 			dev->start_sg = 0;
 			dev->end_sg = blk_rq_map_sg(q, req, dev->sg);
 		}
 
 		req = dev->request;
-		sector = blk_rq_pos(req);
 		while(dev->start_sg < dev->end_sg){
 			struct scatterlist *sg = &dev->sg[dev->start_sg];
 
@@ -1256,10 +1257,9 @@
 				return;
 			}
 			prepare_request(req, io_req,
-					(unsigned long long)sector << 9,
+					(unsigned long long)dev->rq_pos << 9,
 					sg->offset, sg->length, sg_page(sg));
 
-			sector += sg->length >> 9;
 			n = os_write_file(thread_fd, &io_req,
 					  sizeof(struct io_thread_req *));
 			if(n != sizeof(struct io_thread_req *)){
@@ -1272,6 +1272,7 @@
 				return;
 			}
 
+			dev->rq_pos += sg->length >> 9;
 			dev->start_sg++;
 		}
 		dev->end_sg = 0;
diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index a3f0b04..a746e30 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -46,7 +46,7 @@
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
@@ -369,7 +369,7 @@
 
 /* This is used for everything else than the timer. */
 static struct irq_chip normal_irq_type = {
-	.typename = "SIGIO",
+	.name = "SIGIO",
 	.release = free_irq_by_irq_and_dev,
 	.disable = dummy,
 	.enable = dummy,
@@ -378,7 +378,7 @@
 };
 
 static struct irq_chip SIGVTALRM_irq_type = {
-	.typename = "SIGVTALRM",
+	.name = "SIGVTALRM",
 	.release = free_irq_by_irq_and_dev,
 	.shutdown = dummy, /* never called */
 	.disable = dummy,
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index cea0cd9..7ab9db8 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -25,6 +25,7 @@
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_PERF_EVENTS if (!M386 && !M486)
+	select HAVE_IRQ_WORK
 	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
 	select ARCH_WANT_OPTIONAL_GPIOLIB
@@ -33,6 +34,7 @@
 	select HAVE_KRETPROBES
 	select HAVE_OPTPROBES
 	select HAVE_FTRACE_MCOUNT_RECORD
+	select HAVE_C_RECORDMCOUNT
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_GRAPH_TRACER
@@ -59,6 +61,12 @@
 	select ANON_INODES
 	select HAVE_ARCH_KMEMCHECK
 	select HAVE_USER_RETURN_NOTIFIER
+	select HAVE_ARCH_JUMP_LABEL
+	select HAVE_TEXT_POKE_SMP
+	select HAVE_GENERIC_HARDIRQS
+	select HAVE_SPARSE_IRQ
+	select GENERIC_IRQ_PROBE
+	select GENERIC_PENDING_IRQ if SMP
 
 config INSTRUCTION_DECODER
 	def_bool (KPROBES || PERF_EVENTS)
@@ -200,20 +208,6 @@
 	def_bool y
 	depends on EXPERIMENTAL && DMAR && ACPI
 
-# Use the generic interrupt handling code in kernel/irq/:
-config GENERIC_HARDIRQS
-	def_bool y
-
-config GENERIC_HARDIRQS_NO__DO_IRQ
-       def_bool y
-
-config GENERIC_IRQ_PROBE
-	def_bool y
-
-config GENERIC_PENDING_IRQ
-	def_bool y
-	depends on GENERIC_HARDIRQS && SMP
-
 config USE_GENERIC_SMP_HELPERS
 	def_bool y
 	depends on SMP
@@ -296,23 +290,6 @@
 
 	  If you don't know what to do here, say N.
 
-config SPARSE_IRQ
-	bool "Support sparse irq numbering"
-	depends on PCI_MSI || HT_IRQ
-	---help---
-	  This enables support for sparse irqs. This is useful for distro
-	  kernels that want to define a high CONFIG_NR_CPUS value but still
-	  want to have low kernel memory footprint on smaller machines.
-
-	  ( Sparse IRQs can also be beneficial on NUMA boxes, as they spread
-	    out the irq_desc[] array in a more NUMA-friendly way. )
-
-	  If you don't know what to do here, say N.
-
-config NUMA_IRQ_DESC
-	def_bool y
-	depends on SPARSE_IRQ && NUMA
-
 config X86_MPPARSE
 	bool "Enable MPS table" if ACPI
 	default y
@@ -517,25 +494,6 @@
 
 source "arch/x86/xen/Kconfig"
 
-config VMI
-	bool "VMI Guest support (DEPRECATED)"
-	select PARAVIRT
-	depends on X86_32
-	---help---
-	  VMI provides a paravirtualized interface to the VMware ESX server
-	  (it could be used by other hypervisors in theory too, but is not
-	  at the moment), by linking the kernel to a GPL-ed ROM module
-	  provided by the hypervisor.
-
-	  As of September 2009, VMware has started a phased retirement
-	  of this feature from VMware's products. Please see
-	  feature-removal-schedule.txt for details.  If you are
-	  planning to enable this option, please note that you cannot
-	  live migrate a VMI enabled VM to a future VMware product,
-	  which doesn't support VMI. So if you expect your kernel to
-	  seamlessly migrate to newer VMware products, keep this
-	  disabled.
-
 config KVM_CLOCK
 	bool "KVM paravirtualized clock"
 	select PARAVIRT
@@ -670,7 +628,7 @@
 	bool "GART IOMMU support" if EMBEDDED
 	default y
 	select SWIOTLB
-	depends on X86_64 && PCI && K8_NB
+	depends on X86_64 && PCI && AMD_NB
 	---help---
 	  Support for full DMA access of devices with 32bit memory access only
 	  on systems with more than 3GB. This is usually needed for USB,
@@ -795,6 +753,17 @@
 	  making when dealing with multi-core CPU chips at a cost of slightly
 	  increased overhead in some places. If unsure say N here.
 
+config IRQ_TIME_ACCOUNTING
+	bool "Fine granularity task level IRQ time accounting"
+	default n
+	---help---
+	  Select this option to enable fine granularity task irq time
+	  accounting. This is done by reading a timestamp on each
+	  transitions between softirq and hardirq state, so there can be a
+	  small performance impact.
+
+	  If in doubt, say N here.
+
 source "kernel/Kconfig.preempt"
 
 config X86_UP_APIC
@@ -1148,6 +1117,9 @@
 config ARCH_PHYS_ADDR_T_64BIT
 	def_bool X86_64 || X86_PAE
 
+config ARCH_DMA_ADDR_T_64BIT
+	def_bool X86_64 || HIGHMEM64G
+
 config DIRECT_GBPAGES
 	bool "Enable 1GB pages for kernel pagetables" if EMBEDDED
 	default y
@@ -1326,25 +1298,34 @@
 	  Set whether the default state of memory_corruption_check is
 	  on or off.
 
-config X86_RESERVE_LOW_64K
-	bool "Reserve low 64K of RAM on AMI/Phoenix BIOSen"
-	default y
+config X86_RESERVE_LOW
+	int "Amount of low memory, in kilobytes, to reserve for the BIOS"
+	default 64
+	range 4 640
 	---help---
-	  Reserve the first 64K of physical RAM on BIOSes that are known
-	  to potentially corrupt that memory range. A numbers of BIOSes are
-	  known to utilize this area during suspend/resume, so it must not
-	  be used by the kernel.
+	  Specify the amount of low memory to reserve for the BIOS.
 
-	  Set this to N if you are absolutely sure that you trust the BIOS
-	  to get all its memory reservations and usages right.
+	  The first page contains BIOS data structures that the kernel
+	  must not use, so that page must always be reserved.
 
-	  If you have doubts about the BIOS (e.g. suspend/resume does not
-	  work or there's kernel crashes after certain hardware hotplug
-	  events) and it's not AMI or Phoenix, then you might want to enable
-	  X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check typical
-	  corruption patterns.
+	  By default we reserve the first 64K of physical RAM, as a
+	  number of BIOSes are known to corrupt that memory range
+	  during events such as suspend/resume or monitor cable
+	  insertion, so it must not be used by the kernel.
 
-	  Say Y if unsure.
+	  You can set this to 4 if you are absolutely sure that you
+	  trust the BIOS to get all its memory reservations and usages
+	  right.  If you know your BIOS have problems beyond the
+	  default 64K area, you can set this to 640 to avoid using the
+	  entire low memory range.
+
+	  If you have doubts about the BIOS (e.g. suspend/resume does
+	  not work or there's kernel crashes after certain hardware
+	  hotplug events) then you might want to enable
+	  X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check
+	  typical corruption patterns.
+
+	  Leave this to the default value of 64 if you are unsure.
 
 config MATH_EMULATION
 	bool
@@ -1900,7 +1881,7 @@
 	bool "Direct"
 
 config PCI_GOOLPC
-	bool "OLPC"
+	bool "OLPC XO-1"
 	depends on OLPC
 
 config PCI_GOANY
@@ -2061,14 +2042,21 @@
 config OLPC
 	bool "One Laptop Per Child support"
 	select GPIOLIB
+	select OLPC_OPENFIRMWARE
 	---help---
 	  Add support for detecting the unique features of the OLPC
 	  XO hardware.
 
+config OLPC_XO1
+	tristate "OLPC XO-1 support"
+	depends on OLPC && PCI
+	---help---
+	  Add support for non-essential features of the OLPC XO-1 laptop.
+
 config OLPC_OPENFIRMWARE
 	bool "Support for OLPC's Open Firmware"
 	depends on !X86_64 && !X86_PAE
-	default y if OLPC
+	default n
 	help
 	  This option adds support for the implementation of Open Firmware
 	  that is used on the OLPC XO-1 Children's Machine.
@@ -2076,7 +2064,7 @@
 
 endif # X86_32
 
-config K8_NB
+config AMD_NB
 	def_bool y
 	depends on CPU_SUP_AMD && PCI
 
@@ -2125,6 +2113,10 @@
 	def_bool y
 	depends on X86_32
 
+config HAVE_TEXT_POKE_SMP
+	bool
+	select STOP_MACHINE if SMP
+
 source "net/Kconfig"
 
 source "drivers/Kconfig"
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 7508508..e5bb96b 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -43,6 +43,10 @@
 	  with klogd/syslogd or the X server. You should normally N here,
 	  unless you want to debug such a crash.
 
+config EARLY_PRINTK_MRST
+	bool "Early printk for MRST platform support"
+	depends on EARLY_PRINTK && X86_MRST
+
 config EARLY_PRINTK_DBGP
 	bool "Early printk via EHCI debug port"
 	depends on EARLY_PRINTK && PCI
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index e8c8881..b02e509 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -96,8 +96,12 @@
 # is .cfi_signal_frame supported too?
 cfi-sigframe := $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1)
 cfi-sections := $(call as-instr,.cfi_sections .debug_frame,-DCONFIG_AS_CFI_SECTIONS=1)
-KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections)
-KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections)
+
+# does binutils support specific instructions?
+asinstr := $(call as-instr,fxsaveq (%rax),-DCONFIG_AS_FXSAVEQ=1)
+
+KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr)
+KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr)
 
 LDFLAGS := -m elf_$(UTS_MACHINE)
 
diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index 0350311..2d93bdb 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -34,7 +34,7 @@
 #include <asm/ia32.h>
 
 #undef WARN_OLD
-#undef CORE_DUMP /* probably broken */
+#undef CORE_DUMP /* definitely broken */
 
 static int load_aout_binary(struct linux_binprm *, struct pt_regs *regs);
 static int load_aout_library(struct file *);
@@ -131,21 +131,15 @@
  * macros to write out all the necessary info.
  */
 
-static int dump_write(struct file *file, const void *addr, int nr)
-{
-	return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
+#include <linux/coredump.h>
 
 #define DUMP_WRITE(addr, nr)			     \
 	if (!dump_write(file, (void *)(addr), (nr))) \
 		goto end_coredump;
 
-#define DUMP_SEEK(offset)						\
-	if (file->f_op->llseek) {					\
-		if (file->f_op->llseek(file, (offset), 0) != (offset))	\
-			goto end_coredump;				\
-	} else								\
-		file->f_pos = (offset)
+#define DUMP_SEEK(offset)		\
+	if (!dump_seek(file, offset))	\
+		goto end_coredump;
 
 #define START_DATA()	(u.u_tsize << PAGE_SHIFT)
 #define START_STACK(u)	(u.start_stack)
@@ -217,12 +211,6 @@
 		dump_size = dump.u_ssize << PAGE_SHIFT;
 		DUMP_WRITE(dump_start, dump_size);
 	}
-	/*
-	 * Finally dump the task struct.  Not be used by gdb, but
-	 * could be useful
-	 */
-	set_fs(KERNEL_DS);
-	DUMP_WRITE(current, sizeof(*current));
 end_coredump:
 	set_fs(fs);
 	return has_dumped;
diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
index bc6abb7..76561d2 100644
--- a/arch/x86/include/asm/alternative.h
+++ b/arch/x86/include/asm/alternative.h
@@ -4,6 +4,7 @@
 #include <linux/types.h>
 #include <linux/stddef.h>
 #include <linux/stringify.h>
+#include <linux/jump_label.h>
 #include <asm/asm.h>
 
 /*
@@ -160,6 +161,8 @@
 #define __parainstructions_end	NULL
 #endif
 
+extern void *text_poke_early(void *addr, const void *opcode, size_t len);
+
 /*
  * Clear and restore the kernel write-protection flag on the local CPU.
  * Allows the kernel to edit read-only pages.
@@ -180,4 +183,12 @@
 extern void *text_poke(void *addr, const void *opcode, size_t len);
 extern void *text_poke_smp(void *addr, const void *opcode, size_t len);
 
+#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL)
+#define IDEAL_NOP_SIZE_5 5
+extern unsigned char ideal_nop5[IDEAL_NOP_SIZE_5];
+extern void arch_init_ideal_nop5(void);
+#else
+static inline void arch_init_ideal_nop5(void) {}
+#endif
+
 #endif /* _ASM_X86_ALTERNATIVE_H */
diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h
index 5af2982..a6863a2 100644
--- a/arch/x86/include/asm/amd_iommu.h
+++ b/arch/x86/include/asm/amd_iommu.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007-2009 Advanced Micro Devices, Inc.
+ * Copyright (C) 2007-2010 Advanced Micro Devices, Inc.
  * Author: Joerg Roedel <joerg.roedel@amd.com>
  *         Leo Duran <leo.duran@amd.com>
  *
@@ -24,11 +24,11 @@
 
 #ifdef CONFIG_AMD_IOMMU
 
-extern void amd_iommu_detect(void);
+extern int amd_iommu_detect(void);
 
 #else
 
-static inline void amd_iommu_detect(void) { }
+static inline int amd_iommu_detect(void) { return -ENODEV; }
 
 #endif
 
diff --git a/arch/x86/include/asm/amd_iommu_proto.h b/arch/x86/include/asm/amd_iommu_proto.h
index cb03037..916bc81 100644
--- a/arch/x86/include/asm/amd_iommu_proto.h
+++ b/arch/x86/include/asm/amd_iommu_proto.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2009 Advanced Micro Devices, Inc.
+ * Copyright (C) 2009-2010 Advanced Micro Devices, Inc.
  * Author: Joerg Roedel <joerg.roedel@amd.com>
  *
  * This program is free software; you can redistribute it and/or modify it
diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0861618..e3509fc 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007-2009 Advanced Micro Devices, Inc.
+ * Copyright (C) 2007-2010 Advanced Micro Devices, Inc.
  * Author: Joerg Roedel <joerg.roedel@amd.com>
  *         Leo Duran <leo.duran@amd.com>
  *
@@ -416,13 +416,22 @@
 	struct dma_ops_domain *default_dom;
 
 	/*
-	 * This array is required to work around a potential BIOS bug.
-	 * The BIOS may miss to restore parts of the PCI configuration
-	 * space when the system resumes from S3. The result is that the
-	 * IOMMU does not execute commands anymore which leads to system
-	 * failure.
+	 * We can't rely on the BIOS to restore all values on reinit, so we
+	 * need to stash them
 	 */
-	u32 cache_cfg[4];
+
+	/* The iommu BAR */
+	u32 stored_addr_lo;
+	u32 stored_addr_hi;
+
+	/*
+	 * Each iommu has 6 l1s, each of which is documented as having 0x12
+	 * registers
+	 */
+	u32 stored_l1[6][0x12];
+
+	/* The l2 indirect registers */
+	u32 stored_l2[0x83];
 };
 
 /*
diff --git a/arch/x86/include/asm/k8.h b/arch/x86/include/asm/amd_nb.h
similarity index 61%
rename from arch/x86/include/asm/k8.h
rename to arch/x86/include/asm/amd_nb.h
index af00bd1..c8517f8 100644
--- a/arch/x86/include/asm/k8.h
+++ b/arch/x86/include/asm/amd_nb.h
@@ -1,5 +1,5 @@
-#ifndef _ASM_X86_K8_H
-#define _ASM_X86_K8_H
+#ifndef _ASM_X86_AMD_NB_H
+#define _ASM_X86_AMD_NB_H
 
 #include <linux/pci.h>
 
@@ -7,24 +7,27 @@
 struct bootnode;
 
 extern int early_is_k8_nb(u32 value);
-extern struct pci_dev **k8_northbridges;
-extern int num_k8_northbridges;
 extern int cache_k8_northbridges(void);
 extern void k8_flush_garts(void);
 extern int k8_get_nodes(struct bootnode *nodes);
 extern int k8_numa_init(unsigned long start_pfn, unsigned long end_pfn);
 extern int k8_scan_nodes(void);
 
-#ifdef CONFIG_K8_NB
-extern int num_k8_northbridges;
+struct k8_northbridge_info {
+	u16 num;
+	u8 gart_supported;
+	struct pci_dev **nb_misc;
+};
+extern struct k8_northbridge_info k8_northbridges;
+
+#ifdef CONFIG_AMD_NB
 
 static inline struct pci_dev *node_to_k8_nb_misc(int node)
 {
-	return (node < num_k8_northbridges) ? k8_northbridges[node] : NULL;
+	return (node < k8_northbridges.num) ? k8_northbridges.nb_misc[node] : NULL;
 }
 
 #else
-#define num_k8_northbridges 0
 
 static inline struct pci_dev *node_to_k8_nb_misc(int node)
 {
@@ -33,4 +36,4 @@
 #endif
 
 
-#endif /* _ASM_X86_K8_H */
+#endif /* _ASM_X86_AMD_NB_H */
diff --git a/arch/x86/include/asm/apb_timer.h b/arch/x86/include/asm/apb_timer.h
index a69b1ac..2fefa50 100644
--- a/arch/x86/include/asm/apb_timer.h
+++ b/arch/x86/include/asm/apb_timer.h
@@ -54,7 +54,6 @@
 extern unsigned long apbt_quick_calibrate(void);
 extern int arch_setup_apbt_irqs(int irq, int trigger, int mask, int cpu);
 extern void apbt_setup_secondary_clock(void);
-extern unsigned int boot_cpu_id;
 
 extern struct sfi_timer_table_entry *sfi_get_mtmr(int hint);
 extern void sfi_free_mtmr(struct sfi_timer_table_entry *mtmr);
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 1fa03e0..286de34 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -252,9 +252,7 @@
 }
 #endif
 
-extern u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask);
-extern u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask);
-
+extern int setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask);
 
 #else /* !CONFIG_X86_LOCAL_APIC */
 static inline void lapic_shutdown(void) { }
diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h
index 7fe3b30..a859ca4 100644
--- a/arch/x86/include/asm/apicdef.h
+++ b/arch/x86/include/asm/apicdef.h
@@ -131,6 +131,7 @@
 #define APIC_EILVTn(n)	(0x500 + 0x10 * n)
 #define		APIC_EILVT_NR_AMD_K8	1	/* # of extended interrupts */
 #define		APIC_EILVT_NR_AMD_10H	4
+#define		APIC_EILVT_NR_MAX	APIC_EILVT_NR_AMD_10H
 #define		APIC_EILVT_LVTOFF(x)	(((x) >> 4) & 0xF)
 #define		APIC_EILVT_MSG_FIX	0x0
 #define		APIC_EILVT_MSG_SMI	0x2
diff --git a/arch/x86/include/asm/calgary.h b/arch/x86/include/asm/calgary.h
index 0918654..0d467b3 100644
--- a/arch/x86/include/asm/calgary.h
+++ b/arch/x86/include/asm/calgary.h
@@ -62,9 +62,9 @@
 extern int use_calgary;
 
 #ifdef CONFIG_CALGARY_IOMMU
-extern void detect_calgary(void);
+extern int detect_calgary(void);
 #else
-static inline void detect_calgary(void) { return; }
+static inline int detect_calgary(void) { return -ENODEV; }
 #endif
 
 #endif /* _ASM_X86_CALGARY_H */
diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h
index b185091..4fab24d 100644
--- a/arch/x86/include/asm/cpu.h
+++ b/arch/x86/include/asm/cpu.h
@@ -32,6 +32,5 @@
 
 DECLARE_PER_CPU(int, cpu_state);
 
-extern unsigned int boot_cpu_id;
 
 #endif /* _ASM_X86_CPU_H */
diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h
index 3f76523..220e2ea 100644
--- a/arch/x86/include/asm/cpufeature.h
+++ b/arch/x86/include/asm/cpufeature.h
@@ -152,10 +152,14 @@
 #define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */
 #define X86_FEATURE_OSVW	(6*32+ 9) /* OS Visible Workaround */
 #define X86_FEATURE_IBS		(6*32+10) /* Instruction Based Sampling */
-#define X86_FEATURE_SSE5	(6*32+11) /* SSE-5 */
+#define X86_FEATURE_XOP		(6*32+11) /* extended AVX instructions */
 #define X86_FEATURE_SKINIT	(6*32+12) /* SKINIT/STGI instructions */
 #define X86_FEATURE_WDT		(6*32+13) /* Watchdog timer */
+#define X86_FEATURE_LWP		(6*32+15) /* Light Weight Profiling */
+#define X86_FEATURE_FMA4	(6*32+16) /* 4 operands MAC instructions */
 #define X86_FEATURE_NODEID_MSR	(6*32+19) /* NodeId MSR */
+#define X86_FEATURE_TBM		(6*32+21) /* trailing bit manipulations */
+#define X86_FEATURE_TOPOEXT	(6*32+22) /* topology extensions CPUID leafs */
 
 /*
  * Auxiliary flags: Linux defined - For features scattered in various
@@ -180,6 +184,13 @@
 #define X86_FEATURE_LBRV	(8*32+ 6) /* AMD LBR Virtualization support */
 #define X86_FEATURE_SVML	(8*32+ 7) /* "svm_lock" AMD SVM locking MSR */
 #define X86_FEATURE_NRIPS	(8*32+ 8) /* "nrip_save" AMD SVM next_rip save */
+#define X86_FEATURE_TSCRATEMSR  (8*32+ 9) /* "tsc_scale" AMD TSC scaling support */
+#define X86_FEATURE_VMCBCLEAN   (8*32+10) /* "vmcb_clean" AMD VMCB clean bits support */
+#define X86_FEATURE_FLUSHBYASID (8*32+11) /* AMD flush-by-ASID support */
+#define X86_FEATURE_DECODEASSISTS (8*32+12) /* AMD Decode Assists support */
+#define X86_FEATURE_PAUSEFILTER (8*32+13) /* AMD filtered pause intercept */
+#define X86_FEATURE_PFTHRESHOLD (8*32+14) /* AMD pause filter threshold */
+
 
 /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */
 #define X86_FEATURE_FSGSBASE	(9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/
diff --git a/arch/x86/include/asm/dwarf2.h b/arch/x86/include/asm/dwarf2.h
index 733f7e9..3260991 100644
--- a/arch/x86/include/asm/dwarf2.h
+++ b/arch/x86/include/asm/dwarf2.h
@@ -89,6 +89,16 @@
 	CFI_ADJUST_CFA_OFFSET -8
 	.endm
 
+	.macro pushfq_cfi
+	pushfq
+	CFI_ADJUST_CFA_OFFSET 8
+	.endm
+
+	.macro popfq_cfi
+	popfq
+	CFI_ADJUST_CFA_OFFSET -8
+	.endm
+
 	.macro movq_cfi reg offset=0
 	movq %\reg, \offset(%rsp)
 	CFI_REL_OFFSET \reg, \offset
@@ -109,6 +119,16 @@
 	CFI_ADJUST_CFA_OFFSET -4
 	.endm
 
+	.macro pushfl_cfi
+	pushfl
+	CFI_ADJUST_CFA_OFFSET 4
+	.endm
+
+	.macro popfl_cfi
+	popfl
+	CFI_ADJUST_CFA_OFFSET -4
+	.endm
+
 	.macro movl_cfi reg offset=0
 	movl %\reg, \offset(%esp)
 	CFI_REL_OFFSET \reg, \offset
diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h
index 8e8ec66..b8e96a1 100644
--- a/arch/x86/include/asm/entry_arch.h
+++ b/arch/x86/include/asm/entry_arch.h
@@ -49,8 +49,8 @@
 BUILD_INTERRUPT(error_interrupt,ERROR_APIC_VECTOR)
 BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR)
 
-#ifdef CONFIG_PERF_EVENTS
-BUILD_INTERRUPT(perf_pending_interrupt, LOCAL_PENDING_VECTOR)
+#ifdef CONFIG_IRQ_WORK
+BUILD_INTERRUPT(irq_work_interrupt, IRQ_WORK_VECTOR)
 #endif
 
 #ifdef CONFIG_X86_THERMAL_VECTOR
diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h
index d07b44f..4d293dc 100644
--- a/arch/x86/include/asm/fixmap.h
+++ b/arch/x86/include/asm/fixmap.h
@@ -214,5 +214,20 @@
 	BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
 	return __virt_to_fix(vaddr);
 }
+
+/* Return an pointer with offset calculated */
+static inline unsigned long __set_fixmap_offset(enum fixed_addresses idx,
+				phys_addr_t phys, pgprot_t flags)
+{
+	__set_fixmap(idx, phys, flags);
+	return fix_to_virt(idx) + (phys & (PAGE_SIZE - 1));
+}
+
+#define set_fixmap_offset(idx, phys)			\
+	__set_fixmap_offset(idx, phys, PAGE_KERNEL)
+
+#define set_fixmap_offset_nocache(idx, phys)			\
+	__set_fixmap_offset(idx, phys, PAGE_KERNEL_NOCACHE)
+
 #endif /* !__ASSEMBLY__ */
 #endif /* _ASM_X86_FIXMAP_H */
diff --git a/arch/x86/include/asm/gart.h b/arch/x86/include/asm/gart.h
index 4ac5b0f..43085bf 100644
--- a/arch/x86/include/asm/gart.h
+++ b/arch/x86/include/asm/gart.h
@@ -17,6 +17,7 @@
 #define GARTEN		(1<<0)
 #define DISGARTCPU	(1<<4)
 #define DISGARTIO	(1<<5)
+#define DISTLBWALKPRB	(1<<6)
 
 /* GART cache control register bits. */
 #define INVGART		(1<<0)
@@ -27,7 +28,6 @@
 #define AMD64_GARTAPERTUREBASE	0x94
 #define AMD64_GARTTABLEBASE	0x98
 #define AMD64_GARTCACHECTL	0x9c
-#define AMD64_GARTEN		(1<<0)
 
 #ifdef CONFIG_GART_IOMMU
 extern int gart_iommu_aperture;
@@ -37,7 +37,7 @@
 extern void early_gart_iommu_check(void);
 extern int gart_iommu_init(void);
 extern void __init gart_parse_options(char *);
-extern void gart_iommu_hole_init(void);
+extern int gart_iommu_hole_init(void);
 
 #else
 #define gart_iommu_aperture            0
@@ -50,13 +50,27 @@
 static inline void gart_parse_options(char *options)
 {
 }
-static inline void gart_iommu_hole_init(void)
+static inline int gart_iommu_hole_init(void)
 {
+	return -ENODEV;
 }
 #endif
 
 extern int agp_amd64_init(void);
 
+static inline void gart_set_size_and_enable(struct pci_dev *dev, u32 order)
+{
+	u32 ctl;
+
+	/*
+	 * Don't enable translation but enable GART IO and CPU accesses.
+	 * Also, set DISTLBWALKPRB since GART tables memory is UC.
+	 */
+	ctl = DISTLBWALKPRB | order << 1;
+
+	pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, ctl);
+}
+
 static inline void enable_gart_translation(struct pci_dev *dev, u64 addr)
 {
 	u32 tmp, ctl;
diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h
index aeab29a..55e4de6 100644
--- a/arch/x86/include/asm/hardirq.h
+++ b/arch/x86/include/asm/hardirq.h
@@ -14,7 +14,7 @@
 #endif
 	unsigned int x86_platform_ipis;	/* arch dependent */
 	unsigned int apic_perf_irqs;
-	unsigned int apic_pending_irqs;
+	unsigned int apic_irq_work_irqs;
 #ifdef CONFIG_SMP
 	unsigned int irq_resched_count;
 	unsigned int irq_call_count;
diff --git a/arch/x86/include/asm/hpet.h b/arch/x86/include/asm/hpet.h
index 1d5c08a..2c392d6 100644
--- a/arch/x86/include/asm/hpet.h
+++ b/arch/x86/include/asm/hpet.h
@@ -74,10 +74,12 @@
 extern unsigned int hpet_readl(unsigned int a);
 extern void force_hpet_resume(void);
 
-extern void hpet_msi_unmask(unsigned int irq);
-extern void hpet_msi_mask(unsigned int irq);
-extern void hpet_msi_write(unsigned int irq, struct msi_msg *msg);
-extern void hpet_msi_read(unsigned int irq, struct msi_msg *msg);
+struct irq_data;
+extern void hpet_msi_unmask(struct irq_data *data);
+extern void hpet_msi_mask(struct irq_data *data);
+struct hpet_dev;
+extern void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg);
+extern void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg);
 
 #ifdef CONFIG_PCI_MSI
 extern int arch_setup_hpet_msi(unsigned int irq, unsigned int id);
diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 46c0fe0..0274ec5 100644
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -29,7 +29,7 @@
 extern void apic_timer_interrupt(void);
 extern void x86_platform_ipi(void);
 extern void error_interrupt(void);
-extern void perf_pending_interrupt(void);
+extern void irq_work_interrupt(void);
 
 extern void spurious_interrupt(void);
 extern void thermal_interrupt(void);
@@ -78,6 +78,13 @@
 	irq_attr->polarity	= polarity;
 }
 
+struct irq_2_iommu {
+	struct intel_iommu *iommu;
+	u16 irte_index;
+	u16 sub_handle;
+	u8  irte_mask;
+};
+
 /*
  * This is performance-critical, we want to do it O(1)
  *
@@ -89,15 +96,17 @@
 	cpumask_var_t		old_domain;
 	u8			vector;
 	u8			move_in_progress : 1;
+#ifdef CONFIG_INTR_REMAP
+	struct irq_2_iommu	irq_2_iommu;
+#endif
 };
 
-extern struct irq_cfg *irq_cfg(unsigned int);
 extern int assign_irq_vector(int, struct irq_cfg *, const struct cpumask *);
 extern void send_cleanup_vector(struct irq_cfg *);
 
-struct irq_desc;
-extern unsigned int set_desc_affinity(struct irq_desc *, const struct cpumask *,
-				      unsigned int *dest_id);
+struct irq_data;
+int __ioapic_set_affinity(struct irq_data *, const struct cpumask *,
+			  unsigned int *dest_id);
 extern int IO_APIC_get_PCI_irq_vector(int bus, int devfn, int pin, struct io_apic_irq_attr *irq_attr);
 extern void setup_ioapic_dest(void);
 
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index a73a8d5..4aa2bb3 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -55,6 +55,12 @@
 extern int restore_i387_xstate_ia32(void __user *buf);
 #endif
 
+#ifdef CONFIG_MATH_EMULATION
+extern void finit_soft_fpu(struct i387_soft_struct *soft);
+#else
+static inline void finit_soft_fpu(struct i387_soft_struct *soft) {}
+#endif
+
 #define X87_FSW_ES (1 << 7)	/* Exception Summary */
 
 static __always_inline __pure bool use_xsaveopt(void)
@@ -67,6 +73,11 @@
 	return static_cpu_has(X86_FEATURE_XSAVE);
 }
 
+static __always_inline __pure bool use_fxsr(void)
+{
+        return static_cpu_has(X86_FEATURE_FXSR);
+}
+
 extern void __sanitize_i387_state(struct task_struct *);
 
 static inline void sanitize_i387_state(struct task_struct *tsk)
@@ -77,19 +88,11 @@
 }
 
 #ifdef CONFIG_X86_64
-
-/* Ignore delayed exceptions from user space */
-static inline void tolerant_fwait(void)
-{
-	asm volatile("1: fwait\n"
-		     "2:\n"
-		     _ASM_EXTABLE(1b, 2b));
-}
-
 static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
 {
 	int err;
 
+	/* See comment in fxsave() below. */
 	asm volatile("1:  rex64/fxrstor (%[fx])\n\t"
 		     "2:\n"
 		     ".section .fixup,\"ax\"\n"
@@ -98,44 +101,10 @@
 		     ".previous\n"
 		     _ASM_EXTABLE(1b, 3b)
 		     : [err] "=r" (err)
-#if 0 /* See comment in fxsave() below. */
-		     : [fx] "r" (fx), "m" (*fx), "0" (0));
-#else
-		     : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0));
-#endif
+		     : [fx] "R" (fx), "m" (*fx), "0" (0));
 	return err;
 }
 
-/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
-   is pending. Clear the x87 state here by setting it to fixed
-   values. The kernel data segment can be sometimes 0 and sometimes
-   new user value. Both should be ok.
-   Use the PDA as safe address because it should be already in L1. */
-static inline void fpu_clear(struct fpu *fpu)
-{
-	struct xsave_struct *xstate = &fpu->state->xsave;
-	struct i387_fxsave_struct *fx = &fpu->state->fxsave;
-
-	/*
-	 * xsave header may indicate the init state of the FP.
-	 */
-	if (use_xsave() &&
-	    !(xstate->xsave_hdr.xstate_bv & XSTATE_FP))
-		return;
-
-	if (unlikely(fx->swd & X87_FSW_ES))
-		asm volatile("fnclex");
-	alternative_input(ASM_NOP8 ASM_NOP2,
-			  "    emms\n"		/* clear stack tags */
-			  "    fildl %%gs:0",	/* load to clear state */
-			  X86_FEATURE_FXSAVE_LEAK);
-}
-
-static inline void clear_fpu_state(struct task_struct *tsk)
-{
-	fpu_clear(&tsk->thread.fpu);
-}
-
 static inline int fxsave_user(struct i387_fxsave_struct __user *fx)
 {
 	int err;
@@ -149,6 +118,7 @@
 	if (unlikely(err))
 		return -EFAULT;
 
+	/* See comment in fxsave() below. */
 	asm volatile("1:  rex64/fxsave (%[fx])\n\t"
 		     "2:\n"
 		     ".section .fixup,\"ax\"\n"
@@ -157,11 +127,7 @@
 		     ".previous\n"
 		     _ASM_EXTABLE(1b, 3b)
 		     : [err] "=r" (err), "=m" (*fx)
-#if 0 /* See comment in fxsave() below. */
-		     : [fx] "r" (fx), "0" (0));
-#else
-		     : [fx] "cdaSDb" (fx), "0" (0));
-#endif
+		     : [fx] "R" (fx), "0" (0));
 	if (unlikely(err) &&
 	    __clear_user(fx, sizeof(struct i387_fxsave_struct)))
 		err = -EFAULT;
@@ -175,56 +141,29 @@
 	   uses any extended registers for addressing, a second REX prefix
 	   will be generated (to the assembler, rex64 followed by semicolon
 	   is a separate instruction), and hence the 64-bitness is lost. */
-#if 0
+
+#ifdef CONFIG_AS_FXSAVEQ
 	/* Using "fxsaveq %0" would be the ideal choice, but is only supported
 	   starting with gas 2.16. */
 	__asm__ __volatile__("fxsaveq %0"
 			     : "=m" (fpu->state->fxsave));
-#elif 0
+#else
 	/* Using, as a workaround, the properly prefixed form below isn't
 	   accepted by any binutils version so far released, complaining that
 	   the same type of prefix is used twice if an extended register is
-	   needed for addressing (fix submitted to mainline 2005-11-21). */
-	__asm__ __volatile__("rex64/fxsave %0"
-			     : "=m" (fpu->state->fxsave));
-#else
-	/* This, however, we can work around by forcing the compiler to select
+	   needed for addressing (fix submitted to mainline 2005-11-21).
+	asm volatile("rex64/fxsave %0"
+		     : "=m" (fpu->state->fxsave));
+	   This, however, we can work around by forcing the compiler to select
 	   an addressing mode that doesn't require extended registers. */
-	__asm__ __volatile__("rex64/fxsave (%1)"
-			     : "=m" (fpu->state->fxsave)
-			     : "cdaSDb" (&fpu->state->fxsave));
+	asm volatile("rex64/fxsave (%[fx])"
+		     : "=m" (fpu->state->fxsave)
+		     : [fx] "R" (&fpu->state->fxsave));
 #endif
 }
 
-static inline void fpu_save_init(struct fpu *fpu)
-{
-	if (use_xsave())
-		fpu_xsave(fpu);
-	else
-		fpu_fxsave(fpu);
-
-	fpu_clear(fpu);
-}
-
-static inline void __save_init_fpu(struct task_struct *tsk)
-{
-	fpu_save_init(&tsk->thread.fpu);
-	task_thread_info(tsk)->status &= ~TS_USEDFPU;
-}
-
 #else  /* CONFIG_X86_32 */
 
-#ifdef CONFIG_MATH_EMULATION
-extern void finit_soft_fpu(struct i387_soft_struct *soft);
-#else
-static inline void finit_soft_fpu(struct i387_soft_struct *soft) {}
-#endif
-
-static inline void tolerant_fwait(void)
-{
-	asm volatile("fnclex ; fwait");
-}
-
 /* perform fxrstor iff the processor has extended states, otherwise frstor */
 static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
 {
@@ -241,6 +180,14 @@
 	return 0;
 }
 
+static inline void fpu_fxsave(struct fpu *fpu)
+{
+	asm volatile("fxsave %[fx]"
+		     : [fx] "=m" (fpu->state->fxsave));
+}
+
+#endif	/* CONFIG_X86_64 */
+
 /* We need a safe address that is cheap to find and that is already
    in L1 during context switch. The best choices are unfortunately
    different for UP and SMP */
@@ -256,47 +203,33 @@
 static inline void fpu_save_init(struct fpu *fpu)
 {
 	if (use_xsave()) {
-		struct xsave_struct *xstate = &fpu->state->xsave;
-		struct i387_fxsave_struct *fx = &fpu->state->fxsave;
-
 		fpu_xsave(fpu);
 
 		/*
 		 * xsave header may indicate the init state of the FP.
 		 */
-		if (!(xstate->xsave_hdr.xstate_bv & XSTATE_FP))
-			goto end;
-
-		if (unlikely(fx->swd & X87_FSW_ES))
-			asm volatile("fnclex");
-
-		/*
-		 * we can do a simple return here or be paranoid :)
-		 */
-		goto clear_state;
+		if (!(fpu->state->xsave.xsave_hdr.xstate_bv & XSTATE_FP))
+			return;
+	} else if (use_fxsr()) {
+		fpu_fxsave(fpu);
+	} else {
+		asm volatile("fsave %[fx]; fwait"
+			     : [fx] "=m" (fpu->state->fsave));
+		return;
 	}
 
-	/* Use more nops than strictly needed in case the compiler
-	   varies code */
-	alternative_input(
-		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
-		"fxsave %[fx]\n"
-		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
-		X86_FEATURE_FXSR,
-		[fx] "m" (fpu->state->fxsave),
-		[fsw] "m" (fpu->state->fxsave.swd) : "memory");
-clear_state:
+	if (unlikely(fpu->state->fxsave.swd & X87_FSW_ES))
+		asm volatile("fnclex");
+
 	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
 	   is pending.  Clear the x87 state here by setting it to fixed
 	   values. safe_address is a random variable that should be in L1 */
 	alternative_input(
-		GENERIC_NOP8 GENERIC_NOP2,
+		ASM_NOP8 ASM_NOP2,
 		"emms\n\t"	  	/* clear stack tags */
-		"fildl %[addr]", 	/* set F?P to defined value */
+		"fildl %P[addr]",	/* set F?P to defined value */
 		X86_FEATURE_FXSAVE_LEAK,
 		[addr] "m" (safe_address));
-end:
-	;
 }
 
 static inline void __save_init_fpu(struct task_struct *tsk)
@@ -305,9 +238,6 @@
 	task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
-
-#endif	/* CONFIG_X86_64 */
-
 static inline int fpu_fxrstor_checking(struct fpu *fpu)
 {
 	return fxrstor_checking(&fpu->state->fxsave);
@@ -344,7 +274,10 @@
 static inline void __clear_fpu(struct task_struct *tsk)
 {
 	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		tolerant_fwait();
+		/* Ignore delayed exceptions from user space */
+		asm volatile("1: fwait\n"
+			     "2:\n"
+			     _ASM_EXTABLE(1b, 2b));
 		task_thread_info(tsk)->status &= ~TS_USEDFPU;
 		stts();
 	}
@@ -405,19 +338,6 @@
 		stts();
 }
 
-#ifdef CONFIG_X86_64
-
-static inline void save_init_fpu(struct task_struct *tsk)
-{
-	__save_init_fpu(tsk);
-	stts();
-}
-
-#define unlazy_fpu	__unlazy_fpu
-#define clear_fpu	__clear_fpu
-
-#else  /* CONFIG_X86_32 */
-
 /*
  * These disable preemption on their own and are safe
  */
@@ -443,8 +363,6 @@
 	preempt_enable();
 }
 
-#endif	/* CONFIG_X86_64 */
-
 /*
  * i387 state interaction
  */
@@ -508,7 +426,4 @@
 
 #endif /* __ASSEMBLY__ */
 
-#define PSHUFB_XMM5_XMM0 .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
-#define PSHUFB_XMM5_XMM6 .byte 0x66, 0x0f, 0x38, 0x00, 0xf5
-
 #endif /* _ASM_X86_I387_H */
diff --git a/arch/x86/include/asm/i8259.h b/arch/x86/include/asm/i8259.h
index 1655147..a203659 100644
--- a/arch/x86/include/asm/i8259.h
+++ b/arch/x86/include/asm/i8259.h
@@ -55,6 +55,8 @@
 struct legacy_pic {
 	int nr_legacy_irqs;
 	struct irq_chip *chip;
+	void (*mask)(unsigned int irq);
+	void (*unmask)(unsigned int irq);
 	void (*mask_all)(void);
 	void (*restore_mask)(void);
 	void (*init)(int auto_eoi);
diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
index 30a3e97..6a45ec4 100644
--- a/arch/x86/include/asm/io.h
+++ b/arch/x86/include/asm/io.h
@@ -206,6 +206,7 @@
 
 extern void iounmap(volatile void __iomem *addr);
 
+extern void set_iounmap_nonlazy(void);
 
 #ifdef __KERNEL__
 
diff --git a/arch/x86/include/asm/io_apic.h b/arch/x86/include/asm/io_apic.h
index 9cb2edb..c8be456 100644
--- a/arch/x86/include/asm/io_apic.h
+++ b/arch/x86/include/asm/io_apic.h
@@ -170,12 +170,6 @@
 
 extern void probe_nr_irqs_gsi(void);
 
-extern int setup_ioapic_entry(int apic, int irq,
-			      struct IO_APIC_route_entry *entry,
-			      unsigned int destination, int trigger,
-			      int polarity, int vector, int pin);
-extern void ioapic_write_entry(int apic, int pin,
-			       struct IO_APIC_route_entry e);
 extern void setup_ioapic_ids_from_mpc(void);
 
 struct mp_ioapic_gsi{
diff --git a/arch/x86/include/asm/iommu_table.h b/arch/x86/include/asm/iommu_table.h
new file mode 100644
index 0000000..f229b13
--- /dev/null
+++ b/arch/x86/include/asm/iommu_table.h
@@ -0,0 +1,100 @@
+#ifndef _ASM_X86_IOMMU_TABLE_H
+#define _ASM_X86_IOMMU_TABLE_H
+
+#include <asm/swiotlb.h>
+
+/*
+ * History lesson:
+ * The execution chain of IOMMUs in 2.6.36 looks as so:
+ *
+ *            [xen-swiotlb]
+ *                 |
+ *         +----[swiotlb *]--+
+ *        /         |         \
+ *       /          |          \
+ *    [GART]     [Calgary]  [Intel VT-d]
+ *     /
+ *    /
+ * [AMD-Vi]
+ *
+ * *: if SWIOTLB detected 'iommu=soft'/'swiotlb=force' it would skip
+ * over the rest of IOMMUs and unconditionally initialize the SWIOTLB.
+ * Also it would surreptitiously initialize set the swiotlb=1 if there were
+ * more than 4GB and if the user did not pass in 'iommu=off'. The swiotlb
+ * flag would be turned off by all IOMMUs except the Calgary one.
+ *
+ * The IOMMU_INIT* macros allow a similar tree (or more complex if desired)
+ * to be built by defining who we depend on.
+ *
+ * And all that needs to be done is to use one of the macros in the IOMMU
+ * and the pci-dma.c will take care of the rest.
+ */
+
+struct iommu_table_entry {
+	initcall_t	detect;
+	initcall_t	depend;
+	void		(*early_init)(void); /* No memory allocate available. */
+	void		(*late_init)(void); /* Yes, can allocate memory. */
+#define IOMMU_FINISH_IF_DETECTED (1<<0)
+#define IOMMU_DETECTED		 (1<<1)
+	int		flags;
+};
+/*
+ * Macro fills out an entry in the .iommu_table that is equivalent
+ * to the fields that 'struct iommu_table_entry' has. The entries
+ * that are put in the .iommu_table section are not put in any order
+ * hence during boot-time we will have to resort them based on
+ * dependency. */
+
+
+#define __IOMMU_INIT(_detect, _depend, _early_init, _late_init, _finish)\
+	static const struct iommu_table_entry const			\
+		__iommu_entry_##_detect __used				\
+	__attribute__ ((unused, __section__(".iommu_table"),		\
+			aligned((sizeof(void *)))))	\
+	= {_detect, _depend, _early_init, _late_init,			\
+	   _finish ? IOMMU_FINISH_IF_DETECTED : 0}
+/*
+ * The simplest IOMMU definition. Provide the detection routine
+ * and it will be run after the SWIOTLB and the other IOMMUs
+ * that utilize this macro. If the IOMMU is detected (ie, the
+ * detect routine returns a positive value), the other IOMMUs
+ * are also checked. You can use IOMMU_INIT_POST_FINISH if you prefer
+ * to stop detecting the other IOMMUs after yours has been detected.
+ */
+#define IOMMU_INIT_POST(_detect)					\
+	__IOMMU_INIT(_detect, pci_swiotlb_detect_4gb,  0, 0, 0)
+
+#define IOMMU_INIT_POST_FINISH(detect)					\
+	__IOMMU_INIT(_detect, pci_swiotlb_detect_4gb,  0, 0, 1)
+
+/*
+ * A more sophisticated version of IOMMU_INIT. This variant requires:
+ *  a). A detection routine function.
+ *  b). The name of the detection routine we depend on to get called
+ *      before us.
+ *  c). The init routine which gets called if the detection routine
+ *      returns a positive value from the pci_iommu_alloc. This means
+ *      no presence of a memory allocator.
+ *  d). Similar to the 'init', except that this gets called from pci_iommu_init
+ *      where we do have a memory allocator.
+ *
+ * The standard vs the _FINISH differs in that the _FINISH variant will
+ * continue detecting other IOMMUs in the call list after the
+ * the detection routine returns a positive number. The _FINISH will
+ * stop the execution chain. Both will still call the 'init' and
+ * 'late_init' functions if they are set.
+ */
+#define IOMMU_INIT_FINISH(_detect, _depend, _init, _late_init)		\
+	__IOMMU_INIT(_detect, _depend, _init, _late_init, 1)
+
+#define IOMMU_INIT(_detect, _depend, _init, _late_init)			\
+	__IOMMU_INIT(_detect, _depend, _init, _late_init, 0)
+
+void sort_iommu_table(struct iommu_table_entry *start,
+		      struct iommu_table_entry *finish);
+
+void check_iommu_entries(struct iommu_table_entry *start,
+			 struct iommu_table_entry *finish);
+
+#endif /* _ASM_X86_IOMMU_TABLE_H */
diff --git a/arch/x86/include/asm/irq_remapping.h b/arch/x86/include/asm/irq_remapping.h
index f275e22..1c23360 100644
--- a/arch/x86/include/asm/irq_remapping.h
+++ b/arch/x86/include/asm/irq_remapping.h
@@ -3,4 +3,39 @@
 
 #define IRTE_DEST(dest) ((x2apic_mode) ? dest : dest << 8)
 
+#ifdef CONFIG_INTR_REMAP
+static inline void prepare_irte(struct irte *irte, int vector,
+			        unsigned int dest)
+{
+	memset(irte, 0, sizeof(*irte));
+
+	irte->present = 1;
+	irte->dst_mode = apic->irq_dest_mode;
+	/*
+	 * Trigger mode in the IRTE will always be edge, and for IO-APIC, the
+	 * actual level or edge trigger will be setup in the IO-APIC
+	 * RTE. This will help simplify level triggered irq migration.
+	 * For more details, see the comments (in io_apic.c) explainig IO-APIC
+	 * irq migration in the presence of interrupt-remapping.
+	*/
+	irte->trigger_mode = 0;
+	irte->dlvry_mode = apic->irq_delivery_mode;
+	irte->vector = vector;
+	irte->dest_id = IRTE_DEST(dest);
+	irte->redir_hint = 1;
+}
+static inline bool irq_remapped(struct irq_cfg *cfg)
+{
+	return cfg->irq_2_iommu.iommu != NULL;
+}
+#else
+static void prepare_irte(struct irte *irte, int vector, unsigned int dest)
+{
+}
+static inline bool irq_remapped(struct irq_cfg *cfg)
+{
+	return false;
+}
+#endif
+
 #endif	/* _ASM_X86_IRQ_REMAPPING_H */
diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h
index e2ca300..6af0894 100644
--- a/arch/x86/include/asm/irq_vectors.h
+++ b/arch/x86/include/asm/irq_vectors.h
@@ -114,9 +114,9 @@
 #define X86_PLATFORM_IPI_VECTOR		0xed
 
 /*
- * Performance monitoring pending work vector:
+ * IRQ work vector:
  */
-#define LOCAL_PENDING_VECTOR		0xec
+#define IRQ_WORK_VECTOR			0xec
 
 #define UV_BAU_MESSAGE			0xea
 
diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h
new file mode 100644
index 0000000..f52d42e
--- /dev/null
+++ b/arch/x86/include/asm/jump_label.h
@@ -0,0 +1,37 @@
+#ifndef _ASM_X86_JUMP_LABEL_H
+#define _ASM_X86_JUMP_LABEL_H
+
+#ifdef __KERNEL__
+
+#include <linux/types.h>
+#include <asm/nops.h>
+
+#define JUMP_LABEL_NOP_SIZE 5
+
+# define JUMP_LABEL_INITIAL_NOP ".byte 0xe9 \n\t .long 0\n\t"
+
+# define JUMP_LABEL(key, label)					\
+	do {							\
+		asm goto("1:"					\
+			JUMP_LABEL_INITIAL_NOP			\
+			".pushsection __jump_table,  \"a\" \n\t"\
+			_ASM_PTR "1b, %l[" #label "], %c0 \n\t" \
+			".popsection \n\t"			\
+			: :  "i" (key) :  : label);		\
+	} while (0)
+
+#endif /* __KERNEL__ */
+
+#ifdef CONFIG_X86_64
+typedef u64 jump_label_t;
+#else
+typedef u32 jump_label_t;
+#endif
+
+struct jump_entry {
+	jump_label_t code;
+	jump_label_t target;
+	jump_label_t key;
+};
+
+#endif
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 502e53f..c52e2eb 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -652,20 +652,6 @@
 	return (struct kvm_mmu_page *)page_private(page);
 }
 
-static inline u16 kvm_read_fs(void)
-{
-	u16 seg;
-	asm("mov %%fs, %0" : "=g"(seg));
-	return seg;
-}
-
-static inline u16 kvm_read_gs(void)
-{
-	u16 seg;
-	asm("mov %%gs, %0" : "=g"(seg));
-	return seg;
-}
-
 static inline u16 kvm_read_ldt(void)
 {
 	u16 ldt;
@@ -673,16 +659,6 @@
 	return ldt;
 }
 
-static inline void kvm_load_fs(u16 sel)
-{
-	asm("mov %0, %%fs" : : "rm"(sel));
-}
-
-static inline void kvm_load_gs(u16 sel)
-{
-	asm("mov %0, %%gs" : : "rm"(sel));
-}
-
 static inline void kvm_load_ldt(u16 sel)
 {
 	asm("lldt %0" : : "rm"(sel));
diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h
index 1635074..4a711a6 100644
--- a/arch/x86/include/asm/mrst.h
+++ b/arch/x86/include/asm/mrst.h
@@ -10,6 +10,9 @@
  */
 #ifndef _ASM_X86_MRST_H
 #define _ASM_X86_MRST_H
+
+#include <linux/sfi.h>
+
 extern int pci_mrst_init(void);
 int __init sfi_parse_mrtc(struct sfi_table_header *table);
 
@@ -26,7 +29,7 @@
 };
 
 extern enum mrst_cpu_type __mrst_cpu_chip;
-static enum mrst_cpu_type mrst_identify_cpu(void)
+static inline enum mrst_cpu_type mrst_identify_cpu(void)
 {
 	return __mrst_cpu_chip;
 }
@@ -42,4 +45,9 @@
 #define SFI_MTMR_MAX_NUM 8
 #define SFI_MRTC_MAX	8
 
+extern struct console early_mrst_console;
+extern void mrst_early_console_init(void);
+
+extern struct console early_hsu_console;
+extern void hsu_early_console_init(void);
 #endif /* _ASM_X86_MRST_H */
diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h
new file mode 100644
index 0000000..bcdff99
--- /dev/null
+++ b/arch/x86/include/asm/mwait.h
@@ -0,0 +1,15 @@
+#ifndef _ASM_X86_MWAIT_H
+#define _ASM_X86_MWAIT_H
+
+#define MWAIT_SUBSTATE_MASK		0xf
+#define MWAIT_CSTATE_MASK		0xf
+#define MWAIT_SUBSTATE_SIZE		4
+#define MWAIT_MAX_NUM_CSTATES		8
+
+#define CPUID_MWAIT_LEAF		5
+#define CPUID5_ECX_EXTENSIONS_SUPPORTED 0x1
+#define CPUID5_ECX_INTERRUPT_BREAK	0x2
+
+#define MWAIT_ECX_INTERRUPT_BREAK	0x1
+
+#endif /* _ASM_X86_MWAIT_H */
diff --git a/arch/x86/include/asm/olpc_ofw.h b/arch/x86/include/asm/olpc_ofw.h
index 08fde47..2a84781 100644
--- a/arch/x86/include/asm/olpc_ofw.h
+++ b/arch/x86/include/asm/olpc_ofw.h
@@ -21,10 +21,14 @@
 /* install OFW's pde permanently into the kernel's pgtable */
 extern void setup_olpc_ofw_pgd(void);
 
+/* check if OFW was detected during boot */
+extern bool olpc_ofw_present(void);
+
 #else /* !CONFIG_OLPC_OPENFIRMWARE */
 
 static inline void olpc_ofw_detect(void) { }
 static inline void setup_olpc_ofw_pgd(void) { }
+static inline bool olpc_ofw_present(void) { return false; }
 
 #endif /* !CONFIG_OLPC_OPENFIRMWARE */
 
diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h
index a667f24..1df6621 100644
--- a/arch/x86/include/asm/page_types.h
+++ b/arch/x86/include/asm/page_types.h
@@ -8,7 +8,7 @@
 #define PAGE_SIZE	(_AC(1,UL) << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 
-#define __PHYSICAL_MASK		((phys_addr_t)(1ULL << __PHYSICAL_MASK_SHIFT) - 1)
+#define __PHYSICAL_MASK		((phys_addr_t)((1ULL << __PHYSICAL_MASK_SHIFT) - 1))
 #define __VIRTUAL_MASK		((1UL << __VIRTUAL_MASK_SHIFT) - 1)
 
 /* Cast PAGE_MASK to a signed type so that it is sign-extended if
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 499954c..18e3b8a 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -416,11 +416,6 @@
 	PVOP_VCALL2(pv_mmu_ops.alloc_pmd, mm, pfn);
 }
 
-static inline void paravirt_alloc_pmd_clone(unsigned long pfn, unsigned long clonepfn,
-					    unsigned long start, unsigned long count)
-{
-	PVOP_VCALL4(pv_mmu_ops.alloc_pmd_clone, pfn, clonepfn, start, count);
-}
 static inline void paravirt_release_pmd(unsigned long pfn)
 {
 	PVOP_VCALL1(pv_mmu_ops.release_pmd, pfn);
diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h
index db9ef55..b82bac9 100644
--- a/arch/x86/include/asm/paravirt_types.h
+++ b/arch/x86/include/asm/paravirt_types.h
@@ -255,7 +255,6 @@
 	 */
 	void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn);
 	void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn);
-	void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count);
 	void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn);
 	void (*release_pte)(unsigned long pfn);
 	void (*release_pmd)(unsigned long pfn);
diff --git a/arch/x86/include/asm/perf_event_p4.h b/arch/x86/include/asm/perf_event_p4.h
index def5007..a70cd21 100644
--- a/arch/x86/include/asm/perf_event_p4.h
+++ b/arch/x86/include/asm/perf_event_p4.h
@@ -36,19 +36,6 @@
 #define P4_ESCR_EMASK(v)	((v) << P4_ESCR_EVENTMASK_SHIFT)
 #define P4_ESCR_TAG(v)		((v) << P4_ESCR_TAG_SHIFT)
 
-/* Non HT mask */
-#define P4_ESCR_MASK			\
-	(P4_ESCR_EVENT_MASK	|	\
-	P4_ESCR_EVENTMASK_MASK	|	\
-	P4_ESCR_TAG_MASK	|	\
-	P4_ESCR_TAG_ENABLE	|	\
-	P4_ESCR_T0_OS		|	\
-	P4_ESCR_T0_USR)
-
-/* HT mask */
-#define P4_ESCR_MASK_HT			\
-	(P4_ESCR_MASK |	P4_ESCR_T1_OS | P4_ESCR_T1_USR)
-
 #define P4_CCCR_OVF			0x80000000U
 #define P4_CCCR_CASCADE			0x40000000U
 #define P4_CCCR_OVF_PMI_T0		0x04000000U
@@ -70,23 +57,6 @@
 #define P4_CCCR_THRESHOLD(v)		((v) << P4_CCCR_THRESHOLD_SHIFT)
 #define P4_CCCR_ESEL(v)			((v) << P4_CCCR_ESCR_SELECT_SHIFT)
 
-/* Non HT mask */
-#define P4_CCCR_MASK				\
-	(P4_CCCR_OVF			|	\
-	P4_CCCR_CASCADE			|	\
-	P4_CCCR_OVF_PMI_T0		|	\
-	P4_CCCR_FORCE_OVF		|	\
-	P4_CCCR_EDGE			|	\
-	P4_CCCR_THRESHOLD_MASK		|	\
-	P4_CCCR_COMPLEMENT		|	\
-	P4_CCCR_COMPARE			|	\
-	P4_CCCR_ESCR_SELECT_MASK	|	\
-	P4_CCCR_ENABLE)
-
-/* HT mask */
-#define P4_CCCR_MASK_HT				\
-	(P4_CCCR_MASK | P4_CCCR_OVF_PMI_T1 | P4_CCCR_THREAD_ANY)
-
 #define P4_GEN_ESCR_EMASK(class, name, bit)	\
 	class##__##name = ((1 << bit) << P4_ESCR_EVENTMASK_SHIFT)
 #define P4_ESCR_EMASK_BIT(class, name)		class##__##name
@@ -127,6 +97,28 @@
 #define P4_CONFIG_HT_SHIFT		63
 #define P4_CONFIG_HT			(1ULL << P4_CONFIG_HT_SHIFT)
 
+/*
+ * The bits we allow to pass for RAW events
+ */
+#define P4_CONFIG_MASK_ESCR		\
+	P4_ESCR_EVENT_MASK	|	\
+	P4_ESCR_EVENTMASK_MASK	|	\
+	P4_ESCR_TAG_MASK	|	\
+	P4_ESCR_TAG_ENABLE
+
+#define P4_CONFIG_MASK_CCCR		\
+	P4_CCCR_EDGE		|	\
+	P4_CCCR_THRESHOLD_MASK	|	\
+	P4_CCCR_COMPLEMENT	|	\
+	P4_CCCR_COMPARE		|	\
+	P4_CCCR_THREAD_ANY	|	\
+	P4_CCCR_RESERVED
+
+/* some dangerous bits are reserved for kernel internals */
+#define P4_CONFIG_MASK				  	  \
+	(p4_config_pack_escr(P4_CONFIG_MASK_ESCR))	| \
+	(p4_config_pack_cccr(P4_CONFIG_MASK_CCCR))
+
 static inline bool p4_is_event_cascaded(u64 config)
 {
 	u32 cccr = p4_config_unpack_cccr(config);
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index a34c785..ada823a 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -28,6 +28,8 @@
 extern spinlock_t pgd_lock;
 extern struct list_head pgd_list;
 
+extern struct mm_struct *pgd_page_get_mm(struct page *page);
+
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else  /* !CONFIG_PARAVIRT */
@@ -603,6 +605,8 @@
 	pte_update(mm, addr, ptep);
 }
 
+#define flush_tlb_fix_spurious_fault(vma, address)
+
 /*
  * clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
  *
diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
index 076052c..f96ac9b 100644
--- a/arch/x86/include/asm/pgtable_64.h
+++ b/arch/x86/include/asm/pgtable_64.h
@@ -102,6 +102,8 @@
 	native_set_pgd(pgd, native_make_pgd(0));
 }
 
+extern void sync_global_pgds(unsigned long start, unsigned long end);
+
 /*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index 325b7bd..cae9c3c 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -110,6 +110,8 @@
 	u16			phys_proc_id;
 	/* Core id: */
 	u16			cpu_core_id;
+	/* Compute unit id */
+	u8			compute_unit_id;
 	/* Index into per_cpu list: */
 	u16			cpu_index;
 #endif
@@ -602,7 +604,7 @@
 
 static inline void set_in_cr4(unsigned long mask)
 {
-	unsigned cr4;
+	unsigned long cr4;
 
 	mmu_cr4_features |= mask;
 	cr4 = read_cr4();
@@ -612,7 +614,7 @@
 
 static inline void clear_in_cr4(unsigned long mask)
 {
-	unsigned cr4;
+	unsigned long cr4;
 
 	mmu_cr4_features &= ~mask;
 	cr4 = read_cr4();
@@ -764,29 +766,6 @@
 extern unsigned long		idle_nomwait;
 extern bool			c1e_detected;
 
-/*
- * on systems with caches, caches must be flashed as the absolute
- * last instruction before going into a suspended halt.  Otherwise,
- * dirty data can linger in the cache and become stale on resume,
- * leading to strange errors.
- *
- * perform a variety of operations to guarantee that the compiler
- * will not reorder instructions.  wbinvd itself is serializing
- * so the processor will not reorder.
- *
- * Systems without cache can just go into halt.
- */
-static inline void wbinvd_halt(void)
-{
-	mb();
-	/* check for clflush to determine if wbinvd is legal */
-	if (cpu_has_clflush)
-		asm volatile("cli; wbinvd; 1: hlt; jmp 1b" : : : "memory");
-	else
-		while (1)
-			halt();
-}
-
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h
index ef292c7..d6763b139a 100644
--- a/arch/x86/include/asm/setup.h
+++ b/arch/x86/include/asm/setup.h
@@ -93,6 +93,11 @@
 			: : "i" (sz));					\
 	}
 
+/* Helper for reserving space for arrays of things */
+#define RESERVE_BRK_ARRAY(type, name, entries)		\
+	type *name;					\
+	RESERVE_BRK(name, sizeof(type) * entries)
+
 #ifdef __i386__
 
 void __init i386_start_kernel(void);
diff --git a/arch/x86/include/asm/swiotlb.h b/arch/x86/include/asm/swiotlb.h
index 8085277..977f176 100644
--- a/arch/x86/include/asm/swiotlb.h
+++ b/arch/x86/include/asm/swiotlb.h
@@ -5,17 +5,26 @@
 
 #ifdef CONFIG_SWIOTLB
 extern int swiotlb;
-extern int __init pci_swiotlb_detect(void);
+extern int __init pci_swiotlb_detect_override(void);
+extern int __init pci_swiotlb_detect_4gb(void);
 extern void __init pci_swiotlb_init(void);
+extern void __init pci_swiotlb_late_init(void);
 #else
 #define swiotlb 0
-static inline int pci_swiotlb_detect(void)
+static inline int pci_swiotlb_detect_override(void)
+{
+	return 0;
+}
+static inline int pci_swiotlb_detect_4gb(void)
 {
 	return 0;
 }
 static inline void pci_swiotlb_init(void)
 {
 }
+static inline void pci_swiotlb_late_init(void)
+{
+}
 #endif
 
 static inline void dma_mark_clean(void *addr, size_t size) {}
diff --git a/arch/x86/include/asm/vmi.h b/arch/x86/include/asm/vmi.h
deleted file mode 100644
index 61e08c0..0000000
--- a/arch/x86/include/asm/vmi.h
+++ /dev/null
@@ -1,269 +0,0 @@
-/*
- * VMI interface definition
- *
- * Copyright (C) 2005, VMware, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * Maintained by: Zachary Amsden zach@vmware.com
- *
- */
-#include <linux/types.h>
-
-/*
- *---------------------------------------------------------------------
- *
- *  VMI Option ROM API
- *
- *---------------------------------------------------------------------
- */
-#define VMI_SIGNATURE 0x696d5663   /* "cVmi" */
-
-#define PCI_VENDOR_ID_VMWARE            0x15AD
-#define PCI_DEVICE_ID_VMWARE_VMI        0x0801
-
-/*
- * We use two version numbers for compatibility, with the major
- * number signifying interface breakages, and the minor number
- * interface extensions.
- */
-#define VMI_API_REV_MAJOR       3
-#define VMI_API_REV_MINOR       0
-
-#define VMI_CALL_CPUID			0
-#define VMI_CALL_WRMSR			1
-#define VMI_CALL_RDMSR			2
-#define VMI_CALL_SetGDT			3
-#define VMI_CALL_SetLDT			4
-#define VMI_CALL_SetIDT			5
-#define VMI_CALL_SetTR			6
-#define VMI_CALL_GetGDT			7
-#define VMI_CALL_GetLDT			8
-#define VMI_CALL_GetIDT			9
-#define VMI_CALL_GetTR			10
-#define VMI_CALL_WriteGDTEntry		11
-#define VMI_CALL_WriteLDTEntry		12
-#define VMI_CALL_WriteIDTEntry		13
-#define VMI_CALL_UpdateKernelStack	14
-#define VMI_CALL_SetCR0			15
-#define VMI_CALL_SetCR2			16
-#define VMI_CALL_SetCR3			17
-#define VMI_CALL_SetCR4			18
-#define VMI_CALL_GetCR0			19
-#define VMI_CALL_GetCR2			20
-#define VMI_CALL_GetCR3			21
-#define VMI_CALL_GetCR4			22
-#define VMI_CALL_WBINVD			23
-#define VMI_CALL_SetDR			24
-#define VMI_CALL_GetDR			25
-#define VMI_CALL_RDPMC			26
-#define VMI_CALL_RDTSC			27
-#define VMI_CALL_CLTS			28
-#define VMI_CALL_EnableInterrupts	29
-#define VMI_CALL_DisableInterrupts	30
-#define VMI_CALL_GetInterruptMask	31
-#define VMI_CALL_SetInterruptMask	32
-#define VMI_CALL_IRET			33
-#define VMI_CALL_SYSEXIT		34
-#define VMI_CALL_Halt			35
-#define VMI_CALL_Reboot			36
-#define VMI_CALL_Shutdown		37
-#define VMI_CALL_SetPxE			38
-#define VMI_CALL_SetPxELong		39
-#define VMI_CALL_UpdatePxE		40
-#define VMI_CALL_UpdatePxELong		41
-#define VMI_CALL_MachineToPhysical	42
-#define VMI_CALL_PhysicalToMachine	43
-#define VMI_CALL_AllocatePage		44
-#define VMI_CALL_ReleasePage		45
-#define VMI_CALL_InvalPage		46
-#define VMI_CALL_FlushTLB		47
-#define VMI_CALL_SetLinearMapping	48
-
-#define VMI_CALL_SetIOPLMask		61
-#define VMI_CALL_SetInitialAPState	62
-#define VMI_CALL_APICWrite		63
-#define VMI_CALL_APICRead		64
-#define VMI_CALL_IODelay		65
-#define VMI_CALL_SetLazyMode		73
-
-/*
- *---------------------------------------------------------------------
- *
- * MMU operation flags
- *
- *---------------------------------------------------------------------
- */
-
-/* Flags used by VMI_{Allocate|Release}Page call */
-#define VMI_PAGE_PAE             0x10  /* Allocate PAE shadow */
-#define VMI_PAGE_CLONE           0x20  /* Clone from another shadow */
-#define VMI_PAGE_ZEROED          0x40  /* Page is pre-zeroed */
-
-
-/* Flags shared by Allocate|Release Page and PTE updates */
-#define VMI_PAGE_PT              0x01
-#define VMI_PAGE_PD              0x02
-#define VMI_PAGE_PDP             0x04
-#define VMI_PAGE_PML4            0x08
-
-#define VMI_PAGE_NORMAL          0x00 /* for debugging */
-
-/* Flags used by PTE updates */
-#define VMI_PAGE_CURRENT_AS      0x10 /* implies VMI_PAGE_VA_MASK is valid */
-#define VMI_PAGE_DEFER           0x20 /* may queue update until TLB inval */
-#define VMI_PAGE_VA_MASK         0xfffff000
-
-#ifdef CONFIG_X86_PAE
-#define VMI_PAGE_L1		(VMI_PAGE_PT | VMI_PAGE_PAE | VMI_PAGE_ZEROED)
-#define VMI_PAGE_L2		(VMI_PAGE_PD | VMI_PAGE_PAE | VMI_PAGE_ZEROED)
-#else
-#define VMI_PAGE_L1		(VMI_PAGE_PT | VMI_PAGE_ZEROED)
-#define VMI_PAGE_L2		(VMI_PAGE_PD | VMI_PAGE_ZEROED)
-#endif
-
-/* Flags used by VMI_FlushTLB call */
-#define VMI_FLUSH_TLB            0x01
-#define VMI_FLUSH_GLOBAL         0x02
-
-/*
- *---------------------------------------------------------------------
- *
- *  VMI relocation definitions for ROM call get_reloc
- *
- *---------------------------------------------------------------------
- */
-
-/* VMI Relocation types */
-#define VMI_RELOCATION_NONE     0
-#define VMI_RELOCATION_CALL_REL 1
-#define VMI_RELOCATION_JUMP_REL 2
-#define VMI_RELOCATION_NOP	3
-
-#ifndef __ASSEMBLY__
-struct vmi_relocation_info {
-	unsigned char           *eip;
-	unsigned char           type;
-	unsigned char           reserved[3];
-};
-#endif
-
-
-/*
- *---------------------------------------------------------------------
- *
- *  Generic ROM structures and definitions
- *
- *---------------------------------------------------------------------
- */
-
-#ifndef __ASSEMBLY__
-
-struct vrom_header {
-	u16     rom_signature;  /* option ROM signature */
-	u8      rom_length;     /* ROM length in 512 byte chunks */
-	u8      rom_entry[4];   /* 16-bit code entry point */
-	u8      rom_pad0;       /* 4-byte align pad */
-	u32     vrom_signature; /* VROM identification signature */
-	u8      api_version_min;/* Minor version of API */
-	u8      api_version_maj;/* Major version of API */
-	u8      jump_slots;     /* Number of jump slots */
-	u8      reserved1;      /* Reserved for expansion */
-	u32     virtual_top;    /* Hypervisor virtual address start */
-	u16     reserved2;      /* Reserved for expansion */
-	u16	license_offs;	/* Offset to License string */
-	u16     pci_header_offs;/* Offset to PCI OPROM header */
-	u16     pnp_header_offs;/* Offset to PnP OPROM header */
-	u32     rom_pad3;       /* PnP reserverd / VMI reserved */
-	u8      reserved[96];   /* Reserved for headers */
-	char    vmi_init[8];    /* VMI_Init jump point */
-	char    get_reloc[8];   /* VMI_GetRelocationInfo jump point */
-} __attribute__((packed));
-
-struct pnp_header {
-	char sig[4];
-	char rev;
-	char size;
-	short next;
-	short res;
-	long devID;
-	unsigned short manufacturer_offset;
-	unsigned short product_offset;
-} __attribute__((packed));
-
-struct pci_header {
-	char sig[4];
-	short vendorID;
-	short deviceID;
-	short vpdData;
-	short size;
-	char rev;
-	char class;
-	char subclass;
-	char interface;
-	short chunks;
-	char rom_version_min;
-	char rom_version_maj;
-	char codetype;
-	char lastRom;
-	short reserved;
-} __attribute__((packed));
-
-/* Function prototypes for bootstrapping */
-#ifdef CONFIG_VMI
-extern void vmi_init(void);
-extern void vmi_activate(void);
-extern void vmi_bringup(void);
-#else
-static inline void vmi_init(void) {}
-static inline void vmi_activate(void) {}
-static inline void vmi_bringup(void) {}
-#endif
-
-/* State needed to start an application processor in an SMP system. */
-struct vmi_ap_state {
-	u32 cr0;
-	u32 cr2;
-	u32 cr3;
-	u32 cr4;
-
-	u64 efer;
-
-	u32 eip;
-	u32 eflags;
-	u32 eax;
-	u32 ebx;
-	u32 ecx;
-	u32 edx;
-	u32 esp;
-	u32 ebp;
-	u32 esi;
-	u32 edi;
-	u16 cs;
-	u16 ss;
-	u16 ds;
-	u16 es;
-	u16 fs;
-	u16 gs;
-	u16 ldtr;
-
-	u16 gdtr_limit;
-	u32 gdtr_base;
-	u32 idtr_base;
-	u16 idtr_limit;
-};
-
-#endif
diff --git a/arch/x86/include/asm/vmi_time.h b/arch/x86/include/asm/vmi_time.h
deleted file mode 100644
index c6e0bee..0000000
--- a/arch/x86/include/asm/vmi_time.h
+++ /dev/null
@@ -1,98 +0,0 @@
-/*
- * VMI Time wrappers
- *
- * Copyright (C) 2006, VMware, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * Send feedback to dhecht@vmware.com
- *
- */
-
-#ifndef _ASM_X86_VMI_TIME_H
-#define _ASM_X86_VMI_TIME_H
-
-/*
- * Raw VMI call indices for timer functions
- */
-#define VMI_CALL_GetCycleFrequency	66
-#define VMI_CALL_GetCycleCounter	67
-#define VMI_CALL_SetAlarm		68
-#define VMI_CALL_CancelAlarm		69
-#define VMI_CALL_GetWallclockTime	70
-#define VMI_CALL_WallclockUpdated	71
-
-/* Cached VMI timer operations */
-extern struct vmi_timer_ops {
-	u64 (*get_cycle_frequency)(void);
-	u64 (*get_cycle_counter)(int);
-	u64 (*get_wallclock)(void);
-	int (*wallclock_updated)(void);
-	void (*set_alarm)(u32 flags, u64 expiry, u64 period);
-	void (*cancel_alarm)(u32 flags);
-} vmi_timer_ops;
-
-/* Prototypes */
-extern void __init vmi_time_init(void);
-extern unsigned long vmi_get_wallclock(void);
-extern int vmi_set_wallclock(unsigned long now);
-extern unsigned long long vmi_sched_clock(void);
-extern unsigned long vmi_tsc_khz(void);
-
-#ifdef CONFIG_X86_LOCAL_APIC
-extern void __devinit vmi_time_bsp_init(void);
-extern void __devinit vmi_time_ap_init(void);
-#endif
-
-/*
- * When run under a hypervisor, a vcpu is always in one of three states:
- * running, halted, or ready.  The vcpu is in the 'running' state if it
- * is executing.  When the vcpu executes the halt interface, the vcpu
- * enters the 'halted' state and remains halted until there is some work
- * pending for the vcpu (e.g. an alarm expires, host I/O completes on
- * behalf of virtual I/O).  At this point, the vcpu enters the 'ready'
- * state (waiting for the hypervisor to reschedule it).  Finally, at any
- * time when the vcpu is not in the 'running' state nor the 'halted'
- * state, it is in the 'ready' state.
- *
- * Real time is advances while the vcpu is 'running', 'ready', or
- * 'halted'.  Stolen time is the time in which the vcpu is in the
- * 'ready' state.  Available time is the remaining time -- the vcpu is
- * either 'running' or 'halted'.
- *
- * All three views of time are accessible through the VMI cycle
- * counters.
- */
-
-/* The cycle counters. */
-#define VMI_CYCLES_REAL         0
-#define VMI_CYCLES_AVAILABLE    1
-#define VMI_CYCLES_STOLEN       2
-
-/* The alarm interface 'flags' bits */
-#define VMI_ALARM_COUNTERS      2
-
-#define VMI_ALARM_COUNTER_MASK  0x000000ff
-
-#define VMI_ALARM_WIRED_IRQ0    0x00000000
-#define VMI_ALARM_WIRED_LVTT    0x00010000
-
-#define VMI_ALARM_IS_ONESHOT    0x00000000
-#define VMI_ALARM_IS_PERIODIC   0x00000100
-
-#define CONFIG_VMI_ALARM_HZ	100
-
-#endif /* _ASM_X86_VMI_TIME_H */
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index fedf32a..2c833d8 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -34,7 +34,8 @@
 obj-y			:= process_$(BITS).o signal.o entry_$(BITS).o
 obj-y			+= traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o
 obj-y			+= time.o ioport.o ldt.o dumpstack.o
-obj-y			+= setup.o x86_init.o i8259.o irqinit.o
+obj-y			+= setup.o x86_init.o i8259.o irqinit.o jump_label.o
+obj-$(CONFIG_IRQ_WORK)  += irq_work.o
 obj-$(CONFIG_X86_VISWS)	+= visws_quirks.o
 obj-$(CONFIG_X86_32)	+= probe_roms_32.o
 obj-$(CONFIG_X86_32)	+= sys_i386_32.o i386_ksyms_32.o
@@ -44,6 +45,7 @@
 obj-y			+= pci-dma.o quirks.o i8237.o topology.o kdebugfs.o
 obj-y			+= alternative.o i8253.o pci-nommu.o hw_breakpoint.o
 obj-y			+= tsc.o io_delay.o rtc.o
+obj-y			+= pci-iommu_table.o
 
 obj-$(CONFIG_X86_TRAMPOLINE)	+= trampoline.o
 obj-y				+= process.o
@@ -85,15 +87,15 @@
 obj-$(CONFIG_KGDB)		+= kgdb.o
 obj-$(CONFIG_VM86)		+= vm86_32.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
+obj-$(CONFIG_EARLY_PRINTK_MRST)	+= early_printk_mrst.o
 
 obj-$(CONFIG_HPET_TIMER) 	+= hpet.o
 obj-$(CONFIG_APB_TIMER)		+= apb_timer.o
 
-obj-$(CONFIG_K8_NB)		+= k8.o
+obj-$(CONFIG_AMD_NB)		+= amd_nb.o
 obj-$(CONFIG_DEBUG_RODATA_TEST)	+= test_rodata.o
 obj-$(CONFIG_DEBUG_NX_TEST)	+= test_nx.o
 
-obj-$(CONFIG_VMI)		+= vmi_32.o vmiclock_32.o
 obj-$(CONFIG_KVM_GUEST)		+= kvm.o
 obj-$(CONFIG_KVM_CLOCK)		+= kvmclock.o
 obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirt_patch_$(BITS).o
@@ -106,6 +108,7 @@
 scx200-y			+= scx200_32.o
 
 obj-$(CONFIG_OLPC)		+= olpc.o
+obj-$(CONFIG_OLPC_XO1)		+= olpc-xo1.o
 obj-$(CONFIG_OLPC_OPENFIRMWARE)	+= olpc_ofw.o
 obj-$(CONFIG_X86_MRST)		+= mrst.o
 
@@ -122,7 +125,6 @@
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
 	obj-$(CONFIG_X86_UV)		+= tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o
-	obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
 	obj-$(CONFIG_AUDIT)		+= audit_64.o
 
 	obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index fb16f17..5812404 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -13,6 +13,7 @@
 
 #include <acpi/processor.h>
 #include <asm/acpi.h>
+#include <asm/mwait.h>
 
 /*
  * Initialize bm_flags based on the CPU cache properties
@@ -65,16 +66,6 @@
 
 static short mwait_supported[ACPI_PROCESSOR_MAX_POWER];
 
-#define MWAIT_SUBSTATE_MASK	(0xf)
-#define MWAIT_CSTATE_MASK	(0xf)
-#define MWAIT_SUBSTATE_SIZE	(4)
-
-#define CPUID_MWAIT_LEAF (5)
-#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1)
-#define CPUID5_ECX_INTERRUPT_BREAK	(0x2)
-
-#define MWAIT_ECX_INTERRUPT_BREAK	(0x1)
-
 #define NATIVE_CSTATE_BEYOND_HALT	(2)
 
 static long acpi_processor_ffh_cstate_probe_cpu(void *_cx)
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index f65ab8b..a36bb90 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -195,7 +195,7 @@
 
 extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
 extern s32 __smp_locks[], __smp_locks_end[];
-static void *text_poke_early(void *addr, const void *opcode, size_t len);
+void *text_poke_early(void *addr, const void *opcode, size_t len);
 
 /* Replace instructions with better alternatives for this CPU type.
    This runs before SMP is initialized to avoid SMP problems with
@@ -522,7 +522,7 @@
  * instructions. And on the local CPU you need to be protected again NMI or MCE
  * handlers seeing an inconsistent instruction while you patch.
  */
-static void *__init_or_module text_poke_early(void *addr, const void *opcode,
+void *__init_or_module text_poke_early(void *addr, const void *opcode,
 					      size_t len)
 {
 	unsigned long flags;
@@ -637,7 +637,72 @@
 	tpp.len = len;
 	atomic_set(&stop_machine_first, 1);
 	wrote_text = 0;
-	stop_machine(stop_machine_text_poke, (void *)&tpp, NULL);
+	/* Use __stop_machine() because the caller already got online_cpus. */
+	__stop_machine(stop_machine_text_poke, (void *)&tpp, NULL);
 	return addr;
 }
 
+#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL)
+
+unsigned char ideal_nop5[IDEAL_NOP_SIZE_5];
+
+void __init arch_init_ideal_nop5(void)
+{
+	extern const unsigned char ftrace_test_p6nop[];
+	extern const unsigned char ftrace_test_nop5[];
+	extern const unsigned char ftrace_test_jmp[];
+	int faulted = 0;
+
+	/*
+	 * There is no good nop for all x86 archs.
+	 * We will default to using the P6_NOP5, but first we
+	 * will test to make sure that the nop will actually
+	 * work on this CPU. If it faults, we will then
+	 * go to a lesser efficient 5 byte nop. If that fails
+	 * we then just use a jmp as our nop. This isn't the most
+	 * efficient nop, but we can not use a multi part nop
+	 * since we would then risk being preempted in the middle
+	 * of that nop, and if we enabled tracing then, it might
+	 * cause a system crash.
+	 *
+	 * TODO: check the cpuid to determine the best nop.
+	 */
+	asm volatile (
+		"ftrace_test_jmp:"
+		"jmp ftrace_test_p6nop\n"
+		"nop\n"
+		"nop\n"
+		"nop\n"  /* 2 byte jmp + 3 bytes */
+		"ftrace_test_p6nop:"
+		P6_NOP5
+		"jmp 1f\n"
+		"ftrace_test_nop5:"
+		".byte 0x66,0x66,0x66,0x66,0x90\n"
+		"1:"
+		".section .fixup, \"ax\"\n"
+		"2:	movl $1, %0\n"
+		"	jmp ftrace_test_nop5\n"
+		"3:	movl $2, %0\n"
+		"	jmp 1b\n"
+		".previous\n"
+		_ASM_EXTABLE(ftrace_test_p6nop, 2b)
+		_ASM_EXTABLE(ftrace_test_nop5, 3b)
+		: "=r"(faulted) : "0" (faulted));
+
+	switch (faulted) {
+	case 0:
+		pr_info("converting mcount calls to 0f 1f 44 00 00\n");
+		memcpy(ideal_nop5, ftrace_test_p6nop, IDEAL_NOP_SIZE_5);
+		break;
+	case 1:
+		pr_info("converting mcount calls to 66 66 66 66 90\n");
+		memcpy(ideal_nop5, ftrace_test_nop5, IDEAL_NOP_SIZE_5);
+		break;
+	case 2:
+		pr_info("converting mcount calls to jmp . + 5\n");
+		memcpy(ideal_nop5, ftrace_test_jmp, IDEAL_NOP_SIZE_5);
+		break;
+	}
+
+}
+#endif
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 679b645..d2fdb08 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007-2009 Advanced Micro Devices, Inc.
+ * Copyright (C) 2007-2010 Advanced Micro Devices, Inc.
  * Author: Joerg Roedel <joerg.roedel@amd.com>
  *         Leo Duran <leo.duran@amd.com>
  *
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 5a170cb..6e11c81 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007-2009 Advanced Micro Devices, Inc.
+ * Copyright (C) 2007-2010 Advanced Micro Devices, Inc.
  * Author: Joerg Roedel <joerg.roedel@amd.com>
  *         Leo Duran <leo.duran@amd.com>
  *
@@ -31,7 +31,7 @@
 #include <asm/iommu.h>
 #include <asm/gart.h>
 #include <asm/x86_init.h>
-
+#include <asm/iommu_table.h>
 /*
  * definitions for the ACPI scanning code
  */
@@ -194,6 +194,39 @@
 	return 1UL << shift;
 }
 
+/* Access to l1 and l2 indexed register spaces */
+
+static u32 iommu_read_l1(struct amd_iommu *iommu, u16 l1, u8 address)
+{
+	u32 val;
+
+	pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16));
+	pci_read_config_dword(iommu->dev, 0xfc, &val);
+	return val;
+}
+
+static void iommu_write_l1(struct amd_iommu *iommu, u16 l1, u8 address, u32 val)
+{
+	pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16 | 1 << 31));
+	pci_write_config_dword(iommu->dev, 0xfc, val);
+	pci_write_config_dword(iommu->dev, 0xf8, (address | l1 << 16));
+}
+
+static u32 iommu_read_l2(struct amd_iommu *iommu, u8 address)
+{
+	u32 val;
+
+	pci_write_config_dword(iommu->dev, 0xf0, address);
+	pci_read_config_dword(iommu->dev, 0xf4, &val);
+	return val;
+}
+
+static void iommu_write_l2(struct amd_iommu *iommu, u8 address, u32 val)
+{
+	pci_write_config_dword(iommu->dev, 0xf0, (address | 1 << 8));
+	pci_write_config_dword(iommu->dev, 0xf4, val);
+}
+
 /****************************************************************************
  *
  * AMD IOMMU MMIO register space handling functions
@@ -619,6 +652,7 @@
 {
 	int cap_ptr = iommu->cap_ptr;
 	u32 range, misc;
+	int i, j;
 
 	pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET,
 			      &iommu->cap);
@@ -633,12 +667,29 @@
 					MMIO_GET_LD(range));
 	iommu->evt_msi_num = MMIO_MSI_NUM(misc);
 
-	if (is_rd890_iommu(iommu->dev)) {
-		pci_read_config_dword(iommu->dev, 0xf0, &iommu->cache_cfg[0]);
-		pci_read_config_dword(iommu->dev, 0xf4, &iommu->cache_cfg[1]);
-		pci_read_config_dword(iommu->dev, 0xf8, &iommu->cache_cfg[2]);
-		pci_read_config_dword(iommu->dev, 0xfc, &iommu->cache_cfg[3]);
-	}
+	if (!is_rd890_iommu(iommu->dev))
+		return;
+
+	/*
+	 * Some rd890 systems may not be fully reconfigured by the BIOS, so
+	 * it's necessary for us to store this information so it can be
+	 * reprogrammed on resume
+	 */
+
+	pci_read_config_dword(iommu->dev, iommu->cap_ptr + 4,
+			      &iommu->stored_addr_lo);
+	pci_read_config_dword(iommu->dev, iommu->cap_ptr + 8,
+			      &iommu->stored_addr_hi);
+
+	/* Low bit locks writes to configuration space */
+	iommu->stored_addr_lo &= ~1;
+
+	for (i = 0; i < 6; i++)
+		for (j = 0; j < 0x12; j++)
+			iommu->stored_l1[i][j] = iommu_read_l1(iommu, i, j);
+
+	for (i = 0; i < 0x83; i++)
+		iommu->stored_l2[i] = iommu_read_l2(iommu, i);
 }
 
 /*
@@ -1127,14 +1178,53 @@
 	iommu_feature_enable(iommu, CONTROL_COHERENT_EN);
 }
 
-static void iommu_apply_quirks(struct amd_iommu *iommu)
+static void iommu_apply_resume_quirks(struct amd_iommu *iommu)
 {
-	if (is_rd890_iommu(iommu->dev)) {
-		pci_write_config_dword(iommu->dev, 0xf0, iommu->cache_cfg[0]);
-		pci_write_config_dword(iommu->dev, 0xf4, iommu->cache_cfg[1]);
-		pci_write_config_dword(iommu->dev, 0xf8, iommu->cache_cfg[2]);
-		pci_write_config_dword(iommu->dev, 0xfc, iommu->cache_cfg[3]);
-	}
+	int i, j;
+	u32 ioc_feature_control;
+	struct pci_dev *pdev = NULL;
+
+	/* RD890 BIOSes may not have completely reconfigured the iommu */
+	if (!is_rd890_iommu(iommu->dev))
+		return;
+
+	/*
+	 * First, we need to ensure that the iommu is enabled. This is
+	 * controlled by a register in the northbridge
+	 */
+	pdev = pci_get_bus_and_slot(iommu->dev->bus->number, PCI_DEVFN(0, 0));
+
+	if (!pdev)
+		return;
+
+	/* Select Northbridge indirect register 0x75 and enable writing */
+	pci_write_config_dword(pdev, 0x60, 0x75 | (1 << 7));
+	pci_read_config_dword(pdev, 0x64, &ioc_feature_control);
+
+	/* Enable the iommu */
+	if (!(ioc_feature_control & 0x1))
+		pci_write_config_dword(pdev, 0x64, ioc_feature_control | 1);
+
+	pci_dev_put(pdev);
+
+	/* Restore the iommu BAR */
+	pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4,
+			       iommu->stored_addr_lo);
+	pci_write_config_dword(iommu->dev, iommu->cap_ptr + 8,
+			       iommu->stored_addr_hi);
+
+	/* Restore the l1 indirect regs for each of the 6 l1s */
+	for (i = 0; i < 6; i++)
+		for (j = 0; j < 0x12; j++)
+			iommu_write_l1(iommu, i, j, iommu->stored_l1[i][j]);
+
+	/* Restore the l2 indirect regs */
+	for (i = 0; i < 0x83; i++)
+		iommu_write_l2(iommu, i, iommu->stored_l2[i]);
+
+	/* Lock PCI setup registers */
+	pci_write_config_dword(iommu->dev, iommu->cap_ptr + 4,
+			       iommu->stored_addr_lo | 1);
 }
 
 /*
@@ -1147,7 +1237,6 @@
 
 	for_each_iommu(iommu) {
 		iommu_disable(iommu);
-		iommu_apply_quirks(iommu);
 		iommu_init_flags(iommu);
 		iommu_set_device_table(iommu);
 		iommu_enable_command_buffer(iommu);
@@ -1173,6 +1262,11 @@
 
 static int amd_iommu_resume(struct sys_device *dev)
 {
+	struct amd_iommu *iommu;
+
+	for_each_iommu(iommu)
+		iommu_apply_resume_quirks(iommu);
+
 	/* re-load the hardware */
 	enable_iommus();
 
@@ -1405,13 +1499,13 @@
 	return 0;
 }
 
-void __init amd_iommu_detect(void)
+int __init amd_iommu_detect(void)
 {
 	if (no_iommu || (iommu_detected && !gart_iommu_aperture))
-		return;
+		return -ENODEV;
 
 	if (amd_iommu_disabled)
-		return;
+		return -ENODEV;
 
 	if (acpi_table_parse("IVRS", early_amd_iommu_detect) == 0) {
 		iommu_detected = 1;
@@ -1420,7 +1514,9 @@
 
 		/* Make sure ACS will be enabled */
 		pci_request_acs();
+		return 1;
 	}
+	return -ENODEV;
 }
 
 /****************************************************************************
@@ -1451,3 +1547,8 @@
 
 __setup("amd_iommu_dump", parse_amd_iommu_dump);
 __setup("amd_iommu=", parse_amd_iommu_options);
+
+IOMMU_INIT_FINISH(amd_iommu_detect,
+		  gart_iommu_hole_init,
+		  0,
+		  0);
diff --git a/arch/x86/kernel/k8.c b/arch/x86/kernel/amd_nb.c
similarity index 66%
rename from arch/x86/kernel/k8.c
rename to arch/x86/kernel/amd_nb.c
index 0f7bc20..8f6463d 100644
--- a/arch/x86/kernel/k8.c
+++ b/arch/x86/kernel/amd_nb.c
@@ -8,21 +8,19 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>
-#include <asm/k8.h>
-
-int num_k8_northbridges;
-EXPORT_SYMBOL(num_k8_northbridges);
+#include <asm/amd_nb.h>
 
 static u32 *flush_words;
 
 struct pci_device_id k8_nb_ids[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_MISC) },
 	{}
 };
 EXPORT_SYMBOL(k8_nb_ids);
 
-struct pci_dev **k8_northbridges;
+struct k8_northbridge_info k8_northbridges;
 EXPORT_SYMBOL(k8_northbridges);
 
 static struct pci_dev *next_k8_northbridge(struct pci_dev *dev)
@@ -40,36 +38,45 @@
 	int i;
 	struct pci_dev *dev;
 
-	if (num_k8_northbridges)
+	if (k8_northbridges.num)
 		return 0;
 
 	dev = NULL;
 	while ((dev = next_k8_northbridge(dev)) != NULL)
-		num_k8_northbridges++;
+		k8_northbridges.num++;
 
-	k8_northbridges = kmalloc((num_k8_northbridges + 1) * sizeof(void *),
-				  GFP_KERNEL);
-	if (!k8_northbridges)
+	/* some CPU families (e.g. family 0x11) do not support GART */
+	if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 ||
+	    boot_cpu_data.x86 == 0x15)
+		k8_northbridges.gart_supported = 1;
+
+	k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) *
+					  sizeof(void *), GFP_KERNEL);
+	if (!k8_northbridges.nb_misc)
 		return -ENOMEM;
 
-	if (!num_k8_northbridges) {
-		k8_northbridges[0] = NULL;
+	if (!k8_northbridges.num) {
+		k8_northbridges.nb_misc[0] = NULL;
 		return 0;
 	}
 
-	flush_words = kmalloc(num_k8_northbridges * sizeof(u32), GFP_KERNEL);
-	if (!flush_words) {
-		kfree(k8_northbridges);
-		return -ENOMEM;
+	if (k8_northbridges.gart_supported) {
+		flush_words = kmalloc(k8_northbridges.num * sizeof(u32),
+				      GFP_KERNEL);
+		if (!flush_words) {
+			kfree(k8_northbridges.nb_misc);
+			return -ENOMEM;
+		}
 	}
 
 	dev = NULL;
 	i = 0;
 	while ((dev = next_k8_northbridge(dev)) != NULL) {
-		k8_northbridges[i] = dev;
-		pci_read_config_dword(dev, 0x9c, &flush_words[i++]);
+		k8_northbridges.nb_misc[i] = dev;
+		if (k8_northbridges.gart_supported)
+			pci_read_config_dword(dev, 0x9c, &flush_words[i++]);
 	}
-	k8_northbridges[i] = NULL;
+	k8_northbridges.nb_misc[i] = NULL;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(cache_k8_northbridges);
@@ -93,22 +100,25 @@
 	unsigned long flags;
 	static DEFINE_SPINLOCK(gart_lock);
 
+	if (!k8_northbridges.gart_supported)
+		return;
+
 	/* Avoid races between AGP and IOMMU. In theory it's not needed
 	   but I'm not sure if the hardware won't lose flush requests
 	   when another is pending. This whole thing is so expensive anyways
 	   that it doesn't matter to serialize more. -AK */
 	spin_lock_irqsave(&gart_lock, flags);
 	flushed = 0;
-	for (i = 0; i < num_k8_northbridges; i++) {
-		pci_write_config_dword(k8_northbridges[i], 0x9c,
+	for (i = 0; i < k8_northbridges.num; i++) {
+		pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c,
 				       flush_words[i]|1);
 		flushed++;
 	}
-	for (i = 0; i < num_k8_northbridges; i++) {
+	for (i = 0; i < k8_northbridges.num; i++) {
 		u32 w;
 		/* Make sure the hardware actually executed the flush*/
 		for (;;) {
-			pci_read_config_dword(k8_northbridges[i],
+			pci_read_config_dword(k8_northbridges.nb_misc[i],
 					      0x9c, &w);
 			if (!(w & 1))
 				break;
diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c
index 8dd7780..92543c7 100644
--- a/arch/x86/kernel/apb_timer.c
+++ b/arch/x86/kernel/apb_timer.c
@@ -231,34 +231,6 @@
 	apbt_start_counter(phy_cs_timer_id);
 }
 
-/* Setup IRQ routing via IOAPIC */
-#ifdef CONFIG_SMP
-static void apbt_setup_irq(struct apbt_dev *adev)
-{
-	struct irq_chip *chip;
-	struct irq_desc *desc;
-
-	/* timer0 irq has been setup early */
-	if (adev->irq == 0)
-		return;
-	desc = irq_to_desc(adev->irq);
-	chip = get_irq_chip(adev->irq);
-	disable_irq(adev->irq);
-	desc->status |= IRQ_MOVE_PCNTXT;
-	irq_set_affinity(adev->irq, cpumask_of(adev->cpu));
-	/* APB timer irqs are set up as mp_irqs, timer is edge triggerred */
-	set_irq_chip_and_handler_name(adev->irq, chip, handle_edge_irq, "edge");
-	enable_irq(adev->irq);
-	if (system_state == SYSTEM_BOOTING)
-		if (request_irq(adev->irq, apbt_interrupt_handler,
-				IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING,
-				adev->name, adev)) {
-			printk(KERN_ERR "Failed request IRQ for APBT%d\n",
-			       adev->num);
-		}
-}
-#endif
-
 static void apbt_enable_int(int n)
 {
 	unsigned long ctrl = apbt_readl(n, APBTMR_N_CONTROL);
@@ -334,6 +306,27 @@
 }
 
 #ifdef CONFIG_SMP
+
+static void apbt_setup_irq(struct apbt_dev *adev)
+{
+	/* timer0 irq has been setup early */
+	if (adev->irq == 0)
+		return;
+
+	if (system_state == SYSTEM_BOOTING) {
+		irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT);
+		/* APB timer irqs are set up as mp_irqs, timer is edge type */
+		__set_irq_handler(adev->irq, handle_edge_irq, 0, "edge");
+		if (request_irq(adev->irq, apbt_interrupt_handler,
+				IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING,
+				adev->name, adev)) {
+			printk(KERN_ERR "Failed request IRQ for APBT%d\n",
+			       adev->num);
+		}
+	} else
+		enable_irq(adev->irq);
+}
+
 /* Should be called with per cpu */
 void apbt_setup_secondary_clock(void)
 {
@@ -343,7 +336,7 @@
 
 	/* Don't register boot CPU clockevent */
 	cpu = smp_processor_id();
-	if (cpu == boot_cpu_id)
+	if (!cpu)
 		return;
 	/*
 	 * We need to calculate the scaled math multiplication factor for
@@ -389,16 +382,17 @@
 
 	switch (action & 0xf) {
 	case CPU_DEAD:
+		disable_irq(adev->irq);
 		apbt_disable_int(cpu);
-		if (system_state == SYSTEM_RUNNING)
+		if (system_state == SYSTEM_RUNNING) {
 			pr_debug("skipping APBT CPU %lu offline\n", cpu);
-		else if (adev) {
+		} else if (adev) {
 			pr_debug("APBT clockevent for cpu %lu offline\n", cpu);
 			free_irq(adev->irq, adev);
 		}
 		break;
 	default:
-		pr_debug(KERN_INFO "APBT notified %lu, no action\n", action);
+		pr_debug("APBT notified %lu, no action\n", action);
 	}
 	return NOTIFY_OK;
 }
@@ -552,7 +546,7 @@
 		pr_debug("APB CS going back %lx:%lx:%lx ",
 			 t2, last_read, t2 - last_read);
 bad_count_x3:
-		pr_debug(KERN_INFO "tripple check enforced\n");
+		pr_debug("triple check enforced\n");
 		t0 = apbt_readl(phy_cs_timer_id,
 				APBTMR_N_CURRENT_VALUE);
 		udelay(1);
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index a2e0caf..b3a16e8 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -27,7 +27,7 @@
 #include <asm/gart.h>
 #include <asm/pci-direct.h>
 #include <asm/dma.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #include <asm/x86_init.h>
 
 int gart_iommu_aperture;
@@ -307,7 +307,7 @@
 				continue;
 
 			ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL);
-			aper_enabled = ctl & AMD64_GARTEN;
+			aper_enabled = ctl & GARTEN;
 			aper_order = (ctl >> 1) & 7;
 			aper_size = (32 * 1024 * 1024) << aper_order;
 			aper_base = read_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE) & 0x7fff;
@@ -362,7 +362,7 @@
 				continue;
 
 			ctl = read_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL);
-			ctl &= ~AMD64_GARTEN;
+			ctl &= ~GARTEN;
 			write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl);
 		}
 	}
@@ -371,7 +371,7 @@
 
 static int __initdata printed_gart_size_msg;
 
-void __init gart_iommu_hole_init(void)
+int __init gart_iommu_hole_init(void)
 {
 	u32 agp_aper_base = 0, agp_aper_order = 0;
 	u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0;
@@ -381,7 +381,7 @@
 
 	if (gart_iommu_aperture_disabled || !fix_aperture ||
 	    !early_pci_allowed())
-		return;
+		return -ENODEV;
 
 	printk(KERN_INFO  "Checking aperture...\n");
 
@@ -463,8 +463,9 @@
 			unsigned long n = (32 * 1024 * 1024) << last_aper_order;
 
 			insert_aperture_resource((u32)last_aper_base, n);
+			return 1;
 		}
-		return;
+		return 0;
 	}
 
 	if (!fallback_aper_force) {
@@ -500,13 +501,18 @@
 			panic("Not enough memory for aperture");
 		}
 	} else {
-		return;
+		return 0;
 	}
 
 	/* Fix up the north bridges */
 	for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) {
-		int bus;
-		int dev_base, dev_limit;
+		int bus, dev_base, dev_limit;
+
+		/*
+		 * Don't enable translation yet but enable GART IO and CPU
+		 * accesses and set DISTLBWALKPRB since GART table memory is UC.
+		 */
+		u32 ctl = DISTLBWALKPRB | aper_order << 1;
 
 		bus = bus_dev_ranges[i].bus;
 		dev_base = bus_dev_ranges[i].dev_base;
@@ -515,13 +521,12 @@
 			if (!early_is_k8_nb(read_pci_config(bus, slot, 3, 0x00)))
 				continue;
 
-			/* Don't enable translation yet. That is done later.
-			   Assume this BIOS didn't initialise the GART so
-			   just overwrite all previous bits */
-			write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, aper_order << 1);
+			write_pci_config(bus, slot, 3, AMD64_GARTAPERTURECTL, ctl);
 			write_pci_config(bus, slot, 3, AMD64_GARTAPERTUREBASE, aper_alloc >> 25);
 		}
 	}
 
 	set_up_gart_resume(aper_order, aper_alloc);
+
+	return 1;
 }
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index e3b534c..850657d 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -52,6 +52,7 @@
 #include <asm/mce.h>
 #include <asm/kvm_para.h>
 #include <asm/tsc.h>
+#include <asm/atomic.h>
 
 unsigned int num_processors;
 
@@ -370,38 +371,87 @@
 }
 
 /*
- * Setup extended LVT, AMD specific (K8, family 10h)
+ * Setup extended LVT, AMD specific
  *
- * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
- * MCE interrupts are supported. Thus MCE offset must be set to 0.
+ * Software should use the LVT offsets the BIOS provides.  The offsets
+ * are determined by the subsystems using it like those for MCE
+ * threshold or IBS.  On K8 only offset 0 (APIC500) and MCE interrupts
+ * are supported. Beginning with family 10h at least 4 offsets are
+ * available.
  *
+ * Since the offsets must be consistent for all cores, we keep track
+ * of the LVT offsets in software and reserve the offset for the same
+ * vector also to be used on other cores. An offset is freed by
+ * setting the entry to APIC_EILVT_MASKED.
+ *
+ * If the BIOS is right, there should be no conflicts. Otherwise a
+ * "[Firmware Bug]: ..." error message is generated. However, if
+ * software does not properly determines the offsets, it is not
+ * necessarily a BIOS bug.
+ */
+
+static atomic_t eilvt_offsets[APIC_EILVT_NR_MAX];
+
+static inline int eilvt_entry_is_changeable(unsigned int old, unsigned int new)
+{
+	return (old & APIC_EILVT_MASKED)
+		|| (new == APIC_EILVT_MASKED)
+		|| ((new & ~APIC_EILVT_MASKED) == old);
+}
+
+static unsigned int reserve_eilvt_offset(int offset, unsigned int new)
+{
+	unsigned int rsvd;			/* 0: uninitialized */
+
+	if (offset >= APIC_EILVT_NR_MAX)
+		return ~0;
+
+	rsvd = atomic_read(&eilvt_offsets[offset]) & ~APIC_EILVT_MASKED;
+	do {
+		if (rsvd &&
+		    !eilvt_entry_is_changeable(rsvd, new))
+			/* may not change if vectors are different */
+			return rsvd;
+		rsvd = atomic_cmpxchg(&eilvt_offsets[offset], rsvd, new);
+	} while (rsvd != new);
+
+	return new;
+}
+
+/*
  * If mask=1, the LVT entry does not generate interrupts while mask=0
  * enables the vector. See also the BKDGs.
  */
 
-#define APIC_EILVT_LVTOFF_MCE 0
-#define APIC_EILVT_LVTOFF_IBS 1
-
-static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
+int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask)
 {
-	unsigned long reg = (lvt_off << 4) + APIC_EILVTn(0);
-	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
+	unsigned long reg = APIC_EILVTn(offset);
+	unsigned int new, old, reserved;
 
-	apic_write(reg, v);
-}
+	new = (mask << 16) | (msg_type << 8) | vector;
+	old = apic_read(reg);
+	reserved = reserve_eilvt_offset(offset, new);
 
-u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_MCE;
-}
+	if (reserved != new) {
+		pr_err(FW_BUG "cpu %d, try to setup vector 0x%x, but "
+		       "vector 0x%x was already reserved by another core, "
+		       "APIC%lX=0x%x\n",
+		       smp_processor_id(), new, reserved, reg, old);
+		return -EINVAL;
+	}
 
-u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_IBS;
+	if (!eilvt_entry_is_changeable(old, new)) {
+		pr_err(FW_BUG "cpu %d, try to setup vector 0x%x but "
+		       "register already in use, APIC%lX=0x%x\n",
+		       smp_processor_id(), new, reg, old);
+		return -EBUSY;
+	}
+
+	apic_write(reg, new);
+
+	return 0;
 }
-EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs);
+EXPORT_SYMBOL_GPL(setup_APIC_eilvt);
 
 /*
  * Program the next event, relative to now
@@ -1665,10 +1715,7 @@
 	}
 #endif
 
-#ifndef CONFIG_SMP
-	enable_IR_x2apic();
 	default_setup_apic_routing();
-#endif
 
 	verify_local_APIC();
 	connect_bsp_APIC();
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 5c5b8f3..8ae808d 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -131,13 +131,9 @@
 	struct irq_pin_list *next;
 };
 
-static struct irq_pin_list *get_one_free_irq_2_pin(int node)
+static struct irq_pin_list *alloc_irq_pin_list(int node)
 {
-	struct irq_pin_list *pin;
-
-	pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node);
-
-	return pin;
+	return kzalloc_node(sizeof(struct irq_pin_list), GFP_KERNEL, node);
 }
 
 /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
@@ -150,10 +146,7 @@
 int __init arch_early_irq_init(void)
 {
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
-	int count;
-	int node;
-	int i;
+	int count, node, i;
 
 	if (!legacy_pic->nr_legacy_irqs) {
 		nr_irqs_gsi = 0;
@@ -162,13 +155,15 @@
 
 	cfg = irq_cfgx;
 	count = ARRAY_SIZE(irq_cfgx);
-	node= cpu_to_node(boot_cpu_id);
+	node = cpu_to_node(0);
+
+	/* Make sure the legacy interrupts are marked in the bitmap */
+	irq_reserve_irqs(0, legacy_pic->nr_legacy_irqs);
 
 	for (i = 0; i < count; i++) {
-		desc = irq_to_desc(i);
-		desc->chip_data = &cfg[i];
-		zalloc_cpumask_var_node(&cfg[i].domain, GFP_NOWAIT, node);
-		zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_NOWAIT, node);
+		set_irq_chip_data(i, &cfg[i]);
+		zalloc_cpumask_var_node(&cfg[i].domain, GFP_KERNEL, node);
+		zalloc_cpumask_var_node(&cfg[i].old_domain, GFP_KERNEL, node);
 		/*
 		 * For legacy IRQ's, start with assigning irq0 to irq15 to
 		 * IRQ0_VECTOR to IRQ15_VECTOR on cpu 0.
@@ -183,170 +178,88 @@
 }
 
 #ifdef CONFIG_SPARSE_IRQ
-struct irq_cfg *irq_cfg(unsigned int irq)
+static struct irq_cfg *irq_cfg(unsigned int irq)
 {
-	struct irq_cfg *cfg = NULL;
-	struct irq_desc *desc;
+	return get_irq_chip_data(irq);
+}
 
-	desc = irq_to_desc(irq);
-	if (desc)
-		cfg = desc->chip_data;
+static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node)
+{
+	struct irq_cfg *cfg;
 
+	cfg = kzalloc_node(sizeof(*cfg), GFP_KERNEL, node);
+	if (!cfg)
+		return NULL;
+	if (!zalloc_cpumask_var_node(&cfg->domain, GFP_KERNEL, node))
+		goto out_cfg;
+	if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_KERNEL, node))
+		goto out_domain;
 	return cfg;
+out_domain:
+	free_cpumask_var(cfg->domain);
+out_cfg:
+	kfree(cfg);
+	return NULL;
 }
 
-static struct irq_cfg *get_one_free_irq_cfg(int node)
+static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg)
 {
-	struct irq_cfg *cfg;
-
-	cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node);
-	if (cfg) {
-		if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) {
-			kfree(cfg);
-			cfg = NULL;
-		} else if (!zalloc_cpumask_var_node(&cfg->old_domain,
-							  GFP_ATOMIC, node)) {
-			free_cpumask_var(cfg->domain);
-			kfree(cfg);
-			cfg = NULL;
-		}
-	}
-
-	return cfg;
-}
-
-int arch_init_chip_data(struct irq_desc *desc, int node)
-{
-	struct irq_cfg *cfg;
-
-	cfg = desc->chip_data;
-	if (!cfg) {
-		desc->chip_data = get_one_free_irq_cfg(node);
-		if (!desc->chip_data) {
-			printk(KERN_ERR "can not alloc irq_cfg\n");
-			BUG_ON(1);
-		}
-	}
-
-	return 0;
-}
-
-/* for move_irq_desc */
-static void
-init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int node)
-{
-	struct irq_pin_list *old_entry, *head, *tail, *entry;
-
-	cfg->irq_2_pin = NULL;
-	old_entry = old_cfg->irq_2_pin;
-	if (!old_entry)
-		return;
-
-	entry = get_one_free_irq_2_pin(node);
-	if (!entry)
-		return;
-
-	entry->apic	= old_entry->apic;
-	entry->pin	= old_entry->pin;
-	head		= entry;
-	tail		= entry;
-	old_entry	= old_entry->next;
-	while (old_entry) {
-		entry = get_one_free_irq_2_pin(node);
-		if (!entry) {
-			entry = head;
-			while (entry) {
-				head = entry->next;
-				kfree(entry);
-				entry = head;
-			}
-			/* still use the old one */
-			return;
-		}
-		entry->apic	= old_entry->apic;
-		entry->pin	= old_entry->pin;
-		tail->next	= entry;
-		tail		= entry;
-		old_entry	= old_entry->next;
-	}
-
-	tail->next = NULL;
-	cfg->irq_2_pin = head;
-}
-
-static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg)
-{
-	struct irq_pin_list *entry, *next;
-
-	if (old_cfg->irq_2_pin == cfg->irq_2_pin)
-		return;
-
-	entry = old_cfg->irq_2_pin;
-
-	while (entry) {
-		next = entry->next;
-		kfree(entry);
-		entry = next;
-	}
-	old_cfg->irq_2_pin = NULL;
-}
-
-void arch_init_copy_chip_data(struct irq_desc *old_desc,
-				 struct irq_desc *desc, int node)
-{
-	struct irq_cfg *cfg;
-	struct irq_cfg *old_cfg;
-
-	cfg = get_one_free_irq_cfg(node);
-
 	if (!cfg)
 		return;
-
-	desc->chip_data = cfg;
-
-	old_cfg = old_desc->chip_data;
-
-	cfg->vector = old_cfg->vector;
-	cfg->move_in_progress = old_cfg->move_in_progress;
-	cpumask_copy(cfg->domain, old_cfg->domain);
-	cpumask_copy(cfg->old_domain, old_cfg->old_domain);
-
-	init_copy_irq_2_pin(old_cfg, cfg, node);
-}
-
-static void free_irq_cfg(struct irq_cfg *cfg)
-{
+	set_irq_chip_data(at, NULL);
 	free_cpumask_var(cfg->domain);
 	free_cpumask_var(cfg->old_domain);
 	kfree(cfg);
 }
 
-void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc)
-{
-	struct irq_cfg *old_cfg, *cfg;
-
-	old_cfg = old_desc->chip_data;
-	cfg = desc->chip_data;
-
-	if (old_cfg == cfg)
-		return;
-
-	if (old_cfg) {
-		free_irq_2_pin(old_cfg, cfg);
-		free_irq_cfg(old_cfg);
-		old_desc->chip_data = NULL;
-	}
-}
-/* end for move_irq_desc */
-
 #else
+
 struct irq_cfg *irq_cfg(unsigned int irq)
 {
 	return irq < nr_irqs ? irq_cfgx + irq : NULL;
 }
 
+static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node)
+{
+	return irq_cfgx + irq;
+}
+
+static inline void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { }
+
 #endif
 
+static struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node)
+{
+	int res = irq_alloc_desc_at(at, node);
+	struct irq_cfg *cfg;
+
+	if (res < 0) {
+		if (res != -EEXIST)
+			return NULL;
+		cfg = get_irq_chip_data(at);
+		if (cfg)
+			return cfg;
+	}
+
+	cfg = alloc_irq_cfg(at, node);
+	if (cfg)
+		set_irq_chip_data(at, cfg);
+	else
+		irq_free_desc(at);
+	return cfg;
+}
+
+static int alloc_irq_from(unsigned int from, int node)
+{
+	return irq_alloc_desc_from(from, node);
+}
+
+static void free_irq_at(unsigned int at, struct irq_cfg *cfg)
+{
+	free_irq_cfg(at, cfg);
+	irq_free_desc(at);
+}
+
 struct io_apic {
 	unsigned int index;
 	unsigned int unused[3];
@@ -451,7 +364,7 @@
 	io_apic_write(apic, 0x10 + 2*pin, eu.w1);
 }
 
-void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
+static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
 {
 	unsigned long flags;
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
@@ -481,7 +394,7 @@
  * fast in the common case, and fast for shared ISA-space IRQs.
  */
 static int
-add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin)
+__add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
 {
 	struct irq_pin_list **last, *entry;
 
@@ -493,7 +406,7 @@
 		last = &entry->next;
 	}
 
-	entry = get_one_free_irq_2_pin(node);
+	entry = alloc_irq_pin_list(node);
 	if (!entry) {
 		printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n",
 				node, apic, pin);
@@ -508,7 +421,7 @@
 
 static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
 {
-	if (add_pin_to_irq_node_nopanic(cfg, node, apic, pin))
+	if (__add_pin_to_irq_node(cfg, node, apic, pin))
 		panic("IO-APIC: failed to add irq-pin. Can not proceed\n");
 }
 
@@ -571,11 +484,6 @@
 			     IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
 }
 
-static void __unmask_IO_APIC_irq(struct irq_cfg *cfg)
-{
-	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
-}
-
 static void io_apic_sync(struct irq_pin_list *entry)
 {
 	/*
@@ -587,44 +495,37 @@
 	readl(&io_apic->data);
 }
 
-static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
+static void mask_ioapic(struct irq_cfg *cfg)
 {
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&ioapic_lock, flags);
 	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
-}
-
-static void mask_IO_APIC_irq_desc(struct irq_desc *desc)
-{
-	struct irq_cfg *cfg = desc->chip_data;
-	unsigned long flags;
-
-	BUG_ON(!cfg);
-
-	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	__mask_IO_APIC_irq(cfg);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void unmask_IO_APIC_irq_desc(struct irq_desc *desc)
+static void mask_ioapic_irq(struct irq_data *data)
 {
-	struct irq_cfg *cfg = desc->chip_data;
+	mask_ioapic(data->chip_data);
+}
+
+static void __unmask_ioapic(struct irq_cfg *cfg)
+{
+	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
+}
+
+static void unmask_ioapic(struct irq_cfg *cfg)
+{
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	__unmask_IO_APIC_irq(cfg);
+	__unmask_ioapic(cfg);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void mask_IO_APIC_irq(unsigned int irq)
+static void unmask_ioapic_irq(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	mask_IO_APIC_irq_desc(desc);
-}
-static void unmask_IO_APIC_irq(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	unmask_IO_APIC_irq_desc(desc);
+	unmask_ioapic(data->chip_data);
 }
 
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
@@ -694,14 +595,14 @@
 	struct IO_APIC_route_entry **ioapic_entries;
 
 	ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics,
-				GFP_ATOMIC);
+				GFP_KERNEL);
 	if (!ioapic_entries)
 		return 0;
 
 	for (apic = 0; apic < nr_ioapics; apic++) {
 		ioapic_entries[apic] =
 			kzalloc(sizeof(struct IO_APIC_route_entry) *
-				nr_ioapic_registers[apic], GFP_ATOMIC);
+				nr_ioapic_registers[apic], GFP_KERNEL);
 		if (!ioapic_entries[apic])
 			goto nomem;
 	}
@@ -1259,7 +1160,6 @@
 	/* Initialize vector_irq on a new cpu */
 	int irq, vector;
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
 
 	/*
 	 * vector_lock will make sure that we don't run into irq vector
@@ -1268,9 +1168,10 @@
 	 */
 	raw_spin_lock(&vector_lock);
 	/* Mark the inuse vectors */
-	for_each_irq_desc(irq, desc) {
-		cfg = desc->chip_data;
-
+	for_each_active_irq(irq) {
+		cfg = get_irq_chip_data(irq);
+		if (!cfg)
+			continue;
 		/*
 		 * If it is a legacy IRQ handled by the legacy PIC, this cpu
 		 * will be part of the irq_cfg's domain.
@@ -1327,17 +1228,17 @@
 }
 #endif
 
-static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger)
+static void ioapic_register_intr(unsigned int irq, unsigned long trigger)
 {
 
 	if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
 	    trigger == IOAPIC_LEVEL)
-		desc->status |= IRQ_LEVEL;
+		irq_set_status_flags(irq, IRQ_LEVEL);
 	else
-		desc->status &= ~IRQ_LEVEL;
+		irq_clear_status_flags(irq, IRQ_LEVEL);
 
-	if (irq_remapped(irq)) {
-		desc->status |= IRQ_MOVE_PCNTXT;
+	if (irq_remapped(get_irq_chip_data(irq))) {
+		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 		if (trigger)
 			set_irq_chip_and_handler_name(irq, &ir_ioapic_chip,
 						      handle_fasteoi_irq,
@@ -1358,10 +1259,10 @@
 					      handle_edge_irq, "edge");
 }
 
-int setup_ioapic_entry(int apic_id, int irq,
-		       struct IO_APIC_route_entry *entry,
-		       unsigned int destination, int trigger,
-		       int polarity, int vector, int pin)
+static int setup_ioapic_entry(int apic_id, int irq,
+			      struct IO_APIC_route_entry *entry,
+			      unsigned int destination, int trigger,
+			      int polarity, int vector, int pin)
 {
 	/*
 	 * add it to the IO-APIC irq-routing table:
@@ -1382,21 +1283,7 @@
 		if (index < 0)
 			panic("Failed to allocate IRTE for ioapic %d\n", apic_id);
 
-		memset(&irte, 0, sizeof(irte));
-
-		irte.present = 1;
-		irte.dst_mode = apic->irq_dest_mode;
-		/*
-		 * Trigger mode in the IRTE will always be edge, and the
-		 * actual level or edge trigger will be setup in the IO-APIC
-		 * RTE. This will help simplify level triggered irq migration.
-		 * For more details, see the comments above explainig IO-APIC
-		 * irq migration in the presence of interrupt-remapping.
-		 */
-		irte.trigger_mode = 0;
-		irte.dlvry_mode = apic->irq_delivery_mode;
-		irte.vector = vector;
-		irte.dest_id = IRTE_DEST(destination);
+		prepare_irte(&irte, vector, destination);
 
 		/* Set source-id of interrupt request */
 		set_ioapic_sid(&irte, apic_id);
@@ -1431,18 +1318,14 @@
 	return 0;
 }
 
-static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq_desc *desc,
-			      int trigger, int polarity)
+static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq,
+			     struct irq_cfg *cfg, int trigger, int polarity)
 {
-	struct irq_cfg *cfg;
 	struct IO_APIC_route_entry entry;
 	unsigned int dest;
 
 	if (!IO_APIC_IRQ(irq))
 		return;
-
-	cfg = desc->chip_data;
-
 	/*
 	 * For legacy irqs, cfg->domain starts with cpu 0 for legacy
 	 * controllers like 8259. Now that IO-APIC can handle this irq, update
@@ -1471,9 +1354,9 @@
 		return;
 	}
 
-	ioapic_register_intr(irq, desc, trigger);
+	ioapic_register_intr(irq, trigger);
 	if (irq < legacy_pic->nr_legacy_irqs)
-		legacy_pic->chip->mask(irq);
+		legacy_pic->mask(irq);
 
 	ioapic_write_entry(apic_id, pin, entry);
 }
@@ -1484,11 +1367,9 @@
 
 static void __init setup_IO_APIC_irqs(void)
 {
-	int apic_id, pin, idx, irq;
-	int notcon = 0;
-	struct irq_desc *desc;
+	int apic_id, pin, idx, irq, notcon = 0;
+	int node = cpu_to_node(0);
 	struct irq_cfg *cfg;
-	int node = cpu_to_node(boot_cpu_id);
 
 	apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
 
@@ -1525,19 +1406,17 @@
 				apic->multi_timer_check(apic_id, irq))
 			continue;
 
-		desc = irq_to_desc_alloc_node(irq, node);
-		if (!desc) {
-			printk(KERN_INFO "can not get irq_desc for %d\n", irq);
+		cfg = alloc_irq_and_cfg_at(irq, node);
+		if (!cfg)
 			continue;
-		}
-		cfg = desc->chip_data;
+
 		add_pin_to_irq_node(cfg, node, apic_id, pin);
 		/*
 		 * don't mark it in pin_programmed, so later acpi could
 		 * set it correctly when irq < 16
 		 */
-		setup_IO_APIC_irq(apic_id, pin, irq, desc,
-				irq_trigger(idx), irq_polarity(idx));
+		setup_ioapic_irq(apic_id, pin, irq, cfg, irq_trigger(idx),
+				  irq_polarity(idx));
 	}
 
 	if (notcon)
@@ -1552,9 +1431,7 @@
  */
 void setup_IO_APIC_irq_extra(u32 gsi)
 {
-	int apic_id = 0, pin, idx, irq;
-	int node = cpu_to_node(boot_cpu_id);
-	struct irq_desc *desc;
+	int apic_id = 0, pin, idx, irq, node = cpu_to_node(0);
 	struct irq_cfg *cfg;
 
 	/*
@@ -1570,18 +1447,15 @@
 		return;
 
 	irq = pin_2_irq(idx, apic_id, pin);
-#ifdef CONFIG_SPARSE_IRQ
-	desc = irq_to_desc(irq);
-	if (desc)
-		return;
-#endif
-	desc = irq_to_desc_alloc_node(irq, node);
-	if (!desc) {
-		printk(KERN_INFO "can not get irq_desc for %d\n", irq);
-		return;
-	}
 
-	cfg = desc->chip_data;
+	/* Only handle the non legacy irqs on secondary ioapics */
+	if (apic_id == 0 || irq < NR_IRQS_LEGACY)
+		return;
+
+	cfg = alloc_irq_and_cfg_at(irq, node);
+	if (!cfg)
+		return;
+
 	add_pin_to_irq_node(cfg, node, apic_id, pin);
 
 	if (test_bit(pin, mp_ioapic_routing[apic_id].pin_programmed)) {
@@ -1591,7 +1465,7 @@
 	}
 	set_bit(pin, mp_ioapic_routing[apic_id].pin_programmed);
 
-	setup_IO_APIC_irq(apic_id, pin, irq, desc,
+	setup_ioapic_irq(apic_id, pin, irq, cfg,
 			irq_trigger(idx), irq_polarity(idx));
 }
 
@@ -1642,7 +1516,6 @@
 	union IO_APIC_reg_03 reg_03;
 	unsigned long flags;
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
 	unsigned int irq;
 
 	printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
@@ -1729,10 +1602,10 @@
 	}
 	}
 	printk(KERN_DEBUG "IRQ to pin mappings:\n");
-	for_each_irq_desc(irq, desc) {
+	for_each_active_irq(irq) {
 		struct irq_pin_list *entry;
 
-		cfg = desc->chip_data;
+		cfg = get_irq_chip_data(irq);
 		if (!cfg)
 			continue;
 		entry = cfg->irq_2_pin;
@@ -2239,29 +2112,26 @@
  * an edge even if it isn't on the 8259A...
  */
 
-static unsigned int startup_ioapic_irq(unsigned int irq)
+static unsigned int startup_ioapic_irq(struct irq_data *data)
 {
-	int was_pending = 0;
+	int was_pending = 0, irq = data->irq;
 	unsigned long flags;
-	struct irq_cfg *cfg;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
 	if (irq < legacy_pic->nr_legacy_irqs) {
-		legacy_pic->chip->mask(irq);
+		legacy_pic->mask(irq);
 		if (legacy_pic->irq_pending(irq))
 			was_pending = 1;
 	}
-	cfg = irq_cfg(irq);
-	__unmask_IO_APIC_irq(cfg);
+	__unmask_ioapic(data->chip_data);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 
 	return was_pending;
 }
 
-static int ioapic_retrigger_irq(unsigned int irq)
+static int ioapic_retrigger_irq(struct irq_data *data)
 {
-
-	struct irq_cfg *cfg = irq_cfg(irq);
+	struct irq_cfg *cfg = data->chip_data;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&vector_lock, flags);
@@ -2312,7 +2182,7 @@
 		 * With interrupt-remapping, destination information comes
 		 * from interrupt-remapping table entry.
 		 */
-		if (!irq_remapped(irq))
+		if (!irq_remapped(cfg))
 			io_apic_write(apic, 0x11 + pin*2, dest);
 		reg = io_apic_read(apic, 0x10 + pin*2);
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
@@ -2322,65 +2192,46 @@
 }
 
 /*
- * Either sets desc->affinity to a valid value, and returns
+ * Either sets data->affinity to a valid value, and returns
  * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and
- * leaves desc->affinity untouched.
+ * leaves data->affinity untouched.
  */
-unsigned int
-set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask,
-		  unsigned int *dest_id)
+int __ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
+			  unsigned int *dest_id)
 {
-	struct irq_cfg *cfg;
-	unsigned int irq;
+	struct irq_cfg *cfg = data->chip_data;
 
 	if (!cpumask_intersects(mask, cpu_online_mask))
 		return -1;
 
-	irq = desc->irq;
-	cfg = desc->chip_data;
-	if (assign_irq_vector(irq, cfg, mask))
+	if (assign_irq_vector(data->irq, data->chip_data, mask))
 		return -1;
 
-	cpumask_copy(desc->affinity, mask);
+	cpumask_copy(data->affinity, mask);
 
-	*dest_id = apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain);
+	*dest_id = apic->cpu_mask_to_apicid_and(mask, cfg->domain);
 	return 0;
 }
 
 static int
-set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
+ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		    bool force)
 {
-	struct irq_cfg *cfg;
+	unsigned int dest, irq = data->irq;
 	unsigned long flags;
-	unsigned int dest;
-	unsigned int irq;
-	int ret = -1;
-
-	irq = desc->irq;
-	cfg = desc->chip_data;
+	int ret;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	ret = set_desc_affinity(desc, mask, &dest);
+	ret = __ioapic_set_affinity(data, mask, &dest);
 	if (!ret) {
 		/* Only the high 8 bits are valid. */
 		dest = SET_APIC_LOGICAL_ID(dest);
-		__target_IO_APIC_irq(irq, dest, cfg);
+		__target_IO_APIC_irq(irq, dest, data->chip_data);
 	}
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
-
 	return ret;
 }
 
-static int
-set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-
-	return set_ioapic_affinity_irq_desc(desc, mask);
-}
-
 #ifdef CONFIG_INTR_REMAP
 
 /*
@@ -2395,24 +2246,21 @@
  * the interrupt-remapping table entry.
  */
 static int
-migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
+ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		       bool force)
 {
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = data->chip_data;
+	unsigned int dest, irq = data->irq;
 	struct irte irte;
-	unsigned int dest;
-	unsigned int irq;
-	int ret = -1;
 
 	if (!cpumask_intersects(mask, cpu_online_mask))
-		return ret;
+		return -EINVAL;
 
-	irq = desc->irq;
 	if (get_irte(irq, &irte))
-		return ret;
+		return -EBUSY;
 
-	cfg = desc->chip_data;
 	if (assign_irq_vector(irq, cfg, mask))
-		return ret;
+		return -EBUSY;
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask);
 
@@ -2427,29 +2275,14 @@
 	if (cfg->move_in_progress)
 		send_cleanup_vector(cfg);
 
-	cpumask_copy(desc->affinity, mask);
-
+	cpumask_copy(data->affinity, mask);
 	return 0;
 }
 
-/*
- * Migrates the IRQ destination in the process context.
- */
-static int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
-					    const struct cpumask *mask)
-{
-	return migrate_ioapic_irq_desc(desc, mask);
-}
-static int set_ir_ioapic_affinity_irq(unsigned int irq,
-				       const struct cpumask *mask)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	return set_ir_ioapic_affinity_irq_desc(desc, mask);
-}
 #else
-static inline int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
-						   const struct cpumask *mask)
+static inline int
+ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		       bool force)
 {
 	return 0;
 }
@@ -2511,10 +2344,8 @@
 	irq_exit();
 }
 
-static void __irq_complete_move(struct irq_desc **descp, unsigned vector)
+static void __irq_complete_move(struct irq_cfg *cfg, unsigned vector)
 {
-	struct irq_desc *desc = *descp;
-	struct irq_cfg *cfg = desc->chip_data;
 	unsigned me;
 
 	if (likely(!cfg->move_in_progress))
@@ -2526,31 +2357,28 @@
 		send_cleanup_vector(cfg);
 }
 
-static void irq_complete_move(struct irq_desc **descp)
+static void irq_complete_move(struct irq_cfg *cfg)
 {
-	__irq_complete_move(descp, ~get_irq_regs()->orig_ax);
+	__irq_complete_move(cfg, ~get_irq_regs()->orig_ax);
 }
 
 void irq_force_complete_move(int irq)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg = desc->chip_data;
+	struct irq_cfg *cfg = get_irq_chip_data(irq);
 
 	if (!cfg)
 		return;
 
-	__irq_complete_move(&desc, cfg->vector);
+	__irq_complete_move(cfg, cfg->vector);
 }
 #else
-static inline void irq_complete_move(struct irq_desc **descp) {}
+static inline void irq_complete_move(struct irq_cfg *cfg) { }
 #endif
 
-static void ack_apic_edge(unsigned int irq)
+static void ack_apic_edge(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	irq_complete_move(&desc);
-	move_native_irq(irq);
+	irq_complete_move(data->chip_data);
+	move_native_irq(data->irq);
 	ack_APIC_irq();
 }
 
@@ -2572,10 +2400,12 @@
  * Otherwise, we simulate the EOI message manually by changing the trigger
  * mode to edge and then back to level, with RTE being masked during this.
 */
-static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
+static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
 {
 	struct irq_pin_list *entry;
+	unsigned long flags;
 
+	raw_spin_lock_irqsave(&ioapic_lock, flags);
 	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		if (mp_ioapics[entry->apic].apicver >= 0x20) {
 			/*
@@ -2584,7 +2414,7 @@
 			 * intr-remapping table entry. Hence for the io-apic
 			 * EOI we use the pin number.
 			 */
-			if (irq_remapped(irq))
+			if (irq_remapped(cfg))
 				io_apic_eoi(entry->apic, entry->pin);
 			else
 				io_apic_eoi(entry->apic, cfg->vector);
@@ -2593,36 +2423,22 @@
 			__unmask_and_level_IO_APIC_irq(entry);
 		}
 	}
-}
-
-static void eoi_ioapic_irq(struct irq_desc *desc)
-{
-	struct irq_cfg *cfg;
-	unsigned long flags;
-	unsigned int irq;
-
-	irq = desc->irq;
-	cfg = desc->chip_data;
-
-	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	__eoi_ioapic_irq(irq, cfg);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void ack_apic_level(unsigned int irq)
+static void ack_apic_level(struct irq_data *data)
 {
+	struct irq_cfg *cfg = data->chip_data;
+	int i, do_unmask_irq = 0, irq = data->irq;
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long v;
-	int i;
-	struct irq_cfg *cfg;
-	int do_unmask_irq = 0;
 
-	irq_complete_move(&desc);
+	irq_complete_move(cfg);
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	/* If we are moving the irq we need to mask it */
 	if (unlikely(desc->status & IRQ_MOVE_PENDING)) {
 		do_unmask_irq = 1;
-		mask_IO_APIC_irq_desc(desc);
+		mask_ioapic(cfg);
 	}
 #endif
 
@@ -2658,7 +2474,6 @@
 	 * we use the above logic (mask+edge followed by unmask+level) from
 	 * Manfred Spraul to clear the remote IRR.
 	 */
-	cfg = desc->chip_data;
 	i = cfg->vector;
 	v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
 
@@ -2678,7 +2493,7 @@
 	if (!(v & (1 << (i & 0x1f)))) {
 		atomic_inc(&irq_mis_count);
 
-		eoi_ioapic_irq(desc);
+		eoi_ioapic_irq(irq, cfg);
 	}
 
 	/* Now we can move and renable the irq */
@@ -2709,61 +2524,57 @@
 		 * accurate and is causing problems then it is a hardware bug
 		 * and you can go talk to the chipset vendor about it.
 		 */
-		cfg = desc->chip_data;
 		if (!io_apic_level_ack_pending(cfg))
 			move_masked_irq(irq);
-		unmask_IO_APIC_irq_desc(desc);
+		unmask_ioapic(cfg);
 	}
 }
 
 #ifdef CONFIG_INTR_REMAP
-static void ir_ack_apic_edge(unsigned int irq)
+static void ir_ack_apic_edge(struct irq_data *data)
 {
 	ack_APIC_irq();
 }
 
-static void ir_ack_apic_level(unsigned int irq)
+static void ir_ack_apic_level(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-
 	ack_APIC_irq();
-	eoi_ioapic_irq(desc);
+	eoi_ioapic_irq(data->irq, data->chip_data);
 }
 #endif /* CONFIG_INTR_REMAP */
 
 static struct irq_chip ioapic_chip __read_mostly = {
-	.name		= "IO-APIC",
-	.startup	= startup_ioapic_irq,
-	.mask		= mask_IO_APIC_irq,
-	.unmask		= unmask_IO_APIC_irq,
-	.ack		= ack_apic_edge,
-	.eoi		= ack_apic_level,
+	.name			= "IO-APIC",
+	.irq_startup		= startup_ioapic_irq,
+	.irq_mask		= mask_ioapic_irq,
+	.irq_unmask		= unmask_ioapic_irq,
+	.irq_ack		= ack_apic_edge,
+	.irq_eoi		= ack_apic_level,
 #ifdef CONFIG_SMP
-	.set_affinity	= set_ioapic_affinity_irq,
+	.irq_set_affinity	= ioapic_set_affinity,
 #endif
-	.retrigger	= ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 static struct irq_chip ir_ioapic_chip __read_mostly = {
-	.name		= "IR-IO-APIC",
-	.startup	= startup_ioapic_irq,
-	.mask		= mask_IO_APIC_irq,
-	.unmask		= unmask_IO_APIC_irq,
+	.name			= "IR-IO-APIC",
+	.irq_startup		= startup_ioapic_irq,
+	.irq_mask		= mask_ioapic_irq,
+	.irq_unmask		= unmask_ioapic_irq,
 #ifdef CONFIG_INTR_REMAP
-	.ack		= ir_ack_apic_edge,
-	.eoi		= ir_ack_apic_level,
+	.irq_ack		= ir_ack_apic_edge,
+	.irq_eoi		= ir_ack_apic_level,
 #ifdef CONFIG_SMP
-	.set_affinity	= set_ir_ioapic_affinity_irq,
+	.irq_set_affinity	= ir_ioapic_set_affinity,
 #endif
 #endif
-	.retrigger	= ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 static inline void init_IO_APIC_traps(void)
 {
-	int irq;
-	struct irq_desc *desc;
 	struct irq_cfg *cfg;
+	unsigned int irq;
 
 	/*
 	 * NOTE! The local APIC isn't very good at handling
@@ -2776,8 +2587,8 @@
 	 * Also, we've got to be careful not to trash gate
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
-	for_each_irq_desc(irq, desc) {
-		cfg = desc->chip_data;
+	for_each_active_irq(irq) {
+		cfg = get_irq_chip_data(irq);
 		if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
 			/*
 			 * Hmm.. We don't have an entry for this,
@@ -2788,7 +2599,7 @@
 				legacy_pic->make_irq(irq);
 			else
 				/* Strange. Oh, well.. */
-				desc->chip = &no_irq_chip;
+				set_irq_chip(irq, &no_irq_chip);
 		}
 	}
 }
@@ -2797,7 +2608,7 @@
  * The local APIC irq-chip implementation:
  */
 
-static void mask_lapic_irq(unsigned int irq)
+static void mask_lapic_irq(struct irq_data *data)
 {
 	unsigned long v;
 
@@ -2805,7 +2616,7 @@
 	apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
 }
 
-static void unmask_lapic_irq(unsigned int irq)
+static void unmask_lapic_irq(struct irq_data *data)
 {
 	unsigned long v;
 
@@ -2813,21 +2624,21 @@
 	apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
 }
 
-static void ack_lapic_irq(unsigned int irq)
+static void ack_lapic_irq(struct irq_data *data)
 {
 	ack_APIC_irq();
 }
 
 static struct irq_chip lapic_chip __read_mostly = {
 	.name		= "local-APIC",
-	.mask		= mask_lapic_irq,
-	.unmask		= unmask_lapic_irq,
-	.ack		= ack_lapic_irq,
+	.irq_mask	= mask_lapic_irq,
+	.irq_unmask	= unmask_lapic_irq,
+	.irq_ack	= ack_lapic_irq,
 };
 
-static void lapic_register_intr(int irq, struct irq_desc *desc)
+static void lapic_register_intr(int irq)
 {
-	desc->status &= ~IRQ_LEVEL;
+	irq_clear_status_flags(irq, IRQ_LEVEL);
 	set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
 				      "edge");
 }
@@ -2930,9 +2741,8 @@
  */
 static inline void __init check_timer(void)
 {
-	struct irq_desc *desc = irq_to_desc(0);
-	struct irq_cfg *cfg = desc->chip_data;
-	int node = cpu_to_node(boot_cpu_id);
+	struct irq_cfg *cfg = get_irq_chip_data(0);
+	int node = cpu_to_node(0);
 	int apic1, pin1, apic2, pin2;
 	unsigned long flags;
 	int no_pin1 = 0;
@@ -2942,7 +2752,7 @@
 	/*
 	 * get/set the timer IRQ vector:
 	 */
-	legacy_pic->chip->mask(0);
+	legacy_pic->mask(0);
 	assign_irq_vector(0, cfg, apic->target_cpus());
 
 	/*
@@ -3001,7 +2811,7 @@
 			add_pin_to_irq_node(cfg, node, apic1, pin1);
 			setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
 		} else {
-			/* for edge trigger, setup_IO_APIC_irq already
+			/* for edge trigger, setup_ioapic_irq already
 			 * leave it unmasked.
 			 * so only need to unmask if it is level-trigger
 			 * do we really have level trigger timer?
@@ -3009,12 +2819,12 @@
 			int idx;
 			idx = find_irq_entry(apic1, pin1, mp_INT);
 			if (idx != -1 && irq_trigger(idx))
-				unmask_IO_APIC_irq_desc(desc);
+				unmask_ioapic(cfg);
 		}
 		if (timer_irq_works()) {
 			if (nmi_watchdog == NMI_IO_APIC) {
 				setup_nmi();
-				legacy_pic->chip->unmask(0);
+				legacy_pic->unmask(0);
 			}
 			if (disable_timer_pin_1 > 0)
 				clear_IO_APIC_pin(0, pin1);
@@ -3037,14 +2847,14 @@
 		 */
 		replace_pin_at_irq_node(cfg, node, apic1, pin1, apic2, pin2);
 		setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
-		legacy_pic->chip->unmask(0);
+		legacy_pic->unmask(0);
 		if (timer_irq_works()) {
 			apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
 			timer_through_8259 = 1;
 			if (nmi_watchdog == NMI_IO_APIC) {
-				legacy_pic->chip->mask(0);
+				legacy_pic->mask(0);
 				setup_nmi();
-				legacy_pic->chip->unmask(0);
+				legacy_pic->unmask(0);
 			}
 			goto out;
 		}
@@ -3052,7 +2862,7 @@
 		 * Cleanup, just in case ...
 		 */
 		local_irq_disable();
-		legacy_pic->chip->mask(0);
+		legacy_pic->mask(0);
 		clear_IO_APIC_pin(apic2, pin2);
 		apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n");
 	}
@@ -3069,16 +2879,16 @@
 	apic_printk(APIC_QUIET, KERN_INFO
 		    "...trying to set up timer as Virtual Wire IRQ...\n");
 
-	lapic_register_intr(0, desc);
+	lapic_register_intr(0);
 	apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector);	/* Fixed mode */
-	legacy_pic->chip->unmask(0);
+	legacy_pic->unmask(0);
 
 	if (timer_irq_works()) {
 		apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
 		goto out;
 	}
 	local_irq_disable();
-	legacy_pic->chip->mask(0);
+	legacy_pic->mask(0);
 	apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector);
 	apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n");
 
@@ -3244,49 +3054,42 @@
 /*
  * Dynamic irq allocate and deallocation
  */
-unsigned int create_irq_nr(unsigned int irq_want, int node)
+unsigned int create_irq_nr(unsigned int from, int node)
 {
-	/* Allocate an unused irq */
-	unsigned int irq;
-	unsigned int new;
+	struct irq_cfg *cfg;
 	unsigned long flags;
-	struct irq_cfg *cfg_new = NULL;
-	struct irq_desc *desc_new = NULL;
+	unsigned int ret = 0;
+	int irq;
 
-	irq = 0;
-	if (irq_want < nr_irqs_gsi)
-		irq_want = nr_irqs_gsi;
+	if (from < nr_irqs_gsi)
+		from = nr_irqs_gsi;
+
+	irq = alloc_irq_from(from, node);
+	if (irq < 0)
+		return 0;
+	cfg = alloc_irq_cfg(irq, node);
+	if (!cfg) {
+		free_irq_at(irq, NULL);
+		return 0;
+	}
 
 	raw_spin_lock_irqsave(&vector_lock, flags);
-	for (new = irq_want; new < nr_irqs; new++) {
-		desc_new = irq_to_desc_alloc_node(new, node);
-		if (!desc_new) {
-			printk(KERN_INFO "can not get irq_desc for %d\n", new);
-			continue;
-		}
-		cfg_new = desc_new->chip_data;
-
-		if (cfg_new->vector != 0)
-			continue;
-
-		desc_new = move_irq_desc(desc_new, node);
-		cfg_new = desc_new->chip_data;
-
-		if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0)
-			irq = new;
-		break;
-	}
+	if (!__assign_irq_vector(irq, cfg, apic->target_cpus()))
+		ret = irq;
 	raw_spin_unlock_irqrestore(&vector_lock, flags);
 
-	if (irq > 0)
-		dynamic_irq_init_keep_chip_data(irq);
-
-	return irq;
+	if (ret) {
+		set_irq_chip_data(irq, cfg);
+		irq_clear_status_flags(irq, IRQ_NOREQUEST);
+	} else {
+		free_irq_at(irq, cfg);
+	}
+	return ret;
 }
 
 int create_irq(void)
 {
-	int node = cpu_to_node(boot_cpu_id);
+	int node = cpu_to_node(0);
 	unsigned int irq_want;
 	int irq;
 
@@ -3301,14 +3104,17 @@
 
 void destroy_irq(unsigned int irq)
 {
+	struct irq_cfg *cfg = get_irq_chip_data(irq);
 	unsigned long flags;
 
-	dynamic_irq_cleanup_keep_chip_data(irq);
+	irq_set_status_flags(irq, IRQ_NOREQUEST|IRQ_NOPROBE);
 
-	free_irte(irq);
+	if (intr_remapping_enabled)
+		free_irte(irq);
 	raw_spin_lock_irqsave(&vector_lock, flags);
-	__clear_irq_vector(irq, get_irq_chip_data(irq));
+	__clear_irq_vector(irq, cfg);
 	raw_spin_unlock_irqrestore(&vector_lock, flags);
+	free_irq_at(irq, cfg);
 }
 
 /*
@@ -3332,7 +3138,7 @@
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
 
-	if (irq_remapped(irq)) {
+	if (irq_remapped(get_irq_chip_data(irq))) {
 		struct irte irte;
 		int ir_index;
 		u16 sub_handle;
@@ -3340,14 +3146,7 @@
 		ir_index = map_irq_to_irte_handle(irq, &sub_handle);
 		BUG_ON(ir_index == -1);
 
-		memset (&irte, 0, sizeof(irte));
-
-		irte.present = 1;
-		irte.dst_mode = apic->irq_dest_mode;
-		irte.trigger_mode = 0; /* edge */
-		irte.dlvry_mode = apic->irq_delivery_mode;
-		irte.vector = cfg->vector;
-		irte.dest_id = IRTE_DEST(dest);
+		prepare_irte(&irte, cfg->vector, dest);
 
 		/* Set source-id of interrupt request */
 		if (pdev)
@@ -3392,26 +3191,24 @@
 }
 
 #ifdef CONFIG_SMP
-static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
+static int
+msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = data->chip_data;
 	struct msi_msg msg;
 	unsigned int dest;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
-	cfg = desc->chip_data;
-
-	get_cached_msi_msg_desc(desc, &msg);
+	__get_cached_msi_msg(data->msi_desc, &msg);
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
 	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
-	write_msi_msg_desc(desc, &msg);
+	__write_msi_msg(data->msi_desc, &msg);
 
 	return 0;
 }
@@ -3421,17 +3218,17 @@
  * done in the process context using interrupt-remapping hardware.
  */
 static int
-ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
+ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		    bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg = desc->chip_data;
-	unsigned int dest;
+	struct irq_cfg *cfg = data->chip_data;
+	unsigned int dest, irq = data->irq;
 	struct irte irte;
 
 	if (get_irte(irq, &irte))
 		return -1;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
 	irte.vector = cfg->vector;
@@ -3461,27 +3258,27 @@
  * which implement the MSI or MSI-X Capability Structure.
  */
 static struct irq_chip msi_chip = {
-	.name		= "PCI-MSI",
-	.unmask		= unmask_msi_irq,
-	.mask		= mask_msi_irq,
-	.ack		= ack_apic_edge,
+	.name			= "PCI-MSI",
+	.irq_unmask		= unmask_msi_irq,
+	.irq_mask		= mask_msi_irq,
+	.irq_ack		= ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity	= set_msi_irq_affinity,
+	.irq_set_affinity	= msi_set_affinity,
 #endif
-	.retrigger	= ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 static struct irq_chip msi_ir_chip = {
-	.name		= "IR-PCI-MSI",
-	.unmask		= unmask_msi_irq,
-	.mask		= mask_msi_irq,
+	.name			= "IR-PCI-MSI",
+	.irq_unmask		= unmask_msi_irq,
+	.irq_mask		= mask_msi_irq,
 #ifdef CONFIG_INTR_REMAP
-	.ack		= ir_ack_apic_edge,
+	.irq_ack		= ir_ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity	= ir_set_msi_irq_affinity,
+	.irq_set_affinity	= ir_msi_set_affinity,
 #endif
 #endif
-	.retrigger	= ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 /*
@@ -3513,8 +3310,8 @@
 
 static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 {
-	int ret;
 	struct msi_msg msg;
+	int ret;
 
 	ret = msi_compose_msg(dev, irq, &msg, -1);
 	if (ret < 0)
@@ -3523,12 +3320,8 @@
 	set_irq_msi(irq, msidesc);
 	write_msi_msg(irq, &msg);
 
-	if (irq_remapped(irq)) {
-		struct irq_desc *desc = irq_to_desc(irq);
-		/*
-		 * irq migration in process context
-		 */
-		desc->status |= IRQ_MOVE_PCNTXT;
+	if (irq_remapped(get_irq_chip_data(irq))) {
+		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
 	} else
 		set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");
@@ -3540,13 +3333,10 @@
 
 int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 {
-	unsigned int irq;
-	int ret, sub_handle;
+	int node, ret, sub_handle, index = 0;
+	unsigned int irq, irq_want;
 	struct msi_desc *msidesc;
-	unsigned int irq_want;
 	struct intel_iommu *iommu = NULL;
-	int index = 0;
-	int node;
 
 	/* x86 doesn't support multiple MSI yet */
 	if (type == PCI_CAP_ID_MSI && nvec > 1)
@@ -3606,18 +3396,17 @@
 
 #if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
 #ifdef CONFIG_SMP
-static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
+static int
+dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
+		      bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = data->chip_data;
+	unsigned int dest, irq = data->irq;
 	struct msi_msg msg;
-	unsigned int dest;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
-	cfg = desc->chip_data;
-
 	dmar_msi_read(irq, &msg);
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
@@ -3633,14 +3422,14 @@
 #endif /* CONFIG_SMP */
 
 static struct irq_chip dmar_msi_type = {
-	.name = "DMAR_MSI",
-	.unmask = dmar_msi_unmask,
-	.mask = dmar_msi_mask,
-	.ack = ack_apic_edge,
+	.name			= "DMAR_MSI",
+	.irq_unmask		= dmar_msi_unmask,
+	.irq_mask		= dmar_msi_mask,
+	.irq_ack		= ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity = dmar_msi_set_affinity,
+	.irq_set_affinity	= dmar_msi_set_affinity,
 #endif
-	.retrigger = ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 int arch_setup_dmar_msi(unsigned int irq)
@@ -3661,26 +3450,24 @@
 #ifdef CONFIG_HPET_TIMER
 
 #ifdef CONFIG_SMP
-static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
+static int hpet_msi_set_affinity(struct irq_data *data,
+				 const struct cpumask *mask, bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = data->chip_data;
 	struct msi_msg msg;
 	unsigned int dest;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
-	cfg = desc->chip_data;
-
-	hpet_msi_read(irq, &msg);
+	hpet_msi_read(data->handler_data, &msg);
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
 	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
-	hpet_msi_write(irq, &msg);
+	hpet_msi_write(data->handler_data, &msg);
 
 	return 0;
 }
@@ -3688,34 +3475,33 @@
 #endif /* CONFIG_SMP */
 
 static struct irq_chip ir_hpet_msi_type = {
-	.name = "IR-HPET_MSI",
-	.unmask = hpet_msi_unmask,
-	.mask = hpet_msi_mask,
+	.name			= "IR-HPET_MSI",
+	.irq_unmask		= hpet_msi_unmask,
+	.irq_mask		= hpet_msi_mask,
 #ifdef CONFIG_INTR_REMAP
-	.ack = ir_ack_apic_edge,
+	.irq_ack		= ir_ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity = ir_set_msi_irq_affinity,
+	.irq_set_affinity	= ir_msi_set_affinity,
 #endif
 #endif
-	.retrigger = ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 static struct irq_chip hpet_msi_type = {
 	.name = "HPET_MSI",
-	.unmask = hpet_msi_unmask,
-	.mask = hpet_msi_mask,
-	.ack = ack_apic_edge,
+	.irq_unmask = hpet_msi_unmask,
+	.irq_mask = hpet_msi_mask,
+	.irq_ack = ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity = hpet_msi_set_affinity,
+	.irq_set_affinity = hpet_msi_set_affinity,
 #endif
-	.retrigger = ioapic_retrigger_irq,
+	.irq_retrigger = ioapic_retrigger_irq,
 };
 
 int arch_setup_hpet_msi(unsigned int irq, unsigned int id)
 {
-	int ret;
 	struct msi_msg msg;
-	struct irq_desc *desc = irq_to_desc(irq);
+	int ret;
 
 	if (intr_remapping_enabled) {
 		struct intel_iommu *iommu = map_hpet_to_ir(id);
@@ -3733,9 +3519,9 @@
 	if (ret < 0)
 		return ret;
 
-	hpet_msi_write(irq, &msg);
-	desc->status |= IRQ_MOVE_PCNTXT;
-	if (irq_remapped(irq))
+	hpet_msi_write(get_irq_data(irq), &msg);
+	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
+	if (irq_remapped(get_irq_chip_data(irq)))
 		set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type,
 					      handle_edge_irq, "edge");
 	else
@@ -3768,33 +3554,30 @@
 	write_ht_irq_msg(irq, &msg);
 }
 
-static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
+static int
+ht_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
+	struct irq_cfg *cfg = data->chip_data;
 	unsigned int dest;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
-	cfg = desc->chip_data;
-
-	target_ht_irq(irq, dest, cfg->vector);
-
+	target_ht_irq(data->irq, dest, cfg->vector);
 	return 0;
 }
 
 #endif
 
 static struct irq_chip ht_irq_chip = {
-	.name		= "PCI-HT",
-	.mask		= mask_ht_irq,
-	.unmask		= unmask_ht_irq,
-	.ack		= ack_apic_edge,
+	.name			= "PCI-HT",
+	.irq_mask		= mask_ht_irq,
+	.irq_unmask		= unmask_ht_irq,
+	.irq_ack		= ack_apic_edge,
 #ifdef CONFIG_SMP
-	.set_affinity	= set_ht_irq_affinity,
+	.irq_set_affinity	= ht_set_affinity,
 #endif
-	.retrigger	= ioapic_retrigger_irq,
+	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
@@ -3885,14 +3668,13 @@
 	if (nr < nr_irqs)
 		nr_irqs = nr;
 
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 #endif
 
 static int __io_apic_set_pci_routing(struct device *dev, int irq,
 				struct io_apic_irq_attr *irq_attr)
 {
-	struct irq_desc *desc;
 	struct irq_cfg *cfg;
 	int node;
 	int ioapic, pin;
@@ -3908,13 +3690,11 @@
 	if (dev)
 		node = dev_to_node(dev);
 	else
-		node = cpu_to_node(boot_cpu_id);
+		node = cpu_to_node(0);
 
-	desc = irq_to_desc_alloc_node(irq, node);
-	if (!desc) {
-		printk(KERN_INFO "can not get irq_desc %d\n", irq);
+	cfg = alloc_irq_and_cfg_at(irq, node);
+	if (!cfg)
 		return 0;
-	}
 
 	pin = irq_attr->ioapic_pin;
 	trigger = irq_attr->trigger;
@@ -3924,15 +3704,14 @@
 	 * IRQs < 16 are already in the irq_2_pin[] map
 	 */
 	if (irq >= legacy_pic->nr_legacy_irqs) {
-		cfg = desc->chip_data;
-		if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) {
+		if (__add_pin_to_irq_node(cfg, node, ioapic, pin)) {
 			printk(KERN_INFO "can not add pin %d for irq %d\n",
 				pin, irq);
 			return 0;
 		}
 	}
 
-	setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity);
+	setup_ioapic_irq(ioapic, pin, irq, cfg, trigger, polarity);
 
 	return 0;
 }
@@ -4125,14 +3904,14 @@
 		 */
 		if (desc->status &
 		    (IRQ_NO_BALANCING | IRQ_AFFINITY_SET))
-			mask = desc->affinity;
+			mask = desc->irq_data.affinity;
 		else
 			mask = apic->target_cpus();
 
 		if (intr_remapping_enabled)
-			set_ir_ioapic_affinity_irq_desc(desc, mask);
+			ir_ioapic_set_affinity(&desc->irq_data, mask, false);
 		else
-			set_ioapic_affinity_irq_desc(desc, mask);
+			ioapic_set_affinity(&desc->irq_data, mask, false);
 	}
 
 }
@@ -4316,19 +4095,18 @@
 void __init pre_init_apic_IRQ0(void)
 {
 	struct irq_cfg *cfg;
-	struct irq_desc *desc;
 
 	printk(KERN_INFO "Early APIC setup for system timer0\n");
 #ifndef CONFIG_SMP
 	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid);
 #endif
-	desc = irq_to_desc_alloc_node(0, 0);
+	/* Make sure the irq descriptor is set up */
+	cfg = alloc_irq_and_cfg_at(0, 0);
 
 	setup_local_APIC();
 
-	cfg = irq_cfg(0);
 	add_pin_to_irq_node(cfg, 0, 0, 0);
 	set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge");
 
-	setup_IO_APIC_irq(0, 0, 0, desc, 0, 0);
+	setup_ioapic_irq(0, 0, 0, cfg, 0, 0);
 }
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index a43f71c..c90041c 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -178,7 +178,7 @@
 error:
 	if (nmi_watchdog == NMI_IO_APIC) {
 		if (!timer_through_8259)
-			legacy_pic->chip->mask(0);
+			legacy_pic->mask(0);
 		on_each_cpu(__acpi_nmi_disable, NULL, 1);
 	}
 
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index 83e9be4..f9e4e6a 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -54,6 +54,9 @@
  */
 void __init default_setup_apic_routing(void)
 {
+
+	enable_IR_x2apic();
+
 #ifdef CONFIG_X86_X2APIC
 	if (x2apic_mode
 #ifdef CONFIG_X86_UV
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index ba5f62f..9e093f8 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -148,7 +148,7 @@
 {
 #ifdef CONFIG_SMP
 	/* calling is from identify_secondary_cpu() ? */
-	if (c->cpu_index == boot_cpu_id)
+	if (!c->cpu_index)
 		return;
 
 	/*
@@ -253,37 +253,51 @@
 #endif
 
 /*
- * Fixup core topology information for AMD multi-node processors.
- * Assumption: Number of cores in each internal node is the same.
+ * Fixup core topology information for
+ * (1) AMD multi-node processors
+ *     Assumption: Number of cores in each internal node is the same.
+ * (2) AMD processors supporting compute units
  */
 #ifdef CONFIG_X86_HT
-static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c)
+static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c)
 {
-	unsigned long long value;
-	u32 nodes, cores_per_node;
+	u32 nodes;
+	u8 node_id;
 	int cpu = smp_processor_id();
 
-	if (!cpu_has(c, X86_FEATURE_NODEID_MSR))
+	/* get information required for multi-node processors */
+	if (cpu_has(c, X86_FEATURE_TOPOEXT)) {
+		u32 eax, ebx, ecx, edx;
+
+		cpuid(0x8000001e, &eax, &ebx, &ecx, &edx);
+		nodes = ((ecx >> 8) & 7) + 1;
+		node_id = ecx & 7;
+
+		/* get compute unit information */
+		smp_num_siblings = ((ebx >> 8) & 3) + 1;
+		c->compute_unit_id = ebx & 0xff;
+	} else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) {
+		u64 value;
+
+		rdmsrl(MSR_FAM10H_NODE_ID, value);
+		nodes = ((value >> 3) & 7) + 1;
+		node_id = value & 7;
+	} else
 		return;
 
-	/* fixup topology information only once for a core */
-	if (cpu_has(c, X86_FEATURE_AMD_DCM))
-		return;
+	/* fixup multi-node processor information */
+	if (nodes > 1) {
+		u32 cores_per_node;
 
-	rdmsrl(MSR_FAM10H_NODE_ID, value);
+		set_cpu_cap(c, X86_FEATURE_AMD_DCM);
+		cores_per_node = c->x86_max_cores / nodes;
 
-	nodes = ((value >> 3) & 7) + 1;
-	if (nodes == 1)
-		return;
+		/* store NodeID, use llc_shared_map to store sibling info */
+		per_cpu(cpu_llc_id, cpu) = node_id;
 
-	set_cpu_cap(c, X86_FEATURE_AMD_DCM);
-	cores_per_node = c->x86_max_cores / nodes;
-
-	/* store NodeID, use llc_shared_map to store sibling info */
-	per_cpu(cpu_llc_id, cpu) = value & 7;
-
-	/* fixup core id to be in range from 0 to (cores_per_node - 1) */
-	c->cpu_core_id = c->cpu_core_id % cores_per_node;
+		/* core id to be in range from 0 to (cores_per_node - 1) */
+		c->cpu_core_id = c->cpu_core_id % cores_per_node;
+	}
 }
 #endif
 
@@ -304,9 +318,7 @@
 	c->phys_proc_id = c->initial_apicid >> bits;
 	/* use socket ID also for last level cache */
 	per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
-	/* fixup topology information on multi-node processors */
-	if ((c->x86 == 0x10) && (c->x86_model == 9))
-		amd_fixup_dcm(c);
+	amd_get_topology(c);
 #endif
 }
 
@@ -412,6 +424,23 @@
 			set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
 	}
 #endif
+
+	/* We need to do the following only once */
+	if (c != &boot_cpu_data)
+		return;
+
+	if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
+
+		if (c->x86 > 0x10 ||
+		    (c->x86 == 0x10 && c->x86_model >= 0x2)) {
+			u64 val;
+
+			rdmsrl(MSR_K7_HWCR, val);
+			if (!(val & BIT(24)))
+				printk(KERN_WARNING FW_BUG "TSC doesn't count "
+					"with P0 frequency!\n");
+		}
+	}
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
@@ -523,7 +552,7 @@
 #endif
 
 	if (c->extended_cpuid_level >= 0x80000006) {
-		if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000))
+		if (cpuid_edx(0x80000006) & 0xf000)
 			num_cache_leaves = 4;
 		else
 			num_cache_leaves = 3;
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index f2f9ac7..4b68bda 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -665,7 +665,7 @@
 		this_cpu->c_early_init(c);
 
 #ifdef CONFIG_SMP
-	c->cpu_index = boot_cpu_id;
+	c->cpu_index = 0;
 #endif
 	filter_cpuid_features(c, false);
 }
@@ -704,16 +704,21 @@
 }
 
 /*
- * The NOPL instruction is supposed to exist on all CPUs with
- * family >= 6; unfortunately, that's not true in practice because
- * of early VIA chips and (more importantly) broken virtualizers that
- * are not easy to detect.  In the latter case it doesn't even *fail*
- * reliably, so probing for it doesn't even work.  Disable it completely
+ * The NOPL instruction is supposed to exist on all CPUs of family >= 6;
+ * unfortunately, that's not true in practice because of early VIA
+ * chips and (more importantly) broken virtualizers that are not easy
+ * to detect. In the latter case it doesn't even *fail* reliably, so
+ * probing for it doesn't even work. Disable it completely on 32-bit
  * unless we can find a reliable way to detect all the broken cases.
+ * Enable it explicitly on 64-bit for non-constant inputs of cpu_has().
  */
 static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
 {
+#ifdef CONFIG_X86_32
 	clear_cpu_cap(c, X86_FEATURE_NOPL);
+#else
+	set_cpu_cap(c, X86_FEATURE_NOPL);
+#endif
 }
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
@@ -1264,13 +1269,6 @@
 	clear_all_debug_regs();
 	dbg_restore_debug_regs();
 
-	/*
-	 * Force FPU initialization:
-	 */
-	current_thread_info()->status = 0;
-	clear_used_math();
-	mxcsr_feature_mask_init();
-
 	fpu_init();
 	xsave_init();
 }
diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h
index f668bb1..e765633 100644
--- a/arch/x86/kernel/cpu/cpu.h
+++ b/arch/x86/kernel/cpu/cpu.h
@@ -32,6 +32,7 @@
 extern const struct cpu_dev *const __x86_cpu_dev_start[],
 			    *const __x86_cpu_dev_end[];
 
+extern void get_cpu_cap(struct cpuinfo_x86 *c);
 extern void cpu_detect_cache_sizes(struct cpuinfo_x86 *c);
 extern void get_cpu_cap(struct cpuinfo_x86 *c);
 
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index b438944..695f177 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -170,7 +170,7 @@
 {
 #ifdef CONFIG_SMP
 	/* calling is from identify_secondary_cpu() ? */
-	if (c->cpu_index == boot_cpu_id)
+	if (!c->cpu_index)
 		return;
 
 	/*
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 898c2f4..12cd823 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -17,7 +17,7 @@
 
 #include <asm/processor.h>
 #include <linux/smp.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #include <asm/smp.h>
 
 #define LVL_1_INST	1
@@ -306,7 +306,7 @@
 	ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count);
 };
 
-#ifdef CONFIG_CPU_SUP_AMD
+#ifdef CONFIG_AMD_NB
 
 /*
  * L3 cache descriptors
@@ -369,7 +369,7 @@
 			return;
 
 	/* not in virtualized environments */
-	if (num_k8_northbridges == 0)
+	if (k8_northbridges.num == 0)
 		return;
 
 	/*
@@ -377,7 +377,7 @@
 	 * never freed but this is done only on shutdown so it doesn't matter.
 	 */
 	if (!l3_caches) {
-		int size = num_k8_northbridges * sizeof(struct amd_l3_cache *);
+		int size = k8_northbridges.num * sizeof(struct amd_l3_cache *);
 
 		l3_caches = kzalloc(size, GFP_ATOMIC);
 		if (!l3_caches)
@@ -556,12 +556,12 @@
 static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
 		show_cache_disable_1, store_cache_disable_1);
 
-#else	/* CONFIG_CPU_SUP_AMD */
+#else	/* CONFIG_AMD_NB */
 static void __cpuinit
 amd_check_l3_disable(struct _cpuid4_info_regs *this_leaf, int index)
 {
 };
-#endif /* CONFIG_CPU_SUP_AMD */
+#endif /* CONFIG_AMD_NB */
 
 static int
 __cpuinit cpuid4_cache_lookup_regs(int index,
@@ -1000,7 +1000,7 @@
 
 static struct attribute *default_l3_attrs[] = {
 	DEFAULT_SYSFS_CACHE_ATTRS,
-#ifdef CONFIG_CPU_SUP_AMD
+#ifdef CONFIG_AMD_NB
 	&cache_disable_0.attr,
 	&cache_disable_1.attr,
 #endif
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c
index 5e97529..80c4823 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -131,7 +131,8 @@
 	u32 low = 0, high = 0, address = 0;
 	unsigned int bank, block;
 	struct thresh_restart tr;
-	u8 lvt_off;
+	int lvt_off = -1;
+	u8 offset;
 
 	for (bank = 0; bank < NR_BANKS; ++bank) {
 		for (block = 0; block < NR_BLOCKS; ++block) {
@@ -141,6 +142,7 @@
 				address = (low & MASK_BLKPTR_LO) >> 21;
 				if (!address)
 					break;
+
 				address += MCG_XBLK_ADDR;
 			} else
 				++address;
@@ -148,12 +150,8 @@
 			if (rdmsr_safe(address, &low, &high))
 				break;
 
-			if (!(high & MASK_VALID_HI)) {
-				if (block)
-					continue;
-				else
-					break;
-			}
+			if (!(high & MASK_VALID_HI))
+				continue;
 
 			if (!(high & MASK_CNTP_HI)  ||
 			     (high & MASK_LOCKED_HI))
@@ -165,8 +163,28 @@
 			if (shared_bank[bank] && c->cpu_core_id)
 				break;
 #endif
-			lvt_off = setup_APIC_eilvt_mce(THRESHOLD_APIC_VECTOR,
-						       APIC_EILVT_MSG_FIX, 0);
+			offset = (high & MASK_LVTOFF_HI) >> 20;
+			if (lvt_off < 0) {
+				if (setup_APIC_eilvt(offset,
+						     THRESHOLD_APIC_VECTOR,
+						     APIC_EILVT_MSG_FIX, 0)) {
+					pr_err(FW_BUG "cpu %d, failed to "
+					       "setup threshold interrupt "
+					       "for bank %d, block %d "
+					       "(MSR%08X=0x%x%08x)",
+					       smp_processor_id(), bank, block,
+					       address, high, low);
+					continue;
+				}
+				lvt_off = offset;
+			} else if (lvt_off != offset) {
+				pr_err(FW_BUG "cpu %d, invalid threshold "
+				       "interrupt offset %d for bank %d,"
+				       "block %d (MSR%08X=0x%x%08x)",
+				       smp_processor_id(), lvt_off, bank,
+				       block, address, high, low);
+				continue;
+			}
 
 			high &= ~MASK_LVTOFF_HI;
 			high |= lvt_off << 20;
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index d9368ee..4b68326 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -216,7 +216,7 @@
 		err = sysfs_add_file_to_group(&sys_dev->kobj,
 					      &attr_core_power_limit_count.attr,
 					      thermal_attr_group.name);
-	if (cpu_has(c, X86_FEATURE_PTS))
+	if (cpu_has(c, X86_FEATURE_PTS)) {
 		err = sysfs_add_file_to_group(&sys_dev->kobj,
 					      &attr_package_throttle_count.attr,
 					      thermal_attr_group.name);
@@ -224,6 +224,7 @@
 			err = sysfs_add_file_to_group(&sys_dev->kobj,
 					&attr_package_power_limit_count.attr,
 					thermal_attr_group.name);
+	}
 
 	return err;
 }
@@ -349,7 +350,7 @@
 
 static void unexpected_thermal_interrupt(void)
 {
-	printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n",
+	printk(KERN_ERR "CPU%d: Unexpected LVT thermal interrupt!\n",
 			smp_processor_id());
 	add_taint(TAINT_MACHINE_CHECK);
 }
diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c
index c5f59d0..ac140c7 100644
--- a/arch/x86/kernel/cpu/mtrr/cleanup.c
+++ b/arch/x86/kernel/cpu/mtrr/cleanup.c
@@ -827,7 +827,7 @@
 
 	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
 		return 0;
-	if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
+	if (boot_cpu_data.x86 < 0xf)
 		return 0;
 	/* In case some hypervisor doesn't pass SYSCFG through: */
 	if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0)
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 7d28d7d..9f27228 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -64,18 +64,59 @@
 	}
 }
 
+/* Get the size of contiguous MTRR range */
+static u64 get_mtrr_size(u64 mask)
+{
+	u64 size;
+
+	mask >>= PAGE_SHIFT;
+	mask |= size_or_mask;
+	size = -mask;
+	size <<= PAGE_SHIFT;
+	return size;
+}
+
 /*
- * Returns the effective MTRR type for the region
- * Error returns:
- * - 0xFE - when the range is "not entirely covered" by _any_ var range MTRR
- * - 0xFF - when MTRR is not enabled
+ * Check and return the effective type for MTRR-MTRR type overlap.
+ * Returns 1 if the effective type is UNCACHEABLE, else returns 0
  */
-u8 mtrr_type_lookup(u64 start, u64 end)
+static int check_type_overlap(u8 *prev, u8 *curr)
+{
+	if (*prev == MTRR_TYPE_UNCACHABLE || *curr == MTRR_TYPE_UNCACHABLE) {
+		*prev = MTRR_TYPE_UNCACHABLE;
+		*curr = MTRR_TYPE_UNCACHABLE;
+		return 1;
+	}
+
+	if ((*prev == MTRR_TYPE_WRBACK && *curr == MTRR_TYPE_WRTHROUGH) ||
+	    (*prev == MTRR_TYPE_WRTHROUGH && *curr == MTRR_TYPE_WRBACK)) {
+		*prev = MTRR_TYPE_WRTHROUGH;
+		*curr = MTRR_TYPE_WRTHROUGH;
+	}
+
+	if (*prev != *curr) {
+		*prev = MTRR_TYPE_UNCACHABLE;
+		*curr = MTRR_TYPE_UNCACHABLE;
+		return 1;
+	}
+
+	return 0;
+}
+
+/*
+ * Error/Semi-error returns:
+ * 0xFF - when MTRR is not enabled
+ * *repeat == 1 implies [start:end] spanned across MTRR range and type returned
+ *		corresponds only to [start:*partial_end].
+ *		Caller has to lookup again for [*partial_end:end].
+ */
+static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat)
 {
 	int i;
 	u64 base, mask;
 	u8 prev_match, curr_match;
 
+	*repeat = 0;
 	if (!mtrr_state_set)
 		return 0xFF;
 
@@ -126,8 +167,34 @@
 
 		start_state = ((start & mask) == (base & mask));
 		end_state = ((end & mask) == (base & mask));
-		if (start_state != end_state)
-			return 0xFE;
+
+		if (start_state != end_state) {
+			/*
+			 * We have start:end spanning across an MTRR.
+			 * We split the region into
+			 * either
+			 * (start:mtrr_end) (mtrr_end:end)
+			 * or
+			 * (start:mtrr_start) (mtrr_start:end)
+			 * depending on kind of overlap.
+			 * Return the type for first region and a pointer to
+			 * the start of second region so that caller will
+			 * lookup again on the second region.
+			 * Note: This way we handle multiple overlaps as well.
+			 */
+			if (start_state)
+				*partial_end = base + get_mtrr_size(mask);
+			else
+				*partial_end = base;
+
+			if (unlikely(*partial_end <= start)) {
+				WARN_ON(1);
+				*partial_end = start + PAGE_SIZE;
+			}
+
+			end = *partial_end - 1; /* end is inclusive */
+			*repeat = 1;
+		}
 
 		if ((start & mask) != (base & mask))
 			continue;
@@ -138,21 +205,8 @@
 			continue;
 		}
 
-		if (prev_match == MTRR_TYPE_UNCACHABLE ||
-		    curr_match == MTRR_TYPE_UNCACHABLE) {
-			return MTRR_TYPE_UNCACHABLE;
-		}
-
-		if ((prev_match == MTRR_TYPE_WRBACK &&
-		     curr_match == MTRR_TYPE_WRTHROUGH) ||
-		    (prev_match == MTRR_TYPE_WRTHROUGH &&
-		     curr_match == MTRR_TYPE_WRBACK)) {
-			prev_match = MTRR_TYPE_WRTHROUGH;
-			curr_match = MTRR_TYPE_WRTHROUGH;
-		}
-
-		if (prev_match != curr_match)
-			return MTRR_TYPE_UNCACHABLE;
+		if (check_type_overlap(&prev_match, &curr_match))
+			return curr_match;
 	}
 
 	if (mtrr_tom2) {
@@ -166,6 +220,36 @@
 	return mtrr_state.def_type;
 }
 
+/*
+ * Returns the effective MTRR type for the region
+ * Error return:
+ * 0xFF - when MTRR is not enabled
+ */
+u8 mtrr_type_lookup(u64 start, u64 end)
+{
+	u8 type, prev_type;
+	int repeat;
+	u64 partial_end;
+
+	type = __mtrr_type_lookup(start, end, &partial_end, &repeat);
+
+	/*
+	 * Common path is with repeat = 0.
+	 * However, we can have cases where [start:end] spans across some
+	 * MTRR range. Do repeated lookups for that case here.
+	 */
+	while (repeat) {
+		prev_type = type;
+		start = partial_end;
+		type = __mtrr_type_lookup(start, end, &partial_end, &repeat);
+
+		if (check_type_overlap(&prev_type, &type))
+			return type;
+	}
+
+	return type;
+}
+
 /* Get the MSR pair relating to a var range */
 static void
 get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 03a5b03..fe73c18 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -531,7 +531,7 @@
 /*
  * Setup the hardware configuration for a given attr_type
  */
-static int __hw_perf_event_init(struct perf_event *event)
+static int __x86_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
@@ -584,7 +584,7 @@
 	}
 }
 
-void hw_perf_disable(void)
+static void x86_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -619,7 +619,7 @@
 	}
 }
 
-static const struct pmu pmu;
+static struct pmu pmu;
 
 static inline int is_x86_event(struct perf_event *event)
 {
@@ -801,10 +801,10 @@
 		hwc->last_tag == cpuc->tags[i];
 }
 
-static int x86_pmu_start(struct perf_event *event);
-static void x86_pmu_stop(struct perf_event *event);
+static void x86_pmu_start(struct perf_event *event, int flags);
+static void x86_pmu_stop(struct perf_event *event, int flags);
 
-void hw_perf_enable(void)
+static void x86_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct perf_event *event;
@@ -840,7 +840,14 @@
 			    match_prev_assignment(hwc, cpuc, i))
 				continue;
 
-			x86_pmu_stop(event);
+			/*
+			 * Ensure we don't accidentally enable a stopped
+			 * counter simply because we rescheduled.
+			 */
+			if (hwc->state & PERF_HES_STOPPED)
+				hwc->state |= PERF_HES_ARCH;
+
+			x86_pmu_stop(event, PERF_EF_UPDATE);
 		}
 
 		for (i = 0; i < cpuc->n_events; i++) {
@@ -852,7 +859,10 @@
 			else if (i < n_running)
 				continue;
 
-			x86_pmu_start(event);
+			if (hwc->state & PERF_HES_ARCH)
+				continue;
+
+			x86_pmu_start(event, PERF_EF_RELOAD);
 		}
 		cpuc->n_added = 0;
 		perf_events_lapic_init();
@@ -953,15 +963,12 @@
 }
 
 /*
- * activate a single event
+ * Add a single event to the PMU.
  *
  * The event is added to the group of enabled events
  * but only if it can be scehduled with existing events.
- *
- * Called with PMU disabled. If successful and return value 1,
- * then guaranteed to call perf_enable() and hw_perf_enable()
  */
-static int x86_pmu_enable(struct perf_event *event)
+static int x86_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc;
@@ -970,58 +977,67 @@
 
 	hwc = &event->hw;
 
+	perf_pmu_disable(event->pmu);
 	n0 = cpuc->n_events;
-	n = collect_events(cpuc, event, false);
-	if (n < 0)
-		return n;
+	ret = n = collect_events(cpuc, event, false);
+	if (ret < 0)
+		goto out;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
 
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
-	 * at commit time(->commit_txn) as a whole
+	 * at commit time (->commit_txn) as a whole
 	 */
 	if (cpuc->group_flag & PERF_EVENT_TXN)
-		goto out;
+		goto done_collect;
 
 	ret = x86_pmu.schedule_events(cpuc, n, assign);
 	if (ret)
-		return ret;
+		goto out;
 	/*
 	 * copy new assignment, now we know it is possible
 	 * will be used by hw_perf_enable()
 	 */
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
-out:
+done_collect:
 	cpuc->n_events = n;
 	cpuc->n_added += n - n0;
 	cpuc->n_txn += n - n0;
 
-	return 0;
+	ret = 0;
+out:
+	perf_pmu_enable(event->pmu);
+	return ret;
 }
 
-static int x86_pmu_start(struct perf_event *event)
+static void x86_pmu_start(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx = event->hw.idx;
 
-	if (idx == -1)
-		return -EAGAIN;
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
 
-	x86_perf_event_set_period(event);
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		x86_perf_event_set_period(event);
+	}
+
+	event->hw.state = 0;
+
 	cpuc->events[idx] = event;
 	__set_bit(idx, cpuc->active_mask);
 	__set_bit(idx, cpuc->running);
 	x86_pmu.enable(event);
 	perf_event_update_userpage(event);
-
-	return 0;
-}
-
-static void x86_pmu_unthrottle(struct perf_event *event)
-{
-	int ret = x86_pmu_start(event);
-	WARN_ON_ONCE(ret);
 }
 
 void perf_event_print_debug(void)
@@ -1078,27 +1094,29 @@
 	local_irq_restore(flags);
 }
 
-static void x86_pmu_stop(struct perf_event *event)
+static void x86_pmu_stop(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
-	int idx = hwc->idx;
 
-	if (!__test_and_clear_bit(idx, cpuc->active_mask))
-		return;
+	if (__test_and_clear_bit(hwc->idx, cpuc->active_mask)) {
+		x86_pmu.disable(event);
+		cpuc->events[hwc->idx] = NULL;
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
 
-	x86_pmu.disable(event);
-
-	/*
-	 * Drain the remaining delta count out of a event
-	 * that we are disabling:
-	 */
-	x86_perf_event_update(event);
-
-	cpuc->events[idx] = NULL;
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		x86_perf_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
 }
 
-static void x86_pmu_disable(struct perf_event *event)
+static void x86_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int i;
@@ -1111,7 +1129,7 @@
 	if (cpuc->group_flag & PERF_EVENT_TXN)
 		return;
 
-	x86_pmu_stop(event);
+	x86_pmu_stop(event, PERF_EF_UPDATE);
 
 	for (i = 0; i < cpuc->n_events; i++) {
 		if (event == cpuc->event_list[i]) {
@@ -1134,7 +1152,6 @@
 	struct perf_sample_data data;
 	struct cpu_hw_events *cpuc;
 	struct perf_event *event;
-	struct hw_perf_event *hwc;
 	int idx, handled = 0;
 	u64 val;
 
@@ -1155,7 +1172,6 @@
 		}
 
 		event = cpuc->events[idx];
-		hwc = &event->hw;
 
 		val = x86_perf_event_update(event);
 		if (val & (1ULL << (x86_pmu.cntval_bits - 1)))
@@ -1171,7 +1187,7 @@
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	if (handled)
@@ -1180,25 +1196,6 @@
 	return handled;
 }
 
-void smp_perf_pending_interrupt(struct pt_regs *regs)
-{
-	irq_enter();
-	ack_APIC_irq();
-	inc_irq_stat(apic_pending_irqs);
-	perf_event_do_pending();
-	irq_exit();
-}
-
-void set_perf_event_pending(void)
-{
-#ifdef CONFIG_X86_LOCAL_APIC
-	if (!x86_pmu.apic || !x86_pmu_initialized())
-		return;
-
-	apic->send_IPI_self(LOCAL_PENDING_VECTOR);
-#endif
-}
-
 void perf_events_lapic_init(void)
 {
 	if (!x86_pmu.apic || !x86_pmu_initialized())
@@ -1388,7 +1385,6 @@
 		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 	}
 	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
-	perf_max_events = x86_pmu.num_counters;
 
 	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
 		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
@@ -1424,6 +1420,7 @@
 	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_counters_fixed);
 	pr_info("... event mask:             %016Lx\n", x86_pmu.intel_ctrl);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(x86_pmu_notifier);
 }
 
@@ -1437,10 +1434,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void x86_pmu_start_txn(const struct pmu *pmu)
+static void x86_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuc->group_flag |= PERF_EVENT_TXN;
 	cpuc->n_txn = 0;
 }
@@ -1450,7 +1448,7 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void x86_pmu_cancel_txn(const struct pmu *pmu)
+static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -1460,6 +1458,7 @@
 	 */
 	cpuc->n_added -= cpuc->n_txn;
 	cpuc->n_events -= cpuc->n_txn;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1467,7 +1466,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int x86_pmu_commit_txn(const struct pmu *pmu)
+static int x86_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int assign[X86_PMC_IDX_MAX];
@@ -1489,22 +1488,10 @@
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
-
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= x86_pmu_enable,
-	.disable	= x86_pmu_disable,
-	.start		= x86_pmu_start,
-	.stop		= x86_pmu_stop,
-	.read		= x86_pmu_read,
-	.unthrottle	= x86_pmu_unthrottle,
-	.start_txn	= x86_pmu_start_txn,
-	.cancel_txn	= x86_pmu_cancel_txn,
-	.commit_txn	= x86_pmu_commit_txn,
-};
-
 /*
  * validate that we can schedule this event
  */
@@ -1579,12 +1566,22 @@
 	return ret;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+int x86_pmu_event_init(struct perf_event *event)
 {
-	const struct pmu *tmp;
+	struct pmu *tmp;
 	int err;
 
-	err = __hw_perf_event_init(event);
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
+	err = __x86_pmu_event_init(event);
 	if (!err) {
 		/*
 		 * we temporarily connect event to its pmu
@@ -1604,27 +1601,32 @@
 	if (err) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= x86_pmu_enable,
+	.pmu_disable	= x86_pmu_disable,
+
+	.event_init	= x86_pmu_event_init,
+
+	.add		= x86_pmu_add,
+	.del		= x86_pmu_del,
+	.start		= x86_pmu_start,
+	.stop		= x86_pmu_stop,
+	.read		= x86_pmu_read,
+
+	.start_txn	= x86_pmu_start_txn,
+	.cancel_txn	= x86_pmu_cancel_txn,
+	.commit_txn	= x86_pmu_commit_txn,
+};
+
 /*
  * callchain support
  */
 
-static inline
-void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
-
-
 static void
 backtrace_warning_symbol(void *data, char *msg, unsigned long symbol)
 {
@@ -1645,7 +1647,7 @@
 {
 	struct perf_callchain_entry *entry = data;
 
-	callchain_store(entry, addr);
+	perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops backtrace_ops = {
@@ -1656,11 +1658,15 @@
 	.walk_stack		= print_context_stack_bp,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->ip);
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
+	perf_callchain_store(entry, regs->ip);
 
 	dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry);
 }
@@ -1689,7 +1695,7 @@
 		if (fp < compat_ptr(regs->sp))
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = compat_ptr(frame.next_frame);
 	}
 	return 1;
@@ -1702,19 +1708,20 @@
 }
 #endif
 
-static void
-perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stack_frame frame;
 	const void __user *fp;
 
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
 
 	fp = (void __user *)regs->bp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->ip);
+	perf_callchain_store(entry, regs->ip);
 
 	if (perf_callchain_user32(regs, entry))
 		return;
@@ -1731,52 +1738,11 @@
 		if ((unsigned long)fp < regs->sp)
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = frame.next_frame;
 	}
 }
 
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry;
-
-	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
-		/* TODO: We don't support guest os callchain now */
-		return NULL;
-	}
-
-	if (in_nmi())
-		entry = &__get_cpu_var(pmc_nmi_entry);
-	else
-		entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
-
-	return entry;
-}
-
 unsigned long perf_instruction_pointer(struct pt_regs *regs)
 {
 	unsigned long ip;
diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c
index c2897b7..46d5844 100644
--- a/arch/x86/kernel/cpu/perf_event_amd.c
+++ b/arch/x86/kernel/cpu/perf_event_amd.c
@@ -52,7 +52,7 @@
  [ C(DTLB) ] = {
 	[ C(OP_READ) ] = {
 		[ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses        */
-		[ C(RESULT_MISS)   ] = 0x0046, /* L1 DTLB and L2 DLTB Miss   */
+		[ C(RESULT_MISS)   ] = 0x0746, /* L1_DTLB_AND_L2_DLTB_MISS.ALL */
 	},
 	[ C(OP_WRITE) ] = {
 		[ C(RESULT_ACCESS) ] = 0,
@@ -66,7 +66,7 @@
  [ C(ITLB) ] = {
 	[ C(OP_READ) ] = {
 		[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes        */
-		[ C(RESULT_MISS)   ] = 0x0085, /* Instr. fetch ITLB misses   */
+		[ C(RESULT_MISS)   ] = 0x0385, /* L1_ITLB_AND_L2_ITLB_MISS.ALL */
 	},
 	[ C(OP_WRITE) ] = {
 		[ C(RESULT_ACCESS) ] = -1,
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index ee05c90..c8f5c08 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -713,18 +713,18 @@
 	struct cpu_hw_events *cpuc;
 	int bit, loops;
 	u64 status;
-	int handled = 0;
+	int handled;
 
 	perf_sample_data_init(&data, 0);
 
 	cpuc = &__get_cpu_var(cpu_hw_events);
 
 	intel_pmu_disable_all();
-	intel_pmu_drain_bts_buffer();
+	handled = intel_pmu_drain_bts_buffer();
 	status = intel_pmu_get_status();
 	if (!status) {
 		intel_pmu_enable_all(0);
-		return 0;
+		return handled;
 	}
 
 	loops = 0;
@@ -763,7 +763,7 @@
 		data.period = event->hw.last_period;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	/*
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 18018d1..4977f9c 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -214,7 +214,7 @@
 	update_debugctlmsr(debugctlmsr);
 }
 
-static void intel_pmu_drain_bts_buffer(void)
+static int intel_pmu_drain_bts_buffer(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct debug_store *ds = cpuc->ds;
@@ -231,16 +231,16 @@
 	struct pt_regs regs;
 
 	if (!event)
-		return;
+		return 0;
 
 	if (!ds)
-		return;
+		return 0;
 
 	at  = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
 	top = (struct bts_record *)(unsigned long)ds->bts_index;
 
 	if (top <= at)
-		return;
+		return 0;
 
 	ds->bts_index = ds->bts_buffer_base;
 
@@ -256,7 +256,7 @@
 	perf_prepare_sample(&header, &data, event, &regs);
 
 	if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1))
-		return;
+		return 1;
 
 	for (; at < top; at++) {
 		data.ip		= at->from;
@@ -270,6 +270,7 @@
 	/* There's new data available. */
 	event->hw.interrupts++;
 	event->pending_kill = POLL_IN;
+	return 1;
 }
 
 /*
@@ -491,7 +492,7 @@
 		regs.flags &= ~PERF_EFLAGS_EXACT;
 
 	if (perf_event_overflow(event, 1, &data, &regs))
-		x86_pmu_stop(event);
+		x86_pmu_stop(event, 0);
 }
 
 static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index 2490151..81400b9 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -18,6 +18,8 @@
 struct p4_event_bind {
 	unsigned int opcode;			/* Event code and ESCR selector */
 	unsigned int escr_msr[2];		/* ESCR MSR for this event */
+	unsigned int escr_emask;		/* valid ESCR EventMask bits */
+	unsigned int shared;			/* event is shared across threads */
 	char cntr[2][P4_CNTR_LIMIT];		/* counter index (offset), -1 on abscence */
 };
 
@@ -66,231 +68,435 @@
 	[P4_EVENT_TC_DELIVER_MODE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_DELIVER_MODE),
 		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, ID),
+		.shared		= 1,
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_BPU_FETCH_REQUEST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST),
 		.escr_msr	= { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BPU_FETCH_REQUEST, TCMISS),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_ITLB_REFERENCE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_ITLB_REFERENCE),
 		.escr_msr	= { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, MISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT_UK),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_MEMORY_CANCEL] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_CANCEL),
 		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, 64K_CONF),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_MEMORY_COMPLETE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_COMPLETE),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, LSC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, SSC),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_LOAD_PORT_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_STORE_PORT_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0 ,  MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_MOB_LOAD_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY),
 		.escr_msr	= { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STA)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STD)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_PAGE_WALK_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE),
 		.escr_msr	= { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, DTMISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, ITMISS),
+		.shared		= 1,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BSQ_CACHE_REFERENCE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE),
 		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_IOQ_ALLOCATION] = {
 		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ALLOCATION),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_READ)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OWN)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, PREFETCH),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_IOQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
 		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES),
 		.escr_msr	= { MSR_P4_FSB_ESCR1,  MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH),
 		.cntr		= { {2, -1, -1}, {3, -1, -1} },
 	},
 	[P4_EVENT_FSB_DATA_ACTIVITY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER),
+		.shared		= 1,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BSQ_ALLOCATION] = {		/* shared ESCR, broken CCCR1 */
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ALLOCATION),
 		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2),
 		.cntr		= { {0, -1, -1}, {1, -1, -1} },
 	},
 	[P4_EVENT_BSQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES),
 		.escr_msr	= { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2),
 		.cntr		= { {2, -1, -1}, {3, -1, -1} },
 	},
 	[P4_EVENT_SSE_INPUT_ASSIST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SSE_INPUT_ASSIST, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_PACKED_SP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PACKED_SP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_SP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_PACKED_DP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PACKED_DP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_DP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_SCALAR_SP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_SP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_SP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_SCALAR_DP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_DP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_DP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_64BIT_MMX_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_64BIT_MMX_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_64BIT_MMX_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_128BIT_MMX_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_128BIT_MMX_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_128BIT_MMX_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_X87_FP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_X87_FP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_FP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_TC_MISC] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_MISC),
 		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MISC, FLUSH),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_GLOBAL_POWER_EVENTS] = {
 		.opcode		= P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_TC_MS_XFER] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_MS_XFER),
 		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MS_XFER, CISC),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_UOP_QUEUE_WRITES] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES),
 		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE),
 		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RETIRED_BRANCH_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE),
 		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RESOURCE_STALL] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RESOURCE_STALL),
 		.escr_msr	= { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RESOURCE_STALL, SBFULL),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_WC_BUFFER] = {
 		.opcode		= P4_OPCODE(P4_EVENT_WC_BUFFER),
 		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_EVICTS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_B2B_CYCLES] = {
 		.opcode		= P4_OPCODE(P4_EVENT_B2B_CYCLES),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BNR] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BNR),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_SNOOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SNOOP),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_RESPONSE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RESPONSE),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_FRONT_END_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_FRONT_END_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_EXECUTION_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_EXECUTION_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_REPLAY_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_REPLAY_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_INSTR_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_INSTR_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSTAG),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_UOPS_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOPS_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_UOP_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOP_TYPE),
 		.escr_msr	= { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGLOADS)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGSTORES),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_BRANCH_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BRANCH_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNM)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTM),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_MISPRED_BRANCH_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+		P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_X87_ASSIST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_X87_ASSIST),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, PREA),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_MACHINE_CLEAR] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MACHINE_CLEAR),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, CLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, MOCLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, SMCLEAR),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_INSTR_COMPLETED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_INSTR_COMPLETED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 };
@@ -428,29 +634,73 @@
 	return config;
 }
 
-static int p4_validate_raw_event(struct perf_event *event)
+/* check cpu model specifics */
+static bool p4_event_match_cpu_model(unsigned int event_idx)
 {
-	unsigned int v;
-
-	/* user data may have out-of-bound event index */
-	v = p4_config_unpack_event(event->attr.config);
-	if (v >= ARRAY_SIZE(p4_event_bind_map)) {
-		pr_warning("P4 PMU: Unknown event code: %d\n", v);
-		return -EINVAL;
+	/* INSTR_COMPLETED event only exist for model 3, 4, 6 (Prescott) */
+	if (event_idx == P4_EVENT_INSTR_COMPLETED) {
+		if (boot_cpu_data.x86_model != 3 &&
+			boot_cpu_data.x86_model != 4 &&
+			boot_cpu_data.x86_model != 6)
+			return false;
 	}
 
 	/*
-	 * it may have some screwed PEBS bits
+	 * For info
+	 * - IQ_ESCR0, IQ_ESCR1 only for models 1 and 2
 	 */
-	if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) {
-		pr_warning("P4 PMU: PEBS are not supported yet\n");
+
+	return true;
+}
+
+static int p4_validate_raw_event(struct perf_event *event)
+{
+	unsigned int v, emask;
+
+	/* User data may have out-of-bound event index */
+	v = p4_config_unpack_event(event->attr.config);
+	if (v >= ARRAY_SIZE(p4_event_bind_map))
 		return -EINVAL;
+
+	/* It may be unsupported: */
+	if (!p4_event_match_cpu_model(v))
+		return -EINVAL;
+
+	/*
+	 * NOTE: P4_CCCR_THREAD_ANY has not the same meaning as
+	 * in Architectural Performance Monitoring, it means not
+	 * on _which_ logical cpu to count but rather _when_, ie it
+	 * depends on logical cpu state -- count event if one cpu active,
+	 * none, both or any, so we just allow user to pass any value
+	 * desired.
+	 *
+	 * In turn we always set Tx_OS/Tx_USR bits bound to logical
+	 * cpu without their propagation to another cpu
+	 */
+
+	/*
+	 * if an event is shared accross the logical threads
+	 * the user needs special permissions to be able to use it
+	 */
+	if (p4_event_bind_map[v].shared) {
+		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
+			return -EACCES;
 	}
+
+	/* ESCR EventMask bits may be invalid */
+	emask = p4_config_unpack_escr(event->attr.config) & P4_ESCR_EVENTMASK_MASK;
+	if (emask & ~p4_event_bind_map[v].escr_emask)
+		return -EINVAL;
+
+	/*
+	 * it may have some invalid PEBS bits
+	 */
+	if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE))
+		return -EINVAL;
+
 	v = p4_config_unpack_metric(event->attr.config);
-	if (v >= ARRAY_SIZE(p4_pebs_bind_map)) {
-		pr_warning("P4 PMU: Unknown metric code: %d\n", v);
+	if (v >= ARRAY_SIZE(p4_pebs_bind_map))
 		return -EINVAL;
-	}
 
 	return 0;
 }
@@ -478,27 +728,21 @@
 
 	if (event->attr.type == PERF_TYPE_RAW) {
 
+		/*
+		 * Clear bits we reserve to be managed by kernel itself
+		 * and never allowed from a user space
+		 */
+		 event->attr.config &= P4_CONFIG_MASK;
+
 		rc = p4_validate_raw_event(event);
 		if (rc)
 			goto out;
 
 		/*
-		 * We don't control raw events so it's up to the caller
-		 * to pass sane values (and we don't count the thread number
-		 * on HT machine but allow HT-compatible specifics to be
-		 * passed on)
-		 *
 		 * Note that for RAW events we allow user to use P4_CCCR_RESERVED
 		 * bits since we keep additional info here (for cache events and etc)
-		 *
-		 * XXX: HT wide things should check perf_paranoid_cpu() &&
-		 *      CAP_SYS_ADMIN
 		 */
-		event->hw.config |= event->attr.config &
-			(p4_config_pack_escr(P4_ESCR_MASK_HT) |
-			 p4_config_pack_cccr(P4_CCCR_MASK_HT | P4_CCCR_RESERVED));
-
-		event->hw.config &= ~P4_CCCR_FORCE_OVF;
+		event->hw.config |= event->attr.config;
 	}
 
 	rc = x86_setup_perfctr(event);
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index fb329e9..d9f4ff8 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -700,11 +700,10 @@
 {
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
-		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 &&
-		    boot_cpu_data.x86 != 16 && boot_cpu_data.x86 != 17)
-			return;
-		wd_ops = &k7_wd_ops;
-		break;
+		if (boot_cpu_data.x86 == 6 ||
+		    (boot_cpu_data.x86 >= 0xf && boot_cpu_data.x86 <= 0x15))
+			wd_ops = &k7_wd_ops;
+		return;
 	case X86_VENDOR_INTEL:
 		/* Work around where perfctr1 doesn't have a working enable
 		 * bit as described in the following errata:
diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c
index d490795..c7f64e6 100644
--- a/arch/x86/kernel/cpu/scattered.c
+++ b/arch/x86/kernel/cpu/scattered.c
@@ -44,6 +44,12 @@
 		{ X86_FEATURE_LBRV,		CR_EDX, 1, 0x8000000a, 0 },
 		{ X86_FEATURE_SVML,		CR_EDX, 2, 0x8000000a, 0 },
 		{ X86_FEATURE_NRIPS,		CR_EDX, 3, 0x8000000a, 0 },
+		{ X86_FEATURE_TSCRATEMSR,	CR_EDX, 4, 0x8000000a, 0 },
+		{ X86_FEATURE_VMCBCLEAN,	CR_EDX, 5, 0x8000000a, 0 },
+		{ X86_FEATURE_FLUSHBYASID,	CR_EDX, 6, 0x8000000a, 0 },
+		{ X86_FEATURE_DECODEASSISTS,	CR_EDX, 7, 0x8000000a, 0 },
+		{ X86_FEATURE_PAUSEFILTER,	CR_EDX,10, 0x8000000a, 0 },
+		{ X86_FEATURE_PFTHRESHOLD,	CR_EDX,12, 0x8000000a, 0 },
 		{ 0, 0, 0, 0, 0 }
 	};
 
diff --git a/arch/x86/kernel/crash_dump_64.c b/arch/x86/kernel/crash_dump_64.c
index 045b36c..9948288 100644
--- a/arch/x86/kernel/crash_dump_64.c
+++ b/arch/x86/kernel/crash_dump_64.c
@@ -34,7 +34,7 @@
 	if (!csize)
 		return 0;
 
-	vaddr = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE);
+	vaddr = ioremap_cache(pfn << PAGE_SHIFT, PAGE_SIZE);
 	if (!vaddr)
 		return -ENOMEM;
 
@@ -46,6 +46,7 @@
 	} else
 		memcpy(buf, vaddr + offset, csize);
 
+	set_iounmap_nonlazy();
 	iounmap(vaddr);
 	return csize;
 }
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index ebdb85c..76b8cd9 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -97,7 +97,6 @@
 }
 
 #if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC)
-#if defined(CONFIG_ACPI) && defined(CONFIG_X86_IO_APIC)
 static u32 __init ati_ixp4x0_rev(int num, int slot, int func)
 {
 	u32 d;
@@ -115,7 +114,6 @@
 	d &= 0xff;
 	return d;
 }
-#endif
 
 static void __init ati_bugs(int num, int slot, int func)
 {
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index fa99bae..4572f25 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -14,6 +14,7 @@
 #include <xen/hvc-console.h>
 #include <asm/pci-direct.h>
 #include <asm/fixmap.h>
+#include <asm/mrst.h>
 #include <asm/pgtable.h>
 #include <linux/usb/ehci_def.h>
 
@@ -239,6 +240,18 @@
 		if (!strncmp(buf, "xen", 3))
 			early_console_register(&xenboot_console, keep);
 #endif
+#ifdef CONFIG_X86_MRST_EARLY_PRINTK
+		if (!strncmp(buf, "mrst", 4)) {
+			mrst_early_console_init();
+			early_console_register(&early_mrst_console, keep);
+		}
+
+		if (!strncmp(buf, "hsu", 3)) {
+			hsu_early_console_init();
+			early_console_register(&early_hsu_console, keep);
+		}
+
+#endif
 		buf++;
 	}
 	return 0;
diff --git a/arch/x86/kernel/early_printk_mrst.c b/arch/x86/kernel/early_printk_mrst.c
new file mode 100644
index 0000000..65df603
--- /dev/null
+++ b/arch/x86/kernel/early_printk_mrst.c
@@ -0,0 +1,319 @@
+/*
+ * early_printk_mrst.c - early consoles for Intel MID platforms
+ *
+ * Copyright (c) 2008-2010, Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+/*
+ * This file implements two early consoles named mrst and hsu.
+ * mrst is based on Maxim3110 spi-uart device, it exists in both
+ * Moorestown and Medfield platforms, while hsu is based on a High
+ * Speed UART device which only exists in the Medfield platform
+ */
+
+#include <linux/serial_reg.h>
+#include <linux/serial_mfd.h>
+#include <linux/kmsg_dump.h>
+#include <linux/console.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/io.h>
+
+#include <asm/fixmap.h>
+#include <asm/pgtable.h>
+#include <asm/mrst.h>
+
+#define MRST_SPI_TIMEOUT		0x200000
+#define MRST_REGBASE_SPI0		0xff128000
+#define MRST_REGBASE_SPI1		0xff128400
+#define MRST_CLK_SPI0_REG		0xff11d86c
+
+/* Bit fields in CTRLR0 */
+#define SPI_DFS_OFFSET			0
+
+#define SPI_FRF_OFFSET			4
+#define SPI_FRF_SPI			0x0
+#define SPI_FRF_SSP			0x1
+#define SPI_FRF_MICROWIRE		0x2
+#define SPI_FRF_RESV			0x3
+
+#define SPI_MODE_OFFSET			6
+#define SPI_SCPH_OFFSET			6
+#define SPI_SCOL_OFFSET			7
+#define SPI_TMOD_OFFSET			8
+#define	SPI_TMOD_TR			0x0		/* xmit & recv */
+#define SPI_TMOD_TO			0x1		/* xmit only */
+#define SPI_TMOD_RO			0x2		/* recv only */
+#define SPI_TMOD_EPROMREAD		0x3		/* eeprom read mode */
+
+#define SPI_SLVOE_OFFSET		10
+#define SPI_SRL_OFFSET			11
+#define SPI_CFS_OFFSET			12
+
+/* Bit fields in SR, 7 bits */
+#define SR_MASK				0x7f		/* cover 7 bits */
+#define SR_BUSY				(1 << 0)
+#define SR_TF_NOT_FULL			(1 << 1)
+#define SR_TF_EMPT			(1 << 2)
+#define SR_RF_NOT_EMPT			(1 << 3)
+#define SR_RF_FULL			(1 << 4)
+#define SR_TX_ERR			(1 << 5)
+#define SR_DCOL				(1 << 6)
+
+struct dw_spi_reg {
+	u32	ctrl0;
+	u32	ctrl1;
+	u32	ssienr;
+	u32	mwcr;
+	u32	ser;
+	u32	baudr;
+	u32	txfltr;
+	u32	rxfltr;
+	u32	txflr;
+	u32	rxflr;
+	u32	sr;
+	u32	imr;
+	u32	isr;
+	u32	risr;
+	u32	txoicr;
+	u32	rxoicr;
+	u32	rxuicr;
+	u32	msticr;
+	u32	icr;
+	u32	dmacr;
+	u32	dmatdlr;
+	u32	dmardlr;
+	u32	idr;
+	u32	version;
+
+	/* Currently operates as 32 bits, though only the low 16 bits matter */
+	u32	dr;
+} __packed;
+
+#define dw_readl(dw, name)		__raw_readl(&(dw)->name)
+#define dw_writel(dw, name, val)	__raw_writel((val), &(dw)->name)
+
+/* Default use SPI0 register for mrst, we will detect Penwell and use SPI1 */
+static unsigned long mrst_spi_paddr = MRST_REGBASE_SPI0;
+
+static u32 *pclk_spi0;
+/* Always contains an accessable address, start with 0 */
+static struct dw_spi_reg *pspi;
+
+static struct kmsg_dumper dw_dumper;
+static int dumper_registered;
+
+static void dw_kmsg_dump(struct kmsg_dumper *dumper,
+			enum kmsg_dump_reason reason,
+			const char *s1, unsigned long l1,
+			const char *s2, unsigned long l2)
+{
+	int i;
+
+	/* When run to this, we'd better re-init the HW */
+	mrst_early_console_init();
+
+	for (i = 0; i < l1; i++)
+		early_mrst_console.write(&early_mrst_console, s1 + i, 1);
+	for (i = 0; i < l2; i++)
+		early_mrst_console.write(&early_mrst_console, s2 + i, 1);
+}
+
+/* Set the ratio rate to 115200, 8n1, IRQ disabled */
+static void max3110_write_config(void)
+{
+	u16 config;
+
+	config = 0xc001;
+	dw_writel(pspi, dr, config);
+}
+
+/* Translate char to a eligible word and send to max3110 */
+static void max3110_write_data(char c)
+{
+	u16 data;
+
+	data = 0x8000 | c;
+	dw_writel(pspi, dr, data);
+}
+
+void mrst_early_console_init(void)
+{
+	u32 ctrlr0 = 0;
+	u32 spi0_cdiv;
+	u32 freq; /* Freqency info only need be searched once */
+
+	/* Base clk is 100 MHz, the actual clk = 100M / (clk_divider + 1) */
+	pclk_spi0 = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE,
+							MRST_CLK_SPI0_REG);
+	spi0_cdiv = ((*pclk_spi0) & 0xe00) >> 9;
+	freq = 100000000 / (spi0_cdiv + 1);
+
+	if (mrst_identify_cpu() == MRST_CPU_CHIP_PENWELL)
+		mrst_spi_paddr = MRST_REGBASE_SPI1;
+
+	pspi = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE,
+						mrst_spi_paddr);
+
+	/* Disable SPI controller */
+	dw_writel(pspi, ssienr, 0);
+
+	/* Set control param, 8 bits, transmit only mode */
+	ctrlr0 = dw_readl(pspi, ctrl0);
+
+	ctrlr0 &= 0xfcc0;
+	ctrlr0 |= 0xf | (SPI_FRF_SPI << SPI_FRF_OFFSET)
+		      | (SPI_TMOD_TO << SPI_TMOD_OFFSET);
+	dw_writel(pspi, ctrl0, ctrlr0);
+
+	/*
+	 * Change the spi0 clk to comply with 115200 bps, use 100000 to
+	 * calculate the clk dividor to make the clock a little slower
+	 * than real baud rate.
+	 */
+	dw_writel(pspi, baudr, freq/100000);
+
+	/* Disable all INT for early phase */
+	dw_writel(pspi, imr, 0x0);
+
+	/* Set the cs to spi-uart */
+	dw_writel(pspi, ser, 0x2);
+
+	/* Enable the HW, the last step for HW init */
+	dw_writel(pspi, ssienr, 0x1);
+
+	/* Set the default configuration */
+	max3110_write_config();
+
+	/* Register the kmsg dumper */
+	if (!dumper_registered) {
+		dw_dumper.dump = dw_kmsg_dump;
+		kmsg_dump_register(&dw_dumper);
+		dumper_registered = 1;
+	}
+}
+
+/* Slave select should be called in the read/write function */
+static void early_mrst_spi_putc(char c)
+{
+	unsigned int timeout;
+	u32 sr;
+
+	timeout = MRST_SPI_TIMEOUT;
+	/* Early putc needs to make sure the TX FIFO is not full */
+	while (--timeout) {
+		sr = dw_readl(pspi, sr);
+		if (!(sr & SR_TF_NOT_FULL))
+			cpu_relax();
+		else
+			break;
+	}
+
+	if (!timeout)
+		pr_warning("MRST earlycon: timed out\n");
+	else
+		max3110_write_data(c);
+}
+
+/* Early SPI only uses polling mode */
+static void early_mrst_spi_write(struct console *con, const char *str, unsigned n)
+{
+	int i;
+
+	for (i = 0; i < n && *str; i++) {
+		if (*str == '\n')
+			early_mrst_spi_putc('\r');
+		early_mrst_spi_putc(*str);
+		str++;
+	}
+}
+
+struct console early_mrst_console = {
+	.name =		"earlymrst",
+	.write =	early_mrst_spi_write,
+	.flags =	CON_PRINTBUFFER,
+	.index =	-1,
+};
+
+/*
+ * Following is the early console based on Medfield HSU (High
+ * Speed UART) device.
+ */
+#define HSU_PORT2_PADDR		0xffa28180
+
+static void __iomem *phsu;
+
+void hsu_early_console_init(void)
+{
+	u8 lcr;
+
+	phsu = (void *)set_fixmap_offset_nocache(FIX_EARLYCON_MEM_BASE,
+							HSU_PORT2_PADDR);
+
+	/* Disable FIFO */
+	writeb(0x0, phsu + UART_FCR);
+
+	/* Set to default 115200 bps, 8n1 */
+	lcr = readb(phsu + UART_LCR);
+	writeb((0x80 | lcr), phsu + UART_LCR);
+	writeb(0x18, phsu + UART_DLL);
+	writeb(lcr,  phsu + UART_LCR);
+	writel(0x3600, phsu + UART_MUL*4);
+
+	writeb(0x8, phsu + UART_MCR);
+	writeb(0x7, phsu + UART_FCR);
+	writeb(0x3, phsu + UART_LCR);
+
+	/* Clear IRQ status */
+	readb(phsu + UART_LSR);
+	readb(phsu + UART_RX);
+	readb(phsu + UART_IIR);
+	readb(phsu + UART_MSR);
+
+	/* Enable FIFO */
+	writeb(0x7, phsu + UART_FCR);
+}
+
+#define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE)
+
+static void early_hsu_putc(char ch)
+{
+	unsigned int timeout = 10000; /* 10ms */
+	u8 status;
+
+	while (--timeout) {
+		status = readb(phsu + UART_LSR);
+		if (status & BOTH_EMPTY)
+			break;
+		udelay(1);
+	}
+
+	/* Only write the char when there was no timeout */
+	if (timeout)
+		writeb(ch, phsu + UART_TX);
+}
+
+static void early_hsu_write(struct console *con, const char *str, unsigned n)
+{
+	int i;
+
+	for (i = 0; i < n && *str; i++) {
+		if (*str == '\n')
+			early_hsu_putc('\r');
+		early_hsu_putc(*str);
+		str++;
+	}
+}
+
+struct console early_hsu_console = {
+	.name =		"earlyhsu",
+	.write =	early_hsu_write,
+	.flags =	CON_PRINTBUFFER,
+	.index =	-1,
+};
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 227d009..9fb188d 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -115,8 +115,7 @@
 
  /* unfortunately push/pop can't be no-op */
 .macro PUSH_GS
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
 .endm
 .macro POP_GS pop=0
 	addl $(4 + \pop), %esp
@@ -140,14 +139,12 @@
 #else	/* CONFIG_X86_32_LAZY_GS */
 
 .macro PUSH_GS
-	pushl %gs
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %gs
 	/*CFI_REL_OFFSET gs, 0*/
 .endm
 
 .macro POP_GS pop=0
-98:	popl %gs
-	CFI_ADJUST_CFA_OFFSET -4
+98:	popl_cfi %gs
 	/*CFI_RESTORE gs*/
   .if \pop <> 0
 	add $\pop, %esp
@@ -195,35 +192,25 @@
 .macro SAVE_ALL
 	cld
 	PUSH_GS
-	pushl %fs
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %fs
 	/*CFI_REL_OFFSET fs, 0;*/
-	pushl %es
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %es
 	/*CFI_REL_OFFSET es, 0;*/
-	pushl %ds
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ds
 	/*CFI_REL_OFFSET ds, 0;*/
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	CFI_REL_OFFSET eax, 0
-	pushl %ebp
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebp
 	CFI_REL_OFFSET ebp, 0
-	pushl %edi
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %edi
 	CFI_REL_OFFSET edi, 0
-	pushl %esi
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %esi
 	CFI_REL_OFFSET esi, 0
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %edx
 	CFI_REL_OFFSET edx, 0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ecx
 	CFI_REL_OFFSET ecx, 0
-	pushl %ebx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebx
 	CFI_REL_OFFSET ebx, 0
 	movl $(__USER_DS), %edx
 	movl %edx, %ds
@@ -234,39 +221,29 @@
 .endm
 
 .macro RESTORE_INT_REGS
-	popl %ebx
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %ebx
 	CFI_RESTORE ebx
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %ecx
 	CFI_RESTORE ecx
-	popl %edx
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %edx
 	CFI_RESTORE edx
-	popl %esi
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %esi
 	CFI_RESTORE esi
-	popl %edi
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %edi
 	CFI_RESTORE edi
-	popl %ebp
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %ebp
 	CFI_RESTORE ebp
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %eax
 	CFI_RESTORE eax
 .endm
 
 .macro RESTORE_REGS pop=0
 	RESTORE_INT_REGS
-1:	popl %ds
-	CFI_ADJUST_CFA_OFFSET -4
+1:	popl_cfi %ds
 	/*CFI_RESTORE ds;*/
-2:	popl %es
-	CFI_ADJUST_CFA_OFFSET -4
+2:	popl_cfi %es
 	/*CFI_RESTORE es;*/
-3:	popl %fs
-	CFI_ADJUST_CFA_OFFSET -4
+3:	popl_cfi %fs
 	/*CFI_RESTORE fs;*/
 	POP_GS \pop
 .pushsection .fixup, "ax"
@@ -320,16 +297,12 @@
 
 ENTRY(ret_from_fork)
 	CFI_STARTPROC
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	call schedule_tail
 	GET_THREAD_INFO(%ebp)
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
-	pushl $0x0202			# Reset kernel eflags
-	CFI_ADJUST_CFA_OFFSET 4
-	popfl
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %eax
+	pushl_cfi $0x0202		# Reset kernel eflags
+	popfl_cfi
 	jmp syscall_exit
 	CFI_ENDPROC
 END(ret_from_fork)
@@ -409,29 +382,23 @@
 	 * enough kernel state to call TRACE_IRQS_OFF can be called - but
 	 * we immediately enable interrupts at that point anyway.
 	 */
-	pushl $(__USER_DS)
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $(__USER_DS)
 	/*CFI_REL_OFFSET ss, 0*/
-	pushl %ebp
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebp
 	CFI_REL_OFFSET esp, 0
-	pushfl
+	pushfl_cfi
 	orl $X86_EFLAGS_IF, (%esp)
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $(__USER_CS)
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $(__USER_CS)
 	/*CFI_REL_OFFSET cs, 0*/
 	/*
 	 * Push current_thread_info()->sysenter_return to the stack.
 	 * A tiny bit of offset fixup is necessary - 4*4 means the 4 words
 	 * pushed above; +8 corresponds to copy_thread's esp0 setting.
 	 */
-	pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp)
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp)
 	CFI_REL_OFFSET eip, 0
 
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	SAVE_ALL
 	ENABLE_INTERRUPTS(CLBR_NONE)
 
@@ -486,8 +453,7 @@
 	movl %eax,%edx			/* 2nd arg: syscall number */
 	movl $AUDIT_ARCH_I386,%eax	/* 1st arg: audit arch */
 	call audit_syscall_entry
-	pushl %ebx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebx
 	movl PT_EAX(%esp),%eax		/* reload syscall number */
 	jmp sysenter_do_call
 
@@ -529,8 +495,7 @@
 	# system call handler stub
 ENTRY(system_call)
 	RING0_INT_FRAME			# can't unwind into user space anyway
-	pushl %eax			# save orig_eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax			# save orig_eax
 	SAVE_ALL
 	GET_THREAD_INFO(%ebp)
 					# system call tracing in operation / emulation
@@ -566,7 +531,6 @@
 	je ldt_ss			# returning to user-space with LDT SS
 restore_nocheck:
 	RESTORE_REGS 4			# skip orig_eax/error_code
-	CFI_ADJUST_CFA_OFFSET -4
 irq_return:
 	INTERRUPT_RETURN
 .section .fixup,"ax"
@@ -619,10 +583,8 @@
 	shr $16, %edx
 	mov %dl, GDT_ESPFIX_SS + 4 /* bits 16..23 */
 	mov %dh, GDT_ESPFIX_SS + 7 /* bits 24..31 */
-	pushl $__ESPFIX_SS
-	CFI_ADJUST_CFA_OFFSET 4
-	push %eax			/* new kernel esp */
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $__ESPFIX_SS
+	pushl_cfi %eax			/* new kernel esp */
 	/* Disable interrupts, but do not irqtrace this section: we
 	 * will soon execute iret and the tracer was already set to
 	 * the irqstate after the iret */
@@ -666,11 +628,9 @@
 
 	ALIGN
 work_notifysig_v86:
-	pushl %ecx			# save ti_flags for do_notify_resume
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ecx			# save ti_flags for do_notify_resume
 	call save_v86_state		# %eax contains pt_regs pointer
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %ecx
 	movl %eax, %esp
 #else
 	movl %esp, %eax
@@ -750,14 +710,18 @@
 #define PTREGSCALL3(name) \
 	ALIGN; \
 ptregs_##name: \
+	CFI_STARTPROC; \
 	leal 4(%esp),%eax; \
-	pushl %eax; \
+	pushl_cfi %eax; \
 	movl PT_EDX(%eax),%ecx; \
 	movl PT_ECX(%eax),%edx; \
 	movl PT_EBX(%eax),%eax; \
 	call sys_##name; \
 	addl $4,%esp; \
-	ret
+	CFI_ADJUST_CFA_OFFSET -4; \
+	ret; \
+	CFI_ENDPROC; \
+ENDPROC(ptregs_##name)
 
 PTREGSCALL1(iopl)
 PTREGSCALL0(fork)
@@ -772,15 +736,19 @@
 /* Clone is an oddball.  The 4th arg is in %edi */
 	ALIGN;
 ptregs_clone:
+	CFI_STARTPROC
 	leal 4(%esp),%eax
-	pushl %eax
-	pushl PT_EDI(%eax)
+	pushl_cfi %eax
+	pushl_cfi PT_EDI(%eax)
 	movl PT_EDX(%eax),%ecx
 	movl PT_ECX(%eax),%edx
 	movl PT_EBX(%eax),%eax
 	call sys_clone
 	addl $8,%esp
+	CFI_ADJUST_CFA_OFFSET -8
 	ret
+	CFI_ENDPROC
+ENDPROC(ptregs_clone)
 
 .macro FIXUP_ESPFIX_STACK
 /*
@@ -795,10 +763,8 @@
 	mov GDT_ESPFIX_SS + 7, %ah /* bits 24..31 */
 	shl $16, %eax
 	addl %esp, %eax			/* the adjusted stack pointer */
-	pushl $__KERNEL_DS
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $__KERNEL_DS
+	pushl_cfi %eax
 	lss (%esp), %esp		/* switch to the normal stack segment */
 	CFI_ADJUST_CFA_OFFSET -8
 .endm
@@ -835,8 +801,7 @@
       .if vector <> FIRST_EXTERNAL_VECTOR
 	CFI_ADJUST_CFA_OFFSET -4
       .endif
-1:	pushl $(~vector+0x80)	/* Note: always in signed byte range */
-	CFI_ADJUST_CFA_OFFSET 4
+1:	pushl_cfi $(~vector+0x80)	/* Note: always in signed byte range */
       .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6
 	jmp 2f
       .endif
@@ -876,8 +841,7 @@
 #define BUILD_INTERRUPT3(name, nr, fn)	\
 ENTRY(name)				\
 	RING0_INT_FRAME;		\
-	pushl $~(nr);			\
-	CFI_ADJUST_CFA_OFFSET 4;	\
+	pushl_cfi $~(nr);		\
 	SAVE_ALL;			\
 	TRACE_IRQS_OFF			\
 	movl %esp,%eax;			\
@@ -893,21 +857,18 @@
 
 ENTRY(coprocessor_error)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_coprocessor_error
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_coprocessor_error
 	jmp error_code
 	CFI_ENDPROC
 END(coprocessor_error)
 
 ENTRY(simd_coprocessor_error)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
 #ifdef CONFIG_X86_INVD_BUG
 	/* AMD 486 bug: invd from userspace calls exception 19 instead of #GP */
-661:	pushl $do_general_protection
+661:	pushl_cfi $do_general_protection
 662:
 .section .altinstructions,"a"
 	.balign 4
@@ -922,19 +883,16 @@
 664:
 .previous
 #else
-	pushl $do_simd_coprocessor_error
+	pushl_cfi $do_simd_coprocessor_error
 #endif
-	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
 END(simd_coprocessor_error)
 
 ENTRY(device_not_available)
 	RING0_INT_FRAME
-	pushl $-1			# mark this as an int
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_device_not_available
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $-1			# mark this as an int
+	pushl_cfi $do_device_not_available
 	jmp error_code
 	CFI_ENDPROC
 END(device_not_available)
@@ -956,82 +914,68 @@
 
 ENTRY(overflow)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_overflow
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_overflow
 	jmp error_code
 	CFI_ENDPROC
 END(overflow)
 
 ENTRY(bounds)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_bounds
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_bounds
 	jmp error_code
 	CFI_ENDPROC
 END(bounds)
 
 ENTRY(invalid_op)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_invalid_op
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_invalid_op
 	jmp error_code
 	CFI_ENDPROC
 END(invalid_op)
 
 ENTRY(coprocessor_segment_overrun)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_coprocessor_segment_overrun
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_coprocessor_segment_overrun
 	jmp error_code
 	CFI_ENDPROC
 END(coprocessor_segment_overrun)
 
 ENTRY(invalid_TSS)
 	RING0_EC_FRAME
-	pushl $do_invalid_TSS
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_invalid_TSS
 	jmp error_code
 	CFI_ENDPROC
 END(invalid_TSS)
 
 ENTRY(segment_not_present)
 	RING0_EC_FRAME
-	pushl $do_segment_not_present
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_segment_not_present
 	jmp error_code
 	CFI_ENDPROC
 END(segment_not_present)
 
 ENTRY(stack_segment)
 	RING0_EC_FRAME
-	pushl $do_stack_segment
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_stack_segment
 	jmp error_code
 	CFI_ENDPROC
 END(stack_segment)
 
 ENTRY(alignment_check)
 	RING0_EC_FRAME
-	pushl $do_alignment_check
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_alignment_check
 	jmp error_code
 	CFI_ENDPROC
 END(alignment_check)
 
 ENTRY(divide_error)
 	RING0_INT_FRAME
-	pushl $0			# no error code
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_divide_error
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0			# no error code
+	pushl_cfi $do_divide_error
 	jmp error_code
 	CFI_ENDPROC
 END(divide_error)
@@ -1039,10 +983,8 @@
 #ifdef CONFIG_X86_MCE
 ENTRY(machine_check)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl machine_check_vector
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi machine_check_vector
 	jmp error_code
 	CFI_ENDPROC
 END(machine_check)
@@ -1050,10 +992,8 @@
 
 ENTRY(spurious_interrupt_bug)
 	RING0_INT_FRAME
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_spurious_interrupt_bug
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
+	pushl_cfi $do_spurious_interrupt_bug
 	jmp error_code
 	CFI_ENDPROC
 END(spurious_interrupt_bug)
@@ -1084,8 +1024,7 @@
 
 ENTRY(xen_hypervisor_callback)
 	CFI_STARTPROC
-	pushl $0
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $0
 	SAVE_ALL
 	TRACE_IRQS_OFF
 
@@ -1121,23 +1060,20 @@
 # We distinguish between categories by maintaining a status value in EAX.
 ENTRY(xen_failsafe_callback)
 	CFI_STARTPROC
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	movl $1,%eax
 1:	mov 4(%esp),%ds
 2:	mov 8(%esp),%es
 3:	mov 12(%esp),%fs
 4:	mov 16(%esp),%gs
 	testl %eax,%eax
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %eax
 	lea 16(%esp),%esp
 	CFI_ADJUST_CFA_OFFSET -16
 	jz 5f
 	addl $16,%esp
 	jmp iret_exc		# EAX != 0 => Category 2 (Bad IRET)
-5:	pushl $0		# EAX == 0 => Category 1 (Bad segment)
-	CFI_ADJUST_CFA_OFFSET 4
+5:	pushl_cfi $0		# EAX == 0 => Category 1 (Bad segment)
 	SAVE_ALL
 	jmp ret_from_exception
 	CFI_ENDPROC
@@ -1287,40 +1223,29 @@
 
 ENTRY(page_fault)
 	RING0_EC_FRAME
-	pushl $do_page_fault
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_page_fault
 	ALIGN
 error_code:
 	/* the function address is in %gs's slot on the stack */
-	pushl %fs
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %fs
 	/*CFI_REL_OFFSET fs, 0*/
-	pushl %es
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %es
 	/*CFI_REL_OFFSET es, 0*/
-	pushl %ds
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ds
 	/*CFI_REL_OFFSET ds, 0*/
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	CFI_REL_OFFSET eax, 0
-	pushl %ebp
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebp
 	CFI_REL_OFFSET ebp, 0
-	pushl %edi
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %edi
 	CFI_REL_OFFSET edi, 0
-	pushl %esi
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %esi
 	CFI_REL_OFFSET esi, 0
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %edx
 	CFI_REL_OFFSET edx, 0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ecx
 	CFI_REL_OFFSET ecx, 0
-	pushl %ebx
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ebx
 	CFI_REL_OFFSET ebx, 0
 	cld
 	movl $(__KERNEL_PERCPU), %ecx
@@ -1362,12 +1287,9 @@
 	movl TSS_sysenter_sp0 + \offset(%esp), %esp
 	CFI_DEF_CFA esp, 0
 	CFI_UNDEFINED eip
-	pushfl
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $__KERNEL_CS
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $sysenter_past_esp
-	CFI_ADJUST_CFA_OFFSET 4
+	pushfl_cfi
+	pushl_cfi $__KERNEL_CS
+	pushl_cfi $sysenter_past_esp
 	CFI_REL_OFFSET eip, 0
 .endm
 
@@ -1377,8 +1299,7 @@
 	jne debug_stack_correct
 	FIX_STACK 12, debug_stack_correct, debug_esp_fix_insn
 debug_stack_correct:
-	pushl $-1			# mark this as an int
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $-1			# mark this as an int
 	SAVE_ALL
 	TRACE_IRQS_OFF
 	xorl %edx,%edx			# error code 0
@@ -1398,32 +1319,27 @@
  */
 ENTRY(nmi)
 	RING0_INT_FRAME
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	movl %ss, %eax
 	cmpw $__ESPFIX_SS, %ax
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %eax
 	je nmi_espfix_stack
 	cmpl $ia32_sysenter_target,(%esp)
 	je nmi_stack_fixup
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	movl %esp,%eax
 	/* Do not access memory above the end of our stack page,
 	 * it might not exist.
 	 */
 	andl $(THREAD_SIZE-1),%eax
 	cmpl $(THREAD_SIZE-20),%eax
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
+	popl_cfi %eax
 	jae nmi_stack_correct
 	cmpl $ia32_sysenter_target,12(%esp)
 	je nmi_debug_stack_check
 nmi_stack_correct:
 	/* We have a RING0_INT_FRAME here */
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	SAVE_ALL
 	xorl %edx,%edx		# zero error code
 	movl %esp,%eax		# pt_regs pointer
@@ -1452,18 +1368,14 @@
 	 *
 	 * create the pointer to lss back
 	 */
-	pushl %ss
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl %esp
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %ss
+	pushl_cfi %esp
 	addl $4, (%esp)
 	/* copy the iret frame of 12 bytes */
 	.rept 3
-	pushl 16(%esp)
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi 16(%esp)
 	.endr
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi %eax
 	SAVE_ALL
 	FIXUP_ESPFIX_STACK		# %eax == %esp
 	xorl %edx,%edx			# zero error code
@@ -1477,8 +1389,7 @@
 
 ENTRY(int3)
 	RING0_INT_FRAME
-	pushl $-1			# mark this as an int
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $-1			# mark this as an int
 	SAVE_ALL
 	TRACE_IRQS_OFF
 	xorl %edx,%edx		# zero error code
@@ -1490,8 +1401,7 @@
 
 ENTRY(general_protection)
 	RING0_EC_FRAME
-	pushl $do_general_protection
-	CFI_ADJUST_CFA_OFFSET 4
+	pushl_cfi $do_general_protection
 	jmp error_code
 	CFI_ENDPROC
 END(general_protection)
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 17be5ec..a7ae7fd 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -213,23 +213,17 @@
 	.macro FAKE_STACK_FRAME child_rip
 	/* push in order ss, rsp, eflags, cs, rip */
 	xorl %eax, %eax
-	pushq $__KERNEL_DS /* ss */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi $__KERNEL_DS /* ss */
 	/*CFI_REL_OFFSET	ss,0*/
-	pushq %rax /* rsp */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi %rax /* rsp */
 	CFI_REL_OFFSET	rsp,0
-	pushq $X86_EFLAGS_IF /* eflags - interrupts on */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi $X86_EFLAGS_IF /* eflags - interrupts on */
 	/*CFI_REL_OFFSET	rflags,0*/
-	pushq $__KERNEL_CS /* cs */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi $__KERNEL_CS /* cs */
 	/*CFI_REL_OFFSET	cs,0*/
-	pushq \child_rip /* rip */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi \child_rip /* rip */
 	CFI_REL_OFFSET	rip,0
-	pushq	%rax /* orig rax */
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi %rax /* orig rax */
 	.endm
 
 	.macro UNFAKE_STACK_FRAME
@@ -398,10 +392,8 @@
 
 	LOCK ; btr $TIF_FORK,TI_flags(%r8)
 
-	push kernel_eflags(%rip)
-	CFI_ADJUST_CFA_OFFSET 8
-	popf					# reset kernel eflags
-	CFI_ADJUST_CFA_OFFSET -8
+	pushq_cfi kernel_eflags(%rip)
+	popfq_cfi				# reset kernel eflags
 
 	call schedule_tail			# rdi: 'prev' task parameter
 
@@ -521,11 +513,9 @@
 	jnc sysret_signal
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
-	pushq %rdi
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi %rdi
 	call schedule
-	popq  %rdi
-	CFI_ADJUST_CFA_OFFSET -8
+	popq_cfi %rdi
 	jmp sysret_check
 
 	/* Handle a signal */
@@ -634,11 +624,9 @@
 	jnc  int_very_careful
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
-	pushq %rdi
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi %rdi
 	call schedule
-	popq %rdi
-	CFI_ADJUST_CFA_OFFSET -8
+	popq_cfi %rdi
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	jmp int_with_check
@@ -652,12 +640,10 @@
 	/* Check for syscall exit trace */
 	testl $_TIF_WORK_SYSCALL_EXIT,%edx
 	jz int_signal
-	pushq %rdi
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi %rdi
 	leaq 8(%rsp),%rdi	# &ptregs -> arg1
 	call syscall_trace_leave
-	popq %rdi
-	CFI_ADJUST_CFA_OFFSET -8
+	popq_cfi %rdi
 	andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi
 	jmp int_restore_rest
 
@@ -714,9 +700,8 @@
 
 ENTRY(stub_execve)
 	CFI_STARTPROC
-	popq %r11
-	CFI_ADJUST_CFA_OFFSET -8
-	CFI_REGISTER rip, r11
+	addq $8, %rsp
+	PARTIAL_FRAME 0
 	SAVE_REST
 	FIXUP_TOP_OF_STACK %r11
 	movq %rsp, %rcx
@@ -735,7 +720,7 @@
 ENTRY(stub_rt_sigreturn)
 	CFI_STARTPROC
 	addq $8, %rsp
-	CFI_ADJUST_CFA_OFFSET	-8
+	PARTIAL_FRAME 0
 	SAVE_REST
 	movq %rsp,%rdi
 	FIXUP_TOP_OF_STACK %r11
@@ -766,8 +751,7 @@
       .if vector <> FIRST_EXTERNAL_VECTOR
 	CFI_ADJUST_CFA_OFFSET -8
       .endif
-1:	pushq $(~vector+0x80)	/* Note: always in signed byte range */
-	CFI_ADJUST_CFA_OFFSET 8
+1:	pushq_cfi $(~vector+0x80)	/* Note: always in signed byte range */
       .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6
 	jmp 2f
       .endif
@@ -796,8 +780,8 @@
 
 /* 0(%rsp): ~(interrupt number) */
 	.macro interrupt func
-	subq $10*8, %rsp
-	CFI_ADJUST_CFA_OFFSET 10*8
+	subq $ORIG_RAX-ARGOFFSET+8, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-ARGOFFSET+8
 	call save_args
 	PARTIAL_FRAME 0
 	call \func
@@ -822,6 +806,7 @@
 	TRACE_IRQS_OFF
 	decl PER_CPU_VAR(irq_count)
 	leaveq
+	CFI_RESTORE		rbp
 	CFI_DEF_CFA_REGISTER	rsp
 	CFI_ADJUST_CFA_OFFSET	-8
 exit_intr:
@@ -903,11 +888,9 @@
 	jnc   retint_signal
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
-	pushq %rdi
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi %rdi
 	call  schedule
-	popq %rdi
-	CFI_ADJUST_CFA_OFFSET	-8
+	popq_cfi %rdi
 	GET_THREAD_INFO(%rcx)
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
@@ -956,8 +939,7 @@
 .macro apicinterrupt num sym do_sym
 ENTRY(\sym)
 	INTR_FRAME
-	pushq $~(\num)
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi $~(\num)
 	interrupt \do_sym
 	jmp ret_from_intr
 	CFI_ENDPROC
@@ -1023,9 +1005,9 @@
 apicinterrupt SPURIOUS_APIC_VECTOR \
 	spurious_interrupt smp_spurious_interrupt
 
-#ifdef CONFIG_PERF_EVENTS
-apicinterrupt LOCAL_PENDING_VECTOR \
-	perf_pending_interrupt smp_perf_pending_interrupt
+#ifdef CONFIG_IRQ_WORK
+apicinterrupt IRQ_WORK_VECTOR \
+	irq_work_interrupt smp_irq_work_interrupt
 #endif
 
 /*
@@ -1036,8 +1018,8 @@
 	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
 	pushq_cfi $-1		/* ORIG_RAX: no syscall to restart */
-	subq $15*8,%rsp
-	CFI_ADJUST_CFA_OFFSET 15*8
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call error_entry
 	DEFAULT_FRAME 0
 	movq %rsp,%rdi		/* pt_regs pointer */
@@ -1052,9 +1034,9 @@
 ENTRY(\sym)
 	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $-1		/* ORIG_RAX: no syscall to restart */
-	CFI_ADJUST_CFA_OFFSET 8
-	subq $15*8, %rsp
+	pushq_cfi $-1		/* ORIG_RAX: no syscall to restart */
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call save_paranoid
 	TRACE_IRQS_OFF
 	movq %rsp,%rdi		/* pt_regs pointer */
@@ -1070,9 +1052,9 @@
 ENTRY(\sym)
 	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $-1		/* ORIG_RAX: no syscall to restart */
-	CFI_ADJUST_CFA_OFFSET 8
-	subq $15*8, %rsp
+	pushq_cfi $-1		/* ORIG_RAX: no syscall to restart */
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call save_paranoid
 	TRACE_IRQS_OFF
 	movq %rsp,%rdi		/* pt_regs pointer */
@@ -1089,8 +1071,8 @@
 ENTRY(\sym)
 	XCPT_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	subq $15*8,%rsp
-	CFI_ADJUST_CFA_OFFSET 15*8
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call error_entry
 	DEFAULT_FRAME 0
 	movq %rsp,%rdi			/* pt_regs pointer */
@@ -1107,8 +1089,8 @@
 ENTRY(\sym)
 	XCPT_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	subq $15*8,%rsp
-	CFI_ADJUST_CFA_OFFSET 15*8
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call save_paranoid
 	DEFAULT_FRAME 0
 	TRACE_IRQS_OFF
@@ -1139,16 +1121,14 @@
 	/* edi:  new selector */
 ENTRY(native_load_gs_index)
 	CFI_STARTPROC
-	pushf
-	CFI_ADJUST_CFA_OFFSET 8
+	pushfq_cfi
 	DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI)
 	SWAPGS
 gs_change:
 	movl %edi,%gs
 2:	mfence		/* workaround */
 	SWAPGS
-	popf
-	CFI_ADJUST_CFA_OFFSET -8
+	popfq_cfi
 	ret
 	CFI_ENDPROC
 END(native_load_gs_index)
@@ -1215,8 +1195,7 @@
 /* Call softirq on interrupt stack. Interrupts are off. */
 ENTRY(call_softirq)
 	CFI_STARTPROC
-	push %rbp
-	CFI_ADJUST_CFA_OFFSET	8
+	pushq_cfi %rbp
 	CFI_REL_OFFSET rbp,0
 	mov  %rsp,%rbp
 	CFI_DEF_CFA_REGISTER rbp
@@ -1225,6 +1204,7 @@
 	push  %rbp			# backlink for old unwinder
 	call __do_softirq
 	leaveq
+	CFI_RESTORE		rbp
 	CFI_DEF_CFA_REGISTER	rsp
 	CFI_ADJUST_CFA_OFFSET   -8
 	decl PER_CPU_VAR(irq_count)
@@ -1368,7 +1348,7 @@
 
 	/* ebx:	no swapgs flag */
 ENTRY(paranoid_exit)
-	INTR_FRAME
+	DEFAULT_FRAME
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	testl %ebx,%ebx				/* swapgs needed? */
@@ -1445,7 +1425,6 @@
 error_sti:
 	TRACE_IRQS_OFF
 	ret
-	CFI_ENDPROC
 
 /*
  * There are two places in the kernel that can potentially fault with
@@ -1470,6 +1449,7 @@
 	/* Fix truncated RIP */
 	movq %rcx,RIP+8(%rsp)
 	jmp error_swapgs
+	CFI_ENDPROC
 END(error_entry)
 
 
@@ -1498,8 +1478,8 @@
 	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
 	pushq_cfi $-1
-	subq $15*8, %rsp
-	CFI_ADJUST_CFA_OFFSET 15*8
+	subq $ORIG_RAX-R15, %rsp
+	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 	call save_paranoid
 	DEFAULT_FRAME 0
 	/* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index cd37469..3afb33f 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -257,14 +257,9 @@
 	return mod_code_status;
 }
 
-
-
-
-static unsigned char ftrace_nop[MCOUNT_INSN_SIZE];
-
 static unsigned char *ftrace_nop_replace(void)
 {
-	return ftrace_nop;
+	return ideal_nop5;
 }
 
 static int
@@ -338,62 +333,6 @@
 
 int __init ftrace_dyn_arch_init(void *data)
 {
-	extern const unsigned char ftrace_test_p6nop[];
-	extern const unsigned char ftrace_test_nop5[];
-	extern const unsigned char ftrace_test_jmp[];
-	int faulted = 0;
-
-	/*
-	 * There is no good nop for all x86 archs.
-	 * We will default to using the P6_NOP5, but first we
-	 * will test to make sure that the nop will actually
-	 * work on this CPU. If it faults, we will then
-	 * go to a lesser efficient 5 byte nop. If that fails
-	 * we then just use a jmp as our nop. This isn't the most
-	 * efficient nop, but we can not use a multi part nop
-	 * since we would then risk being preempted in the middle
-	 * of that nop, and if we enabled tracing then, it might
-	 * cause a system crash.
-	 *
-	 * TODO: check the cpuid to determine the best nop.
-	 */
-	asm volatile (
-		"ftrace_test_jmp:"
-		"jmp ftrace_test_p6nop\n"
-		"nop\n"
-		"nop\n"
-		"nop\n"  /* 2 byte jmp + 3 bytes */
-		"ftrace_test_p6nop:"
-		P6_NOP5
-		"jmp 1f\n"
-		"ftrace_test_nop5:"
-		".byte 0x66,0x66,0x66,0x66,0x90\n"
-		"1:"
-		".section .fixup, \"ax\"\n"
-		"2:	movl $1, %0\n"
-		"	jmp ftrace_test_nop5\n"
-		"3:	movl $2, %0\n"
-		"	jmp 1b\n"
-		".previous\n"
-		_ASM_EXTABLE(ftrace_test_p6nop, 2b)
-		_ASM_EXTABLE(ftrace_test_nop5, 3b)
-		: "=r"(faulted) : "0" (faulted));
-
-	switch (faulted) {
-	case 0:
-		pr_info("converting mcount calls to 0f 1f 44 00 00\n");
-		memcpy(ftrace_nop, ftrace_test_p6nop, MCOUNT_INSN_SIZE);
-		break;
-	case 1:
-		pr_info("converting mcount calls to 66 66 66 66 90\n");
-		memcpy(ftrace_nop, ftrace_test_nop5, MCOUNT_INSN_SIZE);
-		break;
-	case 2:
-		pr_info("converting mcount calls to jmp . + 5\n");
-		memcpy(ftrace_nop, ftrace_test_jmp, MCOUNT_INSN_SIZE);
-		break;
-	}
-
 	/* The return code is retured via data */
 	*(unsigned long *)data = 0;
 
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 7494999..efaf906 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -440,9 +440,9 @@
 static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev);
 static struct hpet_dev	*hpet_devs;
 
-void hpet_msi_unmask(unsigned int irq)
+void hpet_msi_unmask(struct irq_data *data)
 {
-	struct hpet_dev *hdev = get_irq_data(irq);
+	struct hpet_dev *hdev = data->handler_data;
 	unsigned int cfg;
 
 	/* unmask it */
@@ -451,10 +451,10 @@
 	hpet_writel(cfg, HPET_Tn_CFG(hdev->num));
 }
 
-void hpet_msi_mask(unsigned int irq)
+void hpet_msi_mask(struct irq_data *data)
 {
+	struct hpet_dev *hdev = data->handler_data;
 	unsigned int cfg;
-	struct hpet_dev *hdev = get_irq_data(irq);
 
 	/* mask it */
 	cfg = hpet_readl(HPET_Tn_CFG(hdev->num));
@@ -462,18 +462,14 @@
 	hpet_writel(cfg, HPET_Tn_CFG(hdev->num));
 }
 
-void hpet_msi_write(unsigned int irq, struct msi_msg *msg)
+void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg)
 {
-	struct hpet_dev *hdev = get_irq_data(irq);
-
 	hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num));
 	hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4);
 }
 
-void hpet_msi_read(unsigned int irq, struct msi_msg *msg)
+void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg)
 {
-	struct hpet_dev *hdev = get_irq_data(irq);
-
 	msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num));
 	msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4);
 	msg->address_hi = 0;
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index a46cb35..58bb239 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -68,19 +68,22 @@
 	 */
 
 	if (!HAVE_HWFP) {
+		/*
+		 * Disable xsave as we do not support it if i387
+		 * emulation is enabled.
+		 */
+		setup_clear_cpu_cap(X86_FEATURE_XSAVE);
+		setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT);
 		xstate_size = sizeof(struct i387_soft_struct);
 		return;
 	}
 
 	if (cpu_has_fxsr)
 		xstate_size = sizeof(struct i387_fxsave_struct);
-#ifdef CONFIG_X86_32
 	else
 		xstate_size = sizeof(struct i387_fsave_struct);
-#endif
 }
 
-#ifdef CONFIG_X86_64
 /*
  * Called at bootup to set up the initial FPU state that is later cloned
  * into all processes.
@@ -88,12 +91,21 @@
 
 void __cpuinit fpu_init(void)
 {
-	unsigned long oldcr0 = read_cr0();
+	unsigned long cr0;
+	unsigned long cr4_mask = 0;
 
-	set_in_cr4(X86_CR4_OSFXSR);
-	set_in_cr4(X86_CR4_OSXMMEXCPT);
+	if (cpu_has_fxsr)
+		cr4_mask |= X86_CR4_OSFXSR;
+	if (cpu_has_xmm)
+		cr4_mask |= X86_CR4_OSXMMEXCPT;
+	if (cr4_mask)
+		set_in_cr4(cr4_mask);
 
-	write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */
+	cr0 = read_cr0();
+	cr0 &= ~(X86_CR0_TS|X86_CR0_EM); /* clear TS and EM */
+	if (!HAVE_HWFP)
+		cr0 |= X86_CR0_EM;
+	write_cr0(cr0);
 
 	if (!smp_processor_id())
 		init_thread_xstate();
@@ -104,24 +116,12 @@
 	clear_used_math();
 }
 
-#else	/* CONFIG_X86_64 */
-
-void __cpuinit fpu_init(void)
-{
-	if (!smp_processor_id())
-		init_thread_xstate();
-}
-
-#endif	/* CONFIG_X86_32 */
-
 void fpu_finit(struct fpu *fpu)
 {
-#ifdef CONFIG_X86_32
 	if (!HAVE_HWFP) {
 		finit_soft_fpu(&fpu->state->soft);
 		return;
 	}
-#endif
 
 	if (cpu_has_fxsr) {
 		struct i387_fxsave_struct *fx = &fpu->state->fxsave;
@@ -386,19 +386,17 @@
 #ifdef CONFIG_X86_64
 	env->fip = fxsave->rip;
 	env->foo = fxsave->rdp;
+	/*
+	 * should be actually ds/cs at fpu exception time, but
+	 * that information is not available in 64bit mode.
+	 */
+	env->fcs = task_pt_regs(tsk)->cs;
 	if (tsk == current) {
-		/*
-		 * should be actually ds/cs at fpu exception time, but
-		 * that information is not available in 64bit mode.
-		 */
-		asm("mov %%ds, %[fos]" : [fos] "=r" (env->fos));
-		asm("mov %%cs, %[fcs]" : [fcs] "=r" (env->fcs));
+		savesegment(ds, env->fos);
 	} else {
-		struct pt_regs *regs = task_pt_regs(tsk);
-
-		env->fos = 0xffff0000 | tsk->thread.ds;
-		env->fcs = regs->cs;
+		env->fos = tsk->thread.ds;
 	}
+	env->fos |= 0xffff0000;
 #else
 	env->fip = fxsave->fip;
 	env->fcs = (u16) fxsave->fcs | ((u32) fxsave->fop << 16);
diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c
index cafa7c8..20757cb 100644
--- a/arch/x86/kernel/i8259.c
+++ b/arch/x86/kernel/i8259.c
@@ -29,24 +29,10 @@
  * plus some generic x86 specific things if generic specifics makes
  * any sense at all.
  */
+static void init_8259A(int auto_eoi);
 
 static int i8259A_auto_eoi;
 DEFINE_RAW_SPINLOCK(i8259A_lock);
-static void mask_and_ack_8259A(unsigned int);
-static void mask_8259A(void);
-static void unmask_8259A(void);
-static void disable_8259A_irq(unsigned int irq);
-static void enable_8259A_irq(unsigned int irq);
-static void init_8259A(int auto_eoi);
-static int i8259A_irq_pending(unsigned int irq);
-
-struct irq_chip i8259A_chip = {
-	.name		= "XT-PIC",
-	.mask		= disable_8259A_irq,
-	.disable	= disable_8259A_irq,
-	.unmask		= enable_8259A_irq,
-	.mask_ack	= mask_and_ack_8259A,
-};
 
 /*
  * 8259A PIC functions to handle ISA devices:
@@ -68,7 +54,7 @@
  */
 unsigned long io_apic_irqs;
 
-static void disable_8259A_irq(unsigned int irq)
+static void mask_8259A_irq(unsigned int irq)
 {
 	unsigned int mask = 1 << irq;
 	unsigned long flags;
@@ -82,7 +68,12 @@
 	raw_spin_unlock_irqrestore(&i8259A_lock, flags);
 }
 
-static void enable_8259A_irq(unsigned int irq)
+static void disable_8259A_irq(struct irq_data *data)
+{
+	mask_8259A_irq(data->irq);
+}
+
+static void unmask_8259A_irq(unsigned int irq)
 {
 	unsigned int mask = ~(1 << irq);
 	unsigned long flags;
@@ -96,6 +87,11 @@
 	raw_spin_unlock_irqrestore(&i8259A_lock, flags);
 }
 
+static void enable_8259A_irq(struct irq_data *data)
+{
+	unmask_8259A_irq(data->irq);
+}
+
 static int i8259A_irq_pending(unsigned int irq)
 {
 	unsigned int mask = 1<<irq;
@@ -117,7 +113,7 @@
 	disable_irq_nosync(irq);
 	io_apic_irqs &= ~(1<<irq);
 	set_irq_chip_and_handler_name(irq, &i8259A_chip, handle_level_irq,
-				      "XT");
+				      i8259A_chip.name);
 	enable_irq(irq);
 }
 
@@ -150,8 +146,9 @@
  * first, _then_ send the EOI, and the order of EOI
  * to the two 8259s is important!
  */
-static void mask_and_ack_8259A(unsigned int irq)
+static void mask_and_ack_8259A(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
 	unsigned int irqmask = 1 << irq;
 	unsigned long flags;
 
@@ -223,6 +220,14 @@
 	}
 }
 
+struct irq_chip i8259A_chip = {
+	.name		= "XT-PIC",
+	.irq_mask	= disable_8259A_irq,
+	.irq_disable	= disable_8259A_irq,
+	.irq_unmask	= enable_8259A_irq,
+	.irq_mask_ack	= mask_and_ack_8259A,
+};
+
 static char irq_trigger[2];
 /**
  * ELCR registers (0x4d0, 0x4d1) control edge/level of IRQ
@@ -342,9 +347,9 @@
 		 * In AEOI mode we just have to mask the interrupt
 		 * when acking.
 		 */
-		i8259A_chip.mask_ack = disable_8259A_irq;
+		i8259A_chip.irq_mask_ack = disable_8259A_irq;
 	else
-		i8259A_chip.mask_ack = mask_and_ack_8259A;
+		i8259A_chip.irq_mask_ack = mask_and_ack_8259A;
 
 	udelay(100);		/* wait for 8259A to initialize */
 
@@ -363,14 +368,6 @@
 static void legacy_pic_noop(void) { };
 static void legacy_pic_uint_noop(unsigned int unused) { };
 static void legacy_pic_int_noop(int unused) { };
-
-static struct irq_chip dummy_pic_chip  = {
-	.name = "dummy pic",
-	.mask = legacy_pic_uint_noop,
-	.unmask = legacy_pic_uint_noop,
-	.disable = legacy_pic_uint_noop,
-	.mask_ack = legacy_pic_uint_noop,
-};
 static int legacy_pic_irq_pending_noop(unsigned int irq)
 {
 	return 0;
@@ -378,7 +375,9 @@
 
 struct legacy_pic null_legacy_pic = {
 	.nr_legacy_irqs = 0,
-	.chip = &dummy_pic_chip,
+	.chip = &dummy_irq_chip,
+	.mask = legacy_pic_uint_noop,
+	.unmask = legacy_pic_uint_noop,
 	.mask_all = legacy_pic_noop,
 	.restore_mask = legacy_pic_noop,
 	.init = legacy_pic_int_noop,
@@ -389,7 +388,9 @@
 struct legacy_pic default_legacy_pic = {
 	.nr_legacy_irqs = NR_IRQS_LEGACY,
 	.chip  = &i8259A_chip,
-	.mask_all  = mask_8259A,
+	.mask = mask_8259A_irq,
+	.unmask = unmask_8259A_irq,
+	.mask_all = mask_8259A,
 	.restore_mask = unmask_8259A,
 	.init = init_8259A,
 	.irq_pending = i8259A_irq_pending,
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 91fd0c7..83ec017 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -67,10 +67,10 @@
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs);
 	seq_printf(p, "  Performance monitoring interrupts\n");
-	seq_printf(p, "%*s: ", prec, "PND");
+	seq_printf(p, "%*s: ", prec, "IWI");
 	for_each_online_cpu(j)
-		seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs);
-	seq_printf(p, "  Performance pending work\n");
+		seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs);
+	seq_printf(p, "  IRQ work interrupts\n");
 #endif
 	if (x86_platform_ipi_callback) {
 		seq_printf(p, "%*s: ", prec, "PLT");
@@ -159,7 +159,7 @@
 	seq_printf(p, "%*d: ", prec, i);
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
-	seq_printf(p, " %8s", desc->chip->name);
+	seq_printf(p, " %8s", desc->irq_data.chip->name);
 	seq_printf(p, "-%-8s", desc->name);
 
 	if (action) {
@@ -185,7 +185,7 @@
 	sum += irq_stats(cpu)->apic_timer_irqs;
 	sum += irq_stats(cpu)->irq_spurious_count;
 	sum += irq_stats(cpu)->apic_perf_irqs;
-	sum += irq_stats(cpu)->apic_pending_irqs;
+	sum += irq_stats(cpu)->apic_irq_work_irqs;
 #endif
 	if (x86_platform_ipi_callback)
 		sum += irq_stats(cpu)->x86_platform_ipis;
@@ -282,6 +282,7 @@
 	unsigned int irq, vector;
 	static int warned;
 	struct irq_desc *desc;
+	struct irq_data *data;
 
 	for_each_irq_desc(irq, desc) {
 		int break_affinity = 0;
@@ -296,7 +297,8 @@
 		/* interrupt's are disabled at this point */
 		raw_spin_lock(&desc->lock);
 
-		affinity = desc->affinity;
+		data = &desc->irq_data;
+		affinity = data->affinity;
 		if (!irq_has_action(irq) ||
 		    cpumask_equal(affinity, cpu_online_mask)) {
 			raw_spin_unlock(&desc->lock);
@@ -315,16 +317,16 @@
 			affinity = cpu_all_mask;
 		}
 
-		if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->mask)
-			desc->chip->mask(irq);
+		if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_mask)
+			data->chip->irq_mask(data);
 
-		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, affinity);
+		if (data->chip->irq_set_affinity)
+			data->chip->irq_set_affinity(data, affinity, true);
 		else if (!(warned++))
 			set_affinity = 0;
 
-		if (!(desc->status & IRQ_MOVE_PCNTXT) && desc->chip->unmask)
-			desc->chip->unmask(irq);
+		if (!(desc->status & IRQ_MOVE_PCNTXT) && data->chip->irq_unmask)
+			data->chip->irq_unmask(data);
 
 		raw_spin_unlock(&desc->lock);
 
@@ -355,10 +357,10 @@
 		if (irr  & (1 << (vector % 32))) {
 			irq = __get_cpu_var(vector_irq)[vector];
 
-			desc = irq_to_desc(irq);
+			data = irq_get_irq_data(irq);
 			raw_spin_lock(&desc->lock);
-			if (desc->chip->retrigger)
-				desc->chip->retrigger(irq);
+			if (data->chip->irq_retrigger)
+				data->chip->irq_retrigger(data);
 			raw_spin_unlock(&desc->lock);
 		}
 	}
diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c
new file mode 100644
index 0000000..ca8f703
--- /dev/null
+++ b/arch/x86/kernel/irq_work.c
@@ -0,0 +1,30 @@
+/*
+ * x86 specific code for irq_work
+ *
+ * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/irq_work.h>
+#include <linux/hardirq.h>
+#include <asm/apic.h>
+
+void smp_irq_work_interrupt(struct pt_regs *regs)
+{
+	irq_enter();
+	ack_APIC_irq();
+	inc_irq_stat(apic_irq_work_irqs);
+	irq_work_run();
+	irq_exit();
+}
+
+void arch_irq_work_raise(void)
+{
+#ifdef CONFIG_X86_LOCAL_APIC
+	if (!cpu_has_apic)
+		return;
+
+	apic->send_IPI_self(IRQ_WORK_VECTOR);
+	apic_wait_icr_idle();
+#endif
+}
diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c
index 990ae7c..c752e97 100644
--- a/arch/x86/kernel/irqinit.c
+++ b/arch/x86/kernel/irqinit.c
@@ -100,6 +100,8 @@
 
 void __init init_ISA_irqs(void)
 {
+	struct irq_chip *chip = legacy_pic->chip;
+	const char *name = chip->name;
 	int i;
 
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_LOCAL_APIC)
@@ -107,19 +109,8 @@
 #endif
 	legacy_pic->init(0);
 
-	/*
-	 * 16 old-style INTA-cycle interrupts:
-	 */
-	for (i = 0; i < legacy_pic->nr_legacy_irqs; i++) {
-		struct irq_desc *desc = irq_to_desc(i);
-
-		desc->status = IRQ_DISABLED;
-		desc->action = NULL;
-		desc->depth = 1;
-
-		set_irq_chip_and_handler_name(i, &i8259A_chip,
-					      handle_level_irq, "XT");
-	}
+	for (i = 0; i < legacy_pic->nr_legacy_irqs; i++)
+		set_irq_chip_and_handler_name(i, chip, handle_level_irq, name);
 }
 
 void __init init_IRQ(void)
@@ -224,9 +215,9 @@
 	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
 
-	/* Performance monitoring interrupts: */
-# ifdef CONFIG_PERF_EVENTS
-	alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt);
+	/* IRQ work interrupts: */
+# ifdef CONFIG_IRQ_WORK
+	alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt);
 # endif
 
 #endif
diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
new file mode 100644
index 0000000..961b6b3
--- /dev/null
+++ b/arch/x86/kernel/jump_label.c
@@ -0,0 +1,50 @@
+/*
+ * jump label x86 support
+ *
+ * Copyright (C) 2009 Jason Baron <jbaron@redhat.com>
+ *
+ */
+#include <linux/jump_label.h>
+#include <linux/memory.h>
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/jhash.h>
+#include <linux/cpu.h>
+#include <asm/kprobes.h>
+#include <asm/alternative.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+union jump_code_union {
+	char code[JUMP_LABEL_NOP_SIZE];
+	struct {
+		char jump;
+		int offset;
+	} __attribute__((packed));
+};
+
+void arch_jump_label_transform(struct jump_entry *entry,
+			       enum jump_label_type type)
+{
+	union jump_code_union code;
+
+	if (type == JUMP_LABEL_ENABLE) {
+		code.jump = 0xe9;
+		code.offset = entry->target -
+				(entry->code + JUMP_LABEL_NOP_SIZE);
+	} else
+		memcpy(&code, ideal_nop5, JUMP_LABEL_NOP_SIZE);
+	get_online_cpus();
+	mutex_lock(&text_mutex);
+	text_poke_smp((void *)entry->code, &code, JUMP_LABEL_NOP_SIZE);
+	mutex_unlock(&text_mutex);
+	put_online_cpus();
+}
+
+void arch_jump_label_text_poke_early(jump_label_t addr)
+{
+	text_poke_early((void *)addr, ideal_nop5, JUMP_LABEL_NOP_SIZE);
+}
+
+#endif
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 770ebfb..1cbd54c 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -230,9 +230,6 @@
 	return 0;
 }
 
-/* Dummy buffers for kallsyms_lookup */
-static char __dummy_buf[KSYM_NAME_LEN];
-
 /* Check if paddr is at an instruction boundary */
 static int __kprobes can_probe(unsigned long paddr)
 {
@@ -241,7 +238,7 @@
 	struct insn insn;
 	kprobe_opcode_t buf[MAX_INSN_SIZE];
 
-	if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+	if (!kallsyms_lookup_size_offset(paddr, NULL, &offset))
 		return 0;
 
 	/* Decode instructions */
@@ -1129,7 +1126,7 @@
 	*(unsigned long *)addr = val;
 }
 
-void __kprobes kprobes_optinsn_template_holder(void)
+static void __used __kprobes kprobes_optinsn_template_holder(void)
 {
 	asm volatile (
 			".global optprobe_template_entry\n"
@@ -1221,7 +1218,8 @@
 	}
 	/* Check whether the address range is reserved */
 	if (ftrace_text_reserved(src, src + len - 1) ||
-	    alternatives_text_reserved(src, src + len - 1))
+	    alternatives_text_reserved(src, src + len - 1) ||
+	    jump_label_text_reserved(src, src + len - 1))
 		return -EBUSY;
 
 	return len;
@@ -1269,11 +1267,9 @@
 	unsigned long addr, size = 0, offset = 0;
 	struct insn insn;
 	kprobe_opcode_t buf[MAX_INSN_SIZE];
-	/* Dummy buffers for lookup_symbol_attrs */
-	static char __dummy_buf[KSYM_NAME_LEN];
 
 	/* Lookup symbol including addr */
-	if (!kallsyms_lookup(paddr, &size, &offset, NULL, __dummy_buf))
+	if (!kallsyms_lookup_size_offset(paddr, &size, &offset))
 		return 0;
 
 	/* Check there is enough space for a relative jump. */
diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index 035c8c5..b3ea9db 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -36,7 +36,7 @@
 		if (!page)
 			goto out;
 		pud = (pud_t *)page_address(page);
-		memset(pud, 0, PAGE_SIZE);
+		clear_page(pud);
 		set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE));
 	}
 	pud = pud_offset(pgd, addr);
@@ -45,7 +45,7 @@
 		if (!page)
 			goto out;
 		pmd = (pmd_t *)page_address(page);
-		memset(pmd, 0, PAGE_SIZE);
+		clear_page(pmd);
 		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE));
 	}
 	pmd = pmd_offset(pud, addr);
diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
index 1c355c5..8f29560 100644
--- a/arch/x86/kernel/module.c
+++ b/arch/x86/kernel/module.c
@@ -239,6 +239,9 @@
 		apply_paravirt(pseg, pseg + para->sh_size);
 	}
 
+	/* make jump label nops */
+	jump_label_apply_nops(me);
+
 	return 0;
 }
 
diff --git a/arch/x86/kernel/olpc-xo1.c b/arch/x86/kernel/olpc-xo1.c
new file mode 100644
index 0000000..f5442c0
--- /dev/null
+++ b/arch/x86/kernel/olpc-xo1.c
@@ -0,0 +1,140 @@
+/*
+ * Support for features of the OLPC XO-1 laptop
+ *
+ * Copyright (C) 2010 One Laptop per Child
+ * Copyright (C) 2006 Red Hat, Inc.
+ * Copyright (C) 2006 Advanced Micro Devices, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/pci_ids.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+
+#include <asm/io.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME "olpc-xo1"
+
+#define PMS_BAR		4
+#define ACPI_BAR	5
+
+/* PMC registers (PMS block) */
+#define PM_SCLK		0x10
+#define PM_IN_SLPCTL	0x20
+#define PM_WKXD		0x34
+#define PM_WKD		0x30
+#define PM_SSC		0x54
+
+/* PM registers (ACPI block) */
+#define PM1_CNT		0x08
+#define PM_GPE0_STS	0x18
+
+static unsigned long acpi_base;
+static unsigned long pms_base;
+
+static void xo1_power_off(void)
+{
+	printk(KERN_INFO "OLPC XO-1 power off sequence...\n");
+
+	/* Enable all of these controls with 0 delay */
+	outl(0x40000000, pms_base + PM_SCLK);
+	outl(0x40000000, pms_base + PM_IN_SLPCTL);
+	outl(0x40000000, pms_base + PM_WKXD);
+	outl(0x40000000, pms_base + PM_WKD);
+
+	/* Clear status bits (possibly unnecessary) */
+	outl(0x0002ffff, pms_base  + PM_SSC);
+	outl(0xffffffff, acpi_base + PM_GPE0_STS);
+
+	/* Write SLP_EN bit to start the machinery */
+	outl(0x00002000, acpi_base + PM1_CNT);
+}
+
+/* Read the base addresses from the PCI BAR info */
+static int __devinit setup_bases(struct pci_dev *pdev)
+{
+	int r;
+
+	r = pci_enable_device_io(pdev);
+	if (r) {
+		dev_err(&pdev->dev, "can't enable device IO\n");
+		return r;
+	}
+
+	r = pci_request_region(pdev, ACPI_BAR, DRV_NAME);
+	if (r) {
+		dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", ACPI_BAR);
+		return r;
+	}
+
+	r = pci_request_region(pdev, PMS_BAR, DRV_NAME);
+	if (r) {
+		dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", PMS_BAR);
+		pci_release_region(pdev, ACPI_BAR);
+		return r;
+	}
+
+	acpi_base = pci_resource_start(pdev, ACPI_BAR);
+	pms_base = pci_resource_start(pdev, PMS_BAR);
+
+	return 0;
+}
+
+static int __devinit olpc_xo1_probe(struct platform_device *pdev)
+{
+	struct pci_dev *pcidev;
+	int r;
+
+	pcidev = pci_get_device(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA,
+				NULL);
+	if (!pdev)
+		return -ENODEV;
+
+	r = setup_bases(pcidev);
+	if (r)
+		return r;
+
+	pm_power_off = xo1_power_off;
+
+	printk(KERN_INFO "OLPC XO-1 support registered\n");
+	return 0;
+}
+
+static int __devexit olpc_xo1_remove(struct platform_device *pdev)
+{
+	pm_power_off = NULL;
+	return 0;
+}
+
+static struct platform_driver olpc_xo1_driver = {
+	.driver = {
+		.name = DRV_NAME,
+		.owner = THIS_MODULE,
+	},
+	.probe = olpc_xo1_probe,
+	.remove = __devexit_p(olpc_xo1_remove),
+};
+
+static int __init olpc_xo1_init(void)
+{
+	return platform_driver_register(&olpc_xo1_driver);
+}
+
+static void __exit olpc_xo1_exit(void)
+{
+	platform_driver_unregister(&olpc_xo1_driver);
+}
+
+MODULE_AUTHOR("Daniel Drake <dsd@laptop.org>");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:olpc-xo1");
+
+module_init(olpc_xo1_init);
+module_exit(olpc_xo1_exit);
diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c
index 0e0cdde..edaf3fe 100644
--- a/arch/x86/kernel/olpc.c
+++ b/arch/x86/kernel/olpc.c
@@ -17,6 +17,7 @@
 #include <linux/spinlock.h>
 #include <linux/io.h>
 #include <linux/string.h>
+#include <linux/platform_device.h>
 
 #include <asm/geode.h>
 #include <asm/setup.h>
@@ -114,6 +115,7 @@
 	unsigned long flags;
 	int ret = -EIO;
 	int i;
+	int restarts = 0;
 
 	spin_lock_irqsave(&ec_lock, flags);
 
@@ -169,7 +171,9 @@
 			if (wait_on_obf(0x6c, 1)) {
 				printk(KERN_ERR "olpc-ec:  timeout waiting for"
 						" EC to provide data!\n");
-				goto restart;
+				if (restarts++ < 10)
+					goto restart;
+				goto err;
 			}
 			outbuf[i] = inb(0x68);
 			pr_devel("olpc-ec:  received 0x%x\n", outbuf[i]);
@@ -183,8 +187,21 @@
 }
 EXPORT_SYMBOL_GPL(olpc_ec_cmd);
 
-#ifdef CONFIG_OLPC_OPENFIRMWARE
-static void __init platform_detect(void)
+static bool __init check_ofw_architecture(void)
+{
+	size_t propsize;
+	char olpc_arch[5];
+	const void *args[] = { NULL, "architecture", olpc_arch, (void *)5 };
+	void *res[] = { &propsize };
+
+	if (olpc_ofw("getprop", args, res)) {
+		printk(KERN_ERR "ofw: getprop call failed!\n");
+		return false;
+	}
+	return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0;
+}
+
+static u32 __init get_board_revision(void)
 {
 	size_t propsize;
 	__be32 rev;
@@ -193,46 +210,44 @@
 
 	if (olpc_ofw("getprop", args, res) || propsize != 4) {
 		printk(KERN_ERR "ofw: getprop call failed!\n");
-		rev = cpu_to_be32(0);
+		return cpu_to_be32(0);
 	}
-	olpc_platform_info.boardrev = be32_to_cpu(rev);
+	return be32_to_cpu(rev);
 }
-#else
-static void __init platform_detect(void)
+
+static bool __init platform_detect(void)
 {
-	/* stopgap until OFW support is added to the kernel */
-	olpc_platform_info.boardrev = olpc_board(0xc2);
+	if (!check_ofw_architecture())
+		return false;
+	olpc_platform_info.flags |= OLPC_F_PRESENT;
+	olpc_platform_info.boardrev = get_board_revision();
+	return true;
 }
-#endif
+
+static int __init add_xo1_platform_devices(void)
+{
+	struct platform_device *pdev;
+
+	pdev = platform_device_register_simple("xo1-rfkill", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	pdev = platform_device_register_simple("olpc-xo1", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	return 0;
+}
 
 static int __init olpc_init(void)
 {
-	unsigned char *romsig;
+	int r = 0;
 
-	/* The ioremap check is dangerous; limit what we run it on */
-	if (!is_geode() || cs5535_has_vsa2())
+	if (!olpc_ofw_present() || !platform_detect())
 		return 0;
 
 	spin_lock_init(&ec_lock);
 
-	romsig = ioremap(0xffffffc0, 16);
-	if (!romsig)
-		return 0;
-
-	if (strncmp(romsig, "CL1   Q", 7))
-		goto unmap;
-	if (strncmp(romsig+6, romsig+13, 3)) {
-		printk(KERN_INFO "OLPC BIOS signature looks invalid.  "
-				"Assuming not OLPC\n");
-		goto unmap;
-	}
-
-	printk(KERN_INFO "OLPC board with OpenFirmware %.16s\n", romsig);
-	olpc_platform_info.flags |= OLPC_F_PRESENT;
-
-	/* get the platform revision */
-	platform_detect();
-
 	/* assume B1 and above models always have a DCON */
 	if (olpc_board_at_least(olpc_board(0xb1)))
 		olpc_platform_info.flags |= OLPC_F_DCON;
@@ -242,8 +257,10 @@
 			(unsigned char *) &olpc_platform_info.ecver, 1);
 
 #ifdef CONFIG_PCI_OLPC
-	/* If the VSA exists let it emulate PCI, if not emulate in kernel */
-	if (!cs5535_has_vsa2())
+	/* If the VSA exists let it emulate PCI, if not emulate in kernel.
+	 * XO-1 only. */
+	if (olpc_platform_info.boardrev < olpc_board_pre(0xd0) &&
+			!cs5535_has_vsa2())
 		x86_init.pci.arch_init = pci_olpc_init;
 #endif
 
@@ -252,8 +269,12 @@
 			olpc_platform_info.boardrev >> 4,
 			olpc_platform_info.ecver);
 
-unmap:
-	iounmap(romsig);
+	if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) { /* XO-1 */
+		r = add_xo1_platform_devices();
+		if (r)
+			return r;
+	}
+
 	return 0;
 }
 
diff --git a/arch/x86/kernel/olpc_ofw.c b/arch/x86/kernel/olpc_ofw.c
index 3218aa7..7873204 100644
--- a/arch/x86/kernel/olpc_ofw.c
+++ b/arch/x86/kernel/olpc_ofw.c
@@ -74,6 +74,12 @@
 }
 EXPORT_SYMBOL_GPL(__olpc_ofw);
 
+bool olpc_ofw_present(void)
+{
+	return olpc_ofw_cif != NULL;
+}
+EXPORT_SYMBOL_GPL(olpc_ofw_present);
+
 /* OFW cif _should_ be above this address */
 #define OFW_MIN 0xff000000
 
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 1db183e..c5b2500 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -413,7 +413,6 @@
 
 	.alloc_pte = paravirt_nop,
 	.alloc_pmd = paravirt_nop,
-	.alloc_pmd_clone = paravirt_nop,
 	.alloc_pud = paravirt_nop,
 	.release_pte = paravirt_nop,
 	.release_pmd = paravirt_nop,
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 078d4ec..f56a117 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -47,6 +47,7 @@
 #include <asm/rio.h>
 #include <asm/bios_ebda.h>
 #include <asm/x86_init.h>
+#include <asm/iommu_table.h>
 
 #ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT
 int use_calgary __read_mostly = 1;
@@ -1364,7 +1365,7 @@
 	return 0;
 }
 
-void __init detect_calgary(void)
+int __init detect_calgary(void)
 {
 	int bus;
 	void *tbl;
@@ -1378,13 +1379,13 @@
 	 * another HW IOMMU already, bail out.
 	 */
 	if (no_iommu || iommu_detected)
-		return;
+		return -ENODEV;
 
 	if (!use_calgary)
-		return;
+		return -ENODEV;
 
 	if (!early_pci_allowed())
-		return;
+		return -ENODEV;
 
 	printk(KERN_DEBUG "Calgary: detecting Calgary via BIOS EBDA area\n");
 
@@ -1410,13 +1411,13 @@
 	if (!rio_table_hdr) {
 		printk(KERN_DEBUG "Calgary: Unable to locate Rio Grande table "
 		       "in EBDA - bailing!\n");
-		return;
+		return -ENODEV;
 	}
 
 	ret = build_detail_arrays();
 	if (ret) {
 		printk(KERN_DEBUG "Calgary: build_detail_arrays ret %d\n", ret);
-		return;
+		return -ENOMEM;
 	}
 
 	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
@@ -1464,7 +1465,7 @@
 
 		x86_init.iommu.iommu_init = calgary_iommu_init;
 	}
-	return;
+	return calgary_found;
 
 cleanup:
 	for (--bus; bus >= 0; --bus) {
@@ -1473,6 +1474,7 @@
 		if (info->tce_space)
 			free_tce_table(info->tce_space);
 	}
+	return -ENOMEM;
 }
 
 static int __init calgary_parse_options(char *p)
@@ -1594,3 +1596,5 @@
  * and before device_initcall.
  */
 rootfs_initcall(calgary_fixup_tce_spaces);
+
+IOMMU_INIT_POST(detect_calgary);
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 9f07cfc..9ea999a 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -11,9 +11,8 @@
 #include <asm/iommu.h>
 #include <asm/gart.h>
 #include <asm/calgary.h>
-#include <asm/amd_iommu.h>
 #include <asm/x86_init.h>
-#include <asm/xen/swiotlb-xen.h>
+#include <asm/iommu_table.h>
 
 static int forbid_dac __read_mostly;
 
@@ -45,6 +44,8 @@
  */
 int iommu_pass_through __read_mostly;
 
+extern struct iommu_table_entry __iommu_table[], __iommu_table_end[];
+
 /* Dummy device used for NULL arguments (normally ISA). */
 struct device x86_dma_fallback_dev = {
 	.init_name = "fallback device",
@@ -130,26 +131,24 @@
 
 void __init pci_iommu_alloc(void)
 {
+	struct iommu_table_entry *p;
+
 	/* free the range so iommu could get some range less than 4G */
 	dma32_free_bootmem();
 
-	if (pci_xen_swiotlb_detect() || pci_swiotlb_detect())
-		goto out;
+	sort_iommu_table(__iommu_table, __iommu_table_end);
+	check_iommu_entries(__iommu_table, __iommu_table_end);
 
-	gart_iommu_hole_init();
-
-	detect_calgary();
-
-	detect_intel_iommu();
-
-	/* needs to be called after gart_iommu_hole_init */
-	amd_iommu_detect();
-out:
-	pci_xen_swiotlb_init();
-
-	pci_swiotlb_init();
+	for (p = __iommu_table; p < __iommu_table_end; p++) {
+		if (p && p->detect && p->detect() > 0) {
+			p->flags |= IOMMU_DETECTED;
+			if (p->early_init)
+				p->early_init();
+			if (p->flags & IOMMU_FINISH_IF_DETECTED)
+				break;
+		}
+	}
 }
-
 void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 				 dma_addr_t *dma_addr, gfp_t flag)
 {
@@ -292,6 +291,7 @@
 
 static int __init pci_iommu_init(void)
 {
+	struct iommu_table_entry *p;
 	dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
 
 #ifdef CONFIG_PCI
@@ -299,12 +299,10 @@
 #endif
 	x86_init.iommu.iommu_init();
 
-	if (swiotlb || xen_swiotlb) {
-		printk(KERN_INFO "PCI-DMA: "
-		       "Using software bounce buffering for IO (SWIOTLB)\n");
-		swiotlb_print_info();
-	} else
-		swiotlb_free();
+	for (p = __iommu_table; p < __iommu_table_end; p++) {
+		if (p && (p->flags & IOMMU_DETECTED) && p->late_init)
+			p->late_init();
+	}
 
 	return 0;
 }
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index 0f7f130..ba0f0ca 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -39,8 +39,9 @@
 #include <asm/cacheflush.h>
 #include <asm/swiotlb.h>
 #include <asm/dma.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #include <asm/x86_init.h>
+#include <asm/iommu_table.h>
 
 static unsigned long iommu_bus_base;	/* GART remapping area (physical) */
 static unsigned long iommu_size;	/* size of remapping area bytes */
@@ -560,8 +561,11 @@
 {
 	int i;
 
-	for (i = 0; i < num_k8_northbridges; i++) {
-		struct pci_dev *dev = k8_northbridges[i];
+	if (!k8_northbridges.gart_supported)
+		return;
+
+	for (i = 0; i < k8_northbridges.num; i++) {
+		struct pci_dev *dev = k8_northbridges.nb_misc[i];
 
 		enable_gart_translation(dev, __pa(agp_gatt_table));
 	}
@@ -592,16 +596,19 @@
 	if (!fix_up_north_bridges)
 		return;
 
+	if (!k8_northbridges.gart_supported)
+		return;
+
 	pr_info("PCI-DMA: Restoring GART aperture settings\n");
 
-	for (i = 0; i < num_k8_northbridges; i++) {
-		struct pci_dev *dev = k8_northbridges[i];
+	for (i = 0; i < k8_northbridges.num; i++) {
+		struct pci_dev *dev = k8_northbridges.nb_misc[i];
 
 		/*
 		 * Don't enable translations just yet.  That is the next
 		 * step.  Restore the pre-suspend aperture settings.
 		 */
-		pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, aperture_order << 1);
+		gart_set_size_and_enable(dev, aperture_order);
 		pci_write_config_dword(dev, AMD64_GARTAPERTUREBASE, aperture_alloc >> 25);
 	}
 }
@@ -649,8 +656,8 @@
 
 	aper_size = aper_base = info->aper_size = 0;
 	dev = NULL;
-	for (i = 0; i < num_k8_northbridges; i++) {
-		dev = k8_northbridges[i];
+	for (i = 0; i < k8_northbridges.num; i++) {
+		dev = k8_northbridges.nb_misc[i];
 		new_aper_base = read_aperture(dev, &new_aper_size);
 		if (!new_aper_base)
 			goto nommu;
@@ -718,10 +725,13 @@
 	if (!no_agp)
 		return;
 
-	for (i = 0; i < num_k8_northbridges; i++) {
+	if (!k8_northbridges.gart_supported)
+		return;
+
+	for (i = 0; i < k8_northbridges.num; i++) {
 		u32 ctl;
 
-		dev = k8_northbridges[i];
+		dev = k8_northbridges.nb_misc[i];
 		pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &ctl);
 
 		ctl &= ~GARTEN;
@@ -739,7 +749,7 @@
 	unsigned long scratch;
 	long i;
 
-	if (num_k8_northbridges == 0)
+	if (!k8_northbridges.gart_supported)
 		return 0;
 
 #ifndef CONFIG_AGP_AMD64
@@ -896,3 +906,4 @@
 		}
 	}
 }
+IOMMU_INIT_POST(gart_iommu_hole_init);
diff --git a/arch/x86/kernel/pci-iommu_table.c b/arch/x86/kernel/pci-iommu_table.c
new file mode 100644
index 0000000..55d745e
--- /dev/null
+++ b/arch/x86/kernel/pci-iommu_table.c
@@ -0,0 +1,89 @@
+#include <linux/dma-mapping.h>
+#include <asm/iommu_table.h>
+#include <linux/string.h>
+#include <linux/kallsyms.h>
+
+
+#define DEBUG 1
+
+static struct iommu_table_entry * __init
+find_dependents_of(struct iommu_table_entry *start,
+		   struct iommu_table_entry *finish,
+		   struct iommu_table_entry *q)
+{
+	struct iommu_table_entry *p;
+
+	if (!q)
+		return NULL;
+
+	for (p = start; p < finish; p++)
+		if (p->detect == q->depend)
+			return p;
+
+	return NULL;
+}
+
+
+void __init sort_iommu_table(struct iommu_table_entry *start,
+			     struct iommu_table_entry *finish) {
+
+	struct iommu_table_entry *p, *q, tmp;
+
+	for (p = start; p < finish; p++) {
+again:
+		q = find_dependents_of(start, finish, p);
+		/* We are bit sneaky here. We use the memory address to figure
+		 * out if the node we depend on is past our point, if so, swap.
+		 */
+		if (q > p) {
+			tmp = *p;
+			memmove(p, q, sizeof(*p));
+			*q = tmp;
+			goto again;
+		}
+	}
+
+}
+
+#ifdef DEBUG
+void __init check_iommu_entries(struct iommu_table_entry *start,
+				struct iommu_table_entry *finish)
+{
+	struct iommu_table_entry *p, *q, *x;
+	char sym_p[KSYM_SYMBOL_LEN];
+	char sym_q[KSYM_SYMBOL_LEN];
+
+	/* Simple cyclic dependency checker. */
+	for (p = start; p < finish; p++) {
+		q = find_dependents_of(start, finish, p);
+		x = find_dependents_of(start, finish, q);
+		if (p == x) {
+			sprint_symbol(sym_p, (unsigned long)p->detect);
+			sprint_symbol(sym_q, (unsigned long)q->detect);
+
+			printk(KERN_ERR "CYCLIC DEPENDENCY FOUND! %s depends" \
+					" on %s and vice-versa. BREAKING IT.\n",
+					sym_p, sym_q);
+			/* Heavy handed way..*/
+			x->depend = 0;
+		}
+	}
+
+	for (p = start; p < finish; p++) {
+		q = find_dependents_of(p, finish, p);
+		if (q && q > p) {
+			sprint_symbol(sym_p, (unsigned long)p->detect);
+			sprint_symbol(sym_q, (unsigned long)q->detect);
+
+			printk(KERN_ERR "EXECUTION ORDER INVALID! %s "\
+					"should be called before %s!\n",
+					sym_p, sym_q);
+		}
+	}
+}
+#else
+inline void check_iommu_entries(struct iommu_table_entry *start,
+				       struct iommu_table_entry *finish)
+{
+}
+#endif
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index a5bc528..8f972cb 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -10,7 +10,8 @@
 #include <asm/iommu.h>
 #include <asm/swiotlb.h>
 #include <asm/dma.h>
-
+#include <asm/xen/swiotlb-xen.h>
+#include <asm/iommu_table.h>
 int swiotlb __read_mostly;
 
 static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
@@ -41,25 +42,42 @@
 };
 
 /*
- * pci_swiotlb_detect - set swiotlb to 1 if necessary
+ * pci_swiotlb_detect_override - set swiotlb to 1 if necessary
  *
  * This returns non-zero if we are forced to use swiotlb (by the boot
  * option).
  */
-int __init pci_swiotlb_detect(void)
+int __init pci_swiotlb_detect_override(void)
 {
 	int use_swiotlb = swiotlb | swiotlb_force;
 
-	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
-#ifdef CONFIG_X86_64
-	if (!no_iommu && max_pfn > MAX_DMA32_PFN)
-		swiotlb = 1;
-#endif
 	if (swiotlb_force)
 		swiotlb = 1;
 
 	return use_swiotlb;
 }
+IOMMU_INIT_FINISH(pci_swiotlb_detect_override,
+		  pci_xen_swiotlb_detect,
+		  pci_swiotlb_init,
+		  pci_swiotlb_late_init);
+
+/*
+ * if 4GB or more detected (and iommu=off not set) return 1
+ * and set swiotlb to 1.
+ */
+int __init pci_swiotlb_detect_4gb(void)
+{
+	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
+#ifdef CONFIG_X86_64
+	if (!no_iommu && max_pfn > MAX_DMA32_PFN)
+		swiotlb = 1;
+#endif
+	return swiotlb;
+}
+IOMMU_INIT(pci_swiotlb_detect_4gb,
+	   pci_swiotlb_detect_override,
+	   pci_swiotlb_init,
+	   pci_swiotlb_late_init);
 
 void __init pci_swiotlb_init(void)
 {
@@ -68,3 +86,15 @@
 		dma_ops = &swiotlb_dma_ops;
 	}
 }
+
+void __init pci_swiotlb_late_init(void)
+{
+	/* An IOMMU turned us off. */
+	if (!swiotlb)
+		swiotlb_free();
+	else {
+		printk(KERN_INFO "PCI-DMA: "
+		       "Using software bounce buffering for IO (SWIOTLB)\n");
+		swiotlb_print_info();
+	}
+}
diff --git a/arch/x86/kernel/pmtimer_64.c b/arch/x86/kernel/pmtimer_64.c
deleted file mode 100644
index b112406..0000000
--- a/arch/x86/kernel/pmtimer_64.c
+++ /dev/null
@@ -1,69 +0,0 @@
-/* Ported over from i386 by AK, original copyright was:
- *
- * (C) Dominik Brodowski <linux@brodo.de> 2003
- *
- * Driver to use the Power Management Timer (PMTMR) available in some
- * southbridges as primary timing source for the Linux kernel.
- *
- * Based on parts of linux/drivers/acpi/hardware/hwtimer.c, timer_pit.c,
- * timer_hpet.c, and on Arjan van de Ven's implementation for 2.4.
- *
- * This file is licensed under the GPL v2.
- *
- * Dropped all the hardware bug workarounds for now. Hopefully they
- * are not needed on 64bit chipsets.
- */
-
-#include <linux/jiffies.h>
-#include <linux/kernel.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/cpumask.h>
-#include <linux/acpi_pmtmr.h>
-
-#include <asm/io.h>
-#include <asm/proto.h>
-#include <asm/msr.h>
-#include <asm/vsyscall.h>
-
-static inline u32 cyc2us(u32 cycles)
-{
-	/* The Power Management Timer ticks at 3.579545 ticks per microsecond.
-	 * 1 / PM_TIMER_FREQUENCY == 0.27936511 =~ 286/1024 [error: 0.024%]
-	 *
-	 * Even with HZ = 100, delta is at maximum 35796 ticks, so it can
-	 * easily be multiplied with 286 (=0x11E) without having to fear
-	 * u32 overflows.
-	 */
-	cycles *= 286;
-	return (cycles >> 10);
-}
-
-static unsigned pmtimer_wait_tick(void)
-{
-	u32 a, b;
-	for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK;
-	     a == b;
-	     b = inl(pmtmr_ioport) & ACPI_PM_MASK)
-		cpu_relax();
-	return b;
-}
-
-/* note: wait time is rounded up to one tick */
-void pmtimer_wait(unsigned us)
-{
-	u32 a, b;
-	a = pmtimer_wait_tick();
-	do {
-		b = inl(pmtmr_ioport);
-		cpu_relax();
-	} while (cyc2us(b - a) < us);
-}
-
-static int __init nopmtimer_setup(char *s)
-{
-	pmtmr_ioport = 0;
-	return 1;
-}
-
-__setup("nopmtimer", nopmtimer_setup);
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 3d9ea53..b3d7a3a 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -424,7 +424,7 @@
 	load_TLS(next, cpu);
 
 	/* Must be after DS reload */
-	unlazy_fpu(prev_p);
+	__unlazy_fpu(prev_p);
 
 	/* Make sure cpu is ready for new context */
 	if (preload_fpu)
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index e3af342..7a4cf14 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -84,7 +84,7 @@
 			}
 				/* we will leave sorting out the final value
 				   when we are ready to reboot, since we might not
-				   have set up boot_cpu_id or smp_num_cpu */
+				   have detected BSP APIC ID or smp_num_cpu */
 			break;
 #endif /* CONFIG_SMP */
 
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index c3a4fbb..a59f6a6 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -83,7 +83,6 @@
 #include <asm/dmi.h>
 #include <asm/io_apic.h>
 #include <asm/ist.h>
-#include <asm/vmi.h>
 #include <asm/setup_arch.h>
 #include <asm/bios_ebda.h>
 #include <asm/cacheflush.h>
@@ -107,11 +106,12 @@
 #include <asm/percpu.h>
 #include <asm/topology.h>
 #include <asm/apicdef.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #ifdef CONFIG_X86_64
 #include <asm/numa_64.h>
 #endif
 #include <asm/mce.h>
+#include <asm/alternative.h>
 
 /*
  * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries.
@@ -125,7 +125,6 @@
 RESERVE_BRK(dmi_alloc, 65536);
 #endif
 
-unsigned int boot_cpu_id __read_mostly;
 
 static __initdata unsigned long _brk_start = (unsigned long)__brk_base;
 unsigned long _brk_end = (unsigned long)__brk_base;
@@ -618,79 +617,7 @@
 		reserve_early_overlap_ok(addr, addr + size, "ibft");
 }
 
-#ifdef CONFIG_X86_RESERVE_LOW_64K
-static int __init dmi_low_memory_corruption(const struct dmi_system_id *d)
-{
-	printk(KERN_NOTICE
-		"%s detected: BIOS may corrupt low RAM, working around it.\n",
-		d->ident);
-
-	e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED);
-	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
-
-	return 0;
-}
-#endif
-
-/* List of systems that have known low memory corruption BIOS problems */
-static struct dmi_system_id __initdata bad_bios_dmi_table[] = {
-#ifdef CONFIG_X86_RESERVE_LOW_64K
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "AMI BIOS",
-		.matches = {
-			DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."),
-		},
-	},
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "Phoenix BIOS",
-		.matches = {
-			DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"),
-		},
-	},
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "Phoenix/MSC BIOS",
-		.matches = {
-			DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix/MSC"),
-		},
-	},
-	/*
-	 * AMI BIOS with low memory corruption was found on Intel DG45ID and
-	 * DG45FC boards.
-	 * It has a different DMI_BIOS_VENDOR = "Intel Corp.", for now we will
-	 * match only DMI_BOARD_NAME and see if there is more bad products
-	 * with this vendor.
-	 */
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "AMI BIOS",
-		.matches = {
-			DMI_MATCH(DMI_BOARD_NAME, "DG45ID"),
-		},
-	},
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "AMI BIOS",
-		.matches = {
-			DMI_MATCH(DMI_BOARD_NAME, "DG45FC"),
-		},
-	},
-	/*
-	 * The Dell Inspiron Mini 1012 has DMI_BIOS_VENDOR = "Dell Inc.", so
-	 * match on the product name.
-	 */
-	{
-		.callback = dmi_low_memory_corruption,
-		.ident = "Phoenix BIOS",
-		.matches = {
-			DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 1012"),
-		},
-	},
-#endif
-	{}
-};
+static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10;
 
 static void __init trim_bios_range(void)
 {
@@ -698,8 +625,14 @@
 	 * A special case is the first 4Kb of memory;
 	 * This is a BIOS owned area, not kernel ram, but generally
 	 * not listed as such in the E820 table.
+	 *
+	 * This typically reserves additional memory (64KiB by default)
+	 * since some BIOSes are known to corrupt low memory.  See the
+	 * Kconfig help text for X86_RESERVE_LOW.
 	 */
-	e820_update_range(0, PAGE_SIZE, E820_RAM, E820_RESERVED);
+	e820_update_range(0, ALIGN(reserve_low, PAGE_SIZE),
+			  E820_RAM, E820_RESERVED);
+
 	/*
 	 * special case: Some BIOSen report the PC BIOS
 	 * area (640->1Mb) as ram even though it is not.
@@ -709,6 +642,28 @@
 	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
 }
 
+static int __init parse_reservelow(char *p)
+{
+	unsigned long long size;
+
+	if (!p)
+		return -EINVAL;
+
+	size = memparse(p, &p);
+
+	if (size < 4096)
+		size = 4096;
+
+	if (size > 640*1024)
+		size = 640*1024;
+
+	reserve_low = size;
+
+	return 0;
+}
+
+early_param("reservelow", parse_reservelow);
+
 /*
  * Determine if we were loaded by an EFI loader.  If so, then we have also been
  * passed the efi memmap, systab, etc., so we should use these data structures
@@ -726,6 +681,7 @@
 {
 	int acpi = 0;
 	int k8 = 0;
+	unsigned long flags;
 
 #ifdef CONFIG_X86_32
 	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
@@ -734,10 +690,10 @@
 	printk(KERN_INFO "Command line: %s\n", boot_command_line);
 #endif
 
-	/* VMI may relocate the fixmap; do this before touching ioremap area */
-	vmi_init();
-
-	/* OFW also may relocate the fixmap */
+	/*
+	 * If we have OLPC OFW, we might end up relocating the fixmap due to
+	 * reserve_top(), so do this before touching the ioremap area.
+	 */
 	olpc_ofw_detect();
 
 	early_trap_init();
@@ -838,9 +794,6 @@
 
 	x86_report_nx();
 
-	/* Must be before kernel pagetables are setup */
-	vmi_activate();
-
 	/* after early param, so could get panic from serial */
 	reserve_early_setup_data();
 
@@ -863,8 +816,6 @@
 
 	dmi_scan_machine();
 
-	dmi_check_system(bad_bios_dmi_table);
-
 	/*
 	 * VMware detection requires dmi to be available, so this
 	 * needs to be done after dmi_scan_machine, for the BP.
@@ -1071,6 +1022,10 @@
 	x86_init.oem.banner();
 
 	mcheck_init();
+
+	local_irq_save(flags);
+	arch_init_ideal_nop5();
+	local_irq_restore(flags);
 }
 
 #ifdef CONFIG_X86_32
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index a60df9a..2335c15 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -253,7 +253,7 @@
 		 * Up to this point, the boot CPU has been using .init.data
 		 * area.  Reload any changed state for the boot CPU.
 		 */
-		if (cpu == boot_cpu_id)
+		if (!cpu)
 			switch_to_new_gdt(cpu);
 	}
 
diff --git a/arch/x86/kernel/sfi.c b/arch/x86/kernel/sfi.c
index cb22acf..dd4c281 100644
--- a/arch/x86/kernel/sfi.c
+++ b/arch/x86/kernel/sfi.c
@@ -34,7 +34,7 @@
 #ifdef CONFIG_X86_LOCAL_APIC
 static unsigned long sfi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE;
 
-void __init mp_sfi_register_lapic_address(unsigned long address)
+static void __init mp_sfi_register_lapic_address(unsigned long address)
 {
 	mp_lapic_addr = address;
 
@@ -46,7 +46,7 @@
 }
 
 /* All CPUs enumerated by SFI must be present and enabled */
-void __cpuinit mp_sfi_register_lapic(u8 id)
+static void __cpuinit mp_sfi_register_lapic(u8 id)
 {
 	if (MAX_APICS - id <= 0) {
 		pr_warning("Processor #%d invalid (max %d)\n",
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 8b3bfc4..dfb5089 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -62,7 +62,7 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/mtrr.h>
-#include <asm/vmi.h>
+#include <asm/mwait.h>
 #include <asm/apic.h>
 #include <asm/setup.h>
 #include <asm/uv/uv.h>
@@ -311,7 +311,6 @@
 	__flush_tlb_all();
 #endif
 
-	vmi_bringup();
 	cpu_init();
 	preempt_disable();
 	smp_callin();
@@ -324,9 +323,9 @@
 	check_tsc_sync_target();
 
 	if (nmi_watchdog == NMI_IO_APIC) {
-		legacy_pic->chip->mask(0);
+		legacy_pic->mask(0);
 		enable_NMI_through_LVT0();
-		legacy_pic->chip->unmask(0);
+		legacy_pic->unmask(0);
 	}
 
 	/* This must be done before setting cpu_online_mask */
@@ -397,6 +396,19 @@
 		identify_secondary_cpu(c);
 }
 
+static void __cpuinit link_thread_siblings(int cpu1, int cpu2)
+{
+	struct cpuinfo_x86 *c1 = &cpu_data(cpu1);
+	struct cpuinfo_x86 *c2 = &cpu_data(cpu2);
+
+	cpumask_set_cpu(cpu1, cpu_sibling_mask(cpu2));
+	cpumask_set_cpu(cpu2, cpu_sibling_mask(cpu1));
+	cpumask_set_cpu(cpu1, cpu_core_mask(cpu2));
+	cpumask_set_cpu(cpu2, cpu_core_mask(cpu1));
+	cpumask_set_cpu(cpu1, c2->llc_shared_map);
+	cpumask_set_cpu(cpu2, c1->llc_shared_map);
+}
+
 
 void __cpuinit set_cpu_sibling_map(int cpu)
 {
@@ -409,14 +421,13 @@
 		for_each_cpu(i, cpu_sibling_setup_mask) {
 			struct cpuinfo_x86 *o = &cpu_data(i);
 
-			if (c->phys_proc_id == o->phys_proc_id &&
-			    c->cpu_core_id == o->cpu_core_id) {
-				cpumask_set_cpu(i, cpu_sibling_mask(cpu));
-				cpumask_set_cpu(cpu, cpu_sibling_mask(i));
-				cpumask_set_cpu(i, cpu_core_mask(cpu));
-				cpumask_set_cpu(cpu, cpu_core_mask(i));
-				cpumask_set_cpu(i, c->llc_shared_map);
-				cpumask_set_cpu(cpu, o->llc_shared_map);
+			if (cpu_has(c, X86_FEATURE_TOPOEXT)) {
+				if (c->phys_proc_id == o->phys_proc_id &&
+				    c->compute_unit_id == o->compute_unit_id)
+					link_thread_siblings(cpu, i);
+			} else if (c->phys_proc_id == o->phys_proc_id &&
+				   c->cpu_core_id == o->cpu_core_id) {
+				link_thread_siblings(cpu, i);
 			}
 		}
 	} else {
@@ -1109,8 +1120,6 @@
 	}
 	set_cpu_sibling_map(0);
 
-	enable_IR_x2apic();
-	default_setup_apic_routing();
 
 	if (smp_sanity_check(max_cpus) < 0) {
 		printk(KERN_INFO "SMP disabled\n");
@@ -1118,6 +1127,8 @@
 		goto out;
 	}
 
+	default_setup_apic_routing();
+
 	preempt_disable();
 	if (read_apic_id() != boot_cpu_physical_apicid) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
@@ -1383,11 +1394,88 @@
 	local_irq_disable();
 }
 
+/*
+ * We need to flush the caches before going to sleep, lest we have
+ * dirty data in our caches when we come back up.
+ */
+static inline void mwait_play_dead(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+	unsigned int highest_cstate = 0;
+	unsigned int highest_subcstate = 0;
+	int i;
+	void *mwait_ptr;
+
+	if (!cpu_has(&current_cpu_data, X86_FEATURE_MWAIT))
+		return;
+	if (!cpu_has(&current_cpu_data, X86_FEATURE_CLFLSH))
+		return;
+	if (current_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
+		return;
+
+	eax = CPUID_MWAIT_LEAF;
+	ecx = 0;
+	native_cpuid(&eax, &ebx, &ecx, &edx);
+
+	/*
+	 * eax will be 0 if EDX enumeration is not valid.
+	 * Initialized below to cstate, sub_cstate value when EDX is valid.
+	 */
+	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED)) {
+		eax = 0;
+	} else {
+		edx >>= MWAIT_SUBSTATE_SIZE;
+		for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) {
+			if (edx & MWAIT_SUBSTATE_MASK) {
+				highest_cstate = i;
+				highest_subcstate = edx & MWAIT_SUBSTATE_MASK;
+			}
+		}
+		eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) |
+			(highest_subcstate - 1);
+	}
+
+	/*
+	 * This should be a memory location in a cache line which is
+	 * unlikely to be touched by other processors.  The actual
+	 * content is immaterial as it is not actually modified in any way.
+	 */
+	mwait_ptr = &current_thread_info()->flags;
+
+	wbinvd();
+
+	while (1) {
+		/*
+		 * The CLFLUSH is a workaround for erratum AAI65 for
+		 * the Xeon 7400 series.  It's not clear it is actually
+		 * needed, but it should be harmless in either case.
+		 * The WBINVD is insufficient due to the spurious-wakeup
+		 * case where we return around the loop.
+		 */
+		clflush(mwait_ptr);
+		__monitor(mwait_ptr, 0, 0);
+		mb();
+		__mwait(eax, 0);
+	}
+}
+
+static inline void hlt_play_dead(void)
+{
+	if (current_cpu_data.x86 >= 4)
+		wbinvd();
+
+	while (1) {
+		native_halt();
+	}
+}
+
 void native_play_dead(void)
 {
 	play_dead_common();
 	tboot_shutdown(TB_SHUTDOWN_WFS);
-	wbinvd_halt();
+
+	mwait_play_dead();	/* Only returns on failure */
+	hlt_play_dead();
 }
 
 #else /* ... !CONFIG_HOTPLUG_CPU */
diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c
index d5e0662..0b0cb5f 100644
--- a/arch/x86/kernel/sys_i386_32.c
+++ b/arch/x86/kernel/sys_i386_32.c
@@ -33,8 +33,8 @@
 		  const char *const envp[])
 {
 	long __res;
-	asm volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx"
+	asm volatile ("int $0x80"
 	: "=a" (__res)
-	: "0" (__NR_execve), "ri" (filename), "c" (argv), "d" (envp) : "memory");
+	: "0" (__NR_execve), "b" (filename), "c" (argv), "d" (envp) : "memory");
 	return __res;
 }
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 60788de..d439685 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -776,21 +776,10 @@
 }
 EXPORT_SYMBOL_GPL(math_state_restore);
 
-#ifndef CONFIG_MATH_EMULATION
-void math_emulate(struct math_emu_info *info)
-{
-	printk(KERN_EMERG
-		"math-emulation not enabled and no coprocessor found.\n");
-	printk(KERN_EMERG "killing %s.\n", current->comm);
-	force_sig(SIGFPE, current);
-	schedule();
-}
-#endif /* CONFIG_MATH_EMULATION */
-
 dotraplinkage void __kprobes
 do_device_not_available(struct pt_regs *regs, long error_code)
 {
-#ifdef CONFIG_X86_32
+#ifdef CONFIG_MATH_EMULATION
 	if (read_cr0() & X86_CR0_EM) {
 		struct math_emu_info info = { };
 
@@ -798,12 +787,12 @@
 
 		info.regs = regs;
 		math_emulate(&info);
-	} else {
-		math_state_restore(); /* interrupts still off */
-		conditional_sti(regs);
+		return;
 	}
-#else
-	math_state_restore();
+#endif
+	math_state_restore(); /* interrupts still off */
+#ifdef CONFIG_X86_32
+	conditional_sti(regs);
 #endif
 }
 
@@ -881,18 +870,6 @@
 #endif
 
 #ifdef CONFIG_X86_32
-	if (cpu_has_fxsr) {
-		printk(KERN_INFO "Enabling fast FPU save and restore... ");
-		set_in_cr4(X86_CR4_OSFXSR);
-		printk("done.\n");
-	}
-	if (cpu_has_xmm) {
-		printk(KERN_INFO
-			"Enabling unmasked SIMD FPU exception support... ");
-		set_in_cr4(X86_CR4_OSXMMEXCPT);
-		printk("done.\n");
-	}
-
 	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
 	set_bit(SYSCALL_VECTOR, used_vectors);
 #endif
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 26a863a..0c40d8b 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -104,10 +104,14 @@
 
 __setup("notsc", notsc_setup);
 
+static int no_sched_irq_time;
+
 static int __init tsc_setup(char *str)
 {
 	if (!strcmp(str, "reliable"))
 		tsc_clocksource_reliable = 1;
+	if (!strncmp(str, "noirqtime", 9))
+		no_sched_irq_time = 1;
 	return 1;
 }
 
@@ -801,6 +805,7 @@
 	if (!tsc_unstable) {
 		tsc_unstable = 1;
 		sched_clock_stable = 0;
+		disable_sched_clock_irqtime();
 		printk(KERN_INFO "Marking TSC unstable due to %s\n", reason);
 		/* Change only the rating, when not registered */
 		if (clocksource_tsc.mult)
@@ -892,60 +897,6 @@
 	clocksource_register_khz(&clocksource_tsc, tsc_khz);
 }
 
-#ifdef CONFIG_X86_64
-/*
- * calibrate_cpu is used on systems with fixed rate TSCs to determine
- * processor frequency
- */
-#define TICK_COUNT 100000000
-static unsigned long __init calibrate_cpu(void)
-{
-	int tsc_start, tsc_now;
-	int i, no_ctr_free;
-	unsigned long evntsel3 = 0, pmc3 = 0, pmc_now = 0;
-	unsigned long flags;
-
-	for (i = 0; i < 4; i++)
-		if (avail_to_resrv_perfctr_nmi_bit(i))
-			break;
-	no_ctr_free = (i == 4);
-	if (no_ctr_free) {
-		WARN(1, KERN_WARNING "Warning: AMD perfctrs busy ... "
-		     "cpu_khz value may be incorrect.\n");
-		i = 3;
-		rdmsrl(MSR_K7_EVNTSEL3, evntsel3);
-		wrmsrl(MSR_K7_EVNTSEL3, 0);
-		rdmsrl(MSR_K7_PERFCTR3, pmc3);
-	} else {
-		reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i);
-		reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
-	}
-	local_irq_save(flags);
-	/* start measuring cycles, incrementing from 0 */
-	wrmsrl(MSR_K7_PERFCTR0 + i, 0);
-	wrmsrl(MSR_K7_EVNTSEL0 + i, 1 << 22 | 3 << 16 | 0x76);
-	rdtscl(tsc_start);
-	do {
-		rdmsrl(MSR_K7_PERFCTR0 + i, pmc_now);
-		tsc_now = get_cycles();
-	} while ((tsc_now - tsc_start) < TICK_COUNT);
-
-	local_irq_restore(flags);
-	if (no_ctr_free) {
-		wrmsrl(MSR_K7_EVNTSEL3, 0);
-		wrmsrl(MSR_K7_PERFCTR3, pmc3);
-		wrmsrl(MSR_K7_EVNTSEL3, evntsel3);
-	} else {
-		release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
-		release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
-	}
-
-	return pmc_now * tsc_khz / (tsc_now - tsc_start);
-}
-#else
-static inline unsigned long calibrate_cpu(void) { return cpu_khz; }
-#endif
-
 void __init tsc_init(void)
 {
 	u64 lpj;
@@ -964,10 +915,6 @@
 		return;
 	}
 
-	if (cpu_has(&boot_cpu_data, X86_FEATURE_CONSTANT_TSC) &&
-			(boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
-		cpu_khz = calibrate_cpu();
-
 	printk("Detected %lu.%03lu MHz processor.\n",
 			(unsigned long)cpu_khz / 1000,
 			(unsigned long)cpu_khz % 1000);
@@ -987,6 +934,9 @@
 	/* now allow native_sched_clock() to use rdtsc */
 	tsc_disabled = 0;
 
+	if (!no_sched_irq_time)
+		enable_sched_clock_irqtime();
+
 	lpj = ((u64)tsc_khz * 1000);
 	do_div(lpj, HZ);
 	lpj_fine = lpj;
diff --git a/arch/x86/kernel/uv_irq.c b/arch/x86/kernel/uv_irq.c
index 1132129..7b24460 100644
--- a/arch/x86/kernel/uv_irq.c
+++ b/arch/x86/kernel/uv_irq.c
@@ -28,34 +28,21 @@
 static spinlock_t		uv_irq_lock;
 static struct rb_root		uv_irq_root;
 
-static int uv_set_irq_affinity(unsigned int, const struct cpumask *);
+static int uv_set_irq_affinity(struct irq_data *, const struct cpumask *, bool);
 
-static void uv_noop(unsigned int irq)
-{
-}
+static void uv_noop(struct irq_data *data) { }
 
-static unsigned int uv_noop_ret(unsigned int irq)
-{
-	return 0;
-}
-
-static void uv_ack_apic(unsigned int irq)
+static void uv_ack_apic(struct irq_data *data)
 {
 	ack_APIC_irq();
 }
 
 static struct irq_chip uv_irq_chip = {
-	.name		= "UV-CORE",
-	.startup	= uv_noop_ret,
-	.shutdown	= uv_noop,
-	.enable		= uv_noop,
-	.disable	= uv_noop,
-	.ack		= uv_noop,
-	.mask		= uv_noop,
-	.unmask		= uv_noop,
-	.eoi		= uv_ack_apic,
-	.end		= uv_noop,
-	.set_affinity	= uv_set_irq_affinity,
+	.name			= "UV-CORE",
+	.irq_mask		= uv_noop,
+	.irq_unmask		= uv_noop,
+	.irq_eoi		= uv_ack_apic,
+	.irq_set_affinity	= uv_set_irq_affinity,
 };
 
 /*
@@ -144,26 +131,22 @@
 		       unsigned long mmr_offset, int limit)
 {
 	const struct cpumask *eligible_cpu = cpumask_of(cpu);
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg;
-	int mmr_pnode;
+	struct irq_cfg *cfg = get_irq_chip_data(irq);
 	unsigned long mmr_value;
 	struct uv_IO_APIC_route_entry *entry;
-	int err;
+	int mmr_pnode, err;
 
 	BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
 			sizeof(unsigned long));
 
-	cfg = irq_cfg(irq);
-
 	err = assign_irq_vector(irq, cfg, eligible_cpu);
 	if (err != 0)
 		return err;
 
 	if (limit == UV_AFFINITY_CPU)
-		desc->status |= IRQ_NO_BALANCING;
+		irq_set_status_flags(irq, IRQ_NO_BALANCING);
 	else
-		desc->status |= IRQ_MOVE_PCNTXT;
+		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 
 	set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq,
 				      irq_name);
@@ -206,17 +189,17 @@
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
 }
 
-static int uv_set_irq_affinity(unsigned int irq, const struct cpumask *mask)
+static int
+uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
+		    bool force)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	struct irq_cfg *cfg = desc->chip_data;
+	struct irq_cfg *cfg = data->chip_data;
 	unsigned int dest;
-	unsigned long mmr_value;
+	unsigned long mmr_value, mmr_offset;
 	struct uv_IO_APIC_route_entry *entry;
-	unsigned long mmr_offset;
 	int mmr_pnode;
 
-	if (set_desc_affinity(desc, mask, &dest))
+	if (__ioapic_set_affinity(data, mask, &dest))
 		return -1;
 
 	mmr_value = 0;
@@ -231,7 +214,7 @@
 	entry->dest		= dest;
 
 	/* Get previously stored MMR and pnode of hub sourcing interrupts */
-	if (uv_irq_2_mmr_info(irq, &mmr_offset, &mmr_pnode))
+	if (uv_irq_2_mmr_info(data->irq, &mmr_offset, &mmr_pnode))
 		return -1;
 
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c
index e680ea5..3371bd0 100644
--- a/arch/x86/kernel/visws_quirks.c
+++ b/arch/x86/kernel/visws_quirks.c
@@ -66,10 +66,7 @@
 }
 
 /* Replaces the default init_ISA_irqs in the generic setup */
-static void __init visws_pre_intr_init(void)
-{
-	init_VISWS_APIC_irqs();
-}
+static void __init visws_pre_intr_init(void);
 
 /* Quirk for machine specific memory setup. */
 
@@ -429,67 +426,34 @@
 /*
  * This is the SGI Cobalt (IO-)APIC:
  */
-
-static void enable_cobalt_irq(unsigned int irq)
+static void enable_cobalt_irq(struct irq_data *data)
 {
-	co_apic_set(is_co_apic(irq), irq);
+	co_apic_set(is_co_apic(data->irq), data->irq);
 }
 
-static void disable_cobalt_irq(unsigned int irq)
+static void disable_cobalt_irq(struct irq_data *data)
 {
-	int entry = is_co_apic(irq);
+	int entry = is_co_apic(data->irq);
 
 	co_apic_write(CO_APIC_LO(entry), CO_APIC_MASK);
 	co_apic_read(CO_APIC_LO(entry));
 }
 
-/*
- * "irq" really just serves to identify the device.  Here is where we
- * map this to the Cobalt APIC entry where it's physically wired.
- * This is called via request_irq -> setup_irq -> irq_desc->startup()
- */
-static unsigned int startup_cobalt_irq(unsigned int irq)
-{
-	unsigned long flags;
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	spin_lock_irqsave(&cobalt_lock, flags);
-	if ((desc->status & (IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING)))
-		desc->status &= ~(IRQ_DISABLED | IRQ_INPROGRESS | IRQ_WAITING);
-	enable_cobalt_irq(irq);
-	spin_unlock_irqrestore(&cobalt_lock, flags);
-	return 0;
-}
-
-static void ack_cobalt_irq(unsigned int irq)
+static void ack_cobalt_irq(struct irq_data *data)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&cobalt_lock, flags);
-	disable_cobalt_irq(irq);
+	disable_cobalt_irq(data);
 	apic_write(APIC_EOI, APIC_EIO_ACK);
 	spin_unlock_irqrestore(&cobalt_lock, flags);
 }
 
-static void end_cobalt_irq(unsigned int irq)
-{
-	unsigned long flags;
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	spin_lock_irqsave(&cobalt_lock, flags);
-	if (!(desc->status & (IRQ_DISABLED | IRQ_INPROGRESS)))
-		enable_cobalt_irq(irq);
-	spin_unlock_irqrestore(&cobalt_lock, flags);
-}
-
 static struct irq_chip cobalt_irq_type = {
-	.name =		"Cobalt-APIC",
-	.startup =	startup_cobalt_irq,
-	.shutdown =	disable_cobalt_irq,
-	.enable =	enable_cobalt_irq,
-	.disable =	disable_cobalt_irq,
-	.ack =		ack_cobalt_irq,
-	.end =		end_cobalt_irq,
+	.name		= "Cobalt-APIC",
+	.irq_enable	= enable_cobalt_irq,
+	.irq_disable	= disable_cobalt_irq,
+	.irq_ack	= ack_cobalt_irq,
 };
 
 
@@ -503,35 +467,34 @@
  * interrupt controller type, and through a special virtual interrupt-
  * controller. Device drivers only see the virtual interrupt sources.
  */
-static unsigned int startup_piix4_master_irq(unsigned int irq)
+static unsigned int startup_piix4_master_irq(struct irq_data *data)
 {
 	legacy_pic->init(0);
-
-	return startup_cobalt_irq(irq);
+	enable_cobalt_irq(data);
 }
 
-static void end_piix4_master_irq(unsigned int irq)
+static void end_piix4_master_irq(struct irq_data *data)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&cobalt_lock, flags);
-	enable_cobalt_irq(irq);
+	enable_cobalt_irq(data);
 	spin_unlock_irqrestore(&cobalt_lock, flags);
 }
 
 static struct irq_chip piix4_master_irq_type = {
-	.name =		"PIIX4-master",
-	.startup =	startup_piix4_master_irq,
-	.ack =		ack_cobalt_irq,
-	.end =		end_piix4_master_irq,
+	.name		= "PIIX4-master",
+	.irq_startup	= startup_piix4_master_irq,
+	.irq_ack	= ack_cobalt_irq,
 };
 
+static void pii4_mask(struct irq_data *data) { }
 
 static struct irq_chip piix4_virtual_irq_type = {
-	.name =		"PIIX4-virtual",
+	.name		= "PIIX4-virtual",
+	.mask		= pii4_mask,
 };
 
-
 /*
  * PIIX4-8259 master/virtual functions to handle interrupt requests
  * from legacy devices: floppy, parallel, serial, rtc.
@@ -549,9 +512,8 @@
  */
 static irqreturn_t piix4_master_intr(int irq, void *dev_id)
 {
-	int realirq;
-	struct irq_desc *desc;
 	unsigned long flags;
+	int realirq;
 
 	raw_spin_lock_irqsave(&i8259A_lock, flags);
 
@@ -592,18 +554,10 @@
 
 	raw_spin_unlock_irqrestore(&i8259A_lock, flags);
 
-	desc = irq_to_desc(realirq);
-
 	/*
 	 * handle this 'virtual interrupt' as a Cobalt one now.
 	 */
-	kstat_incr_irqs_this_cpu(realirq, desc);
-
-	if (likely(desc->action != NULL))
-		handle_IRQ_event(realirq, desc->action);
-
-	if (!(desc->status & IRQ_DISABLED))
-		legacy_pic->chip->unmask(realirq);
+	generic_handle_irq(realirq);
 
 	return IRQ_HANDLED;
 
@@ -624,41 +578,35 @@
 
 static inline void set_piix4_virtual_irq_type(void)
 {
-	piix4_virtual_irq_type.shutdown = i8259A_chip.mask;
 	piix4_virtual_irq_type.enable =	i8259A_chip.unmask;
 	piix4_virtual_irq_type.disable = i8259A_chip.mask;
+	piix4_virtual_irq_type.unmask =	i8259A_chip.unmask;
 }
 
-void init_VISWS_APIC_irqs(void)
+static void __init visws_pre_intr_init(void)
 {
 	int i;
 
+	set_piix4_virtual_irq_type();
+
 	for (i = 0; i < CO_IRQ_APIC0 + CO_APIC_LAST + 1; i++) {
-		struct irq_desc *desc = irq_to_desc(i);
+		struct irq_chip *chip = NULL;
 
-		desc->status = IRQ_DISABLED;
-		desc->action = 0;
-		desc->depth = 1;
+		if (i == 0)
+			chip = &cobalt_irq_type;
+		else if (i == CO_IRQ_IDE0)
+			chip = &cobalt_irq_type;
+		else if (i == CO_IRQ_IDE1)
+			>chip = &cobalt_irq_type;
+		else if (i == CO_IRQ_8259)
+			chip = &piix4_master_irq_type;
+		else if (i < CO_IRQ_APIC0)
+			chip = &piix4_virtual_irq_type;
+		else if (IS_CO_APIC(i))
+			chip = &cobalt_irq_type;
 
-		if (i == 0) {
-			desc->chip = &cobalt_irq_type;
-		}
-		else if (i == CO_IRQ_IDE0) {
-			desc->chip = &cobalt_irq_type;
-		}
-		else if (i == CO_IRQ_IDE1) {
-			desc->chip = &cobalt_irq_type;
-		}
-		else if (i == CO_IRQ_8259) {
-			desc->chip = &piix4_master_irq_type;
-		}
-		else if (i < CO_IRQ_APIC0) {
-			set_piix4_virtual_irq_type();
-			desc->chip = &piix4_virtual_irq_type;
-		}
-		else if (IS_CO_APIC(i)) {
-			desc->chip = &cobalt_irq_type;
-		}
+		if (chip)
+			set_irq_chip(i, chip);
 	}
 
 	setup_irq(CO_IRQ_8259, &master_action);
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
deleted file mode 100644
index ce9fbac..0000000
--- a/arch/x86/kernel/vmi_32.c
+++ /dev/null
@@ -1,893 +0,0 @@
-/*
- * VMI specific paravirt-ops implementation
- *
- * Copyright (C) 2005, VMware, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * Send feedback to zach@vmware.com
- *
- */
-
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <linux/bootmem.h>
-#include <linux/mm.h>
-#include <linux/highmem.h>
-#include <linux/sched.h>
-#include <linux/gfp.h>
-#include <asm/vmi.h>
-#include <asm/io.h>
-#include <asm/fixmap.h>
-#include <asm/apicdef.h>
-#include <asm/apic.h>
-#include <asm/pgalloc.h>
-#include <asm/processor.h>
-#include <asm/timer.h>
-#include <asm/vmi_time.h>
-#include <asm/kmap_types.h>
-#include <asm/setup.h>
-
-/* Convenient for calling VMI functions indirectly in the ROM */
-typedef u32 __attribute__((regparm(1))) (VROMFUNC)(void);
-typedef u64 __attribute__((regparm(2))) (VROMLONGFUNC)(int);
-
-#define call_vrom_func(rom,func) \
-   (((VROMFUNC *)(rom->func))())
-
-#define call_vrom_long_func(rom,func,arg) \
-   (((VROMLONGFUNC *)(rom->func)) (arg))
-
-static struct vrom_header *vmi_rom;
-static int disable_pge;
-static int disable_pse;
-static int disable_sep;
-static int disable_tsc;
-static int disable_mtrr;
-static int disable_noidle;
-static int disable_vmi_timer;
-
-/* Cached VMI operations */
-static struct {
-	void (*cpuid)(void /* non-c */);
-	void (*_set_ldt)(u32 selector);
-	void (*set_tr)(u32 selector);
-	void (*write_idt_entry)(struct desc_struct *, int, u32, u32);
-	void (*write_gdt_entry)(struct desc_struct *, int, u32, u32);
-	void (*write_ldt_entry)(struct desc_struct *, int, u32, u32);
-	void (*set_kernel_stack)(u32 selector, u32 sp0);
-	void (*allocate_page)(u32, u32, u32, u32, u32);
-	void (*release_page)(u32, u32);
-	void (*set_pte)(pte_t, pte_t *, unsigned);
-	void (*update_pte)(pte_t *, unsigned);
-	void (*set_linear_mapping)(int, void *, u32, u32);
-	void (*_flush_tlb)(int);
-	void (*set_initial_ap_state)(int, int);
-	void (*halt)(void);
-  	void (*set_lazy_mode)(int mode);
-} vmi_ops;
-
-/* Cached VMI operations */
-struct vmi_timer_ops vmi_timer_ops;
-
-/*
- * VMI patching routines.
- */
-#define MNEM_CALL 0xe8
-#define MNEM_JMP  0xe9
-#define MNEM_RET  0xc3
-
-#define IRQ_PATCH_INT_MASK 0
-#define IRQ_PATCH_DISABLE  5
-
-static inline void patch_offset(void *insnbuf,
-				unsigned long ip, unsigned long dest)
-{
-        *(unsigned long *)(insnbuf+1) = dest-ip-5;
-}
-
-static unsigned patch_internal(int call, unsigned len, void *insnbuf,
-			       unsigned long ip)
-{
-	u64 reloc;
-	struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc;
-	reloc = call_vrom_long_func(vmi_rom, get_reloc,	call);
-	switch(rel->type) {
-		case VMI_RELOCATION_CALL_REL:
-			BUG_ON(len < 5);
-			*(char *)insnbuf = MNEM_CALL;
-			patch_offset(insnbuf, ip, (unsigned long)rel->eip);
-			return 5;
-
-		case VMI_RELOCATION_JUMP_REL:
-			BUG_ON(len < 5);
-			*(char *)insnbuf = MNEM_JMP;
-			patch_offset(insnbuf, ip, (unsigned long)rel->eip);
-			return 5;
-
-		case VMI_RELOCATION_NOP:
-			/* obliterate the whole thing */
-			return 0;
-
-		case VMI_RELOCATION_NONE:
-			/* leave native code in place */
-			break;
-
-		default:
-			BUG();
-	}
-	return len;
-}
-
-/*
- * Apply patch if appropriate, return length of new instruction
- * sequence.  The callee does nop padding for us.
- */
-static unsigned vmi_patch(u8 type, u16 clobbers, void *insns,
-			  unsigned long ip, unsigned len)
-{
-	switch (type) {
-		case PARAVIRT_PATCH(pv_irq_ops.irq_disable):
-			return patch_internal(VMI_CALL_DisableInterrupts, len,
-					      insns, ip);
-		case PARAVIRT_PATCH(pv_irq_ops.irq_enable):
-			return patch_internal(VMI_CALL_EnableInterrupts, len,
-					      insns, ip);
-		case PARAVIRT_PATCH(pv_irq_ops.restore_fl):
-			return patch_internal(VMI_CALL_SetInterruptMask, len,
-					      insns, ip);
-		case PARAVIRT_PATCH(pv_irq_ops.save_fl):
-			return patch_internal(VMI_CALL_GetInterruptMask, len,
-					      insns, ip);
-		case PARAVIRT_PATCH(pv_cpu_ops.iret):
-			return patch_internal(VMI_CALL_IRET, len, insns, ip);
-		case PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit):
-			return patch_internal(VMI_CALL_SYSEXIT, len, insns, ip);
-		default:
-			break;
-	}
-	return len;
-}
-
-/* CPUID has non-C semantics, and paravirt-ops API doesn't match hardware ISA */
-static void vmi_cpuid(unsigned int *ax, unsigned int *bx,
-                               unsigned int *cx, unsigned int *dx)
-{
-	int override = 0;
-	if (*ax == 1)
-		override = 1;
-        asm volatile ("call *%6"
-                      : "=a" (*ax),
-                        "=b" (*bx),
-                        "=c" (*cx),
-                        "=d" (*dx)
-                      : "0" (*ax), "2" (*cx), "r" (vmi_ops.cpuid));
-	if (override) {
-		if (disable_pse)
-			*dx &= ~X86_FEATURE_PSE;
-		if (disable_pge)
-			*dx &= ~X86_FEATURE_PGE;
-		if (disable_sep)
-			*dx &= ~X86_FEATURE_SEP;
-		if (disable_tsc)
-			*dx &= ~X86_FEATURE_TSC;
-		if (disable_mtrr)
-			*dx &= ~X86_FEATURE_MTRR;
-	}
-}
-
-static inline void vmi_maybe_load_tls(struct desc_struct *gdt, int nr, struct desc_struct *new)
-{
-	if (gdt[nr].a != new->a || gdt[nr].b != new->b)
-		write_gdt_entry(gdt, nr, new, 0);
-}
-
-static void vmi_load_tls(struct thread_struct *t, unsigned int cpu)
-{
-	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
-	vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 0, &t->tls_array[0]);
-	vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 1, &t->tls_array[1]);
-	vmi_maybe_load_tls(gdt, GDT_ENTRY_TLS_MIN + 2, &t->tls_array[2]);
-}
-
-static void vmi_set_ldt(const void *addr, unsigned entries)
-{
-	unsigned cpu = smp_processor_id();
-	struct desc_struct desc;
-
-	pack_descriptor(&desc, (unsigned long)addr,
-			entries * sizeof(struct desc_struct) - 1,
-			DESC_LDT, 0);
-	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, &desc, DESC_LDT);
-	vmi_ops._set_ldt(entries ? GDT_ENTRY_LDT*sizeof(struct desc_struct) : 0);
-}
-
-static void vmi_set_tr(void)
-{
-	vmi_ops.set_tr(GDT_ENTRY_TSS*sizeof(struct desc_struct));
-}
-
-static void vmi_write_idt_entry(gate_desc *dt, int entry, const gate_desc *g)
-{
-	u32 *idt_entry = (u32 *)g;
-	vmi_ops.write_idt_entry(dt, entry, idt_entry[0], idt_entry[1]);
-}
-
-static void vmi_write_gdt_entry(struct desc_struct *dt, int entry,
-				const void *desc, int type)
-{
-	u32 *gdt_entry = (u32 *)desc;
-	vmi_ops.write_gdt_entry(dt, entry, gdt_entry[0], gdt_entry[1]);
-}
-
-static void vmi_write_ldt_entry(struct desc_struct *dt, int entry,
-				const void *desc)
-{
-	u32 *ldt_entry = (u32 *)desc;
-	vmi_ops.write_ldt_entry(dt, entry, ldt_entry[0], ldt_entry[1]);
-}
-
-static void vmi_load_sp0(struct tss_struct *tss,
-				   struct thread_struct *thread)
-{
-	tss->x86_tss.sp0 = thread->sp0;
-
-	/* This can only happen when SEP is enabled, no need to test "SEP"arately */
-	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
-		tss->x86_tss.ss1 = thread->sysenter_cs;
-		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
-	}
-	vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.sp0);
-}
-
-static void vmi_flush_tlb_user(void)
-{
-	vmi_ops._flush_tlb(VMI_FLUSH_TLB);
-}
-
-static void vmi_flush_tlb_kernel(void)
-{
-	vmi_ops._flush_tlb(VMI_FLUSH_TLB | VMI_FLUSH_GLOBAL);
-}
-
-/* Stub to do nothing at all; used for delays and unimplemented calls */
-static void vmi_nop(void)
-{
-}
-
-static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn)
-{
-	vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0);
-}
-
-static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn)
-{
- 	/*
-	 * This call comes in very early, before mem_map is setup.
-	 * It is called only for swapper_pg_dir, which already has
-	 * data on it.
-	 */
-	vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0);
-}
-
-static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count)
-{
-	vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count);
-}
-
-static void vmi_release_pte(unsigned long pfn)
-{
-	vmi_ops.release_page(pfn, VMI_PAGE_L1);
-}
-
-static void vmi_release_pmd(unsigned long pfn)
-{
-	vmi_ops.release_page(pfn, VMI_PAGE_L2);
-}
-
-/*
- * We use the pgd_free hook for releasing the pgd page:
- */
-static void vmi_pgd_free(struct mm_struct *mm, pgd_t *pgd)
-{
-	unsigned long pfn = __pa(pgd) >> PAGE_SHIFT;
-
-	vmi_ops.release_page(pfn, VMI_PAGE_L2);
-}
-
-/*
- * Helper macros for MMU update flags.  We can defer updates until a flush
- * or page invalidation only if the update is to the current address space
- * (otherwise, there is no flush).  We must check against init_mm, since
- * this could be a kernel update, which usually passes init_mm, although
- * sometimes this check can be skipped if we know the particular function
- * is only called on user mode PTEs.  We could change the kernel to pass
- * current->active_mm here, but in particular, I was unsure if changing
- * mm/highmem.c to do this would still be correct on other architectures.
- */
-#define is_current_as(mm, mustbeuser) ((mm) == current->active_mm ||    \
-                                       (!mustbeuser && (mm) == &init_mm))
-#define vmi_flags_addr(mm, addr, level, user)                           \
-        ((level) | (is_current_as(mm, user) ?                           \
-                (VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0))
-#define vmi_flags_addr_defer(mm, addr, level, user)                     \
-        ((level) | (is_current_as(mm, user) ?                           \
-                (VMI_PAGE_DEFER | VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0))
-
-static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	vmi_ops.update_pte(ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
-}
-
-static void vmi_update_pte_defer(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	vmi_ops.update_pte(ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 0));
-}
-
-static void vmi_set_pte(pte_t *ptep, pte_t pte)
-{
-	/* XXX because of set_pmd_pte, this can be called on PT or PD layers */
-	vmi_ops.set_pte(pte, ptep, VMI_PAGE_PT);
-}
-
-static void vmi_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
-{
-	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
-}
-
-static void vmi_set_pmd(pmd_t *pmdp, pmd_t pmdval)
-{
-#ifdef CONFIG_X86_PAE
-	const pte_t pte = { .pte = pmdval.pmd };
-#else
-	const pte_t pte = { pmdval.pud.pgd.pgd };
-#endif
-	vmi_ops.set_pte(pte, (pte_t *)pmdp, VMI_PAGE_PD);
-}
-
-#ifdef CONFIG_X86_PAE
-
-static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval)
-{
-	/*
-	 * XXX This is called from set_pmd_pte, but at both PT
-	 * and PD layers so the VMI_PAGE_PT flag is wrong.  But
-	 * it is only called for large page mapping changes,
-	 * the Xen backend, doesn't support large pages, and the
-	 * ESX backend doesn't depend on the flag.
-	 */
-	set_64bit((unsigned long long *)ptep,pte_val(pteval));
-	vmi_ops.update_pte(ptep, VMI_PAGE_PT);
-}
-
-static void vmi_set_pud(pud_t *pudp, pud_t pudval)
-{
-	/* Um, eww */
-	const pte_t pte = { .pte = pudval.pgd.pgd };
-	vmi_ops.set_pte(pte, (pte_t *)pudp, VMI_PAGE_PDP);
-}
-
-static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
-{
-	const pte_t pte = { .pte = 0 };
-	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
-}
-
-static void vmi_pmd_clear(pmd_t *pmd)
-{
-	const pte_t pte = { .pte = 0 };
-	vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD);
-}
-#endif
-
-#ifdef CONFIG_SMP
-static void __devinit
-vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
-		     unsigned long start_esp)
-{
-	struct vmi_ap_state ap;
-
-	/* Default everything to zero.  This is fine for most GPRs. */
-	memset(&ap, 0, sizeof(struct vmi_ap_state));
-
-	ap.gdtr_limit = GDT_SIZE - 1;
-	ap.gdtr_base = (unsigned long) get_cpu_gdt_table(phys_apicid);
-
-	ap.idtr_limit = IDT_ENTRIES * 8 - 1;
-	ap.idtr_base = (unsigned long) idt_table;
-
-	ap.ldtr = 0;
-
-	ap.cs = __KERNEL_CS;
-	ap.eip = (unsigned long) start_eip;
-	ap.ss = __KERNEL_DS;
-	ap.esp = (unsigned long) start_esp;
-
-	ap.ds = __USER_DS;
-	ap.es = __USER_DS;
-	ap.fs = __KERNEL_PERCPU;
-	ap.gs = __KERNEL_STACK_CANARY;
-
-	ap.eflags = 0;
-
-#ifdef CONFIG_X86_PAE
-	/* efer should match BSP efer. */
-	if (cpu_has_nx) {
-		unsigned l, h;
-		rdmsr(MSR_EFER, l, h);
-		ap.efer = (unsigned long long) h << 32 | l;
-	}
-#endif
-
-	ap.cr3 = __pa(swapper_pg_dir);
-	/* Protected mode, paging, AM, WP, NE, MP. */
-	ap.cr0 = 0x80050023;
-	ap.cr4 = mmu_cr4_features;
-	vmi_ops.set_initial_ap_state((u32)&ap, phys_apicid);
-}
-#endif
-
-static void vmi_start_context_switch(struct task_struct *prev)
-{
-	paravirt_start_context_switch(prev);
-	vmi_ops.set_lazy_mode(2);
-}
-
-static void vmi_end_context_switch(struct task_struct *next)
-{
-	vmi_ops.set_lazy_mode(0);
-	paravirt_end_context_switch(next);
-}
-
-static void vmi_enter_lazy_mmu(void)
-{
-	paravirt_enter_lazy_mmu();
-	vmi_ops.set_lazy_mode(1);
-}
-
-static void vmi_leave_lazy_mmu(void)
-{
-	vmi_ops.set_lazy_mode(0);
-	paravirt_leave_lazy_mmu();
-}
-
-static inline int __init check_vmi_rom(struct vrom_header *rom)
-{
-	struct pci_header *pci;
-	struct pnp_header *pnp;
-	const char *manufacturer = "UNKNOWN";
-	const char *product = "UNKNOWN";
-	const char *license = "unspecified";
-
-	if (rom->rom_signature != 0xaa55)
-		return 0;
-	if (rom->vrom_signature != VMI_SIGNATURE)
-		return 0;
-	if (rom->api_version_maj != VMI_API_REV_MAJOR ||
-	    rom->api_version_min+1 < VMI_API_REV_MINOR+1) {
-		printk(KERN_WARNING "VMI: Found mismatched rom version %d.%d\n",
-				rom->api_version_maj,
-				rom->api_version_min);
-		return 0;
-	}
-
-	/*
-	 * Relying on the VMI_SIGNATURE field is not 100% safe, so check
-	 * the PCI header and device type to make sure this is really a
-	 * VMI device.
-	 */
-	if (!rom->pci_header_offs) {
-		printk(KERN_WARNING "VMI: ROM does not contain PCI header.\n");
-		return 0;
-	}
-
-	pci = (struct pci_header *)((char *)rom+rom->pci_header_offs);
-	if (pci->vendorID != PCI_VENDOR_ID_VMWARE ||
-	    pci->deviceID != PCI_DEVICE_ID_VMWARE_VMI) {
-		/* Allow it to run... anyways, but warn */
-		printk(KERN_WARNING "VMI: ROM from unknown manufacturer\n");
-	}
-
-	if (rom->pnp_header_offs) {
-		pnp = (struct pnp_header *)((char *)rom+rom->pnp_header_offs);
-		if (pnp->manufacturer_offset)
-			manufacturer = (const char *)rom+pnp->manufacturer_offset;
-		if (pnp->product_offset)
-			product = (const char *)rom+pnp->product_offset;
-	}
-
-	if (rom->license_offs)
-		license = (char *)rom+rom->license_offs;
-
-	printk(KERN_INFO "VMI: Found %s %s, API version %d.%d, ROM version %d.%d\n",
-		manufacturer, product,
-		rom->api_version_maj, rom->api_version_min,
-		pci->rom_version_maj, pci->rom_version_min);
-
-	/* Don't allow BSD/MIT here for now because we don't want to end up
-	   with any binary only shim layers */
-	if (strcmp(license, "GPL") && strcmp(license, "GPL v2")) {
-		printk(KERN_WARNING "VMI: Non GPL license `%s' found for ROM. Not used.\n",
-			license);
-		return 0;
-	}
-
-	return 1;
-}
-
-/*
- * Probe for the VMI option ROM
- */
-static inline int __init probe_vmi_rom(void)
-{
-	unsigned long base;
-
-	/* VMI ROM is in option ROM area, check signature */
-	for (base = 0xC0000; base < 0xE0000; base += 2048) {
-		struct vrom_header *romstart;
-		romstart = (struct vrom_header *)isa_bus_to_virt(base);
-		if (check_vmi_rom(romstart)) {
-			vmi_rom = romstart;
-			return 1;
-		}
-	}
-	return 0;
-}
-
-/*
- * VMI setup common to all processors
- */
-void vmi_bringup(void)
-{
- 	/* We must establish the lowmem mapping for MMU ops to work */
-	if (vmi_ops.set_linear_mapping)
-		vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, MAXMEM_PFN, 0);
-}
-
-/*
- * Return a pointer to a VMI function or NULL if unimplemented
- */
-static void *vmi_get_function(int vmicall)
-{
-	u64 reloc;
-	const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc;
-	reloc = call_vrom_long_func(vmi_rom, get_reloc,	vmicall);
-	BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL);
-	if (rel->type == VMI_RELOCATION_CALL_REL)
-		return (void *)rel->eip;
-	else
-		return NULL;
-}
-
-/*
- * Helper macro for making the VMI paravirt-ops fill code readable.
- * For unimplemented operations, fall back to default, unless nop
- * is returned by the ROM.
- */
-#define para_fill(opname, vmicall)				\
-do {								\
-	reloc = call_vrom_long_func(vmi_rom, get_reloc,		\
-				    VMI_CALL_##vmicall);	\
-	if (rel->type == VMI_RELOCATION_CALL_REL) 		\
-		opname = (void *)rel->eip;			\
-	else if (rel->type == VMI_RELOCATION_NOP) 		\
-		opname = (void *)vmi_nop;			\
-	else if (rel->type != VMI_RELOCATION_NONE)		\
-		printk(KERN_WARNING "VMI: Unknown relocation "	\
-				    "type %d for " #vmicall"\n",\
-					rel->type);		\
-} while (0)
-
-/*
- * Helper macro for making the VMI paravirt-ops fill code readable.
- * For cached operations which do not match the VMI ROM ABI and must
- * go through a tranlation stub.  Ignore NOPs, since it is not clear
- * a NOP * VMI function corresponds to a NOP paravirt-op when the
- * functions are not in 1-1 correspondence.
- */
-#define para_wrap(opname, wrapper, cache, vmicall)		\
-do {								\
-	reloc = call_vrom_long_func(vmi_rom, get_reloc,		\
-				    VMI_CALL_##vmicall);	\
-	BUG_ON(rel->type == VMI_RELOCATION_JUMP_REL);		\
-	if (rel->type == VMI_RELOCATION_CALL_REL) {		\
-		opname = wrapper;				\
-		vmi_ops.cache = (void *)rel->eip;		\
-	}							\
-} while (0)
-
-/*
- * Activate the VMI interface and switch into paravirtualized mode
- */
-static inline int __init activate_vmi(void)
-{
-	short kernel_cs;
-	u64 reloc;
-	const struct vmi_relocation_info *rel = (struct vmi_relocation_info *)&reloc;
-
-	/*
-	 * Prevent page tables from being allocated in highmem, even if
-	 * CONFIG_HIGHPTE is enabled.
-	 */
-	__userpte_alloc_gfp &= ~__GFP_HIGHMEM;
-
-	if (call_vrom_func(vmi_rom, vmi_init) != 0) {
-		printk(KERN_ERR "VMI ROM failed to initialize!");
-		return 0;
-	}
-	savesegment(cs, kernel_cs);
-
-	pv_info.paravirt_enabled = 1;
-	pv_info.kernel_rpl = kernel_cs & SEGMENT_RPL_MASK;
-	pv_info.name = "vmi [deprecated]";
-
-	pv_init_ops.patch = vmi_patch;
-
-	/*
-	 * Many of these operations are ABI compatible with VMI.
-	 * This means we can fill in the paravirt-ops with direct
-	 * pointers into the VMI ROM.  If the calling convention for
-	 * these operations changes, this code needs to be updated.
-	 *
-	 * Exceptions
-	 *  CPUID paravirt-op uses pointers, not the native ISA
-	 *  halt has no VMI equivalent; all VMI halts are "safe"
-	 *  no MSR support yet - just trap and emulate.  VMI uses the
-	 *    same ABI as the native ISA, but Linux wants exceptions
-	 *    from bogus MSR read / write handled
-	 *  rdpmc is not yet used in Linux
-	 */
-
-	/* CPUID is special, so very special it gets wrapped like a present */
-	para_wrap(pv_cpu_ops.cpuid, vmi_cpuid, cpuid, CPUID);
-
-	para_fill(pv_cpu_ops.clts, CLTS);
-	para_fill(pv_cpu_ops.get_debugreg, GetDR);
-	para_fill(pv_cpu_ops.set_debugreg, SetDR);
-	para_fill(pv_cpu_ops.read_cr0, GetCR0);
-	para_fill(pv_mmu_ops.read_cr2, GetCR2);
-	para_fill(pv_mmu_ops.read_cr3, GetCR3);
-	para_fill(pv_cpu_ops.read_cr4, GetCR4);
-	para_fill(pv_cpu_ops.write_cr0, SetCR0);
-	para_fill(pv_mmu_ops.write_cr2, SetCR2);
-	para_fill(pv_mmu_ops.write_cr3, SetCR3);
-	para_fill(pv_cpu_ops.write_cr4, SetCR4);
-
-	para_fill(pv_irq_ops.save_fl.func, GetInterruptMask);
-	para_fill(pv_irq_ops.restore_fl.func, SetInterruptMask);
-	para_fill(pv_irq_ops.irq_disable.func, DisableInterrupts);
-	para_fill(pv_irq_ops.irq_enable.func, EnableInterrupts);
-
-	para_fill(pv_cpu_ops.wbinvd, WBINVD);
-	para_fill(pv_cpu_ops.read_tsc, RDTSC);
-
-	/* The following we emulate with trap and emulate for now */
-	/* paravirt_ops.read_msr = vmi_rdmsr */
-	/* paravirt_ops.write_msr = vmi_wrmsr */
-	/* paravirt_ops.rdpmc = vmi_rdpmc */
-
-	/* TR interface doesn't pass TR value, wrap */
-	para_wrap(pv_cpu_ops.load_tr_desc, vmi_set_tr, set_tr, SetTR);
-
-	/* LDT is special, too */
-	para_wrap(pv_cpu_ops.set_ldt, vmi_set_ldt, _set_ldt, SetLDT);
-
-	para_fill(pv_cpu_ops.load_gdt, SetGDT);
-	para_fill(pv_cpu_ops.load_idt, SetIDT);
-	para_fill(pv_cpu_ops.store_gdt, GetGDT);
-	para_fill(pv_cpu_ops.store_idt, GetIDT);
-	para_fill(pv_cpu_ops.store_tr, GetTR);
-	pv_cpu_ops.load_tls = vmi_load_tls;
-	para_wrap(pv_cpu_ops.write_ldt_entry, vmi_write_ldt_entry,
-		  write_ldt_entry, WriteLDTEntry);
-	para_wrap(pv_cpu_ops.write_gdt_entry, vmi_write_gdt_entry,
-		  write_gdt_entry, WriteGDTEntry);
-	para_wrap(pv_cpu_ops.write_idt_entry, vmi_write_idt_entry,
-		  write_idt_entry, WriteIDTEntry);
-	para_wrap(pv_cpu_ops.load_sp0, vmi_load_sp0, set_kernel_stack, UpdateKernelStack);
-	para_fill(pv_cpu_ops.set_iopl_mask, SetIOPLMask);
-	para_fill(pv_cpu_ops.io_delay, IODelay);
-
-	para_wrap(pv_cpu_ops.start_context_switch, vmi_start_context_switch,
-		  set_lazy_mode, SetLazyMode);
-	para_wrap(pv_cpu_ops.end_context_switch, vmi_end_context_switch,
-		  set_lazy_mode, SetLazyMode);
-
-	para_wrap(pv_mmu_ops.lazy_mode.enter, vmi_enter_lazy_mmu,
-		  set_lazy_mode, SetLazyMode);
-	para_wrap(pv_mmu_ops.lazy_mode.leave, vmi_leave_lazy_mmu,
-		  set_lazy_mode, SetLazyMode);
-
-	/* user and kernel flush are just handled with different flags to FlushTLB */
-	para_wrap(pv_mmu_ops.flush_tlb_user, vmi_flush_tlb_user, _flush_tlb, FlushTLB);
-	para_wrap(pv_mmu_ops.flush_tlb_kernel, vmi_flush_tlb_kernel, _flush_tlb, FlushTLB);
-	para_fill(pv_mmu_ops.flush_tlb_single, InvalPage);
-
-	/*
-	 * Until a standard flag format can be agreed on, we need to
-	 * implement these as wrappers in Linux.  Get the VMI ROM
-	 * function pointers for the two backend calls.
-	 */
-#ifdef CONFIG_X86_PAE
-	vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxELong);
-	vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxELong);
-#else
-	vmi_ops.set_pte = vmi_get_function(VMI_CALL_SetPxE);
-	vmi_ops.update_pte = vmi_get_function(VMI_CALL_UpdatePxE);
-#endif
-
-	if (vmi_ops.set_pte) {
-		pv_mmu_ops.set_pte = vmi_set_pte;
-		pv_mmu_ops.set_pte_at = vmi_set_pte_at;
-		pv_mmu_ops.set_pmd = vmi_set_pmd;
-#ifdef CONFIG_X86_PAE
-		pv_mmu_ops.set_pte_atomic = vmi_set_pte_atomic;
-		pv_mmu_ops.set_pud = vmi_set_pud;
-		pv_mmu_ops.pte_clear = vmi_pte_clear;
-		pv_mmu_ops.pmd_clear = vmi_pmd_clear;
-#endif
-	}
-
-	if (vmi_ops.update_pte) {
-		pv_mmu_ops.pte_update = vmi_update_pte;
-		pv_mmu_ops.pte_update_defer = vmi_update_pte_defer;
-	}
-
-	vmi_ops.allocate_page = vmi_get_function(VMI_CALL_AllocatePage);
-	if (vmi_ops.allocate_page) {
-		pv_mmu_ops.alloc_pte = vmi_allocate_pte;
-		pv_mmu_ops.alloc_pmd = vmi_allocate_pmd;
-		pv_mmu_ops.alloc_pmd_clone = vmi_allocate_pmd_clone;
-	}
-
-	vmi_ops.release_page = vmi_get_function(VMI_CALL_ReleasePage);
-	if (vmi_ops.release_page) {
-		pv_mmu_ops.release_pte = vmi_release_pte;
-		pv_mmu_ops.release_pmd = vmi_release_pmd;
-		pv_mmu_ops.pgd_free = vmi_pgd_free;
-	}
-
-	/* Set linear is needed in all cases */
-	vmi_ops.set_linear_mapping = vmi_get_function(VMI_CALL_SetLinearMapping);
-
-	/*
-	 * These MUST always be patched.  Don't support indirect jumps
-	 * through these operations, as the VMI interface may use either
-	 * a jump or a call to get to these operations, depending on
-	 * the backend.  They are performance critical anyway, so requiring
-	 * a patch is not a big problem.
-	 */
-	pv_cpu_ops.irq_enable_sysexit = (void *)0xfeedbab0;
-	pv_cpu_ops.iret = (void *)0xbadbab0;
-
-#ifdef CONFIG_SMP
-	para_wrap(pv_apic_ops.startup_ipi_hook, vmi_startup_ipi_hook, set_initial_ap_state, SetInitialAPState);
-#endif
-
-#ifdef CONFIG_X86_LOCAL_APIC
-       para_fill(apic->read, APICRead);
-       para_fill(apic->write, APICWrite);
-#endif
-
-	/*
-	 * Check for VMI timer functionality by probing for a cycle frequency method
-	 */
-	reloc = call_vrom_long_func(vmi_rom, get_reloc, VMI_CALL_GetCycleFrequency);
-	if (!disable_vmi_timer && rel->type != VMI_RELOCATION_NONE) {
-		vmi_timer_ops.get_cycle_frequency = (void *)rel->eip;
-		vmi_timer_ops.get_cycle_counter =
-			vmi_get_function(VMI_CALL_GetCycleCounter);
-		vmi_timer_ops.get_wallclock =
-			vmi_get_function(VMI_CALL_GetWallclockTime);
-		vmi_timer_ops.wallclock_updated =
-			vmi_get_function(VMI_CALL_WallclockUpdated);
-		vmi_timer_ops.set_alarm = vmi_get_function(VMI_CALL_SetAlarm);
-		vmi_timer_ops.cancel_alarm =
-			 vmi_get_function(VMI_CALL_CancelAlarm);
-		x86_init.timers.timer_init = vmi_time_init;
-#ifdef CONFIG_X86_LOCAL_APIC
-		x86_init.timers.setup_percpu_clockev = vmi_time_bsp_init;
-		x86_cpuinit.setup_percpu_clockev = vmi_time_ap_init;
-#endif
-		pv_time_ops.sched_clock = vmi_sched_clock;
-		x86_platform.calibrate_tsc = vmi_tsc_khz;
-		x86_platform.get_wallclock = vmi_get_wallclock;
-		x86_platform.set_wallclock = vmi_set_wallclock;
-
-		/* We have true wallclock functions; disable CMOS clock sync */
-		no_sync_cmos_clock = 1;
-	} else {
-		disable_noidle = 1;
-		disable_vmi_timer = 1;
-	}
-
-	para_fill(pv_irq_ops.safe_halt, Halt);
-
-	/*
-	 * Alternative instruction rewriting doesn't happen soon enough
-	 * to convert VMI_IRET to a call instead of a jump; so we have
-	 * to do this before IRQs get reenabled.  Fortunately, it is
-	 * idempotent.
-	 */
-	apply_paravirt(__parainstructions, __parainstructions_end);
-
-	vmi_bringup();
-
-	return 1;
-}
-
-#undef para_fill
-
-void __init vmi_init(void)
-{
-	if (!vmi_rom)
-		probe_vmi_rom();
-	else
-		check_vmi_rom(vmi_rom);
-
-	/* In case probing for or validating the ROM failed, basil */
-	if (!vmi_rom)
-		return;
-
-	reserve_top_address(-vmi_rom->virtual_top);
-
-#ifdef CONFIG_X86_IO_APIC
-	/* This is virtual hardware; timer routing is wired correctly */
-	no_timer_check = 1;
-#endif
-}
-
-void __init vmi_activate(void)
-{
-	unsigned long flags;
-
-	if (!vmi_rom)
-		return;
-
-	local_irq_save(flags);
-	activate_vmi();
-	local_irq_restore(flags & X86_EFLAGS_IF);
-}
-
-static int __init parse_vmi(char *arg)
-{
-	if (!arg)
-		return -EINVAL;
-
-	if (!strcmp(arg, "disable_pge")) {
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PGE);
-		disable_pge = 1;
-	} else if (!strcmp(arg, "disable_pse")) {
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_PSE);
-		disable_pse = 1;
-	} else if (!strcmp(arg, "disable_sep")) {
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_SEP);
-		disable_sep = 1;
-	} else if (!strcmp(arg, "disable_tsc")) {
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC);
-		disable_tsc = 1;
-	} else if (!strcmp(arg, "disable_mtrr")) {
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_MTRR);
-		disable_mtrr = 1;
-	} else if (!strcmp(arg, "disable_timer")) {
-		disable_vmi_timer = 1;
-		disable_noidle = 1;
-	} else if (!strcmp(arg, "disable_noidle"))
-		disable_noidle = 1;
-	return 0;
-}
-
-early_param("vmi", parse_vmi);
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
deleted file mode 100644
index 5e1ff66..0000000
--- a/arch/x86/kernel/vmiclock_32.c
+++ /dev/null
@@ -1,317 +0,0 @@
-/*
- * VMI paravirtual timer support routines.
- *
- * Copyright (C) 2007, VMware, Inc.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- */
-
-#include <linux/smp.h>
-#include <linux/interrupt.h>
-#include <linux/cpumask.h>
-#include <linux/clocksource.h>
-#include <linux/clockchips.h>
-
-#include <asm/vmi.h>
-#include <asm/vmi_time.h>
-#include <asm/apicdef.h>
-#include <asm/apic.h>
-#include <asm/timer.h>
-#include <asm/i8253.h>
-#include <asm/irq_vectors.h>
-
-#define VMI_ONESHOT  (VMI_ALARM_IS_ONESHOT  | VMI_CYCLES_REAL | vmi_get_alarm_wiring())
-#define VMI_PERIODIC (VMI_ALARM_IS_PERIODIC | VMI_CYCLES_REAL | vmi_get_alarm_wiring())
-
-static DEFINE_PER_CPU(struct clock_event_device, local_events);
-
-static inline u32 vmi_counter(u32 flags)
-{
-	/* Given VMI_ONESHOT or VMI_PERIODIC, return the corresponding
-	 * cycle counter. */
-	return flags & VMI_ALARM_COUNTER_MASK;
-}
-
-/* paravirt_ops.get_wallclock = vmi_get_wallclock */
-unsigned long vmi_get_wallclock(void)
-{
-	unsigned long long wallclock;
-	wallclock = vmi_timer_ops.get_wallclock(); // nsec
-	(void)do_div(wallclock, 1000000000);       // sec
-
-	return wallclock;
-}
-
-/* paravirt_ops.set_wallclock = vmi_set_wallclock */
-int vmi_set_wallclock(unsigned long now)
-{
-	return 0;
-}
-
-/* paravirt_ops.sched_clock = vmi_sched_clock */
-unsigned long long vmi_sched_clock(void)
-{
-	return cycles_2_ns(vmi_timer_ops.get_cycle_counter(VMI_CYCLES_AVAILABLE));
-}
-
-/* x86_platform.calibrate_tsc = vmi_tsc_khz */
-unsigned long vmi_tsc_khz(void)
-{
-	unsigned long long khz;
-	khz = vmi_timer_ops.get_cycle_frequency();
-	(void)do_div(khz, 1000);
-	return khz;
-}
-
-static inline unsigned int vmi_get_timer_vector(void)
-{
-	return IRQ0_VECTOR;
-}
-
-/** vmi clockchip */
-#ifdef CONFIG_X86_LOCAL_APIC
-static unsigned int startup_timer_irq(unsigned int irq)
-{
-	unsigned long val = apic_read(APIC_LVTT);
-	apic_write(APIC_LVTT, vmi_get_timer_vector());
-
-	return (val & APIC_SEND_PENDING);
-}
-
-static void mask_timer_irq(unsigned int irq)
-{
-	unsigned long val = apic_read(APIC_LVTT);
-	apic_write(APIC_LVTT, val | APIC_LVT_MASKED);
-}
-
-static void unmask_timer_irq(unsigned int irq)
-{
-	unsigned long val = apic_read(APIC_LVTT);
-	apic_write(APIC_LVTT, val & ~APIC_LVT_MASKED);
-}
-
-static void ack_timer_irq(unsigned int irq)
-{
-	ack_APIC_irq();
-}
-
-static struct irq_chip vmi_chip __read_mostly = {
-	.name 		= "VMI-LOCAL",
-	.startup 	= startup_timer_irq,
-	.mask	 	= mask_timer_irq,
-	.unmask	 	= unmask_timer_irq,
-	.ack 		= ack_timer_irq
-};
-#endif
-
-/** vmi clockevent */
-#define VMI_ALARM_WIRED_IRQ0    0x00000000
-#define VMI_ALARM_WIRED_LVTT    0x00010000
-static int vmi_wiring = VMI_ALARM_WIRED_IRQ0;
-
-static inline int vmi_get_alarm_wiring(void)
-{
-	return vmi_wiring;
-}
-
-static void vmi_timer_set_mode(enum clock_event_mode mode,
-			       struct clock_event_device *evt)
-{
-	cycle_t now, cycles_per_hz;
-	BUG_ON(!irqs_disabled());
-
-	switch (mode) {
-	case CLOCK_EVT_MODE_ONESHOT:
-	case CLOCK_EVT_MODE_RESUME:
-		break;
-	case CLOCK_EVT_MODE_PERIODIC:
-		cycles_per_hz = vmi_timer_ops.get_cycle_frequency();
-		(void)do_div(cycles_per_hz, HZ);
-		now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_PERIODIC));
-		vmi_timer_ops.set_alarm(VMI_PERIODIC, now, cycles_per_hz);
-		break;
-	case CLOCK_EVT_MODE_UNUSED:
-	case CLOCK_EVT_MODE_SHUTDOWN:
-		switch (evt->mode) {
-		case CLOCK_EVT_MODE_ONESHOT:
-			vmi_timer_ops.cancel_alarm(VMI_ONESHOT);
-			break;
-		case CLOCK_EVT_MODE_PERIODIC:
-			vmi_timer_ops.cancel_alarm(VMI_PERIODIC);
-			break;
-		default:
-			break;
-		}
-		break;
-	default:
-		break;
-	}
-}
-
-static int vmi_timer_next_event(unsigned long delta,
-				struct clock_event_device *evt)
-{
-	/* Unfortunately, set_next_event interface only passes relative
-	 * expiry, but we want absolute expiry.  It'd be better if were
-	 * were passed an absolute expiry, since a bunch of time may
-	 * have been stolen between the time the delta is computed and
-	 * when we set the alarm below. */
-	cycle_t now = vmi_timer_ops.get_cycle_counter(vmi_counter(VMI_ONESHOT));
-
-	BUG_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT);
-	vmi_timer_ops.set_alarm(VMI_ONESHOT, now + delta, 0);
-	return 0;
-}
-
-static struct clock_event_device vmi_clockevent = {
-	.name		= "vmi-timer",
-	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
-	.shift		= 22,
-	.set_mode	= vmi_timer_set_mode,
-	.set_next_event = vmi_timer_next_event,
-	.rating         = 1000,
-	.irq		= 0,
-};
-
-static irqreturn_t vmi_timer_interrupt(int irq, void *dev_id)
-{
-	struct clock_event_device *evt = &__get_cpu_var(local_events);
-	evt->event_handler(evt);
-	return IRQ_HANDLED;
-}
-
-static struct irqaction vmi_clock_action  = {
-	.name 		= "vmi-timer",
-	.handler 	= vmi_timer_interrupt,
-	.flags 		= IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER,
-};
-
-static void __devinit vmi_time_init_clockevent(void)
-{
-	cycle_t cycles_per_msec;
-	struct clock_event_device *evt;
-
-	int cpu = smp_processor_id();
-	evt = &__get_cpu_var(local_events);
-
-	/* Use cycles_per_msec since div_sc params are 32-bits. */
-	cycles_per_msec = vmi_timer_ops.get_cycle_frequency();
-	(void)do_div(cycles_per_msec, 1000);
-
-	memcpy(evt, &vmi_clockevent, sizeof(*evt));
-	/* Must pick .shift such that .mult fits in 32-bits.  Choosing
-	 * .shift to be 22 allows 2^(32-22) cycles per nano-seconds
-	 * before overflow. */
-	evt->mult = div_sc(cycles_per_msec, NSEC_PER_MSEC, evt->shift);
-	/* Upper bound is clockevent's use of ulong for cycle deltas. */
-	evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt);
-	evt->min_delta_ns = clockevent_delta2ns(1, evt);
-	evt->cpumask = cpumask_of(cpu);
-
-	printk(KERN_WARNING "vmi: registering clock event %s. mult=%u shift=%u\n",
-	       evt->name, evt->mult, evt->shift);
-	clockevents_register_device(evt);
-}
-
-void __init vmi_time_init(void)
-{
-	unsigned int cpu;
-	/* Disable PIT: BIOSes start PIT CH0 with 18.2hz peridic. */
-	outb_pit(0x3a, PIT_MODE); /* binary, mode 5, LSB/MSB, ch 0 */
-
-	vmi_time_init_clockevent();
-	setup_irq(0, &vmi_clock_action);
-	for_each_possible_cpu(cpu)
-		per_cpu(vector_irq, cpu)[vmi_get_timer_vector()] = 0;
-}
-
-#ifdef CONFIG_X86_LOCAL_APIC
-void __devinit vmi_time_bsp_init(void)
-{
-	/*
-	 * On APIC systems, we want local timers to fire on each cpu.  We do
-	 * this by programming LVTT to deliver timer events to the IRQ handler
-	 * for IRQ-0, since we can't re-use the APIC local timer handler
-	 * without interfering with that code.
-	 */
-	clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL);
-	local_irq_disable();
-#ifdef CONFIG_SMP
-	/*
-	 * XXX handle_percpu_irq only defined for SMP; we need to switch over
-	 * to using it, since this is a local interrupt, which each CPU must
-	 * handle individually without locking out or dropping simultaneous
-	 * local timers on other CPUs.  We also don't want to trigger the
-	 * quirk workaround code for interrupts which gets invoked from
-	 * handle_percpu_irq via eoi, so we use our own IRQ chip.
-	 */
-	set_irq_chip_and_handler_name(0, &vmi_chip, handle_percpu_irq, "lvtt");
-#else
-	set_irq_chip_and_handler_name(0, &vmi_chip, handle_edge_irq, "lvtt");
-#endif
-	vmi_wiring = VMI_ALARM_WIRED_LVTT;
-	apic_write(APIC_LVTT, vmi_get_timer_vector());
-	local_irq_enable();
-	clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL);
-}
-
-void __devinit vmi_time_ap_init(void)
-{
-	vmi_time_init_clockevent();
-	apic_write(APIC_LVTT, vmi_get_timer_vector());
-}
-#endif
-
-/** vmi clocksource */
-static struct clocksource clocksource_vmi;
-
-static cycle_t read_real_cycles(struct clocksource *cs)
-{
-	cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL);
-	return max(ret, clocksource_vmi.cycle_last);
-}
-
-static struct clocksource clocksource_vmi = {
-	.name			= "vmi-timer",
-	.rating			= 450,
-	.read			= read_real_cycles,
-	.mask			= CLOCKSOURCE_MASK(64),
-	.mult			= 0, /* to be set */
-	.shift			= 22,
-	.flags			= CLOCK_SOURCE_IS_CONTINUOUS,
-};
-
-static int __init init_vmi_clocksource(void)
-{
-	cycle_t cycles_per_msec;
-
-	if (!vmi_timer_ops.get_cycle_frequency)
-		return 0;
-	/* Use khz2mult rather than hz2mult since hz arg is only 32-bits. */
-	cycles_per_msec = vmi_timer_ops.get_cycle_frequency();
-	(void)do_div(cycles_per_msec, 1000);
-
-	/* Note that clocksource.{mult, shift} converts in the opposite direction
-	 * as clockevents.  */
-	clocksource_vmi.mult = clocksource_khz2mult(cycles_per_msec,
-						    clocksource_vmi.shift);
-
-	printk(KERN_WARNING "vmi: registering clock source khz=%lld\n", cycles_per_msec);
-	return clocksource_register(&clocksource_vmi);
-
-}
-module_init(init_vmi_clocksource);
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index d0bb522..38e2b67 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -242,6 +242,12 @@
 		__x86_cpu_dev_end = .;
 	}
 
+	/*
+	 * start address and size of operations which during runtime
+	 * can be patched with virtualization friendly instructions or
+	 * baremetal native ones. Think page table operations.
+	 * Details in paravirt_types.h
+	 */
 	. = ALIGN(8);
 	.parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) {
 		__parainstructions = .;
@@ -249,6 +255,11 @@
 		__parainstructions_end = .;
 	}
 
+	/*
+	 * struct alt_inst entries. From the header (alternative.h):
+	 * "Alternative instructions for different CPU types or capabilities"
+	 * Think locking instructions on spinlocks.
+	 */
 	. = ALIGN(8);
 	.altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) {
 		__alt_instructions = .;
@@ -256,11 +267,28 @@
 		__alt_instructions_end = .;
 	}
 
+	/*
+	 * And here are the replacement instructions. The linker sticks
+	 * them as binary blobs. The .altinstructions has enough data to
+	 * get the address and the length of them to patch the kernel safely.
+	 */
 	.altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) {
 		*(.altinstr_replacement)
 	}
 
 	/*
+	 * struct iommu_table_entry entries are injected in this section.
+	 * It is an array of IOMMUs which during run time gets sorted depending
+	 * on its dependency order. After rootfs_initcall is complete
+	 * this section can be safely removed.
+	 */
+	.iommu_table : AT(ADDR(.iommu_table) - LOAD_OFFSET) {
+		__iommu_table = .;
+		*(.iommu_table)
+		__iommu_table_end = .;
+	}
+	. = ALIGN(8);
+	/*
 	 * .exit.text is discard at runtime, not link time, to deal with
 	 *  references from .altinstructions and .eh_frame
 	 */
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 77d8c0f..22b06f7 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -1056,14 +1056,13 @@
 
 	vcpu->arch.apic = apic;
 
-	apic->regs_page = alloc_page(GFP_KERNEL);
+	apic->regs_page = alloc_page(GFP_KERNEL|__GFP_ZERO);
 	if (apic->regs_page == NULL) {
 		printk(KERN_ERR "malloc apic regs error for vcpu %x\n",
 		       vcpu->vcpu_id);
 		goto nomem_free_apic;
 	}
 	apic->regs = page_address(apic->regs_page);
-	memset(apic->regs, 0, PAGE_SIZE);
 	apic->vcpu = vcpu;
 
 	hrtimer_init(&apic->lapic_timer.timer, CLOCK_MONOTONIC,
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index bc5b9b8..8a3f9f6 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -766,7 +766,6 @@
 
 	control->iopm_base_pa = iopm_base;
 	control->msrpm_base_pa = __pa(svm->msrpm);
-	control->tsc_offset = 0;
 	control->int_ctl = V_INTR_MASKING_MASK;
 
 	init_seg(&save->es);
@@ -902,6 +901,7 @@
 	svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT;
 	svm->asid_generation = 0;
 	init_vmcb(svm);
+	svm->vmcb->control.tsc_offset = 0-native_read_tsc();
 
 	err = fx_init(&svm->vcpu);
 	if (err)
@@ -3163,8 +3163,8 @@
 	sync_lapic_to_cr8(vcpu);
 
 	save_host_msrs(vcpu);
-	fs_selector = kvm_read_fs();
-	gs_selector = kvm_read_gs();
+	savesegment(fs, fs_selector);
+	savesegment(gs, gs_selector);
 	ldt_selector = kvm_read_ldt();
 	svm->vmcb->save.cr2 = vcpu->arch.cr2;
 	/* required for live migration with NPT */
@@ -3251,10 +3251,15 @@
 	vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
 	vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip;
 
-	kvm_load_fs(fs_selector);
-	kvm_load_gs(gs_selector);
-	kvm_load_ldt(ldt_selector);
 	load_host_msrs(vcpu);
+	loadsegment(fs, fs_selector);
+#ifdef CONFIG_X86_64
+	load_gs_index(gs_selector);
+	wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs);
+#else
+	loadsegment(gs, gs_selector);
+#endif
+	kvm_load_ldt(ldt_selector);
 
 	reload_tss(vcpu);
 
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 49b25ee..7bddfab 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -803,7 +803,7 @@
 	 */
 	vmx->host_state.ldt_sel = kvm_read_ldt();
 	vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel;
-	vmx->host_state.fs_sel = kvm_read_fs();
+	savesegment(fs, vmx->host_state.fs_sel);
 	if (!(vmx->host_state.fs_sel & 7)) {
 		vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel);
 		vmx->host_state.fs_reload_needed = 0;
@@ -811,7 +811,7 @@
 		vmcs_write16(HOST_FS_SELECTOR, 0);
 		vmx->host_state.fs_reload_needed = 1;
 	}
-	vmx->host_state.gs_sel = kvm_read_gs();
+	savesegment(gs, vmx->host_state.gs_sel);
 	if (!(vmx->host_state.gs_sel & 7))
 		vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel);
 	else {
@@ -841,27 +841,21 @@
 
 static void __vmx_load_host_state(struct vcpu_vmx *vmx)
 {
-	unsigned long flags;
-
 	if (!vmx->host_state.loaded)
 		return;
 
 	++vmx->vcpu.stat.host_state_reload;
 	vmx->host_state.loaded = 0;
 	if (vmx->host_state.fs_reload_needed)
-		kvm_load_fs(vmx->host_state.fs_sel);
+		loadsegment(fs, vmx->host_state.fs_sel);
 	if (vmx->host_state.gs_ldt_reload_needed) {
 		kvm_load_ldt(vmx->host_state.ldt_sel);
-		/*
-		 * If we have to reload gs, we must take care to
-		 * preserve our gs base.
-		 */
-		local_irq_save(flags);
-		kvm_load_gs(vmx->host_state.gs_sel);
 #ifdef CONFIG_X86_64
-		wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE));
+		load_gs_index(vmx->host_state.gs_sel);
+		wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs);
+#else
+		loadsegment(gs, vmx->host_state.gs_sel);
 #endif
-		local_irq_restore(flags);
 	}
 	reload_tss();
 #ifdef CONFIG_X86_64
@@ -2589,8 +2583,8 @@
 	vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS);  /* 22.2.4 */
 	vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
 	vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
-	vmcs_write16(HOST_FS_SELECTOR, kvm_read_fs());    /* 22.2.4 */
-	vmcs_write16(HOST_GS_SELECTOR, kvm_read_gs());    /* 22.2.4 */
+	vmcs_write16(HOST_FS_SELECTOR, 0);            /* 22.2.4 */
+	vmcs_write16(HOST_GS_SELECTOR, 0);            /* 22.2.4 */
 	vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS);  /* 22.2.4 */
 #ifdef CONFIG_X86_64
 	rdmsrl(MSR_FS_BASE, a);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 3a09c62..6c2ecf0 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1991,13 +1991,14 @@
 		0 /* Reserved */ | F(CX16) | 0 /* xTPR Update, PDCM */ |
 		0 /* Reserved, DCA */ | F(XMM4_1) |
 		F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) |
-		0 /* Reserved, AES */ | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX);
+		0 /* Reserved*/ | F(AES) | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX) |
+		F(F16C);
 	/* cpuid 0x80000001.ecx */
 	const u32 kvm_supported_word6_x86_features =
 		F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ |
 		F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) |
-		F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(SSE5) |
-		0 /* SKINIT */ | 0 /* WDT */;
+		F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) |
+		0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM);
 
 	/* all calls to cpuid_count() should be made on the same cpu */
 	get_cpu();
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index 9d5f558..73b1e1a 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -791,22 +791,22 @@
  * simple as setting a bit.  We don't actually "ack" interrupts as such, we
  * just mask and unmask them.  I wonder if we should be cleverer?
  */
-static void disable_lguest_irq(unsigned int irq)
+static void disable_lguest_irq(struct irq_data *data)
 {
-	set_bit(irq, lguest_data.blocked_interrupts);
+	set_bit(data->irq, lguest_data.blocked_interrupts);
 }
 
-static void enable_lguest_irq(unsigned int irq)
+static void enable_lguest_irq(struct irq_data *data)
 {
-	clear_bit(irq, lguest_data.blocked_interrupts);
+	clear_bit(data->irq, lguest_data.blocked_interrupts);
 }
 
 /* This structure describes the lguest IRQ controller. */
 static struct irq_chip lguest_irq_controller = {
 	.name		= "lguest",
-	.mask		= disable_lguest_irq,
-	.mask_ack	= disable_lguest_irq,
-	.unmask		= enable_lguest_irq,
+	.irq_mask	= disable_lguest_irq,
+	.irq_mask_ack	= disable_lguest_irq,
+	.irq_unmask	= enable_lguest_irq,
 };
 
 /*
@@ -838,12 +838,12 @@
  * rather than set them in lguest_init_IRQ we are called here every time an
  * lguest device needs an interrupt.
  *
- * FIXME: irq_to_desc_alloc_node() can fail due to lack of memory, we should
+ * FIXME: irq_alloc_desc_at() can fail due to lack of memory, we should
  * pass that up!
  */
 void lguest_setup_irq(unsigned int irq)
 {
-	irq_to_desc_alloc_node(irq, 0);
+	irq_alloc_desc_at(irq, 0);
 	set_irq_chip_and_handler_name(irq, &lguest_irq_controller,
 				      handle_level_irq, "level");
 }
diff --git a/arch/x86/lib/memcpy_32.c b/arch/x86/lib/memcpy_32.c
index 5415a9d..b908a59 100644
--- a/arch/x86/lib/memcpy_32.c
+++ b/arch/x86/lib/memcpy_32.c
@@ -22,22 +22,187 @@
 
 void *memmove(void *dest, const void *src, size_t n)
 {
-	int d0, d1, d2;
+	int d0,d1,d2,d3,d4,d5;
+	char *ret = dest;
 
-	if (dest < src) {
-		memcpy(dest, src, n);
-	} else {
-		__asm__ __volatile__(
-			"std\n\t"
-			"rep\n\t"
-			"movsb\n\t"
-			"cld"
-			: "=&c" (d0), "=&S" (d1), "=&D" (d2)
-			:"0" (n),
-			 "1" (n-1+src),
-			 "2" (n-1+dest)
-			:"memory");
-	}
-	return dest;
+	__asm__ __volatile__(
+		/* Handle more 16bytes in loop */
+		"cmp $0x10, %0\n\t"
+		"jb	1f\n\t"
+
+		/* Decide forward/backward copy mode */
+		"cmp %2, %1\n\t"
+		"jb	2f\n\t"
+
+		/*
+		 * movs instruction have many startup latency
+		 * so we handle small size by general register.
+		 */
+		"cmp  $680, %0\n\t"
+		"jb 3f\n\t"
+		/*
+		 * movs instruction is only good for aligned case.
+		 */
+		"mov %1, %3\n\t"
+		"xor %2, %3\n\t"
+		"and $0xff, %3\n\t"
+		"jz 4f\n\t"
+		"3:\n\t"
+		"sub $0x10, %0\n\t"
+
+		/*
+		 * We gobble 16byts forward in each loop.
+		 */
+		"3:\n\t"
+		"sub $0x10, %0\n\t"
+		"mov 0*4(%1), %3\n\t"
+		"mov 1*4(%1), %4\n\t"
+		"mov  %3, 0*4(%2)\n\t"
+		"mov  %4, 1*4(%2)\n\t"
+		"mov 2*4(%1), %3\n\t"
+		"mov 3*4(%1), %4\n\t"
+		"mov  %3, 2*4(%2)\n\t"
+		"mov  %4, 3*4(%2)\n\t"
+		"lea  0x10(%1), %1\n\t"
+		"lea  0x10(%2), %2\n\t"
+		"jae 3b\n\t"
+		"add $0x10, %0\n\t"
+		"jmp 1f\n\t"
+
+		/*
+		 * Handle data forward by movs.
+		 */
+		".p2align 4\n\t"
+		"4:\n\t"
+		"mov -4(%1, %0), %3\n\t"
+		"lea -4(%2, %0), %4\n\t"
+		"shr $2, %0\n\t"
+		"rep movsl\n\t"
+		"mov %3, (%4)\n\t"
+		"jmp 11f\n\t"
+		/*
+		 * Handle data backward by movs.
+		 */
+		".p2align 4\n\t"
+		"6:\n\t"
+		"mov (%1), %3\n\t"
+		"mov %2, %4\n\t"
+		"lea -4(%1, %0), %1\n\t"
+		"lea -4(%2, %0), %2\n\t"
+		"shr $2, %0\n\t"
+		"std\n\t"
+		"rep movsl\n\t"
+		"mov %3,(%4)\n\t"
+		"cld\n\t"
+		"jmp 11f\n\t"
+
+		/*
+		 * Start to prepare for backward copy.
+		 */
+		".p2align 4\n\t"
+		"2:\n\t"
+		"cmp  $680, %0\n\t"
+		"jb 5f\n\t"
+		"mov %1, %3\n\t"
+		"xor %2, %3\n\t"
+		"and $0xff, %3\n\t"
+		"jz 6b\n\t"
+
+		/*
+		 * Calculate copy position to tail.
+		 */
+		"5:\n\t"
+		"add %0, %1\n\t"
+		"add %0, %2\n\t"
+		"sub $0x10, %0\n\t"
+
+		/*
+		 * We gobble 16byts backward in each loop.
+		 */
+		"7:\n\t"
+		"sub $0x10, %0\n\t"
+
+		"mov -1*4(%1), %3\n\t"
+		"mov -2*4(%1), %4\n\t"
+		"mov  %3, -1*4(%2)\n\t"
+		"mov  %4, -2*4(%2)\n\t"
+		"mov -3*4(%1), %3\n\t"
+		"mov -4*4(%1), %4\n\t"
+		"mov  %3, -3*4(%2)\n\t"
+		"mov  %4, -4*4(%2)\n\t"
+		"lea  -0x10(%1), %1\n\t"
+		"lea  -0x10(%2), %2\n\t"
+		"jae 7b\n\t"
+		/*
+		 * Calculate copy position to head.
+		 */
+		"add $0x10, %0\n\t"
+		"sub %0, %1\n\t"
+		"sub %0, %2\n\t"
+
+		/*
+		 * Move data from 8 bytes to 15 bytes.
+		 */
+		".p2align 4\n\t"
+		"1:\n\t"
+		"cmp $8, %0\n\t"
+		"jb 8f\n\t"
+		"mov 0*4(%1), %3\n\t"
+		"mov 1*4(%1), %4\n\t"
+		"mov -2*4(%1, %0), %5\n\t"
+		"mov -1*4(%1, %0), %1\n\t"
+
+		"mov  %3, 0*4(%2)\n\t"
+		"mov  %4, 1*4(%2)\n\t"
+		"mov  %5, -2*4(%2, %0)\n\t"
+		"mov  %1, -1*4(%2, %0)\n\t"
+		"jmp 11f\n\t"
+
+		/*
+		 * Move data from 4 bytes to 7 bytes.
+		 */
+		".p2align 4\n\t"
+		"8:\n\t"
+		"cmp $4, %0\n\t"
+		"jb 9f\n\t"
+		"mov 0*4(%1), %3\n\t"
+		"mov -1*4(%1, %0), %4\n\t"
+		"mov  %3, 0*4(%2)\n\t"
+		"mov  %4, -1*4(%2, %0)\n\t"
+		"jmp 11f\n\t"
+
+		/*
+		 * Move data from 2 bytes to 3 bytes.
+		 */
+		".p2align 4\n\t"
+		"9:\n\t"
+		"cmp $2, %0\n\t"
+		"jb 10f\n\t"
+		"movw 0*2(%1), %%dx\n\t"
+		"movw -1*2(%1, %0), %%bx\n\t"
+		"movw %%dx, 0*2(%2)\n\t"
+		"movw %%bx, -1*2(%2, %0)\n\t"
+		"jmp 11f\n\t"
+
+		/*
+		 * Move data for 1 byte.
+		 */
+		".p2align 4\n\t"
+		"10:\n\t"
+		"cmp $1, %0\n\t"
+		"jb 11f\n\t"
+		"movb (%1), %%cl\n\t"
+		"movb %%cl, (%2)\n\t"
+		".p2align 4\n\t"
+		"11:"
+		: "=&c" (d0), "=&S" (d1), "=&D" (d2),
+		  "=r" (d3),"=r" (d4), "=r"(d5)
+		:"0" (n),
+		 "1" (src),
+		 "2" (dest)
+		:"memory");
+
+	return ret;
+
 }
 EXPORT_SYMBOL(memmove);
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index bcbcd1e..75ef61e 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -40,84 +40,132 @@
 ENTRY(__memcpy)
 ENTRY(memcpy)
 	CFI_STARTPROC
-
-	/*
-	 * Put the number of full 64-byte blocks into %ecx.
-	 * Tail portion is handled at the end:
-	 */
 	movq %rdi, %rax
-	movl %edx, %ecx
-	shrl   $6, %ecx
-	jz .Lhandle_tail
 
+	/*
+	 * Use 32bit CMP here to avoid long NOP padding.
+	 */
+	cmp  $0x20, %edx
+	jb .Lhandle_tail
+
+	/*
+	 * We check whether memory false dependece could occur,
+	 * then jump to corresponding copy mode.
+	 */
+	cmp  %dil, %sil
+	jl .Lcopy_backward
+	subl $0x20, %edx
+.Lcopy_forward_loop:
+	subq $0x20,	%rdx
+
+	/*
+	 * Move in blocks of 4x8 bytes:
+	 */
+	movq 0*8(%rsi),	%r8
+	movq 1*8(%rsi),	%r9
+	movq 2*8(%rsi),	%r10
+	movq 3*8(%rsi),	%r11
+	leaq 4*8(%rsi),	%rsi
+
+	movq %r8,	0*8(%rdi)
+	movq %r9,	1*8(%rdi)
+	movq %r10,	2*8(%rdi)
+	movq %r11,	3*8(%rdi)
+	leaq 4*8(%rdi),	%rdi
+	jae  .Lcopy_forward_loop
+	addq $0x20,	%rdx
+	jmp  .Lhandle_tail
+
+.Lcopy_backward:
+	/*
+	 * Calculate copy position to tail.
+	 */
+	addq %rdx,	%rsi
+	addq %rdx,	%rdi
+	subq $0x20,	%rdx
+	/*
+	 * At most 3 ALU operations in one cycle,
+	 * so append NOPS in the same 16bytes trunk.
+	 */
 	.p2align 4
-.Lloop_64:
-	/*
-	 * We decrement the loop index here - and the zero-flag is
-	 * checked at the end of the loop (instructions inbetween do
-	 * not change the zero flag):
-	 */
-	decl %ecx
+.Lcopy_backward_loop:
+	subq $0x20,	%rdx
+	movq -1*8(%rsi),	%r8
+	movq -2*8(%rsi),	%r9
+	movq -3*8(%rsi),	%r10
+	movq -4*8(%rsi),	%r11
+	leaq -4*8(%rsi),	%rsi
+	movq %r8,		-1*8(%rdi)
+	movq %r9,		-2*8(%rdi)
+	movq %r10,		-3*8(%rdi)
+	movq %r11,		-4*8(%rdi)
+	leaq -4*8(%rdi),	%rdi
+	jae  .Lcopy_backward_loop
 
 	/*
-	 * Move in blocks of 4x16 bytes:
+	 * Calculate copy position to head.
 	 */
-	movq 0*8(%rsi),		%r11
-	movq 1*8(%rsi),		%r8
-	movq %r11,		0*8(%rdi)
-	movq %r8,		1*8(%rdi)
-
-	movq 2*8(%rsi),		%r9
-	movq 3*8(%rsi),		%r10
-	movq %r9,		2*8(%rdi)
-	movq %r10,		3*8(%rdi)
-
-	movq 4*8(%rsi),		%r11
-	movq 5*8(%rsi),		%r8
-	movq %r11,		4*8(%rdi)
-	movq %r8,		5*8(%rdi)
-
-	movq 6*8(%rsi),		%r9
-	movq 7*8(%rsi),		%r10
-	movq %r9,		6*8(%rdi)
-	movq %r10,		7*8(%rdi)
-
-	leaq 64(%rsi), %rsi
-	leaq 64(%rdi), %rdi
-
-	jnz  .Lloop_64
-
+	addq $0x20,	%rdx
+	subq %rdx,	%rsi
+	subq %rdx,	%rdi
 .Lhandle_tail:
-	movl %edx, %ecx
-	andl  $63, %ecx
-	shrl   $3, %ecx
-	jz   .Lhandle_7
+	cmpq $16,	%rdx
+	jb   .Lless_16bytes
 
+	/*
+	 * Move data from 16 bytes to 31 bytes.
+	 */
+	movq 0*8(%rsi), %r8
+	movq 1*8(%rsi),	%r9
+	movq -2*8(%rsi, %rdx),	%r10
+	movq -1*8(%rsi, %rdx),	%r11
+	movq %r8,	0*8(%rdi)
+	movq %r9,	1*8(%rdi)
+	movq %r10,	-2*8(%rdi, %rdx)
+	movq %r11,	-1*8(%rdi, %rdx)
+	retq
 	.p2align 4
-.Lloop_8:
-	decl %ecx
-	movq (%rsi),		%r8
-	movq %r8,		(%rdi)
-	leaq 8(%rdi),		%rdi
-	leaq 8(%rsi),		%rsi
-	jnz  .Lloop_8
-
-.Lhandle_7:
-	movl %edx, %ecx
-	andl $7, %ecx
-	jz .Lend
-
+.Lless_16bytes:
+	cmpq $8,	%rdx
+	jb   .Lless_8bytes
+	/*
+	 * Move data from 8 bytes to 15 bytes.
+	 */
+	movq 0*8(%rsi),	%r8
+	movq -1*8(%rsi, %rdx),	%r9
+	movq %r8,	0*8(%rdi)
+	movq %r9,	-1*8(%rdi, %rdx)
+	retq
 	.p2align 4
+.Lless_8bytes:
+	cmpq $4,	%rdx
+	jb   .Lless_3bytes
+
+	/*
+	 * Move data from 4 bytes to 7 bytes.
+	 */
+	movl (%rsi), %ecx
+	movl -4(%rsi, %rdx), %r8d
+	movl %ecx, (%rdi)
+	movl %r8d, -4(%rdi, %rdx)
+	retq
+	.p2align 4
+.Lless_3bytes:
+	cmpl $0, %edx
+	je .Lend
+	/*
+	 * Move data from 1 bytes to 3 bytes.
+	 */
 .Lloop_1:
 	movb (%rsi), %r8b
 	movb %r8b, (%rdi)
 	incq %rdi
 	incq %rsi
-	decl %ecx
+	decl %edx
 	jnz .Lloop_1
 
 .Lend:
-	ret
+	retq
 	CFI_ENDPROC
 ENDPROC(memcpy)
 ENDPROC(__memcpy)
diff --git a/arch/x86/lib/memmove_64.c b/arch/x86/lib/memmove_64.c
index 0a33909..6d0f0ec 100644
--- a/arch/x86/lib/memmove_64.c
+++ b/arch/x86/lib/memmove_64.c
@@ -8,14 +8,185 @@
 #undef memmove
 void *memmove(void *dest, const void *src, size_t count)
 {
-	if (dest < src) {
-		return memcpy(dest, src, count);
-	} else {
-		char *p = dest + count;
-		const char *s = src + count;
-		while (count--)
-			*--p = *--s;
-	}
-	return dest;
+	unsigned long d0,d1,d2,d3,d4,d5,d6,d7;
+	char *ret;
+
+	__asm__ __volatile__(
+		/* Handle more 32bytes in loop */
+		"mov %2, %3\n\t"
+		"cmp $0x20, %0\n\t"
+		"jb	1f\n\t"
+
+		/* Decide forward/backward copy mode */
+		"cmp %2, %1\n\t"
+		"jb	2f\n\t"
+
+		/*
+		 * movsq instruction have many startup latency
+		 * so we handle small size by general register.
+		 */
+		"cmp  $680, %0\n\t"
+		"jb 3f\n\t"
+		/*
+		 * movsq instruction is only good for aligned case.
+		 */
+		"cmpb %%dil, %%sil\n\t"
+		"je 4f\n\t"
+		"3:\n\t"
+		"sub $0x20, %0\n\t"
+		/*
+		 * We gobble 32byts forward in each loop.
+		 */
+		"5:\n\t"
+		"sub $0x20, %0\n\t"
+		"movq 0*8(%1), %4\n\t"
+		"movq 1*8(%1), %5\n\t"
+		"movq 2*8(%1), %6\n\t"
+		"movq 3*8(%1), %7\n\t"
+		"leaq 4*8(%1), %1\n\t"
+
+		"movq %4, 0*8(%2)\n\t"
+		"movq %5, 1*8(%2)\n\t"
+		"movq %6, 2*8(%2)\n\t"
+		"movq %7, 3*8(%2)\n\t"
+		"leaq 4*8(%2), %2\n\t"
+		"jae 5b\n\t"
+		"addq $0x20, %0\n\t"
+		"jmp 1f\n\t"
+		/*
+		 * Handle data forward by movsq.
+		 */
+		".p2align 4\n\t"
+		"4:\n\t"
+		"movq %0, %8\n\t"
+		"movq -8(%1, %0), %4\n\t"
+		"lea -8(%2, %0), %5\n\t"
+		"shrq $3, %8\n\t"
+		"rep movsq\n\t"
+		"movq %4, (%5)\n\t"
+		"jmp 13f\n\t"
+		/*
+		 * Handle data backward by movsq.
+		 */
+		".p2align 4\n\t"
+		"7:\n\t"
+		"movq %0, %8\n\t"
+		"movq (%1), %4\n\t"
+		"movq %2, %5\n\t"
+		"leaq -8(%1, %0), %1\n\t"
+		"leaq -8(%2, %0), %2\n\t"
+		"shrq $3, %8\n\t"
+		"std\n\t"
+		"rep movsq\n\t"
+		"cld\n\t"
+		"movq %4, (%5)\n\t"
+		"jmp 13f\n\t"
+
+		/*
+		 * Start to prepare for backward copy.
+		 */
+		".p2align 4\n\t"
+		"2:\n\t"
+		"cmp $680, %0\n\t"
+		"jb 6f \n\t"
+		"cmp %%dil, %%sil\n\t"
+		"je 7b \n\t"
+		"6:\n\t"
+		/*
+		 * Calculate copy position to tail.
+		 */
+		"addq %0, %1\n\t"
+		"addq %0, %2\n\t"
+		"subq $0x20, %0\n\t"
+		/*
+		 * We gobble 32byts backward in each loop.
+		 */
+		"8:\n\t"
+		"subq $0x20, %0\n\t"
+		"movq -1*8(%1), %4\n\t"
+		"movq -2*8(%1), %5\n\t"
+		"movq -3*8(%1), %6\n\t"
+		"movq -4*8(%1), %7\n\t"
+		"leaq -4*8(%1), %1\n\t"
+
+		"movq %4, -1*8(%2)\n\t"
+		"movq %5, -2*8(%2)\n\t"
+		"movq %6, -3*8(%2)\n\t"
+		"movq %7, -4*8(%2)\n\t"
+		"leaq -4*8(%2), %2\n\t"
+		"jae 8b\n\t"
+		/*
+		 * Calculate copy position to head.
+		 */
+		"addq $0x20, %0\n\t"
+		"subq %0, %1\n\t"
+		"subq %0, %2\n\t"
+		"1:\n\t"
+		"cmpq $16, %0\n\t"
+		"jb 9f\n\t"
+		/*
+		 * Move data from 16 bytes to 31 bytes.
+		 */
+		"movq 0*8(%1), %4\n\t"
+		"movq 1*8(%1), %5\n\t"
+		"movq -2*8(%1, %0), %6\n\t"
+		"movq -1*8(%1, %0), %7\n\t"
+		"movq %4, 0*8(%2)\n\t"
+		"movq %5, 1*8(%2)\n\t"
+		"movq %6, -2*8(%2, %0)\n\t"
+		"movq %7, -1*8(%2, %0)\n\t"
+		"jmp 13f\n\t"
+		".p2align 4\n\t"
+		"9:\n\t"
+		"cmpq $8, %0\n\t"
+		"jb 10f\n\t"
+		/*
+		 * Move data from 8 bytes to 15 bytes.
+		 */
+		"movq 0*8(%1), %4\n\t"
+		"movq -1*8(%1, %0), %5\n\t"
+		"movq %4, 0*8(%2)\n\t"
+		"movq %5, -1*8(%2, %0)\n\t"
+		"jmp 13f\n\t"
+		"10:\n\t"
+		"cmpq $4, %0\n\t"
+		"jb 11f\n\t"
+		/*
+		 * Move data from 4 bytes to 7 bytes.
+		 */
+		"movl (%1), %4d\n\t"
+		"movl -4(%1, %0), %5d\n\t"
+		"movl %4d, (%2)\n\t"
+		"movl %5d, -4(%2, %0)\n\t"
+		"jmp 13f\n\t"
+		"11:\n\t"
+		"cmp $2, %0\n\t"
+		"jb 12f\n\t"
+		/*
+		 * Move data from 2 bytes to 3 bytes.
+		 */
+		"movw (%1), %4w\n\t"
+		"movw -2(%1, %0), %5w\n\t"
+		"movw %4w, (%2)\n\t"
+		"movw %5w, -2(%2, %0)\n\t"
+		"jmp 13f\n\t"
+		"12:\n\t"
+		"cmp $1, %0\n\t"
+		"jb 13f\n\t"
+		/*
+		 * Move data for 1 byte.
+		 */
+		"movb (%1), %4b\n\t"
+		"movb %4b, (%2)\n\t"
+		"13:\n\t"
+		: "=&d" (d0), "=&S" (d1), "=&D" (d2), "=&a" (ret) ,
+		  "=r"(d3), "=r"(d4), "=r"(d5), "=r"(d6), "=&c" (d7)
+		:"0" (count),
+		 "1" (src),
+		 "2" (dest)
+		:"memory");
+
+		return ret;
+
 }
 EXPORT_SYMBOL(memmove);
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 4c4508e..79b0b37 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -229,7 +229,16 @@
 
 		spin_lock_irqsave(&pgd_lock, flags);
 		list_for_each_entry(page, &pgd_list, lru) {
-			if (!vmalloc_sync_one(page_address(page), address))
+			spinlock_t *pgt_lock;
+			pmd_t *ret;
+
+			pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
+
+			spin_lock(pgt_lock);
+			ret = vmalloc_sync_one(page_address(page), address);
+			spin_unlock(pgt_lock);
+
+			if (!ret)
 				break;
 		}
 		spin_unlock_irqrestore(&pgd_lock, flags);
@@ -251,6 +260,8 @@
 	if (!(address >= VMALLOC_START && address < VMALLOC_END))
 		return -1;
 
+	WARN_ON_ONCE(in_nmi());
+
 	/*
 	 * Synchronize this task's top level page-table
 	 * with the 'reference' page table.
@@ -326,29 +337,7 @@
 
 void vmalloc_sync_all(void)
 {
-	unsigned long address;
-
-	for (address = VMALLOC_START & PGDIR_MASK; address <= VMALLOC_END;
-	     address += PGDIR_SIZE) {
-
-		const pgd_t *pgd_ref = pgd_offset_k(address);
-		unsigned long flags;
-		struct page *page;
-
-		if (pgd_none(*pgd_ref))
-			continue;
-
-		spin_lock_irqsave(&pgd_lock, flags);
-		list_for_each_entry(page, &pgd_list, lru) {
-			pgd_t *pgd;
-			pgd = (pgd_t *)page_address(page) + pgd_index(address);
-			if (pgd_none(*pgd))
-				set_pgd(pgd, *pgd_ref);
-			else
-				BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
-		}
-		spin_unlock_irqrestore(&pgd_lock, flags);
-	}
+	sync_global_pgds(VMALLOC_START & PGDIR_MASK, VMALLOC_END);
 }
 
 /*
@@ -369,6 +358,8 @@
 	if (!(address >= VMALLOC_START && address < VMALLOC_END))
 		return -1;
 
+	WARN_ON_ONCE(in_nmi());
+
 	/*
 	 * Copy kernel mappings over when needed. This can also
 	 * happen within a race in page table update. In the later
@@ -894,8 +885,14 @@
 	if (pmd_large(*pmd))
 		return spurious_fault_check(error_code, (pte_t *) pmd);
 
+	/*
+	 * Note: don't use pte_present() here, since it returns true
+	 * if the _PAGE_PROTNONE bit is set.  However, this aliases the
+	 * _PAGE_GLOBAL bit, which for kernel pages give false positives
+	 * when CONFIG_DEBUG_PAGEALLOC is used.
+	 */
 	pte = pte_offset_kernel(pmd, address);
-	if (!pte_present(*pte))
+	if (!(pte_flags(*pte) & _PAGE_PRESENT))
 		return 0;
 
 	ret = spurious_fault_check(error_code, pte);
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index bca7909..558f2d3 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -67,7 +67,7 @@
 		panic("alloc_low_page: ran out of memory");
 
 	adr = __va(pfn * PAGE_SIZE);
-	memset(adr, 0, PAGE_SIZE);
+	clear_page(adr);
 	return adr;
 }
 
@@ -558,7 +558,7 @@
 
 static inline void save_pg_dir(void)
 {
-	memcpy(swsusp_pg_dir, swapper_pg_dir, PAGE_SIZE);
+	copy_page(swsusp_pg_dir, swapper_pg_dir);
 }
 #else /* !CONFIG_ACPI_SLEEP */
 static inline void save_pg_dir(void)
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 9a66746..c55f900 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -98,6 +98,43 @@
 __setup("noexec32=", nonx32_setup);
 
 /*
+ * When memory was added/removed make sure all the processes MM have
+ * suitable PGD entries in the local PGD level page.
+ */
+void sync_global_pgds(unsigned long start, unsigned long end)
+{
+	unsigned long address;
+
+	for (address = start; address <= end; address += PGDIR_SIZE) {
+		const pgd_t *pgd_ref = pgd_offset_k(address);
+		unsigned long flags;
+		struct page *page;
+
+		if (pgd_none(*pgd_ref))
+			continue;
+
+		spin_lock_irqsave(&pgd_lock, flags);
+		list_for_each_entry(page, &pgd_list, lru) {
+			pgd_t *pgd;
+			spinlock_t *pgt_lock;
+
+			pgd = (pgd_t *)page_address(page) + pgd_index(address);
+			pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
+			spin_lock(pgt_lock);
+
+			if (pgd_none(*pgd))
+				set_pgd(pgd, *pgd_ref);
+			else
+				BUG_ON(pgd_page_vaddr(*pgd)
+				       != pgd_page_vaddr(*pgd_ref));
+
+			spin_unlock(pgt_lock);
+		}
+		spin_unlock_irqrestore(&pgd_lock, flags);
+	}
+}
+
+/*
  * NOTE: This function is marked __ref because it calls __init function
  * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0.
  */
@@ -293,7 +330,7 @@
 		panic("alloc_low_page: ran out of memory");
 
 	adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE);
-	memset(adr, 0, PAGE_SIZE);
+	clear_page(adr);
 	*phys  = pfn * PAGE_SIZE;
 	return adr;
 }
@@ -534,11 +571,13 @@
 			     unsigned long end,
 			     unsigned long page_size_mask)
 {
-
+	bool pgd_changed = false;
 	unsigned long next, last_map_addr = end;
+	unsigned long addr;
 
 	start = (unsigned long)__va(start);
 	end = (unsigned long)__va(end);
+	addr = start;
 
 	for (; start < end; start = next) {
 		pgd_t *pgd = pgd_offset_k(start);
@@ -563,7 +602,12 @@
 		spin_lock(&init_mm.page_table_lock);
 		pgd_populate(&init_mm, pgd, __va(pud_phys));
 		spin_unlock(&init_mm.page_table_lock);
+		pgd_changed = true;
 	}
+
+	if (pgd_changed)
+		sync_global_pgds(addr, end);
+
 	__flush_tlb_all();
 
 	return last_map_addr;
@@ -1003,6 +1047,7 @@
 		}
 
 	}
+	sync_global_pgds((unsigned long)start_page, end);
 	return 0;
 }
 
diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c
index 970ed57..52d54bf 100644
--- a/arch/x86/mm/k8topology_64.c
+++ b/arch/x86/mm/k8topology_64.c
@@ -22,7 +22,7 @@
 #include <asm/numa.h>
 #include <asm/mpspec.h>
 #include <asm/apic.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 
 static struct bootnode __initdata nodes[8];
 static nodemask_t __initdata nodes_parsed = NODE_MASK_NONE;
@@ -54,8 +54,8 @@
 static __init void early_get_boot_cpu_id(void)
 {
 	/*
-	 * need to get boot_cpu_id so can use that to create apicid_to_node
-	 * in k8_scan_nodes()
+	 * need to get the APIC ID of the BSP so can use that to
+	 * create apicid_to_node in k8_scan_nodes()
 	 */
 #ifdef CONFIG_X86_MPPARSE
 	/*
@@ -212,7 +212,7 @@
 	bits = boot_cpu_data.x86_coreid_bits;
 	cores = (1<<bits);
 	apicid_base = 0;
-	/* need to get boot_cpu_id early for system with apicid lifting */
+	/* get the APIC ID of the BSP early for systems with apicid lifting */
 	early_get_boot_cpu_id();
 	if (boot_cpu_physical_apicid > 0) {
 		pr_info("BSP APIC ID: %02x\n", boot_cpu_physical_apicid);
diff --git a/arch/x86/mm/kmemcheck/kmemcheck.c b/arch/x86/mm/kmemcheck/kmemcheck.c
index b3b531a..d87dd6d 100644
--- a/arch/x86/mm/kmemcheck/kmemcheck.c
+++ b/arch/x86/mm/kmemcheck/kmemcheck.c
@@ -631,6 +631,8 @@
 	if (!pte)
 		return false;
 
+	WARN_ON_ONCE(in_nmi());
+
 	if (error_code & 2)
 		kmemcheck_access(regs, address, KMEMCHECK_WRITE);
 	else
diff --git a/arch/x86/mm/kmemcheck/opcode.c b/arch/x86/mm/kmemcheck/opcode.c
index 63c19e2..324aa3f 100644
--- a/arch/x86/mm/kmemcheck/opcode.c
+++ b/arch/x86/mm/kmemcheck/opcode.c
@@ -9,7 +9,7 @@
 		b == 0xf0 || b == 0xf2 || b == 0xf3
 		/* Group 2 */
 		|| b == 0x2e || b == 0x36 || b == 0x3e || b == 0x26
-		|| b == 0x64 || b == 0x65 || b == 0x2e || b == 0x3e
+		|| b == 0x64 || b == 0x65
 		/* Group 3 */
 		|| b == 0x66
 		/* Group 4 */
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index a7bcc23..4962f1a 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -18,7 +18,7 @@
 #include <asm/dma.h>
 #include <asm/numa.h>
 #include <asm/acpi.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 5c4ee42..8be8c7d 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -87,7 +87,19 @@
 #define UNSHARED_PTRS_PER_PGD				\
 	(SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD)
 
-static void pgd_ctor(pgd_t *pgd)
+
+static void pgd_set_mm(pgd_t *pgd, struct mm_struct *mm)
+{
+	BUILD_BUG_ON(sizeof(virt_to_page(pgd)->index) < sizeof(mm));
+	virt_to_page(pgd)->index = (pgoff_t)mm;
+}
+
+struct mm_struct *pgd_page_get_mm(struct page *page)
+{
+	return (struct mm_struct *)page->index;
+}
+
+static void pgd_ctor(struct mm_struct *mm, pgd_t *pgd)
 {
 	/* If the pgd points to a shared pagetable level (either the
 	   ptes in non-PAE, or shared PMD in PAE), then just copy the
@@ -98,15 +110,13 @@
 		clone_pgd_range(pgd + KERNEL_PGD_BOUNDARY,
 				swapper_pg_dir + KERNEL_PGD_BOUNDARY,
 				KERNEL_PGD_PTRS);
-		paravirt_alloc_pmd_clone(__pa(pgd) >> PAGE_SHIFT,
-					 __pa(swapper_pg_dir) >> PAGE_SHIFT,
-					 KERNEL_PGD_BOUNDARY,
-					 KERNEL_PGD_PTRS);
 	}
 
 	/* list required to sync kernel mapping updates */
-	if (!SHARED_KERNEL_PMD)
+	if (!SHARED_KERNEL_PMD) {
+		pgd_set_mm(pgd, mm);
 		pgd_list_add(pgd);
+	}
 }
 
 static void pgd_dtor(pgd_t *pgd)
@@ -272,7 +282,7 @@
 	 */
 	spin_lock_irqsave(&pgd_lock, flags);
 
-	pgd_ctor(pgd);
+	pgd_ctor(mm, pgd);
 	pgd_prepopulate_pmd(mm, pgd, pmds);
 
 	spin_unlock_irqrestore(&pgd_lock, flags);
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index f9897f7..9c0d0d3 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -420,9 +420,11 @@
 		return -1;
 	}
 
-	for_each_node_mask(i, nodes_parsed)
-		e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT,
-						nodes[i].end >> PAGE_SHIFT);
+	for (i = 0; i < num_node_memblks; i++)
+		e820_register_active_regions(memblk_nodeid[i],
+				node_memblk_range[i].start >> PAGE_SHIFT,
+				node_memblk_range[i].end >> PAGE_SHIFT);
+
 	/* for out of order entries in SRAT */
 	sort_node_map();
 	if (!nodes_cover_memory(nodes)) {
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index c03f14a..4935848 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -5,6 +5,7 @@
 #include <linux/smp.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
+#include <linux/cpu.h>
 
 #include <asm/tlbflush.h>
 #include <asm/mmu_context.h>
@@ -52,6 +53,8 @@
    want false sharing in the per cpu data segment. */
 static union smp_flush_state flush_state[NUM_INVALIDATE_TLB_VECTORS];
 
+static DEFINE_PER_CPU_READ_MOSTLY(int, tlb_vector_offset);
+
 /*
  * We cannot call mmdrop() because we are in interrupt context,
  * instead update mm->cpu_vm_mask.
@@ -173,7 +176,7 @@
 	union smp_flush_state *f;
 
 	/* Caller has disabled preemption */
-	sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS;
+	sender = this_cpu_read(tlb_vector_offset);
 	f = &flush_state[sender];
 
 	/*
@@ -218,6 +221,47 @@
 	flush_tlb_others_ipi(cpumask, mm, va);
 }
 
+static void __cpuinit calculate_tlb_offset(void)
+{
+	int cpu, node, nr_node_vecs;
+	/*
+	 * we are changing tlb_vector_offset for each CPU in runtime, but this
+	 * will not cause inconsistency, as the write is atomic under X86. we
+	 * might see more lock contentions in a short time, but after all CPU's
+	 * tlb_vector_offset are changed, everything should go normal
+	 *
+	 * Note: if NUM_INVALIDATE_TLB_VECTORS % nr_online_nodes !=0, we might
+	 * waste some vectors.
+	 **/
+	if (nr_online_nodes > NUM_INVALIDATE_TLB_VECTORS)
+		nr_node_vecs = 1;
+	else
+		nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
+
+	for_each_online_node(node) {
+		int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
+			nr_node_vecs;
+		int cpu_offset = 0;
+		for_each_cpu(cpu, cpumask_of_node(node)) {
+			per_cpu(tlb_vector_offset, cpu) = node_offset +
+				cpu_offset;
+			cpu_offset++;
+			cpu_offset = cpu_offset % nr_node_vecs;
+		}
+	}
+}
+
+static int tlb_cpuhp_notify(struct notifier_block *n,
+		unsigned long action, void *hcpu)
+{
+	switch (action & 0xf) {
+	case CPU_ONLINE:
+	case CPU_DEAD:
+		calculate_tlb_offset();
+	}
+	return NOTIFY_OK;
+}
+
 static int __cpuinit init_smp_flush(void)
 {
 	int i;
@@ -225,6 +269,8 @@
 	for (i = 0; i < ARRAY_SIZE(flush_state); i++)
 		raw_spin_lock_init(&flush_state[i].tlbstate_lock);
 
+	calculate_tlb_offset();
+	hotcpu_notifier(tlb_cpuhp_notify, 0);
 	return 0;
 }
 core_initcall(init_smp_flush);
diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c
index 3855096..2d49d4e 100644
--- a/arch/x86/oprofile/backtrace.c
+++ b/arch/x86/oprofile/backtrace.c
@@ -14,6 +14,7 @@
 #include <asm/ptrace.h>
 #include <asm/uaccess.h>
 #include <asm/stacktrace.h>
+#include <linux/compat.h>
 
 static void backtrace_warning_symbol(void *data, char *msg,
 				     unsigned long symbol)
@@ -48,14 +49,12 @@
 	.walk_stack	= print_context_stack,
 };
 
-struct frame_head {
-	struct frame_head *bp;
-	unsigned long ret;
-} __attribute__((packed));
-
-static struct frame_head *dump_user_backtrace(struct frame_head *head)
+#ifdef CONFIG_COMPAT
+static struct stack_frame_ia32 *
+dump_user_backtrace_32(struct stack_frame_ia32 *head)
 {
-	struct frame_head bufhead[2];
+	struct stack_frame_ia32 bufhead[2];
+	struct stack_frame_ia32 *fp;
 
 	/* Also check accessibility of one struct frame_head beyond */
 	if (!access_ok(VERIFY_READ, head, sizeof(bufhead)))
@@ -63,20 +62,66 @@
 	if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead)))
 		return NULL;
 
-	oprofile_add_trace(bufhead[0].ret);
+	fp = (struct stack_frame_ia32 *) compat_ptr(bufhead[0].next_frame);
+
+	oprofile_add_trace(bufhead[0].return_address);
+
+	/* frame pointers should strictly progress back up the stack
+	* (towards higher addresses) */
+	if (head >= fp)
+		return NULL;
+
+	return fp;
+}
+
+static inline int
+x86_backtrace_32(struct pt_regs * const regs, unsigned int depth)
+{
+	struct stack_frame_ia32 *head;
+
+	/* User process is 32-bit */
+	if (!current || !test_thread_flag(TIF_IA32))
+		return 0;
+
+	head = (struct stack_frame_ia32 *) regs->bp;
+	while (depth-- && head)
+		head = dump_user_backtrace_32(head);
+
+	return 1;
+}
+
+#else
+static inline int
+x86_backtrace_32(struct pt_regs * const regs, unsigned int depth)
+{
+	return 0;
+}
+#endif /* CONFIG_COMPAT */
+
+static struct stack_frame *dump_user_backtrace(struct stack_frame *head)
+{
+	struct stack_frame bufhead[2];
+
+	/* Also check accessibility of one struct stack_frame beyond */
+	if (!access_ok(VERIFY_READ, head, sizeof(bufhead)))
+		return NULL;
+	if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead)))
+		return NULL;
+
+	oprofile_add_trace(bufhead[0].return_address);
 
 	/* frame pointers should strictly progress back up the stack
 	 * (towards higher addresses) */
-	if (head >= bufhead[0].bp)
+	if (head >= bufhead[0].next_frame)
 		return NULL;
 
-	return bufhead[0].bp;
+	return bufhead[0].next_frame;
 }
 
 void
 x86_backtrace(struct pt_regs * const regs, unsigned int depth)
 {
-	struct frame_head *head = (struct frame_head *)frame_pointer(regs);
+	struct stack_frame *head = (struct stack_frame *)frame_pointer(regs);
 
 	if (!user_mode_vm(regs)) {
 		unsigned long stack = kernel_stack_pointer(regs);
@@ -86,6 +131,9 @@
 		return;
 	}
 
+	if (x86_backtrace_32(regs, depth))
+		return;
+
 	while (depth-- && head)
 		head = dump_user_backtrace(head);
 }
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index f1575c9..bd1489c 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -695,9 +695,6 @@
 	return 1;
 }
 
-/* in order to get sysfs right */
-static int using_nmi;
-
 int __init op_nmi_init(struct oprofile_operations *ops)
 {
 	__u8 vendor = boot_cpu_data.x86_vendor;
@@ -705,8 +702,6 @@
 	char *cpu_type = NULL;
 	int ret = 0;
 
-	using_nmi = 0;
-
 	if (!cpu_has_apic)
 		return -ENODEV;
 
@@ -790,13 +785,11 @@
 	if (ret)
 		return ret;
 
-	using_nmi = 1;
 	printk(KERN_INFO "oprofile: using NMI interrupt.\n");
 	return 0;
 }
 
 void op_nmi_exit(void)
 {
-	if (using_nmi)
-		exit_sysfs();
+	exit_sysfs();
 }
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index b67a6b5..42fb46f 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -64,15 +64,22 @@
  * IBS cpuid feature detection
  */
 
-#define IBS_CPUID_FEATURES      0x8000001b
+#define IBS_CPUID_FEATURES		0x8000001b
 
 /*
  * Same bit mask as for IBS cpuid feature flags (Fn8000_001B_EAX), but
  * bit 0 is used to indicate the existence of IBS.
  */
-#define IBS_CAPS_AVAIL			(1LL<<0)
-#define IBS_CAPS_RDWROPCNT		(1LL<<3)
-#define IBS_CAPS_OPCNT			(1LL<<4)
+#define IBS_CAPS_AVAIL			(1U<<0)
+#define IBS_CAPS_RDWROPCNT		(1U<<3)
+#define IBS_CAPS_OPCNT			(1U<<4)
+
+/*
+ * IBS APIC setup
+ */
+#define IBSCTL				0x1cc
+#define IBSCTL_LVT_OFFSET_VALID		(1ULL<<8)
+#define IBSCTL_LVT_OFFSET_MASK		0x0F
 
 /*
  * IBS randomization macros
@@ -266,6 +273,74 @@
 		wrmsrl(MSR_AMD64_IBSOPCTL, 0);
 }
 
+static inline int eilvt_is_available(int offset)
+{
+	/* check if we may assign a vector */
+	return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1);
+}
+
+static inline int ibs_eilvt_valid(void)
+{
+	u64 val;
+	int offset;
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	if (!(val & IBSCTL_LVT_OFFSET_VALID)) {
+		pr_err(FW_BUG "cpu %d, invalid IBS "
+		       "interrupt offset %d (MSR%08X=0x%016llx)",
+		       smp_processor_id(), offset,
+		       MSR_AMD64_IBSCTL, val);
+		return 0;
+	}
+
+	offset = val & IBSCTL_LVT_OFFSET_MASK;
+
+	if (eilvt_is_available(offset))
+		return !0;
+
+	pr_err(FW_BUG "cpu %d, IBS interrupt offset %d "
+	       "not available (MSR%08X=0x%016llx)",
+	       smp_processor_id(), offset,
+	       MSR_AMD64_IBSCTL, val);
+
+	return 0;
+}
+
+static inline int get_ibs_offset(void)
+{
+	u64 val;
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	if (!(val & IBSCTL_LVT_OFFSET_VALID))
+		return -EINVAL;
+
+	return val & IBSCTL_LVT_OFFSET_MASK;
+}
+
+static void setup_APIC_ibs(void)
+{
+	int offset;
+
+	offset = get_ibs_offset();
+	if (offset < 0)
+		goto failed;
+
+	if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0))
+		return;
+failed:
+	pr_warn("oprofile: IBS APIC setup failed on cpu #%d\n",
+		smp_processor_id());
+}
+
+static void clear_APIC_ibs(void)
+{
+	int offset;
+
+	offset = get_ibs_offset();
+	if (offset >= 0)
+		setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
+}
+
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
 
 static void op_mux_switch_ctrl(struct op_x86_model_spec const *model,
@@ -376,13 +451,13 @@
 	}
 
 	if (ibs_caps)
-		setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_NMI, 0);
+		setup_APIC_ibs();
 }
 
 static void op_amd_cpu_shutdown(void)
 {
 	if (ibs_caps)
-		setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1);
+		clear_APIC_ibs();
 }
 
 static int op_amd_check_ctrs(struct pt_regs * const regs,
@@ -445,16 +520,11 @@
 	op_amd_stop_ibs();
 }
 
-static int __init_ibs_nmi(void)
+static int setup_ibs_ctl(int ibs_eilvt_off)
 {
-#define IBSCTL_LVTOFFSETVAL		(1 << 8)
-#define IBSCTL				0x1cc
 	struct pci_dev *cpu_cfg;
 	int nodes;
 	u32 value = 0;
-	u8 ibs_eilvt_off;
-
-	ibs_eilvt_off = setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1);
 
 	nodes = 0;
 	cpu_cfg = NULL;
@@ -466,24 +536,63 @@
 			break;
 		++nodes;
 		pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
-				       | IBSCTL_LVTOFFSETVAL);
+				       | IBSCTL_LVT_OFFSET_VALID);
 		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
-		if (value != (ibs_eilvt_off | IBSCTL_LVTOFFSETVAL)) {
+		if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) {
 			pci_dev_put(cpu_cfg);
 			printk(KERN_DEBUG "Failed to setup IBS LVT offset, "
-				"IBSCTL = 0x%08x", value);
-			return 1;
+			       "IBSCTL = 0x%08x\n", value);
+			return -EINVAL;
 		}
 	} while (1);
 
 	if (!nodes) {
-		printk(KERN_DEBUG "No CPU node configured for IBS");
-		return 1;
+		printk(KERN_DEBUG "No CPU node configured for IBS\n");
+		return -ENODEV;
 	}
 
 	return 0;
 }
 
+static int force_ibs_eilvt_setup(void)
+{
+	int i;
+	int ret;
+
+	/* find the next free available EILVT entry */
+	for (i = 1; i < 4; i++) {
+		if (!eilvt_is_available(i))
+			continue;
+		ret = setup_ibs_ctl(i);
+		if (ret)
+			return ret;
+		return 0;
+	}
+
+	printk(KERN_DEBUG "No EILVT entry available\n");
+
+	return -EBUSY;
+}
+
+static int __init_ibs_nmi(void)
+{
+	int ret;
+
+	if (ibs_eilvt_valid())
+		return 0;
+
+	ret = force_ibs_eilvt_setup();
+	if (ret)
+		return ret;
+
+	if (!ibs_eilvt_valid())
+		return -EFAULT;
+
+	pr_err(FW_BUG "workaround enabled for IBS LVT offset\n");
+
+	return 0;
+}
+
 /* initialize the APIC for the IBS interrupts if available */
 static void init_ibs(void)
 {
diff --git a/arch/x86/pci/olpc.c b/arch/x86/pci/olpc.c
index b348154..13700ec 100644
--- a/arch/x86/pci/olpc.c
+++ b/arch/x86/pci/olpc.c
@@ -304,7 +304,7 @@
 
 int __init pci_olpc_init(void)
 {
-	printk(KERN_INFO "PCI: Using configuration type OLPC\n");
+	printk(KERN_INFO "PCI: Using configuration type OLPC XO-1\n");
 	raw_pci_ops = &pci_olpc_conf;
 	is_lx = is_geode_lx();
 	return 0;
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 42086ac..b2363fc 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -1969,7 +1969,6 @@
 	.alloc_pte = xen_alloc_pte_init,
 	.release_pte = xen_release_pte_init,
 	.alloc_pmd = xen_alloc_pmd_init,
-	.alloc_pmd_clone = paravirt_nop,
 	.release_pmd = xen_release_pmd_init,
 
 #ifdef CONFIG_X86_64
diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c
index a013ec9..2247100 100644
--- a/arch/x86/xen/pci-swiotlb-xen.c
+++ b/arch/x86/xen/pci-swiotlb-xen.c
@@ -5,6 +5,7 @@
 
 #include <asm/xen/hypervisor.h>
 #include <xen/xen.h>
+#include <asm/iommu_table.h>
 
 int xen_swiotlb __read_mostly;
 
@@ -56,3 +57,7 @@
 		dma_ops = &xen_swiotlb_dma_ops;
 	}
 }
+IOMMU_INIT_FINISH(pci_xen_swiotlb_detect,
+		  0,
+		  pci_xen_swiotlb_init,
+		  0);
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
index c64a5d3..8750888 100644
--- a/arch/xtensa/kernel/irq.c
+++ b/arch/xtensa/kernel/irq.c
@@ -92,7 +92,7 @@
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
diff --git a/block/bsg.c b/block/bsg.c
index 82d5882..0c00870 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -426,7 +426,7 @@
 	/*
 	 * fill in all the output members
 	 */
-	hdr->device_status = status_byte(rq->errors);
+	hdr->device_status = rq->errors & 0xff;
 	hdr->transport_status = host_byte(rq->errors);
 	hdr->driver_status = driver_byte(rq->errors);
 	hdr->info = 0;
diff --git a/block/elevator.c b/block/elevator.c
index 205b09a..4e11559 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -938,6 +938,7 @@
 			}
 		}
 		kobject_uevent(&e->kobj, KOBJ_ADD);
+		e->registered = 1;
 	}
 	return error;
 }
@@ -947,6 +948,7 @@
 {
 	kobject_uevent(&e->kobj, KOBJ_REMOVE);
 	kobject_del(&e->kobj);
+	e->registered = 0;
 }
 
 void elv_unregister_queue(struct request_queue *q)
@@ -1042,11 +1044,13 @@
 
 	spin_unlock_irq(q->queue_lock);
 
-	__elv_unregister_queue(old_elevator);
+	if (old_elevator->registered) {
+		__elv_unregister_queue(old_elevator);
 
-	err = elv_register_queue(q);
-	if (err)
-		goto fail_register;
+		err = elv_register_queue(q);
+		if (err)
+			goto fail_register;
+	}
 
 	/*
 	 * finally exit old elevator and turn off BYPASS.
diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c
index 6b115f6..6afceb3 100644
--- a/drivers/acpi/acpi_pad.c
+++ b/drivers/acpi/acpi_pad.c
@@ -30,18 +30,13 @@
 #include <linux/slab.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
+#include <asm/mwait.h>
 
 #define ACPI_PROCESSOR_AGGREGATOR_CLASS	"acpi_pad"
 #define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator"
 #define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80
 static DEFINE_MUTEX(isolated_cpus_lock);
 
-#define MWAIT_SUBSTATE_MASK	(0xf)
-#define MWAIT_CSTATE_MASK	(0xf)
-#define MWAIT_SUBSTATE_SIZE	(4)
-#define CPUID_MWAIT_LEAF (5)
-#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1)
-#define CPUID5_ECX_INTERRUPT_BREAK	(0x2)
 static unsigned long power_saving_mwait_eax;
 
 static unsigned char tsc_detected_unstable;
diff --git a/drivers/acpi/blacklist.c b/drivers/acpi/blacklist.c
index f761960..af308d0 100644
--- a/drivers/acpi/blacklist.c
+++ b/drivers/acpi/blacklist.c
@@ -204,6 +204,23 @@
 		},
 	},
 	{
+	/*
+	 * There have a NVIF method in MSI GX723 DSDT need call by Nvidia
+	 * driver (e.g. nouveau) when user press brightness hotkey.
+	 * Currently, nouveau driver didn't do the job and it causes there
+	 * have a infinite while loop in DSDT when user press hotkey.
+	 * We add MSI GX723's dmi information to this table for workaround
+	 * this issue.
+	 * Will remove MSI GX723 from the table after nouveau grows support.
+	 */
+	.callback = dmi_disable_osi_vista,
+	.ident = "MSI GX723",
+	.matches = {
+		     DMI_MATCH(DMI_SYS_VENDOR, "Micro-Star International"),
+		     DMI_MATCH(DMI_PRODUCT_NAME, "GX723"),
+		},
+	},
+	{
 	.callback = dmi_disable_osi_vista,
 	.ident = "Sony VGN-NS10J_S",
 	.matches = {
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index b618f88..bec561c 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -346,4 +346,5 @@
 	acpi_walk_namespace(ACPI_TYPE_PROCESSOR, ACPI_ROOT_OBJECT,
 			    ACPI_UINT32_MAX,
 			    early_init_pdc, NULL, NULL, NULL);
+	acpi_get_devices("ACPI0007", early_init_pdc, NULL, NULL);
 }
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index e944aa0..8062921 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -34,7 +34,6 @@
 #include <linux/ata.h>
 #include <linux/libata.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/cisreg.h>
@@ -168,63 +167,26 @@
 };
 
 
-struct pcmcia_config_check {
-	unsigned long ctl_base;
-	int skip_vcc;
-	int is_kme;
-};
-
-static int pcmcia_check_one_config(struct pcmcia_device *pdev,
-				   cistpl_cftable_entry_t *cfg,
-				   cistpl_cftable_entry_t *dflt,
-				   unsigned int vcc,
-				   void *priv_data)
+static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
 {
-	struct pcmcia_config_check *stk = priv_data;
+	int *is_kme = priv_data;
 
-	/* Check for matching Vcc, unless we're desperate */
-	if (!stk->skip_vcc) {
-		if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-			if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000)
-				return -ENODEV;
-		} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-			if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] / 10000)
-				return -ENODEV;
-		}
+	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	}
+	pdev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	pdev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
 
-	if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		pdev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		pdev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		pdev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		pdev->resource[0]->start = io->win[0].base;
-		if (!(io->flags & CISTPL_IO_16BIT)) {
-			pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-			pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-		}
-		if (io->nwin == 2) {
-			pdev->resource[0]->end = 8;
-			pdev->resource[1]->start = io->win[1].base;
-			pdev->resource[1]->end = (stk->is_kme) ? 2 : 1;
-			if (pcmcia_request_io(pdev) != 0)
-				return -ENODEV;
-			stk->ctl_base = pdev->resource[1]->start;
-		} else if ((io->nwin == 1) && (io->win[0].len >= 16)) {
-			pdev->resource[0]->end = io->win[0].len;
-			pdev->resource[1]->end = 0;
-			if (pcmcia_request_io(pdev) != 0)
-				return -ENODEV;
-			stk->ctl_base = pdev->resource[0]->start + 0x0e;
-		} else
+	if (pdev->resource[1]->end) {
+		pdev->resource[0]->end = 8;
+		pdev->resource[1]->end = (*is_kme) ? 2 : 1;
+	} else {
+		if (pdev->resource[0]->end < 16)
 			return -ENODEV;
-		/* If we've got this far, we're done */
-		return 0;
 	}
-	return -ENODEV;
+
+	return pcmcia_request_io(pdev);
 }
 
 /**
@@ -239,7 +201,6 @@
 {
 	struct ata_host *host;
 	struct ata_port *ap;
-	struct pcmcia_config_check *stk = NULL;
 	int is_kme = 0, ret = -ENOMEM, p;
 	unsigned long io_base, ctl_base;
 	void __iomem *io_addr, *ctl_addr;
@@ -247,10 +208,8 @@
 	struct ata_port_operations *ops = &pcmcia_port_ops;
 
 	/* Set up attributes in order to probe card and get resources */
-	pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	pdev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
-	pdev->conf.Attributes = CONF_ENABLE_IRQ;
-	pdev->conf.IntType = INT_MEMORY_AND_IO;
+	pdev->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO |
+		CONF_AUTO_SET_VPP | CONF_AUTO_CHECK_VCC;
 
 	/* See if we have a manufacturer identifier. Use it to set is_kme for
 	   vendor quirks */
@@ -258,25 +217,21 @@
 		  ((pdev->card_id == PRODID_KME_KXLC005_A) ||
 		   (pdev->card_id == PRODID_KME_KXLC005_B)));
 
-	/* Allocate resoure probing structures */
-
-	stk = kzalloc(sizeof(*stk), GFP_KERNEL);
-	if (!stk)
-		goto out1;
-	stk->is_kme = is_kme;
-	stk->skip_vcc = io_base = ctl_base = 0;
-
-	if (pcmcia_loop_config(pdev, pcmcia_check_one_config, stk)) {
-		stk->skip_vcc = 1;
-		if (pcmcia_loop_config(pdev, pcmcia_check_one_config, stk))
+	if (pcmcia_loop_config(pdev, pcmcia_check_one_config, &is_kme)) {
+		pdev->config_flags &= ~CONF_AUTO_CHECK_VCC;
+		if (pcmcia_loop_config(pdev, pcmcia_check_one_config, &is_kme))
 			goto failed; /* No suitable config found */
 	}
 	io_base = pdev->resource[0]->start;
-	ctl_base = stk->ctl_base;
+	if (pdev->resource[1]->end)
+		ctl_base = pdev->resource[1]->start;
+	else
+		ctl_base = pdev->resource[0]->start + 0x0e;
+
 	if (!pdev->irq)
 		goto failed;
 
-	ret = pcmcia_request_configuration(pdev, &pdev->conf);
+	ret = pcmcia_enable_device(pdev);
 	if (ret)
 		goto failed;
 
@@ -329,13 +284,10 @@
 		goto failed;
 
 	pdev->priv = host;
-	kfree(stk);
 	return 0;
 
 failed:
-	kfree(stk);
 	pcmcia_disable_device(pdev);
-out1:
 	return ret;
 }
 
@@ -430,9 +382,7 @@
 
 static struct pcmcia_driver pcmcia_driver = {
 	.owner		= THIS_MODULE,
-	.drv = {
-		.name		= DRV_NAME,
-	},
+	.name		= DRV_NAME,
 	.id_table	= pcmcia_devices,
 	.probe		= pcmcia_init_one,
 	.remove		= pcmcia_remove_one,
diff --git a/drivers/atm/iphase.c b/drivers/atm/iphase.c
index ee9ddeb..8cb0347 100644
--- a/drivers/atm/iphase.c
+++ b/drivers/atm/iphase.c
@@ -3156,7 +3156,6 @@
 {  
 	struct atm_dev *dev;  
 	IADEV *iadev;  
-        unsigned long flags;
 	int ret;
 
 	iadev = kzalloc(sizeof(*iadev), GFP_KERNEL);
@@ -3188,19 +3187,14 @@
 	ia_dev[iadev_count] = iadev;
 	_ia_dev[iadev_count] = dev;
 	iadev_count++;
-	spin_lock_init(&iadev->misc_lock);
-	/* First fixes first. I don't want to think about this now. */
-	spin_lock_irqsave(&iadev->misc_lock, flags); 
 	if (ia_init(dev) || ia_start(dev)) {  
 		IF_INIT(printk("IA register failed!\n");)
 		iadev_count--;
 		ia_dev[iadev_count] = NULL;
 		_ia_dev[iadev_count] = NULL;
-		spin_unlock_irqrestore(&iadev->misc_lock, flags); 
 		ret = -EINVAL;
 		goto err_out_deregister_dev;
 	}
-	spin_unlock_irqrestore(&iadev->misc_lock, flags); 
 	IF_EVENT(printk("iadev_count = %d\n", iadev_count);)
 
 	iadev->next_board = ia_boards;  
diff --git a/drivers/atm/iphase.h b/drivers/atm/iphase.h
index b2cd20f..077735e 100644
--- a/drivers/atm/iphase.h
+++ b/drivers/atm/iphase.h
@@ -1022,7 +1022,7 @@
 	struct dle_q rx_dle_q;  
 	struct free_desc_q *rx_free_desc_qhead;  
 	struct sk_buff_head rx_dma_q;  
-        spinlock_t rx_lock, misc_lock;
+	spinlock_t rx_lock;
 	struct atm_vcc **rx_open;	/* list of all open VCs */  
         u16 num_rx_desc, rx_buf_sz, rxing;
         u32 rx_pkt_ram, rx_tmp_cnt;
diff --git a/drivers/atm/solos-pci.c b/drivers/atm/solos-pci.c
index f916ddf..f46138a 100644
--- a/drivers/atm/solos-pci.c
+++ b/drivers/atm/solos-pci.c
@@ -444,6 +444,7 @@
 	struct atm_dev *atmdev = container_of(dev, struct atm_dev, class_dev);
 	struct solos_card *card = atmdev->dev_data;
 	struct sk_buff *skb;
+	unsigned int len;
 
 	spin_lock(&card->cli_queue_lock);
 	skb = skb_dequeue(&card->cli_queue[SOLOS_CHAN(atmdev)]);
@@ -451,11 +452,12 @@
 	if(skb == NULL)
 		return sprintf(buf, "No data.\n");
 
-	memcpy(buf, skb->data, skb->len);
-	dev_dbg(&card->dev->dev, "len: %d\n", skb->len);
+	len = skb->len;
+	memcpy(buf, skb->data, len);
+	dev_dbg(&card->dev->dev, "len: %d\n", len);
 
 	kfree_skb(skb);
-	return skb->len;
+	return len;
 }
 
 static int send_command(struct solos_card *card, int dev, const char *buf, size_t size)
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 9fc630c..f6f37a0 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -45,7 +45,8 @@
 	return sprintf(buf, "%d\n", topology_##name(cpu));	\
 }
 
-#if defined(topology_thread_cpumask) || defined(topology_core_cpumask)
+#if defined(topology_thread_cpumask) || defined(topology_core_cpumask) || \
+    defined(topology_book_cpumask)
 static ssize_t show_cpumap(int type, const struct cpumask *mask, char *buf)
 {
 	ptrdiff_t len = PTR_ALIGN(buf + PAGE_SIZE - 1, PAGE_SIZE) - buf;
@@ -114,6 +115,14 @@
 define_one_ro_named(core_siblings, show_core_cpumask);
 define_one_ro_named(core_siblings_list, show_core_cpumask_list);
 
+#ifdef CONFIG_SCHED_BOOK
+define_id_show_func(book_id);
+define_one_ro(book_id);
+define_siblings_show_func(book_cpumask);
+define_one_ro_named(book_siblings, show_book_cpumask);
+define_one_ro_named(book_siblings_list, show_book_cpumask_list);
+#endif
+
 static struct attribute *default_attrs[] = {
 	&attr_physical_package_id.attr,
 	&attr_core_id.attr,
@@ -121,6 +130,11 @@
 	&attr_thread_siblings_list.attr,
 	&attr_core_siblings.attr,
 	&attr_core_siblings_list.attr,
+#ifdef CONFIG_SCHED_BOOK
+	&attr_book_id.attr,
+	&attr_book_siblings.attr,
+	&attr_book_siblings_list.attr,
+#endif
 	NULL
 };
 
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index de27768..4b9359a 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -488,4 +488,21 @@
 
 	  If unsure, say N.
 
+config BLK_DEV_RBD
+	tristate "Rados block device (RBD)"
+	depends on INET && EXPERIMENTAL && BLOCK
+	select CEPH_LIB
+	select LIBCRC32C
+	select CRYPTO_AES
+	select CRYPTO
+	default n
+	help
+	  Say Y here if you want include the Rados block device, which stripes
+	  a block device over objects stored in the Ceph distributed object
+	  store.
+
+	  More information at http://ceph.newdream.net/.
+
+	  If unsure, say N.
+
 endif # BLK_DEV
diff --git a/drivers/block/Makefile b/drivers/block/Makefile
index aff5ac9..d7f463d 100644
--- a/drivers/block/Makefile
+++ b/drivers/block/Makefile
@@ -37,5 +37,6 @@
 
 obj-$(CONFIG_XEN_BLKDEV_FRONTEND)	+= xen-blkfront.o
 obj-$(CONFIG_BLK_DEV_DRBD)     += drbd/
+obj-$(CONFIG_BLK_DEV_RBD)     += rbd.o
 
 swim_mod-objs	:= swim.o swim_asm.o
diff --git a/drivers/block/ps3disk.c b/drivers/block/ps3disk.c
index e9da874..03688c2 100644
--- a/drivers/block/ps3disk.c
+++ b/drivers/block/ps3disk.c
@@ -113,7 +113,7 @@
 			memcpy(buf, dev->bounce_buf+offset, size);
 		offset += size;
 		flush_kernel_dcache_page(bvec->bv_page);
-		bvec_kunmap_irq(bvec, &flags);
+		bvec_kunmap_irq(buf, &flags);
 		i++;
 	}
 }
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
new file mode 100644
index 0000000..6ec9d53
--- /dev/null
+++ b/drivers/block/rbd.c
@@ -0,0 +1,1841 @@
+/*
+   rbd.c -- Export ceph rados objects as a Linux block device
+
+
+   based on drivers/block/osdblk.c:
+
+   Copyright 2009 Red Hat, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; see the file COPYING.  If not, write to
+   the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
+
+
+
+   Instructions for use
+   --------------------
+
+   1) Map a Linux block device to an existing rbd image.
+
+      Usage: <mon ip addr> <options> <pool name> <rbd image name> [snap name]
+
+      $ echo "192.168.0.1 name=admin rbd foo" > /sys/class/rbd/add
+
+      The snapshot name can be "-" or omitted to map the image read/write.
+
+   2) List all active blkdev<->object mappings.
+
+      In this example, we have performed step #1 twice, creating two blkdevs,
+      mapped to two separate rados objects in the rados rbd pool
+
+      $ cat /sys/class/rbd/list
+      #id     major   client_name     pool    name    snap    KB
+      0       254     client4143      rbd     foo     -      1024000
+
+      The columns, in order, are:
+      - blkdev unique id
+      - blkdev assigned major
+      - rados client id
+      - rados pool name
+      - rados block device name
+      - mapped snapshot ("-" if none)
+      - device size in KB
+
+
+   3) Create a snapshot.
+
+      Usage: <blkdev id> <snapname>
+
+      $ echo "0 mysnap" > /sys/class/rbd/snap_create
+
+
+   4) Listing a snapshot.
+
+      $ cat /sys/class/rbd/snaps_list
+      #id     snap    KB
+      0       -       1024000 (*)
+      0       foo     1024000
+
+      The columns, in order, are:
+      - blkdev unique id
+      - snapshot name, '-' means none (active read/write version)
+      - size of device at time of snapshot
+      - the (*) indicates this is the active version
+
+   5) Rollback to snapshot.
+
+      Usage: <blkdev id> <snapname>
+
+      $ echo "0 mysnap" > /sys/class/rbd/snap_rollback
+
+
+   6) Mapping an image using snapshot.
+
+      A snapshot mapping is read-only. This is being done by passing
+      snap=<snapname> to the options when adding a device.
+
+      $ echo "192.168.0.1 name=admin,snap=mysnap rbd foo" > /sys/class/rbd/add
+
+
+   7) Remove an active blkdev<->rbd image mapping.
+
+      In this example, we remove the mapping with blkdev unique id 1.
+
+      $ echo 1 > /sys/class/rbd/remove
+
+
+   NOTE:  The actual creation and deletion of rados objects is outside the scope
+   of this driver.
+
+ */
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/osd_client.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/decode.h>
+
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/blkdev.h>
+
+#include "rbd_types.h"
+
+#define DRV_NAME "rbd"
+#define DRV_NAME_LONG "rbd (rados block device)"
+
+#define RBD_MINORS_PER_MAJOR	256		/* max minors per blkdev */
+
+#define RBD_MAX_MD_NAME_LEN	(96 + sizeof(RBD_SUFFIX))
+#define RBD_MAX_POOL_NAME_LEN	64
+#define RBD_MAX_SNAP_NAME_LEN	32
+#define RBD_MAX_OPT_LEN		1024
+
+#define RBD_SNAP_HEAD_NAME	"-"
+
+#define DEV_NAME_LEN		32
+
+/*
+ * block device image metadata (in-memory version)
+ */
+struct rbd_image_header {
+	u64 image_size;
+	char block_name[32];
+	__u8 obj_order;
+	__u8 crypt_type;
+	__u8 comp_type;
+	struct rw_semaphore snap_rwsem;
+	struct ceph_snap_context *snapc;
+	size_t snap_names_len;
+	u64 snap_seq;
+	u32 total_snaps;
+
+	char *snap_names;
+	u64 *snap_sizes;
+};
+
+/*
+ * an instance of the client.  multiple devices may share a client.
+ */
+struct rbd_client {
+	struct ceph_client	*client;
+	struct kref		kref;
+	struct list_head	node;
+};
+
+/*
+ * a single io request
+ */
+struct rbd_request {
+	struct request		*rq;		/* blk layer request */
+	struct bio		*bio;		/* cloned bio */
+	struct page		**pages;	/* list of used pages */
+	u64			len;
+};
+
+/*
+ * a single device
+ */
+struct rbd_device {
+	int			id;		/* blkdev unique id */
+
+	int			major;		/* blkdev assigned major */
+	struct gendisk		*disk;		/* blkdev's gendisk and rq */
+	struct request_queue	*q;
+
+	struct ceph_client	*client;
+	struct rbd_client	*rbd_client;
+
+	char			name[DEV_NAME_LEN]; /* blkdev name, e.g. rbd3 */
+
+	spinlock_t		lock;		/* queue lock */
+
+	struct rbd_image_header	header;
+	char			obj[RBD_MAX_OBJ_NAME_LEN]; /* rbd image name */
+	int			obj_len;
+	char			obj_md_name[RBD_MAX_MD_NAME_LEN]; /* hdr nm. */
+	char			pool_name[RBD_MAX_POOL_NAME_LEN];
+	int			poolid;
+
+	char                    snap_name[RBD_MAX_SNAP_NAME_LEN];
+	u32 cur_snap;	/* index+1 of current snapshot within snap context
+			   0 - for the head */
+	int read_only;
+
+	struct list_head	node;
+};
+
+static spinlock_t node_lock;      /* protects client get/put */
+
+static struct class *class_rbd;	  /* /sys/class/rbd */
+static DEFINE_MUTEX(ctl_mutex);	  /* Serialize open/close/setup/teardown */
+static LIST_HEAD(rbd_dev_list);    /* devices */
+static LIST_HEAD(rbd_client_list);      /* clients */
+
+
+static int rbd_open(struct block_device *bdev, fmode_t mode)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	struct rbd_device *rbd_dev = disk->private_data;
+
+	set_device_ro(bdev, rbd_dev->read_only);
+
+	if ((mode & FMODE_WRITE) && rbd_dev->read_only)
+		return -EROFS;
+
+	return 0;
+}
+
+static const struct block_device_operations rbd_bd_ops = {
+	.owner			= THIS_MODULE,
+	.open			= rbd_open,
+};
+
+/*
+ * Initialize an rbd client instance.
+ * We own *opt.
+ */
+static struct rbd_client *rbd_client_create(struct ceph_options *opt)
+{
+	struct rbd_client *rbdc;
+	int ret = -ENOMEM;
+
+	dout("rbd_client_create\n");
+	rbdc = kmalloc(sizeof(struct rbd_client), GFP_KERNEL);
+	if (!rbdc)
+		goto out_opt;
+
+	kref_init(&rbdc->kref);
+	INIT_LIST_HEAD(&rbdc->node);
+
+	rbdc->client = ceph_create_client(opt, rbdc);
+	if (IS_ERR(rbdc->client))
+		goto out_rbdc;
+	opt = NULL; /* Now rbdc->client is responsible for opt */
+
+	ret = ceph_open_session(rbdc->client);
+	if (ret < 0)
+		goto out_err;
+
+	spin_lock(&node_lock);
+	list_add_tail(&rbdc->node, &rbd_client_list);
+	spin_unlock(&node_lock);
+
+	dout("rbd_client_create created %p\n", rbdc);
+	return rbdc;
+
+out_err:
+	ceph_destroy_client(rbdc->client);
+out_rbdc:
+	kfree(rbdc);
+out_opt:
+	if (opt)
+		ceph_destroy_options(opt);
+	return ERR_PTR(ret);
+}
+
+/*
+ * Find a ceph client with specific addr and configuration.
+ */
+static struct rbd_client *__rbd_client_find(struct ceph_options *opt)
+{
+	struct rbd_client *client_node;
+
+	if (opt->flags & CEPH_OPT_NOSHARE)
+		return NULL;
+
+	list_for_each_entry(client_node, &rbd_client_list, node)
+		if (ceph_compare_options(opt, client_node->client) == 0)
+			return client_node;
+	return NULL;
+}
+
+/*
+ * Get a ceph client with specific addr and configuration, if one does
+ * not exist create it.
+ */
+static int rbd_get_client(struct rbd_device *rbd_dev, const char *mon_addr,
+			  char *options)
+{
+	struct rbd_client *rbdc;
+	struct ceph_options *opt;
+	int ret;
+
+	ret = ceph_parse_options(&opt, options, mon_addr,
+				 mon_addr + strlen(mon_addr), NULL, NULL);
+	if (ret < 0)
+		return ret;
+
+	spin_lock(&node_lock);
+	rbdc = __rbd_client_find(opt);
+	if (rbdc) {
+		ceph_destroy_options(opt);
+
+		/* using an existing client */
+		kref_get(&rbdc->kref);
+		rbd_dev->rbd_client = rbdc;
+		rbd_dev->client = rbdc->client;
+		spin_unlock(&node_lock);
+		return 0;
+	}
+	spin_unlock(&node_lock);
+
+	rbdc = rbd_client_create(opt);
+	if (IS_ERR(rbdc))
+		return PTR_ERR(rbdc);
+
+	rbd_dev->rbd_client = rbdc;
+	rbd_dev->client = rbdc->client;
+	return 0;
+}
+
+/*
+ * Destroy ceph client
+ */
+static void rbd_client_release(struct kref *kref)
+{
+	struct rbd_client *rbdc = container_of(kref, struct rbd_client, kref);
+
+	dout("rbd_release_client %p\n", rbdc);
+	spin_lock(&node_lock);
+	list_del(&rbdc->node);
+	spin_unlock(&node_lock);
+
+	ceph_destroy_client(rbdc->client);
+	kfree(rbdc);
+}
+
+/*
+ * Drop reference to ceph client node. If it's not referenced anymore, release
+ * it.
+ */
+static void rbd_put_client(struct rbd_device *rbd_dev)
+{
+	kref_put(&rbd_dev->rbd_client->kref, rbd_client_release);
+	rbd_dev->rbd_client = NULL;
+	rbd_dev->client = NULL;
+}
+
+
+/*
+ * Create a new header structure, translate header format from the on-disk
+ * header.
+ */
+static int rbd_header_from_disk(struct rbd_image_header *header,
+				 struct rbd_image_header_ondisk *ondisk,
+				 int allocated_snaps,
+				 gfp_t gfp_flags)
+{
+	int i;
+	u32 snap_count = le32_to_cpu(ondisk->snap_count);
+	int ret = -ENOMEM;
+
+	init_rwsem(&header->snap_rwsem);
+
+	header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
+	header->snapc = kmalloc(sizeof(struct ceph_snap_context) +
+				snap_count *
+				 sizeof(struct rbd_image_snap_ondisk),
+				gfp_flags);
+	if (!header->snapc)
+		return -ENOMEM;
+	if (snap_count) {
+		header->snap_names = kmalloc(header->snap_names_len,
+					     GFP_KERNEL);
+		if (!header->snap_names)
+			goto err_snapc;
+		header->snap_sizes = kmalloc(snap_count * sizeof(u64),
+					     GFP_KERNEL);
+		if (!header->snap_sizes)
+			goto err_names;
+	} else {
+		header->snap_names = NULL;
+		header->snap_sizes = NULL;
+	}
+	memcpy(header->block_name, ondisk->block_name,
+	       sizeof(ondisk->block_name));
+
+	header->image_size = le64_to_cpu(ondisk->image_size);
+	header->obj_order = ondisk->options.order;
+	header->crypt_type = ondisk->options.crypt_type;
+	header->comp_type = ondisk->options.comp_type;
+
+	atomic_set(&header->snapc->nref, 1);
+	header->snap_seq = le64_to_cpu(ondisk->snap_seq);
+	header->snapc->num_snaps = snap_count;
+	header->total_snaps = snap_count;
+
+	if (snap_count &&
+	    allocated_snaps == snap_count) {
+		for (i = 0; i < snap_count; i++) {
+			header->snapc->snaps[i] =
+				le64_to_cpu(ondisk->snaps[i].id);
+			header->snap_sizes[i] =
+				le64_to_cpu(ondisk->snaps[i].image_size);
+		}
+
+		/* copy snapshot names */
+		memcpy(header->snap_names, &ondisk->snaps[i],
+			header->snap_names_len);
+	}
+
+	return 0;
+
+err_names:
+	kfree(header->snap_names);
+err_snapc:
+	kfree(header->snapc);
+	return ret;
+}
+
+static int snap_index(struct rbd_image_header *header, int snap_num)
+{
+	return header->total_snaps - snap_num;
+}
+
+static u64 cur_snap_id(struct rbd_device *rbd_dev)
+{
+	struct rbd_image_header *header = &rbd_dev->header;
+
+	if (!rbd_dev->cur_snap)
+		return 0;
+
+	return header->snapc->snaps[snap_index(header, rbd_dev->cur_snap)];
+}
+
+static int snap_by_name(struct rbd_image_header *header, const char *snap_name,
+			u64 *seq, u64 *size)
+{
+	int i;
+	char *p = header->snap_names;
+
+	for (i = 0; i < header->total_snaps; i++, p += strlen(p) + 1) {
+		if (strcmp(snap_name, p) == 0)
+			break;
+	}
+	if (i == header->total_snaps)
+		return -ENOENT;
+	if (seq)
+		*seq = header->snapc->snaps[i];
+
+	if (size)
+		*size = header->snap_sizes[i];
+
+	return i;
+}
+
+static int rbd_header_set_snap(struct rbd_device *dev,
+			       const char *snap_name,
+			       u64 *size)
+{
+	struct rbd_image_header *header = &dev->header;
+	struct ceph_snap_context *snapc = header->snapc;
+	int ret = -ENOENT;
+
+	down_write(&header->snap_rwsem);
+
+	if (!snap_name ||
+	    !*snap_name ||
+	    strcmp(snap_name, "-") == 0 ||
+	    strcmp(snap_name, RBD_SNAP_HEAD_NAME) == 0) {
+		if (header->total_snaps)
+			snapc->seq = header->snap_seq;
+		else
+			snapc->seq = 0;
+		dev->cur_snap = 0;
+		dev->read_only = 0;
+		if (size)
+			*size = header->image_size;
+	} else {
+		ret = snap_by_name(header, snap_name, &snapc->seq, size);
+		if (ret < 0)
+			goto done;
+
+		dev->cur_snap = header->total_snaps - ret;
+		dev->read_only = 1;
+	}
+
+	ret = 0;
+done:
+	up_write(&header->snap_rwsem);
+	return ret;
+}
+
+static void rbd_header_free(struct rbd_image_header *header)
+{
+	kfree(header->snapc);
+	kfree(header->snap_names);
+	kfree(header->snap_sizes);
+}
+
+/*
+ * get the actual striped segment name, offset and length
+ */
+static u64 rbd_get_segment(struct rbd_image_header *header,
+			   const char *block_name,
+			   u64 ofs, u64 len,
+			   char *seg_name, u64 *segofs)
+{
+	u64 seg = ofs >> header->obj_order;
+
+	if (seg_name)
+		snprintf(seg_name, RBD_MAX_SEG_NAME_LEN,
+			 "%s.%012llx", block_name, seg);
+
+	ofs = ofs & ((1 << header->obj_order) - 1);
+	len = min_t(u64, len, (1 << header->obj_order) - ofs);
+
+	if (segofs)
+		*segofs = ofs;
+
+	return len;
+}
+
+/*
+ * bio helpers
+ */
+
+static void bio_chain_put(struct bio *chain)
+{
+	struct bio *tmp;
+
+	while (chain) {
+		tmp = chain;
+		chain = chain->bi_next;
+		bio_put(tmp);
+	}
+}
+
+/*
+ * zeros a bio chain, starting at specific offset
+ */
+static void zero_bio_chain(struct bio *chain, int start_ofs)
+{
+	struct bio_vec *bv;
+	unsigned long flags;
+	void *buf;
+	int i;
+	int pos = 0;
+
+	while (chain) {
+		bio_for_each_segment(bv, chain, i) {
+			if (pos + bv->bv_len > start_ofs) {
+				int remainder = max(start_ofs - pos, 0);
+				buf = bvec_kmap_irq(bv, &flags);
+				memset(buf + remainder, 0,
+				       bv->bv_len - remainder);
+				bvec_kunmap_irq(buf, &flags);
+			}
+			pos += bv->bv_len;
+		}
+
+		chain = chain->bi_next;
+	}
+}
+
+/*
+ * bio_chain_clone - clone a chain of bios up to a certain length.
+ * might return a bio_pair that will need to be released.
+ */
+static struct bio *bio_chain_clone(struct bio **old, struct bio **next,
+				   struct bio_pair **bp,
+				   int len, gfp_t gfpmask)
+{
+	struct bio *tmp, *old_chain = *old, *new_chain = NULL, *tail = NULL;
+	int total = 0;
+
+	if (*bp) {
+		bio_pair_release(*bp);
+		*bp = NULL;
+	}
+
+	while (old_chain && (total < len)) {
+		tmp = bio_kmalloc(gfpmask, old_chain->bi_max_vecs);
+		if (!tmp)
+			goto err_out;
+
+		if (total + old_chain->bi_size > len) {
+			struct bio_pair *bp;
+
+			/*
+			 * this split can only happen with a single paged bio,
+			 * split_bio will BUG_ON if this is not the case
+			 */
+			dout("bio_chain_clone split! total=%d remaining=%d"
+			     "bi_size=%d\n",
+			     (int)total, (int)len-total,
+			     (int)old_chain->bi_size);
+
+			/* split the bio. We'll release it either in the next
+			   call, or it will have to be released outside */
+			bp = bio_split(old_chain, (len - total) / 512ULL);
+			if (!bp)
+				goto err_out;
+
+			__bio_clone(tmp, &bp->bio1);
+
+			*next = &bp->bio2;
+		} else {
+			__bio_clone(tmp, old_chain);
+			*next = old_chain->bi_next;
+		}
+
+		tmp->bi_bdev = NULL;
+		gfpmask &= ~__GFP_WAIT;
+		tmp->bi_next = NULL;
+
+		if (!new_chain) {
+			new_chain = tail = tmp;
+		} else {
+			tail->bi_next = tmp;
+			tail = tmp;
+		}
+		old_chain = old_chain->bi_next;
+
+		total += tmp->bi_size;
+	}
+
+	BUG_ON(total < len);
+
+	if (tail)
+		tail->bi_next = NULL;
+
+	*old = old_chain;
+
+	return new_chain;
+
+err_out:
+	dout("bio_chain_clone with err\n");
+	bio_chain_put(new_chain);
+	return NULL;
+}
+
+/*
+ * helpers for osd request op vectors.
+ */
+static int rbd_create_rw_ops(struct ceph_osd_req_op **ops,
+			    int num_ops,
+			    int opcode,
+			    u32 payload_len)
+{
+	*ops = kzalloc(sizeof(struct ceph_osd_req_op) * (num_ops + 1),
+		       GFP_NOIO);
+	if (!*ops)
+		return -ENOMEM;
+	(*ops)[0].op = opcode;
+	/*
+	 * op extent offset and length will be set later on
+	 * in calc_raw_layout()
+	 */
+	(*ops)[0].payload_len = payload_len;
+	return 0;
+}
+
+static void rbd_destroy_ops(struct ceph_osd_req_op *ops)
+{
+	kfree(ops);
+}
+
+/*
+ * Send ceph osd request
+ */
+static int rbd_do_request(struct request *rq,
+			  struct rbd_device *dev,
+			  struct ceph_snap_context *snapc,
+			  u64 snapid,
+			  const char *obj, u64 ofs, u64 len,
+			  struct bio *bio,
+			  struct page **pages,
+			  int num_pages,
+			  int flags,
+			  struct ceph_osd_req_op *ops,
+			  int num_reply,
+			  void (*rbd_cb)(struct ceph_osd_request *req,
+					 struct ceph_msg *msg))
+{
+	struct ceph_osd_request *req;
+	struct ceph_file_layout *layout;
+	int ret;
+	u64 bno;
+	struct timespec mtime = CURRENT_TIME;
+	struct rbd_request *req_data;
+	struct ceph_osd_request_head *reqhead;
+	struct rbd_image_header *header = &dev->header;
+
+	ret = -ENOMEM;
+	req_data = kzalloc(sizeof(*req_data), GFP_NOIO);
+	if (!req_data)
+		goto done;
+
+	dout("rbd_do_request len=%lld ofs=%lld\n", len, ofs);
+
+	down_read(&header->snap_rwsem);
+
+	req = ceph_osdc_alloc_request(&dev->client->osdc, flags,
+				      snapc,
+				      ops,
+				      false,
+				      GFP_NOIO, pages, bio);
+	if (IS_ERR(req)) {
+		up_read(&header->snap_rwsem);
+		ret = PTR_ERR(req);
+		goto done_pages;
+	}
+
+	req->r_callback = rbd_cb;
+
+	req_data->rq = rq;
+	req_data->bio = bio;
+	req_data->pages = pages;
+	req_data->len = len;
+
+	req->r_priv = req_data;
+
+	reqhead = req->r_request->front.iov_base;
+	reqhead->snapid = cpu_to_le64(CEPH_NOSNAP);
+
+	strncpy(req->r_oid, obj, sizeof(req->r_oid));
+	req->r_oid_len = strlen(req->r_oid);
+
+	layout = &req->r_file_layout;
+	memset(layout, 0, sizeof(*layout));
+	layout->fl_stripe_unit = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
+	layout->fl_stripe_count = cpu_to_le32(1);
+	layout->fl_object_size = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
+	layout->fl_pg_preferred = cpu_to_le32(-1);
+	layout->fl_pg_pool = cpu_to_le32(dev->poolid);
+	ceph_calc_raw_layout(&dev->client->osdc, layout, snapid,
+			     ofs, &len, &bno, req, ops);
+
+	ceph_osdc_build_request(req, ofs, &len,
+				ops,
+				snapc,
+				&mtime,
+				req->r_oid, req->r_oid_len);
+	up_read(&header->snap_rwsem);
+
+	ret = ceph_osdc_start_request(&dev->client->osdc, req, false);
+	if (ret < 0)
+		goto done_err;
+
+	if (!rbd_cb) {
+		ret = ceph_osdc_wait_request(&dev->client->osdc, req);
+		ceph_osdc_put_request(req);
+	}
+	return ret;
+
+done_err:
+	bio_chain_put(req_data->bio);
+	ceph_osdc_put_request(req);
+done_pages:
+	kfree(req_data);
+done:
+	if (rq)
+		blk_end_request(rq, ret, len);
+	return ret;
+}
+
+/*
+ * Ceph osd op callback
+ */
+static void rbd_req_cb(struct ceph_osd_request *req, struct ceph_msg *msg)
+{
+	struct rbd_request *req_data = req->r_priv;
+	struct ceph_osd_reply_head *replyhead;
+	struct ceph_osd_op *op;
+	__s32 rc;
+	u64 bytes;
+	int read_op;
+
+	/* parse reply */
+	replyhead = msg->front.iov_base;
+	WARN_ON(le32_to_cpu(replyhead->num_ops) == 0);
+	op = (void *)(replyhead + 1);
+	rc = le32_to_cpu(replyhead->result);
+	bytes = le64_to_cpu(op->extent.length);
+	read_op = (le32_to_cpu(op->op) == CEPH_OSD_OP_READ);
+
+	dout("rbd_req_cb bytes=%lld readop=%d rc=%d\n", bytes, read_op, rc);
+
+	if (rc == -ENOENT && read_op) {
+		zero_bio_chain(req_data->bio, 0);
+		rc = 0;
+	} else if (rc == 0 && read_op && bytes < req_data->len) {
+		zero_bio_chain(req_data->bio, bytes);
+		bytes = req_data->len;
+	}
+
+	blk_end_request(req_data->rq, rc, bytes);
+
+	if (req_data->bio)
+		bio_chain_put(req_data->bio);
+
+	ceph_osdc_put_request(req);
+	kfree(req_data);
+}
+
+/*
+ * Do a synchronous ceph osd operation
+ */
+static int rbd_req_sync_op(struct rbd_device *dev,
+			   struct ceph_snap_context *snapc,
+			   u64 snapid,
+			   int opcode,
+			   int flags,
+			   struct ceph_osd_req_op *orig_ops,
+			   int num_reply,
+			   const char *obj,
+			   u64 ofs, u64 len,
+			   char *buf)
+{
+	int ret;
+	struct page **pages;
+	int num_pages;
+	struct ceph_osd_req_op *ops = orig_ops;
+	u32 payload_len;
+
+	num_pages = calc_pages_for(ofs , len);
+	pages = ceph_alloc_page_vector(num_pages, GFP_KERNEL);
+	if (IS_ERR(pages))
+		return PTR_ERR(pages);
+
+	if (!orig_ops) {
+		payload_len = (flags & CEPH_OSD_FLAG_WRITE ? len : 0);
+		ret = rbd_create_rw_ops(&ops, 1, opcode, payload_len);
+		if (ret < 0)
+			goto done;
+
+		if ((flags & CEPH_OSD_FLAG_WRITE) && buf) {
+			ret = ceph_copy_to_page_vector(pages, buf, ofs, len);
+			if (ret < 0)
+				goto done_ops;
+		}
+	}
+
+	ret = rbd_do_request(NULL, dev, snapc, snapid,
+			  obj, ofs, len, NULL,
+			  pages, num_pages,
+			  flags,
+			  ops,
+			  2,
+			  NULL);
+	if (ret < 0)
+		goto done_ops;
+
+	if ((flags & CEPH_OSD_FLAG_READ) && buf)
+		ret = ceph_copy_from_page_vector(pages, buf, ofs, ret);
+
+done_ops:
+	if (!orig_ops)
+		rbd_destroy_ops(ops);
+done:
+	ceph_release_page_vector(pages, num_pages);
+	return ret;
+}
+
+/*
+ * Do an asynchronous ceph osd operation
+ */
+static int rbd_do_op(struct request *rq,
+		     struct rbd_device *rbd_dev ,
+		     struct ceph_snap_context *snapc,
+		     u64 snapid,
+		     int opcode, int flags, int num_reply,
+		     u64 ofs, u64 len,
+		     struct bio *bio)
+{
+	char *seg_name;
+	u64 seg_ofs;
+	u64 seg_len;
+	int ret;
+	struct ceph_osd_req_op *ops;
+	u32 payload_len;
+
+	seg_name = kmalloc(RBD_MAX_SEG_NAME_LEN + 1, GFP_NOIO);
+	if (!seg_name)
+		return -ENOMEM;
+
+	seg_len = rbd_get_segment(&rbd_dev->header,
+				  rbd_dev->header.block_name,
+				  ofs, len,
+				  seg_name, &seg_ofs);
+
+	payload_len = (flags & CEPH_OSD_FLAG_WRITE ? seg_len : 0);
+
+	ret = rbd_create_rw_ops(&ops, 1, opcode, payload_len);
+	if (ret < 0)
+		goto done;
+
+	/* we've taken care of segment sizes earlier when we
+	   cloned the bios. We should never have a segment
+	   truncated at this point */
+	BUG_ON(seg_len < len);
+
+	ret = rbd_do_request(rq, rbd_dev, snapc, snapid,
+			     seg_name, seg_ofs, seg_len,
+			     bio,
+			     NULL, 0,
+			     flags,
+			     ops,
+			     num_reply,
+			     rbd_req_cb);
+done:
+	kfree(seg_name);
+	return ret;
+}
+
+/*
+ * Request async osd write
+ */
+static int rbd_req_write(struct request *rq,
+			 struct rbd_device *rbd_dev,
+			 struct ceph_snap_context *snapc,
+			 u64 ofs, u64 len,
+			 struct bio *bio)
+{
+	return rbd_do_op(rq, rbd_dev, snapc, CEPH_NOSNAP,
+			 CEPH_OSD_OP_WRITE,
+			 CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
+			 2,
+			 ofs, len, bio);
+}
+
+/*
+ * Request async osd read
+ */
+static int rbd_req_read(struct request *rq,
+			 struct rbd_device *rbd_dev,
+			 u64 snapid,
+			 u64 ofs, u64 len,
+			 struct bio *bio)
+{
+	return rbd_do_op(rq, rbd_dev, NULL,
+			 (snapid ? snapid : CEPH_NOSNAP),
+			 CEPH_OSD_OP_READ,
+			 CEPH_OSD_FLAG_READ,
+			 2,
+			 ofs, len, bio);
+}
+
+/*
+ * Request sync osd read
+ */
+static int rbd_req_sync_read(struct rbd_device *dev,
+			  struct ceph_snap_context *snapc,
+			  u64 snapid,
+			  const char *obj,
+			  u64 ofs, u64 len,
+			  char *buf)
+{
+	return rbd_req_sync_op(dev, NULL,
+			       (snapid ? snapid : CEPH_NOSNAP),
+			       CEPH_OSD_OP_READ,
+			       CEPH_OSD_FLAG_READ,
+			       NULL,
+			       1, obj, ofs, len, buf);
+}
+
+/*
+ * Request sync osd read
+ */
+static int rbd_req_sync_rollback_obj(struct rbd_device *dev,
+				     u64 snapid,
+				     const char *obj)
+{
+	struct ceph_osd_req_op *ops;
+	int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_ROLLBACK, 0);
+	if (ret < 0)
+		return ret;
+
+	ops[0].snap.snapid = snapid;
+
+	ret = rbd_req_sync_op(dev, NULL,
+			       CEPH_NOSNAP,
+			       0,
+			       CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
+			       ops,
+			       1, obj, 0, 0, NULL);
+
+	rbd_destroy_ops(ops);
+
+	if (ret < 0)
+		return ret;
+
+	return ret;
+}
+
+/*
+ * Request sync osd read
+ */
+static int rbd_req_sync_exec(struct rbd_device *dev,
+			     const char *obj,
+			     const char *cls,
+			     const char *method,
+			     const char *data,
+			     int len)
+{
+	struct ceph_osd_req_op *ops;
+	int cls_len = strlen(cls);
+	int method_len = strlen(method);
+	int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_CALL,
+				    cls_len + method_len + len);
+	if (ret < 0)
+		return ret;
+
+	ops[0].cls.class_name = cls;
+	ops[0].cls.class_len = (__u8)cls_len;
+	ops[0].cls.method_name = method;
+	ops[0].cls.method_len = (__u8)method_len;
+	ops[0].cls.argc = 0;
+	ops[0].cls.indata = data;
+	ops[0].cls.indata_len = len;
+
+	ret = rbd_req_sync_op(dev, NULL,
+			       CEPH_NOSNAP,
+			       0,
+			       CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
+			       ops,
+			       1, obj, 0, 0, NULL);
+
+	rbd_destroy_ops(ops);
+
+	dout("cls_exec returned %d\n", ret);
+	return ret;
+}
+
+/*
+ * block device queue callback
+ */
+static void rbd_rq_fn(struct request_queue *q)
+{
+	struct rbd_device *rbd_dev = q->queuedata;
+	struct request *rq;
+	struct bio_pair *bp = NULL;
+
+	rq = blk_fetch_request(q);
+
+	while (1) {
+		struct bio *bio;
+		struct bio *rq_bio, *next_bio = NULL;
+		bool do_write;
+		int size, op_size = 0;
+		u64 ofs;
+
+		/* peek at request from block layer */
+		if (!rq)
+			break;
+
+		dout("fetched request\n");
+
+		/* filter out block requests we don't understand */
+		if ((rq->cmd_type != REQ_TYPE_FS)) {
+			__blk_end_request_all(rq, 0);
+			goto next;
+		}
+
+		/* deduce our operation (read, write) */
+		do_write = (rq_data_dir(rq) == WRITE);
+
+		size = blk_rq_bytes(rq);
+		ofs = blk_rq_pos(rq) * 512ULL;
+		rq_bio = rq->bio;
+		if (do_write && rbd_dev->read_only) {
+			__blk_end_request_all(rq, -EROFS);
+			goto next;
+		}
+
+		spin_unlock_irq(q->queue_lock);
+
+		dout("%s 0x%x bytes at 0x%llx\n",
+		     do_write ? "write" : "read",
+		     size, blk_rq_pos(rq) * 512ULL);
+
+		do {
+			/* a bio clone to be passed down to OSD req */
+			dout("rq->bio->bi_vcnt=%d\n", rq->bio->bi_vcnt);
+			op_size = rbd_get_segment(&rbd_dev->header,
+						  rbd_dev->header.block_name,
+						  ofs, size,
+						  NULL, NULL);
+			bio = bio_chain_clone(&rq_bio, &next_bio, &bp,
+					      op_size, GFP_ATOMIC);
+			if (!bio) {
+				spin_lock_irq(q->queue_lock);
+				__blk_end_request_all(rq, -ENOMEM);
+				goto next;
+			}
+
+			/* init OSD command: write or read */
+			if (do_write)
+				rbd_req_write(rq, rbd_dev,
+					      rbd_dev->header.snapc,
+					      ofs,
+					      op_size, bio);
+			else
+				rbd_req_read(rq, rbd_dev,
+					     cur_snap_id(rbd_dev),
+					     ofs,
+					     op_size, bio);
+
+			size -= op_size;
+			ofs += op_size;
+
+			rq_bio = next_bio;
+		} while (size > 0);
+
+		if (bp)
+			bio_pair_release(bp);
+
+		spin_lock_irq(q->queue_lock);
+next:
+		rq = blk_fetch_request(q);
+	}
+}
+
+/*
+ * a queue callback. Makes sure that we don't create a bio that spans across
+ * multiple osd objects. One exception would be with a single page bios,
+ * which we handle later at bio_chain_clone
+ */
+static int rbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bmd,
+			  struct bio_vec *bvec)
+{
+	struct rbd_device *rbd_dev = q->queuedata;
+	unsigned int chunk_sectors = 1 << (rbd_dev->header.obj_order - 9);
+	sector_t sector = bmd->bi_sector + get_start_sect(bmd->bi_bdev);
+	unsigned int bio_sectors = bmd->bi_size >> 9;
+	int max;
+
+	max =  (chunk_sectors - ((sector & (chunk_sectors - 1))
+				 + bio_sectors)) << 9;
+	if (max < 0)
+		max = 0; /* bio_add cannot handle a negative return */
+	if (max <= bvec->bv_len && bio_sectors == 0)
+		return bvec->bv_len;
+	return max;
+}
+
+static void rbd_free_disk(struct rbd_device *rbd_dev)
+{
+	struct gendisk *disk = rbd_dev->disk;
+
+	if (!disk)
+		return;
+
+	rbd_header_free(&rbd_dev->header);
+
+	if (disk->flags & GENHD_FL_UP)
+		del_gendisk(disk);
+	if (disk->queue)
+		blk_cleanup_queue(disk->queue);
+	put_disk(disk);
+}
+
+/*
+ * reload the ondisk the header 
+ */
+static int rbd_read_header(struct rbd_device *rbd_dev,
+			   struct rbd_image_header *header)
+{
+	ssize_t rc;
+	struct rbd_image_header_ondisk *dh;
+	int snap_count = 0;
+	u64 snap_names_len = 0;
+
+	while (1) {
+		int len = sizeof(*dh) +
+			  snap_count * sizeof(struct rbd_image_snap_ondisk) +
+			  snap_names_len;
+
+		rc = -ENOMEM;
+		dh = kmalloc(len, GFP_KERNEL);
+		if (!dh)
+			return -ENOMEM;
+
+		rc = rbd_req_sync_read(rbd_dev,
+				       NULL, CEPH_NOSNAP,
+				       rbd_dev->obj_md_name,
+				       0, len,
+				       (char *)dh);
+		if (rc < 0)
+			goto out_dh;
+
+		rc = rbd_header_from_disk(header, dh, snap_count, GFP_KERNEL);
+		if (rc < 0)
+			goto out_dh;
+
+		if (snap_count != header->total_snaps) {
+			snap_count = header->total_snaps;
+			snap_names_len = header->snap_names_len;
+			rbd_header_free(header);
+			kfree(dh);
+			continue;
+		}
+		break;
+	}
+
+out_dh:
+	kfree(dh);
+	return rc;
+}
+
+/*
+ * create a snapshot
+ */
+static int rbd_header_add_snap(struct rbd_device *dev,
+			       const char *snap_name,
+			       gfp_t gfp_flags)
+{
+	int name_len = strlen(snap_name);
+	u64 new_snapid;
+	int ret;
+	void *data, *data_start, *data_end;
+
+	/* we should create a snapshot only if we're pointing at the head */
+	if (dev->cur_snap)
+		return -EINVAL;
+
+	ret = ceph_monc_create_snapid(&dev->client->monc, dev->poolid,
+				      &new_snapid);
+	dout("created snapid=%lld\n", new_snapid);
+	if (ret < 0)
+		return ret;
+
+	data = kmalloc(name_len + 16, gfp_flags);
+	if (!data)
+		return -ENOMEM;
+
+	data_start = data;
+	data_end = data + name_len + 16;
+
+	ceph_encode_string_safe(&data, data_end, snap_name, name_len, bad);
+	ceph_encode_64_safe(&data, data_end, new_snapid, bad);
+
+	ret = rbd_req_sync_exec(dev, dev->obj_md_name, "rbd", "snap_add",
+				data_start, data - data_start);
+
+	kfree(data_start);
+
+	if (ret < 0)
+		return ret;
+
+	dev->header.snapc->seq =  new_snapid;
+
+	return 0;
+bad:
+	return -ERANGE;
+}
+
+/*
+ * only read the first part of the ondisk header, without the snaps info
+ */
+static int rbd_update_snaps(struct rbd_device *rbd_dev)
+{
+	int ret;
+	struct rbd_image_header h;
+	u64 snap_seq;
+
+	ret = rbd_read_header(rbd_dev, &h);
+	if (ret < 0)
+		return ret;
+
+	down_write(&rbd_dev->header.snap_rwsem);
+
+	snap_seq = rbd_dev->header.snapc->seq;
+
+	kfree(rbd_dev->header.snapc);
+	kfree(rbd_dev->header.snap_names);
+	kfree(rbd_dev->header.snap_sizes);
+
+	rbd_dev->header.total_snaps = h.total_snaps;
+	rbd_dev->header.snapc = h.snapc;
+	rbd_dev->header.snap_names = h.snap_names;
+	rbd_dev->header.snap_sizes = h.snap_sizes;
+	rbd_dev->header.snapc->seq = snap_seq;
+
+	up_write(&rbd_dev->header.snap_rwsem);
+
+	return 0;
+}
+
+static int rbd_init_disk(struct rbd_device *rbd_dev)
+{
+	struct gendisk *disk;
+	struct request_queue *q;
+	int rc;
+	u64 total_size = 0;
+
+	/* contact OSD, request size info about the object being mapped */
+	rc = rbd_read_header(rbd_dev, &rbd_dev->header);
+	if (rc)
+		return rc;
+
+	rc = rbd_header_set_snap(rbd_dev, rbd_dev->snap_name, &total_size);
+	if (rc)
+		return rc;
+
+	/* create gendisk info */
+	rc = -ENOMEM;
+	disk = alloc_disk(RBD_MINORS_PER_MAJOR);
+	if (!disk)
+		goto out;
+
+	sprintf(disk->disk_name, DRV_NAME "%d", rbd_dev->id);
+	disk->major = rbd_dev->major;
+	disk->first_minor = 0;
+	disk->fops = &rbd_bd_ops;
+	disk->private_data = rbd_dev;
+
+	/* init rq */
+	rc = -ENOMEM;
+	q = blk_init_queue(rbd_rq_fn, &rbd_dev->lock);
+	if (!q)
+		goto out_disk;
+	blk_queue_merge_bvec(q, rbd_merge_bvec);
+	disk->queue = q;
+
+	q->queuedata = rbd_dev;
+
+	rbd_dev->disk = disk;
+	rbd_dev->q = q;
+
+	/* finally, announce the disk to the world */
+	set_capacity(disk, total_size / 512ULL);
+	add_disk(disk);
+
+	pr_info("%s: added with size 0x%llx\n",
+		disk->disk_name, (unsigned long long)total_size);
+	return 0;
+
+out_disk:
+	put_disk(disk);
+out:
+	return rc;
+}
+
+/********************************************************************
+ * /sys/class/rbd/
+ *                   add	map rados objects to blkdev
+ *                   remove	unmap rados objects
+ *                   list	show mappings
+ *******************************************************************/
+
+static void class_rbd_release(struct class *cls)
+{
+	kfree(cls);
+}
+
+static ssize_t class_rbd_list(struct class *c,
+			      struct class_attribute *attr,
+			      char *data)
+{
+	int n = 0;
+	struct list_head *tmp;
+	int max = PAGE_SIZE;
+
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	n += snprintf(data, max,
+		      "#id\tmajor\tclient_name\tpool\tname\tsnap\tKB\n");
+
+	list_for_each(tmp, &rbd_dev_list) {
+		struct rbd_device *rbd_dev;
+
+		rbd_dev = list_entry(tmp, struct rbd_device, node);
+		n += snprintf(data+n, max-n,
+			      "%d\t%d\tclient%lld\t%s\t%s\t%s\t%lld\n",
+			      rbd_dev->id,
+			      rbd_dev->major,
+			      ceph_client_id(rbd_dev->client),
+			      rbd_dev->pool_name,
+			      rbd_dev->obj, rbd_dev->snap_name,
+			      rbd_dev->header.image_size >> 10);
+		if (n == max)
+			break;
+	}
+
+	mutex_unlock(&ctl_mutex);
+	return n;
+}
+
+static ssize_t class_rbd_add(struct class *c,
+			     struct class_attribute *attr,
+			     const char *buf, size_t count)
+{
+	struct ceph_osd_client *osdc;
+	struct rbd_device *rbd_dev;
+	ssize_t rc = -ENOMEM;
+	int irc, new_id = 0;
+	struct list_head *tmp;
+	char *mon_dev_name;
+	char *options;
+
+	if (!try_module_get(THIS_MODULE))
+		return -ENODEV;
+
+	mon_dev_name = kmalloc(RBD_MAX_OPT_LEN, GFP_KERNEL);
+	if (!mon_dev_name)
+		goto err_out_mod;
+
+	options = kmalloc(RBD_MAX_OPT_LEN, GFP_KERNEL);
+	if (!options)
+		goto err_mon_dev;
+
+	/* new rbd_device object */
+	rbd_dev = kzalloc(sizeof(*rbd_dev), GFP_KERNEL);
+	if (!rbd_dev)
+		goto err_out_opt;
+
+	/* static rbd_device initialization */
+	spin_lock_init(&rbd_dev->lock);
+	INIT_LIST_HEAD(&rbd_dev->node);
+
+	/* generate unique id: find highest unique id, add one */
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	list_for_each(tmp, &rbd_dev_list) {
+		struct rbd_device *rbd_dev;
+
+		rbd_dev = list_entry(tmp, struct rbd_device, node);
+		if (rbd_dev->id >= new_id)
+			new_id = rbd_dev->id + 1;
+	}
+
+	rbd_dev->id = new_id;
+
+	/* add to global list */
+	list_add_tail(&rbd_dev->node, &rbd_dev_list);
+
+	/* parse add command */
+	if (sscanf(buf, "%" __stringify(RBD_MAX_OPT_LEN) "s "
+		   "%" __stringify(RBD_MAX_OPT_LEN) "s "
+		   "%" __stringify(RBD_MAX_POOL_NAME_LEN) "s "
+		   "%" __stringify(RBD_MAX_OBJ_NAME_LEN) "s"
+		   "%" __stringify(RBD_MAX_SNAP_NAME_LEN) "s",
+		   mon_dev_name, options, rbd_dev->pool_name,
+		   rbd_dev->obj, rbd_dev->snap_name) < 4) {
+		rc = -EINVAL;
+		goto err_out_slot;
+	}
+
+	if (rbd_dev->snap_name[0] == 0)
+		rbd_dev->snap_name[0] = '-';
+
+	rbd_dev->obj_len = strlen(rbd_dev->obj);
+	snprintf(rbd_dev->obj_md_name, sizeof(rbd_dev->obj_md_name), "%s%s",
+		 rbd_dev->obj, RBD_SUFFIX);
+
+	/* initialize rest of new object */
+	snprintf(rbd_dev->name, DEV_NAME_LEN, DRV_NAME "%d", rbd_dev->id);
+	rc = rbd_get_client(rbd_dev, mon_dev_name, options);
+	if (rc < 0)
+		goto err_out_slot;
+
+	mutex_unlock(&ctl_mutex);
+
+	/* pick the pool */
+	osdc = &rbd_dev->client->osdc;
+	rc = ceph_pg_poolid_by_name(osdc->osdmap, rbd_dev->pool_name);
+	if (rc < 0)
+		goto err_out_client;
+	rbd_dev->poolid = rc;
+
+	/* register our block device */
+	irc = register_blkdev(0, rbd_dev->name);
+	if (irc < 0) {
+		rc = irc;
+		goto err_out_client;
+	}
+	rbd_dev->major = irc;
+
+	/* set up and announce blkdev mapping */
+	rc = rbd_init_disk(rbd_dev);
+	if (rc)
+		goto err_out_blkdev;
+
+	return count;
+
+err_out_blkdev:
+	unregister_blkdev(rbd_dev->major, rbd_dev->name);
+err_out_client:
+	rbd_put_client(rbd_dev);
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+err_out_slot:
+	list_del_init(&rbd_dev->node);
+	mutex_unlock(&ctl_mutex);
+
+	kfree(rbd_dev);
+err_out_opt:
+	kfree(options);
+err_mon_dev:
+	kfree(mon_dev_name);
+err_out_mod:
+	dout("Error adding device %s\n", buf);
+	module_put(THIS_MODULE);
+	return rc;
+}
+
+static struct rbd_device *__rbd_get_dev(unsigned long id)
+{
+	struct list_head *tmp;
+	struct rbd_device *rbd_dev;
+
+	list_for_each(tmp, &rbd_dev_list) {
+		rbd_dev = list_entry(tmp, struct rbd_device, node);
+		if (rbd_dev->id == id)
+			return rbd_dev;
+	}
+	return NULL;
+}
+
+static ssize_t class_rbd_remove(struct class *c,
+				struct class_attribute *attr,
+				const char *buf,
+				size_t count)
+{
+	struct rbd_device *rbd_dev = NULL;
+	int target_id, rc;
+	unsigned long ul;
+
+	rc = strict_strtoul(buf, 10, &ul);
+	if (rc)
+		return rc;
+
+	/* convert to int; abort if we lost anything in the conversion */
+	target_id = (int) ul;
+	if (target_id != ul)
+		return -EINVAL;
+
+	/* remove object from list immediately */
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	rbd_dev = __rbd_get_dev(target_id);
+	if (rbd_dev)
+		list_del_init(&rbd_dev->node);
+
+	mutex_unlock(&ctl_mutex);
+
+	if (!rbd_dev)
+		return -ENOENT;
+
+	rbd_put_client(rbd_dev);
+
+	/* clean up and free blkdev */
+	rbd_free_disk(rbd_dev);
+	unregister_blkdev(rbd_dev->major, rbd_dev->name);
+	kfree(rbd_dev);
+
+	/* release module ref */
+	module_put(THIS_MODULE);
+
+	return count;
+}
+
+static ssize_t class_rbd_snaps_list(struct class *c,
+			      struct class_attribute *attr,
+			      char *data)
+{
+	struct rbd_device *rbd_dev = NULL;
+	struct list_head *tmp;
+	struct rbd_image_header *header;
+	int i, n = 0, max = PAGE_SIZE;
+	int ret;
+
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	n += snprintf(data, max, "#id\tsnap\tKB\n");
+
+	list_for_each(tmp, &rbd_dev_list) {
+		char *names, *p;
+		struct ceph_snap_context *snapc;
+
+		rbd_dev = list_entry(tmp, struct rbd_device, node);
+		header = &rbd_dev->header;
+
+		down_read(&header->snap_rwsem);
+
+		names = header->snap_names;
+		snapc = header->snapc;
+
+		n += snprintf(data + n, max - n, "%d\t%s\t%lld%s\n",
+			      rbd_dev->id, RBD_SNAP_HEAD_NAME,
+			      header->image_size >> 10,
+			      (!rbd_dev->cur_snap ? " (*)" : ""));
+		if (n == max)
+			break;
+
+		p = names;
+		for (i = 0; i < header->total_snaps; i++, p += strlen(p) + 1) {
+			n += snprintf(data + n, max - n, "%d\t%s\t%lld%s\n",
+			      rbd_dev->id, p, header->snap_sizes[i] >> 10,
+			      (rbd_dev->cur_snap &&
+			       (snap_index(header, i) == rbd_dev->cur_snap) ?
+			       " (*)" : ""));
+			if (n == max)
+				break;
+		}
+
+		up_read(&header->snap_rwsem);
+	}
+
+
+	ret = n;
+	mutex_unlock(&ctl_mutex);
+	return ret;
+}
+
+static ssize_t class_rbd_snaps_refresh(struct class *c,
+				struct class_attribute *attr,
+				const char *buf,
+				size_t count)
+{
+	struct rbd_device *rbd_dev = NULL;
+	int target_id, rc;
+	unsigned long ul;
+	int ret = count;
+
+	rc = strict_strtoul(buf, 10, &ul);
+	if (rc)
+		return rc;
+
+	/* convert to int; abort if we lost anything in the conversion */
+	target_id = (int) ul;
+	if (target_id != ul)
+		return -EINVAL;
+
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	rbd_dev = __rbd_get_dev(target_id);
+	if (!rbd_dev) {
+		ret = -ENOENT;
+		goto done;
+	}
+
+	rc = rbd_update_snaps(rbd_dev);
+	if (rc < 0)
+		ret = rc;
+
+done:
+	mutex_unlock(&ctl_mutex);
+	return ret;
+}
+
+static ssize_t class_rbd_snap_create(struct class *c,
+				struct class_attribute *attr,
+				const char *buf,
+				size_t count)
+{
+	struct rbd_device *rbd_dev = NULL;
+	int target_id, ret;
+	char *name;
+
+	name = kmalloc(RBD_MAX_SNAP_NAME_LEN + 1, GFP_KERNEL);
+	if (!name)
+		return -ENOMEM;
+
+	/* parse snaps add command */
+	if (sscanf(buf, "%d "
+		   "%" __stringify(RBD_MAX_SNAP_NAME_LEN) "s",
+		   &target_id,
+		   name) != 2) {
+		ret = -EINVAL;
+		goto done;
+	}
+
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	rbd_dev = __rbd_get_dev(target_id);
+	if (!rbd_dev) {
+		ret = -ENOENT;
+		goto done_unlock;
+	}
+
+	ret = rbd_header_add_snap(rbd_dev,
+				  name, GFP_KERNEL);
+	if (ret < 0)
+		goto done_unlock;
+
+	ret = rbd_update_snaps(rbd_dev);
+	if (ret < 0)
+		goto done_unlock;
+
+	ret = count;
+done_unlock:
+	mutex_unlock(&ctl_mutex);
+done:
+	kfree(name);
+	return ret;
+}
+
+static ssize_t class_rbd_rollback(struct class *c,
+				struct class_attribute *attr,
+				const char *buf,
+				size_t count)
+{
+	struct rbd_device *rbd_dev = NULL;
+	int target_id, ret;
+	u64 snapid;
+	char snap_name[RBD_MAX_SNAP_NAME_LEN];
+	u64 cur_ofs;
+	char *seg_name;
+
+	/* parse snaps add command */
+	if (sscanf(buf, "%d "
+		   "%" __stringify(RBD_MAX_SNAP_NAME_LEN) "s",
+		   &target_id,
+		   snap_name) != 2) {
+		return -EINVAL;
+	}
+
+	ret = -ENOMEM;
+	seg_name = kmalloc(RBD_MAX_SEG_NAME_LEN + 1, GFP_NOIO);
+	if (!seg_name)
+		return ret;
+
+	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
+	rbd_dev = __rbd_get_dev(target_id);
+	if (!rbd_dev) {
+		ret = -ENOENT;
+		goto done_unlock;
+	}
+
+	ret = snap_by_name(&rbd_dev->header, snap_name, &snapid, NULL);
+	if (ret < 0)
+		goto done_unlock;
+
+	dout("snapid=%lld\n", snapid);
+
+	cur_ofs = 0;
+	while (cur_ofs < rbd_dev->header.image_size) {
+		cur_ofs += rbd_get_segment(&rbd_dev->header,
+					   rbd_dev->obj,
+					   cur_ofs, (u64)-1,
+					   seg_name, NULL);
+		dout("seg_name=%s\n", seg_name);
+
+		ret = rbd_req_sync_rollback_obj(rbd_dev, snapid, seg_name);
+		if (ret < 0)
+			pr_warning("could not roll back obj %s err=%d\n",
+				   seg_name, ret);
+	}
+
+	ret = rbd_update_snaps(rbd_dev);
+	if (ret < 0)
+		goto done_unlock;
+
+	ret = count;
+
+done_unlock:
+	mutex_unlock(&ctl_mutex);
+	kfree(seg_name);
+
+	return ret;
+}
+
+static struct class_attribute class_rbd_attrs[] = {
+	__ATTR(add,		0200, NULL, class_rbd_add),
+	__ATTR(remove,		0200, NULL, class_rbd_remove),
+	__ATTR(list,		0444, class_rbd_list, NULL),
+	__ATTR(snaps_refresh,	0200, NULL, class_rbd_snaps_refresh),
+	__ATTR(snap_create,	0200, NULL, class_rbd_snap_create),
+	__ATTR(snaps_list,	0444, class_rbd_snaps_list, NULL),
+	__ATTR(snap_rollback,	0200, NULL, class_rbd_rollback),
+	__ATTR_NULL
+};
+
+/*
+ * create control files in sysfs
+ * /sys/class/rbd/...
+ */
+static int rbd_sysfs_init(void)
+{
+	int ret = -ENOMEM;
+
+	class_rbd = kzalloc(sizeof(*class_rbd), GFP_KERNEL);
+	if (!class_rbd)
+		goto out;
+
+	class_rbd->name = DRV_NAME;
+	class_rbd->owner = THIS_MODULE;
+	class_rbd->class_release = class_rbd_release;
+	class_rbd->class_attrs = class_rbd_attrs;
+
+	ret = class_register(class_rbd);
+	if (ret)
+		goto out_class;
+	return 0;
+
+out_class:
+	kfree(class_rbd);
+	class_rbd = NULL;
+	pr_err(DRV_NAME ": failed to create class rbd\n");
+out:
+	return ret;
+}
+
+static void rbd_sysfs_cleanup(void)
+{
+	if (class_rbd)
+		class_destroy(class_rbd);
+	class_rbd = NULL;
+}
+
+int __init rbd_init(void)
+{
+	int rc;
+
+	rc = rbd_sysfs_init();
+	if (rc)
+		return rc;
+	spin_lock_init(&node_lock);
+	pr_info("loaded " DRV_NAME_LONG "\n");
+	return 0;
+}
+
+void __exit rbd_exit(void)
+{
+	rbd_sysfs_cleanup();
+}
+
+module_init(rbd_init);
+module_exit(rbd_exit);
+
+MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
+MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
+MODULE_DESCRIPTION("rados block device");
+
+/* following authorship retained from original osdblk.c */
+MODULE_AUTHOR("Jeff Garzik <jeff@garzik.org>");
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/block/rbd_types.h b/drivers/block/rbd_types.h
new file mode 100644
index 0000000..fc6c678
--- /dev/null
+++ b/drivers/block/rbd_types.h
@@ -0,0 +1,73 @@
+/*
+ * Ceph - scalable distributed file system
+ *
+ * Copyright (C) 2004-2010 Sage Weil <sage@newdream.net>
+ *
+ * This is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License version 2.1, as published by the Free Software
+ * Foundation.  See file COPYING.
+ *
+ */
+
+#ifndef CEPH_RBD_TYPES_H
+#define CEPH_RBD_TYPES_H
+
+#include <linux/types.h>
+
+/*
+ * rbd image 'foo' consists of objects
+ *   foo.rbd      - image metadata
+ *   foo.00000000
+ *   foo.00000001
+ *   ...          - data
+ */
+
+#define RBD_SUFFIX		".rbd"
+#define RBD_DIRECTORY           "rbd_directory"
+#define RBD_INFO                "rbd_info"
+
+#define RBD_DEFAULT_OBJ_ORDER	22   /* 4MB */
+#define RBD_MIN_OBJ_ORDER       16
+#define RBD_MAX_OBJ_ORDER       30
+
+#define RBD_MAX_OBJ_NAME_LEN	96
+#define RBD_MAX_SEG_NAME_LEN	128
+
+#define RBD_COMP_NONE		0
+#define RBD_CRYPT_NONE		0
+
+#define RBD_HEADER_TEXT		"<<< Rados Block Device Image >>>\n"
+#define RBD_HEADER_SIGNATURE	"RBD"
+#define RBD_HEADER_VERSION	"001.005"
+
+struct rbd_info {
+	__le64 max_id;
+} __attribute__ ((packed));
+
+struct rbd_image_snap_ondisk {
+	__le64 id;
+	__le64 image_size;
+} __attribute__((packed));
+
+struct rbd_image_header_ondisk {
+	char text[40];
+	char block_name[24];
+	char signature[4];
+	char version[8];
+	struct {
+		__u8 order;
+		__u8 crypt_type;
+		__u8 comp_type;
+		__u8 unused;
+	} __attribute__((packed)) options;
+	__le64 image_size;
+	__le64 snap_seq;
+	__le32 snap_count;
+	__le32 reserved;
+	__le64 snap_names_len;
+	struct rbd_image_snap_ondisk snaps[0];
+} __attribute__((packed));
+
+
+#endif
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 2aafafc..8320490 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -2,7 +2,6 @@
 #include <linux/spinlock.h>
 #include <linux/slab.h>
 #include <linux/blkdev.h>
-#include <linux/smp_lock.h>
 #include <linux/hdreg.h>
 #include <linux/virtio.h>
 #include <linux/virtio_blk.h>
@@ -202,6 +201,7 @@
 	struct virtio_blk *vblk = disk->private_data;
 	struct request *req;
 	struct bio *bio;
+	int err;
 
 	bio = bio_map_kern(vblk->disk->queue, id_str, VIRTIO_BLK_ID_BYTES,
 			   GFP_KERNEL);
@@ -215,11 +215,14 @@
 	}
 
 	req->cmd_type = REQ_TYPE_SPECIAL;
-	return blk_execute_rq(vblk->disk->queue, vblk->disk, req, false);
+	err = blk_execute_rq(vblk->disk->queue, vblk->disk, req, false);
+	blk_put_request(req);
+
+	return err;
 }
 
-static int virtblk_locked_ioctl(struct block_device *bdev, fmode_t mode,
-			 unsigned cmd, unsigned long data)
+static int virtblk_ioctl(struct block_device *bdev, fmode_t mode,
+			     unsigned int cmd, unsigned long data)
 {
 	struct gendisk *disk = bdev->bd_disk;
 	struct virtio_blk *vblk = disk->private_data;
@@ -234,18 +237,6 @@
 			      (void __user *)data);
 }
 
-static int virtblk_ioctl(struct block_device *bdev, fmode_t mode,
-			     unsigned int cmd, unsigned long param)
-{
-	int ret;
-
-	lock_kernel();
-	ret = virtblk_locked_ioctl(bdev, mode, cmd, param);
-	unlock_kernel();
-
-	return ret;
-}
-
 /* We provide getgeo only to please some old bootloader/partitioning tools */
 static int virtblk_getgeo(struct block_device *bd, struct hd_geometry *geo)
 {
diff --git a/drivers/bluetooth/bluecard_cs.c b/drivers/bluetooth/bluecard_cs.c
index d52e90a5..4104b7f 100644
--- a/drivers/bluetooth/bluecard_cs.c
+++ b/drivers/bluetooth/bluecard_cs.c
@@ -39,7 +39,6 @@
 #include <linux/skbuff.h>
 #include <linux/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -865,8 +864,7 @@
 	info->p_dev = link;
 	link->priv = info;
 
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ;
 
 	return bluecard_config(link);
 }
@@ -886,7 +884,7 @@
 	bluecard_info_t *info = link->priv;
 	int i, n;
 
-	link->conf.ConfigIndex = 0x20;
+	link->config_index = 0x20;
 
 	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
 	link->resource[0]->end = 64;
@@ -906,7 +904,7 @@
 	if (i != 0)
 		goto failed;
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		goto failed;
 
@@ -942,9 +940,7 @@
 
 static struct pcmcia_driver bluecard_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "bluecard_cs",
-	},
+	.name		= "bluecard_cs",
 	.probe		= bluecard_probe,
 	.remove		= bluecard_detach,
 	.id_table	= bluecard_ids,
diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c
index 7ab8f29..0c8a655 100644
--- a/drivers/bluetooth/bt3c_cs.c
+++ b/drivers/bluetooth/bt3c_cs.c
@@ -45,7 +45,6 @@
 #include <linux/device.h>
 #include <linux/firmware.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -657,11 +656,8 @@
 	info->p_dev = link;
 	link->priv = info;
 
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	link->resource[0]->end = 8;
-
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_VPP |
+		CONF_AUTO_SET_IO;
 
 	return bt3c_config(link);
 }
@@ -675,43 +671,41 @@
 	kfree(info);
 }
 
-static int bt3c_check_config(struct pcmcia_device *p_dev,
-			     cistpl_cftable_entry_t *cf,
-			     cistpl_cftable_entry_t *dflt,
-			     unsigned int vcc,
-			     void *priv_data)
+static int bt3c_check_config(struct pcmcia_device *p_dev, void *priv_data)
 {
-	unsigned long try = (unsigned long) priv_data;
+	int *try = priv_data;
 
-	p_dev->io_lines = (try == 0) ? 16 : cf->io.flags & CISTPL_IO_LINES_MASK;
+	if (try == 0)
+		p_dev->io_lines = 16;
 
-	if (cf->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cf->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	if ((cf->io.nwin > 0) && (cf->io.win[0].len == 8) &&
-	    (cf->io.win[0].base != 0)) {
-		p_dev->resource[0]->start = cf->io.win[0].base;
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -ENODEV;
+	if ((p_dev->resource[0]->end != 8) || (p_dev->resource[0]->start == 0))
+		return -EINVAL;
+
+	p_dev->resource[0]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int bt3c_check_config_notpicky(struct pcmcia_device *p_dev,
-				      cistpl_cftable_entry_t *cf,
-				      cistpl_cftable_entry_t *dflt,
-				      unsigned int vcc,
 				      void *priv_data)
 {
 	static unsigned int base[5] = { 0x3f8, 0x2f8, 0x3e8, 0x2e8, 0x0 };
 	int j;
 
-	if ((cf->io.nwin > 0) && ((cf->io.flags & CISTPL_IO_LINES_MASK) <= 3)) {
-		for (j = 0; j < 5; j++) {
-			p_dev->resource[0]->start = base[j];
-			p_dev->io_lines = base[j] ? 16 : 3;
-			if (!pcmcia_request_io(p_dev))
-				return 0;
-		}
+	if (p_dev->io_lines > 3)
+		return -ENODEV;
+
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[0]->end = 8;
+
+	for (j = 0; j < 5; j++) {
+		p_dev->resource[0]->start = base[j];
+		p_dev->io_lines = base[j] ? 16 : 3;
+		if (!pcmcia_request_io(p_dev))
+			return 0;
 	}
 	return -ENODEV;
 }
@@ -742,7 +736,7 @@
 	if (i != 0)
 		goto failed;
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		goto failed;
 
@@ -775,9 +769,7 @@
 
 static struct pcmcia_driver bt3c_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "bt3c_cs",
-	},
+	.name		= "bt3c_cs",
 	.probe		= bt3c_probe,
 	.remove		= bt3c_detach,
 	.id_table	= bt3c_ids,
diff --git a/drivers/bluetooth/btuart_cs.c b/drivers/bluetooth/btuart_cs.c
index 1c4f5e8..f8a0708 100644
--- a/drivers/bluetooth/btuart_cs.c
+++ b/drivers/bluetooth/btuart_cs.c
@@ -41,7 +41,6 @@
 #include <asm/system.h>
 #include <asm/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -586,11 +585,8 @@
 	info->p_dev = link;
 	link->priv = info;
 
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	link->resource[0]->end = 8;
-
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_VPP |
+		CONF_AUTO_SET_IO;
 
 	return btuart_config(link);
 }
@@ -604,43 +600,41 @@
 	kfree(info);
 }
 
-static int btuart_check_config(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cf,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int btuart_check_config(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int *try = priv_data;
 
-	p_dev->io_lines = (try == 0) ? 16 : cf->io.flags & CISTPL_IO_LINES_MASK;
+	if (try == 0)
+		p_dev->io_lines = 16;
 
-	if (cf->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cf->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	if ((cf->io.nwin > 0) && (cf->io.win[0].len == 8) &&
-	    (cf->io.win[0].base != 0)) {
-		p_dev->resource[0]->start = cf->io.win[0].base;
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -ENODEV;
+	if ((p_dev->resource[0]->end != 8) || (p_dev->resource[0]->start == 0))
+		return -EINVAL;
+
+	p_dev->resource[0]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int btuart_check_config_notpicky(struct pcmcia_device *p_dev,
-					cistpl_cftable_entry_t *cf,
-					cistpl_cftable_entry_t *dflt,
-					unsigned int vcc,
 					void *priv_data)
 {
 	static unsigned int base[5] = { 0x3f8, 0x2f8, 0x3e8, 0x2e8, 0x0 };
 	int j;
 
-	if ((cf->io.nwin > 0) && ((cf->io.flags & CISTPL_IO_LINES_MASK) <= 3)) {
-		for (j = 0; j < 5; j++) {
-			p_dev->resource[0]->start = base[j];
-			p_dev->io_lines = base[j] ? 16 : 3;
-			if (!pcmcia_request_io(p_dev))
-				return 0;
-		}
+	if (p_dev->io_lines > 3)
+		return -ENODEV;
+
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[0]->end = 8;
+
+	for (j = 0; j < 5; j++) {
+		p_dev->resource[0]->start = base[j];
+		p_dev->io_lines = base[j] ? 16 : 3;
+		if (!pcmcia_request_io(p_dev))
+			return 0;
 	}
 	return -ENODEV;
 }
@@ -671,7 +665,7 @@
 	if (i != 0)
 		goto failed;
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		goto failed;
 
@@ -703,9 +697,7 @@
 
 static struct pcmcia_driver btuart_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "btuart_cs",
-	},
+	.name		= "btuart_cs",
 	.probe		= btuart_probe,
 	.remove		= btuart_detach,
 	.id_table	= btuart_ids,
diff --git a/drivers/bluetooth/dtl1_cs.c b/drivers/bluetooth/dtl1_cs.c
index db7c8db..26ee0cf 100644
--- a/drivers/bluetooth/dtl1_cs.c
+++ b/drivers/bluetooth/dtl1_cs.c
@@ -41,7 +41,6 @@
 #include <asm/system.h>
 #include <asm/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -572,11 +571,7 @@
 	info->p_dev = link;
 	link->priv = info;
 
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	link->resource[0]->end = 8;
-
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
 	return dtl1_config(link);
 }
@@ -591,18 +586,14 @@
 	kfree(info);
 }
 
-static int dtl1_confcheck(struct pcmcia_device *p_dev,
-			  cistpl_cftable_entry_t *cf,
-			  cistpl_cftable_entry_t *dflt,
-			  unsigned int vcc,
-			  void *priv_data)
+static int dtl1_confcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if ((cf->io.nwin != 1) || (cf->io.win[0].len <= 8))
+	if ((p_dev->resource[1]->end) || (p_dev->resource[1]->end < 8))
 		return -ENODEV;
 
-	p_dev->resource[0]->start = cf->io.win[0].base;
-	p_dev->resource[0]->end = cf->io.win[0].len;	/*yo */
-	p_dev->io_lines = cf->io.flags & CISTPL_IO_LINES_MASK;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
 	return pcmcia_request_io(p_dev);
 }
 
@@ -620,7 +611,7 @@
 	if (i != 0)
 		goto failed;
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		goto failed;
 
@@ -656,9 +647,7 @@
 
 static struct pcmcia_driver dtl1_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "dtl1_cs",
-	},
+	.name		= "dtl1_cs",
 	.probe		= dtl1_probe,
 	.remove		= dtl1_detach,
 	.id_table	= dtl1_ids,
diff --git a/drivers/char/agp/Kconfig b/drivers/char/agp/Kconfig
index 4b66c69..5ddf67e 100644
--- a/drivers/char/agp/Kconfig
+++ b/drivers/char/agp/Kconfig
@@ -57,7 +57,7 @@
 
 config AGP_AMD64
 	tristate "AMD Opteron/Athlon64 on-CPU GART support"
-	depends on AGP && X86 && K8_NB
+	depends on AGP && X86 && AMD_NB
 	help
 	  This option gives you AGP support for the GLX component of
 	  X using the on-CPU northbridge of the AMD Athlon64/Opteron CPUs.
diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c
index 70312da..42396df 100644
--- a/drivers/char/agp/amd64-agp.c
+++ b/drivers/char/agp/amd64-agp.c
@@ -15,7 +15,7 @@
 #include <linux/mmzone.h>
 #include <asm/page.h>		/* PAGE_SIZE */
 #include <asm/e820.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #include <asm/gart.h>
 #include "agp.h"
 
@@ -124,7 +124,7 @@
 	u32 temp;
 	struct aper_size_info_32 *values;
 
-	dev = k8_northbridges[0];
+	dev = k8_northbridges.nb_misc[0];
 	if (dev==NULL)
 		return 0;
 
@@ -181,10 +181,14 @@
 	unsigned long gatt_bus = virt_to_phys(agp_bridge->gatt_table_real);
 	int i;
 
+	if (!k8_northbridges.gart_supported)
+		return 0;
+
 	/* Configure AGP regs in each x86-64 host bridge. */
-        for (i = 0; i < num_k8_northbridges; i++) {
+	for (i = 0; i < k8_northbridges.num; i++) {
 		agp_bridge->gart_bus_addr =
-				amd64_configure(k8_northbridges[i], gatt_bus);
+				amd64_configure(k8_northbridges.nb_misc[i],
+						gatt_bus);
 	}
 	k8_flush_garts();
 	return 0;
@@ -195,11 +199,15 @@
 {
 	u32 tmp;
 	int i;
-        for (i = 0; i < num_k8_northbridges; i++) {
-		struct pci_dev *dev = k8_northbridges[i];
+
+	if (!k8_northbridges.gart_supported)
+		return;
+
+	for (i = 0; i < k8_northbridges.num; i++) {
+		struct pci_dev *dev = k8_northbridges.nb_misc[i];
 		/* disable gart translation */
 		pci_read_config_dword(dev, AMD64_GARTAPERTURECTL, &tmp);
-		tmp &= ~AMD64_GARTEN;
+		tmp &= ~GARTEN;
 		pci_write_config_dword(dev, AMD64_GARTAPERTURECTL, tmp);
 	}
 }
@@ -313,22 +321,25 @@
 	if (order < 0 || !agp_aperture_valid(aper, (32*1024*1024)<<order))
 		return -1;
 
-	pci_write_config_dword(nb, AMD64_GARTAPERTURECTL, order << 1);
+	gart_set_size_and_enable(nb, order);
 	pci_write_config_dword(nb, AMD64_GARTAPERTUREBASE, aper >> 25);
 
 	return 0;
 }
 
-static __devinit int cache_nbs (struct pci_dev *pdev, u32 cap_ptr)
+static __devinit int cache_nbs(struct pci_dev *pdev, u32 cap_ptr)
 {
 	int i;
 
 	if (cache_k8_northbridges() < 0)
 		return -ENODEV;
 
+	if (!k8_northbridges.gart_supported)
+		return -ENODEV;
+
 	i = 0;
-	for (i = 0; i < num_k8_northbridges; i++) {
-		struct pci_dev *dev = k8_northbridges[i];
+	for (i = 0; i < k8_northbridges.num; i++) {
+		struct pci_dev *dev = k8_northbridges.nb_misc[i];
 		if (fix_northbridge(dev, pdev, cap_ptr) < 0) {
 			dev_err(&dev->dev, "no usable aperture found\n");
 #ifdef __x86_64__
@@ -405,7 +416,8 @@
 	}
 
 	/* shadow x86-64 registers into ULi registers */
-	pci_read_config_dword (k8_northbridges[0], AMD64_GARTAPERTUREBASE, &httfea);
+	pci_read_config_dword (k8_northbridges.nb_misc[0], AMD64_GARTAPERTUREBASE,
+			       &httfea);
 
 	/* if x86-64 aperture base is beyond 4G, exit here */
 	if ((httfea & 0x7fff) >> (32 - 25)) {
@@ -472,7 +484,8 @@
 	pci_write_config_dword(dev1, NVIDIA_X86_64_1_APSIZE, tmp);
 
 	/* shadow x86-64 registers into NVIDIA registers */
-	pci_read_config_dword (k8_northbridges[0], AMD64_GARTAPERTUREBASE, &apbase);
+	pci_read_config_dword (k8_northbridges.nb_misc[0], AMD64_GARTAPERTUREBASE,
+			       &apbase);
 
 	/* if x86-64 aperture base is beyond 4G, exit here */
 	if ( (apbase & 0x7fff) >> (32 - 25) ) {
diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c
index d2abf51..64255ce 100644
--- a/drivers/char/agp/generic.c
+++ b/drivers/char/agp/generic.c
@@ -984,7 +984,9 @@
 
 	bridge->driver->cache_flush();
 #ifdef CONFIG_X86
-	set_memory_uc((unsigned long)table, 1 << page_order);
+	if (set_memory_uc((unsigned long)table, 1 << page_order))
+		printk(KERN_WARNING "Could not set GATT table memory to UC!");
+
 	bridge->gatt_table = (void *)table;
 #else
 	bridge->gatt_table = ioremap_nocache(virt_to_phys(table),
diff --git a/drivers/char/pcmcia/cm4000_cs.c b/drivers/char/pcmcia/cm4000_cs.c
index ec73d9f..c7b482d 100644
--- a/drivers/char/pcmcia/cm4000_cs.c
+++ b/drivers/char/pcmcia/cm4000_cs.c
@@ -34,7 +34,6 @@
 #include <linux/uaccess.h>
 #include <linux/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -55,8 +54,6 @@
 			   __func__ , ## args);		\
 	} while (0)
 
-static char *version = "cm4000_cs.c v2.4.0gm6 - All bugs added by Harald Welte";
-
 #define	T_1SEC		(HZ)
 #define	T_10MSEC	msecs_to_jiffies(10)
 #define	T_20MSEC	msecs_to_jiffies(20)
@@ -1742,20 +1739,8 @@
 
 /*==== Interface to PCMCIA Layer =======================================*/
 
-static int cm4000_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cfg,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int cm4000_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (!cfg->io.nwin)
-		return -ENODEV;
-
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
-	p_dev->resource[0]->flags |= pcmcia_io_cfg_data_width(cfg->io.flags);
-	p_dev->io_lines = cfg->io.flags & CISTPL_IO_LINES_MASK;
-
 	return pcmcia_request_io(p_dev);
 }
 
@@ -1763,13 +1748,13 @@
 {
 	struct cm4000_dev *dev;
 
+	link->config_flags |= CONF_AUTO_SET_IO;
+
 	/* read the config-tuples */
 	if (pcmcia_loop_config(link, cm4000_config_check, NULL))
 		goto cs_release;
 
-	link->conf.IntType = 00000002;
-
-	if (pcmcia_request_configuration(link, &link->conf))
+	if (pcmcia_enable_device(link))
 		goto cs_release;
 
 	dev = link->priv;
@@ -1829,7 +1814,6 @@
 
 	dev->p_dev = link;
 	link->priv = dev;
-	link->conf.IntType = INT_MEMORY_AND_IO;
 	dev_table[i] = link;
 
 	init_waitqueue_head(&dev->devq);
@@ -1891,9 +1875,7 @@
 
 static struct pcmcia_driver cm4000_driver = {
 	.owner	  = THIS_MODULE,
-	.drv	  = {
-		.name = "cm4000_cs",
-		},
+	.name	  = "cm4000_cs",
 	.probe    = cm4000_probe,
 	.remove   = cm4000_detach,
 	.suspend  = cm4000_suspend,
@@ -1905,8 +1887,6 @@
 {
 	int rc;
 
-	printk(KERN_INFO "%s\n", version);
-
 	cmm_class = class_create(THIS_MODULE, "cardman_4000");
 	if (IS_ERR(cmm_class))
 		return PTR_ERR(cmm_class);
@@ -1931,7 +1911,6 @@
 
 static void __exit cmm_exit(void)
 {
-	printk(KERN_INFO MODULE_NAME ": unloading\n");
 	pcmcia_unregister_driver(&cm4000_driver);
 	unregister_chrdev(major, DEVICE_NAME);
 	class_destroy(cmm_class);
diff --git a/drivers/char/pcmcia/cm4040_cs.c b/drivers/char/pcmcia/cm4040_cs.c
index 815cde1..bf2f046 100644
--- a/drivers/char/pcmcia/cm4040_cs.c
+++ b/drivers/char/pcmcia/cm4040_cs.c
@@ -29,7 +29,6 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -49,9 +48,6 @@
 			   __func__ , ## args);		\
 	} while (0)
 
-static char *version =
-"OMNIKEY CardMan 4040 v1.1.0gm5 - All bugs added by Harald Welte";
-
 #define	CCID_DRIVER_BULK_DEFAULT_TIMEOUT  	(150*HZ)
 #define	CCID_DRIVER_ASYNC_POWERUP_TIMEOUT 	(35*HZ)
 #define	CCID_DRIVER_MINIMUM_TIMEOUT 		(3*HZ)
@@ -516,26 +512,9 @@
 	return;
 }
 
-static int cm4040_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cfg,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int cm4040_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	int rc;
-	if (!cfg->io.nwin)
-		return -ENODEV;
-
-	/* Get the IOaddr */
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
-	p_dev->resource[0]->flags |= pcmcia_io_cfg_data_width(cfg->io.flags);
-	p_dev->io_lines = cfg->io.flags & CISTPL_IO_LINES_MASK;
-	rc = pcmcia_request_io(p_dev);
-
-	dev_printk(KERN_INFO, &p_dev->dev,
-		   "pcmcia_request_io returned 0x%x\n", rc);
-	return rc;
+	return pcmcia_request_io(p_dev);
 }
 
 
@@ -544,15 +523,15 @@
 	struct reader_dev *dev;
 	int fail_rc;
 
+	link->config_flags |= CONF_AUTO_SET_IO;
+
 	if (pcmcia_loop_config(link, cm4040_config_check, NULL))
 		goto cs_release;
 
-	link->conf.IntType = 00000002;
-
-	fail_rc = pcmcia_request_configuration(link, &link->conf);
+	fail_rc = pcmcia_enable_device(link);
 	if (fail_rc != 0) {
 		dev_printk(KERN_INFO, &link->dev,
-			   "pcmcia_request_configuration failed 0x%x\n",
+			   "pcmcia_enable_device failed 0x%x\n",
 			   fail_rc);
 		goto cs_release;
 	}
@@ -599,7 +578,6 @@
 	link->priv = dev;
 	dev->p_dev = link;
 
-	link->conf.IntType = INT_MEMORY_AND_IO;
 	dev_table[i] = link;
 
 	init_waitqueue_head(&dev->devq);
@@ -662,9 +640,7 @@
 
 static struct pcmcia_driver reader_driver = {
   	.owner		= THIS_MODULE,
-  	.drv		= {
-		.name	= "cm4040_cs",
-	},
+	.name		= "cm4040_cs",
 	.probe		= reader_probe,
 	.remove		= reader_detach,
 	.id_table	= cm4040_ids,
@@ -674,7 +650,6 @@
 {
 	int rc;
 
-	printk(KERN_INFO "%s\n", version);
 	cmx_class = class_create(THIS_MODULE, "cardman_4040");
 	if (IS_ERR(cmx_class))
 		return PTR_ERR(cmx_class);
@@ -699,7 +674,6 @@
 
 static void __exit cm4040_exit(void)
 {
-	printk(KERN_INFO MODULE_NAME ": unloading\n");
 	pcmcia_unregister_driver(&reader_driver);
 	unregister_chrdev(major, DEVICE_NAME);
 	class_destroy(cmx_class);
diff --git a/drivers/char/pcmcia/ipwireless/main.c b/drivers/char/pcmcia/ipwireless/main.c
index 67bdb05..94b8eb4 100644
--- a/drivers/char/pcmcia/ipwireless/main.c
+++ b/drivers/char/pcmcia/ipwireless/main.c
@@ -32,7 +32,6 @@
 #include <pcmcia/device_id.h>
 #include <pcmcia/ss.h>
 #include <pcmcia/ds.h>
-#include <pcmcia/cs.h>
 
 static struct pcmcia_device_id ipw_ids[] = {
 	PCMCIA_DEVICE_MANF_CARD(0x02f2, 0x0100),
@@ -76,23 +75,18 @@
 	schedule_work(&ipw->work_reboot);
 }
 
-static int ipwireless_probe(struct pcmcia_device *p_dev,
-			    cistpl_cftable_entry_t *cfg,
-			    cistpl_cftable_entry_t *dflt,
-			    unsigned int vcc,
-			    void *priv_data)
+static int ipwireless_probe(struct pcmcia_device *p_dev, void *priv_data)
 {
 	struct ipw_dev *ipw = priv_data;
 	struct resource *io_resource;
 	int ret;
 
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
 	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
 
 	/* 0x40 causes it to generate level mode interrupts. */
 	/* 0x04 enables IREQ pin. */
-	p_dev->conf.ConfigIndex = cfg->index | 0x44;
+	p_dev->config_index |= 0x44;
 	p_dev->io_lines = 16;
 	ret = pcmcia_request_io(p_dev);
 	if (ret)
@@ -102,65 +96,49 @@
 				resource_size(p_dev->resource[0]),
 				IPWIRELESS_PCCARD_NAME);
 
-	if (cfg->mem.nwin == 0)
-		return 0;
-
-	ipw->request_common_memory.Attributes =
+	p_dev->resource[2]->flags |=
 		WIN_DATA_WIDTH_16 | WIN_MEMORY_TYPE_CM | WIN_ENABLE;
-	ipw->request_common_memory.Base = cfg->mem.win[0].host_addr;
-	ipw->request_common_memory.Size = cfg->mem.win[0].len;
-	if (ipw->request_common_memory.Size < 0x1000)
-		ipw->request_common_memory.Size = 0x1000;
-	ipw->request_common_memory.AccessSpeed = 0;
 
-	ret = pcmcia_request_window(p_dev, &ipw->request_common_memory,
-				&ipw->handle_common_memory);
-
+	ret = pcmcia_request_window(p_dev, p_dev->resource[2], 0);
 	if (ret != 0)
 		goto exit1;
 
-	ret = pcmcia_map_mem_page(p_dev, ipw->handle_common_memory,
-				cfg->mem.win[0].card_addr);
-
+	ret = pcmcia_map_mem_page(p_dev, p_dev->resource[2], p_dev->card_addr);
 	if (ret != 0)
 		goto exit2;
 
-	ipw->is_v2_card = cfg->mem.win[0].len == 0x100;
+	ipw->is_v2_card = resource_size(p_dev->resource[2]) == 0x100;
 
-	ipw->common_memory = ioremap(ipw->request_common_memory.Base,
-				ipw->request_common_memory.Size);
-	request_mem_region(ipw->request_common_memory.Base,
-			ipw->request_common_memory.Size,
+	ipw->attr_memory = ioremap(p_dev->resource[2]->start,
+				resource_size(p_dev->resource[2]));
+	request_mem_region(p_dev->resource[2]->start,
+			resource_size(p_dev->resource[2]),
 			IPWIRELESS_PCCARD_NAME);
 
-	ipw->request_attr_memory.Attributes =
-		WIN_DATA_WIDTH_16 | WIN_MEMORY_TYPE_AM | WIN_ENABLE;
-	ipw->request_attr_memory.Base = 0;
-	ipw->request_attr_memory.Size = 0;	/* this used to be 0x1000 */
-	ipw->request_attr_memory.AccessSpeed = 0;
-
-	ret = pcmcia_request_window(p_dev, &ipw->request_attr_memory,
-				&ipw->handle_attr_memory);
-
+	p_dev->resource[3]->flags |= WIN_DATA_WIDTH_16 | WIN_MEMORY_TYPE_AM |
+					WIN_ENABLE;
+	p_dev->resource[3]->end = 0; /* this used to be 0x1000 */
+	ret = pcmcia_request_window(p_dev, p_dev->resource[3], 0);
 	if (ret != 0)
 		goto exit2;
 
-	ret = pcmcia_map_mem_page(p_dev, ipw->handle_attr_memory, 0);
+	ret = pcmcia_map_mem_page(p_dev, p_dev->resource[3], 0);
 	if (ret != 0)
 		goto exit3;
 
-	ipw->attr_memory = ioremap(ipw->request_attr_memory.Base,
-				ipw->request_attr_memory.Size);
-	request_mem_region(ipw->request_attr_memory.Base,
-			ipw->request_attr_memory.Size, IPWIRELESS_PCCARD_NAME);
+	ipw->attr_memory = ioremap(p_dev->resource[3]->start,
+				resource_size(p_dev->resource[3]));
+	request_mem_region(p_dev->resource[3]->start,
+			resource_size(p_dev->resource[3]),
+			IPWIRELESS_PCCARD_NAME);
 
 	return 0;
 
 exit3:
 exit2:
 	if (ipw->common_memory) {
-		release_mem_region(ipw->request_common_memory.Base,
-				ipw->request_common_memory.Size);
+		release_mem_region(p_dev->resource[2]->start,
+				resource_size(p_dev->resource[2]));
 		iounmap(ipw->common_memory);
 	}
 exit1:
@@ -175,14 +153,13 @@
 	int ret = 0;
 
 	ipw->is_v2_card = 0;
+	link->config_flags |= CONF_AUTO_SET_IO | CONF_AUTO_SET_IOMEM |
+		CONF_ENABLE_IRQ;
 
 	ret = pcmcia_loop_config(link, ipwireless_probe, ipw);
 	if (ret != 0)
 		return ret;
 
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	INIT_WORK(&ipw->work_reboot, signalled_reboot_work);
 
 	ipwireless_init_hardware_v1(ipw->hardware, link->resource[0]->start,
@@ -201,13 +178,9 @@
 			(unsigned int) link->irq);
 	if (ipw->attr_memory && ipw->common_memory)
 		printk(KERN_INFO IPWIRELESS_PCCARD_NAME
-			": attr memory 0x%08lx-0x%08lx, common memory 0x%08lx-0x%08lx\n",
-			ipw->request_attr_memory.Base,
-			ipw->request_attr_memory.Base
-			+ ipw->request_attr_memory.Size - 1,
-			ipw->request_common_memory.Base,
-			ipw->request_common_memory.Base
-			+ ipw->request_common_memory.Size - 1);
+			": attr memory %pR, common memory %pR\n",
+			link->resource[3],
+			link->resource[2]);
 
 	ipw->network = ipwireless_network_create(ipw->hardware);
 	if (!ipw->network)
@@ -223,25 +196,23 @@
 	 * Do the RequestConfiguration last, because it enables interrupts.
 	 * Then we don't get any interrupts before we're ready for them.
 	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
-
+	ret = pcmcia_enable_device(link);
 	if (ret != 0)
 		goto exit;
 
 	return 0;
 
 exit:
-	if (ipw->attr_memory) {
-		release_mem_region(ipw->request_attr_memory.Base,
-				ipw->request_attr_memory.Size);
-		iounmap(ipw->attr_memory);
-
-	}
 	if (ipw->common_memory) {
-		release_mem_region(ipw->request_common_memory.Base,
-				ipw->request_common_memory.Size);
+		release_mem_region(link->resource[2]->start,
+				resource_size(link->resource[2]));
 		iounmap(ipw->common_memory);
 	}
+	if (ipw->attr_memory) {
+		release_mem_region(link->resource[3]->start,
+				resource_size(link->resource[3]));
+		iounmap(ipw->attr_memory);
+	}
 	pcmcia_disable_device(link);
 	return -1;
 }
@@ -249,13 +220,13 @@
 static void release_ipwireless(struct ipw_dev *ipw)
 {
 	if (ipw->common_memory) {
-		release_mem_region(ipw->request_common_memory.Base,
-				ipw->request_common_memory.Size);
+		release_mem_region(ipw->link->resource[2]->start,
+				resource_size(ipw->link->resource[2]));
 		iounmap(ipw->common_memory);
 	}
 	if (ipw->attr_memory) {
-		release_mem_region(ipw->request_attr_memory.Base,
-				ipw->request_attr_memory.Size);
+		release_mem_region(ipw->link->resource[3]->start,
+				resource_size(ipw->link->resource[3]));
 		iounmap(ipw->attr_memory);
 	}
 	pcmcia_disable_device(ipw->link);
@@ -324,7 +295,7 @@
 	.owner		= THIS_MODULE,
 	.probe          = ipwireless_attach,
 	.remove         = ipwireless_detach,
-	.drv = { .name  = IPWIRELESS_PCCARD_NAME },
+	.name		= IPWIRELESS_PCCARD_NAME,
 	.id_table       = ipw_ids
 };
 
@@ -336,9 +307,6 @@
 {
 	int ret;
 
-	printk(KERN_INFO IPWIRELESS_PCCARD_NAME " "
-	       IPWIRELESS_PCMCIA_VERSION " by " IPWIRELESS_PCMCIA_AUTHOR "\n");
-
 	ret = ipwireless_tty_init();
 	if (ret != 0)
 		return ret;
@@ -355,9 +323,6 @@
  */
 static void __exit exit_ipwireless(void)
 {
-	printk(KERN_INFO IPWIRELESS_PCCARD_NAME " "
-			IPWIRELESS_PCMCIA_VERSION " removed\n");
-
 	pcmcia_unregister_driver(&me);
 	ipwireless_tty_release();
 }
diff --git a/drivers/char/pcmcia/ipwireless/main.h b/drivers/char/pcmcia/ipwireless/main.h
index c207be8..f2cbb11 100644
--- a/drivers/char/pcmcia/ipwireless/main.h
+++ b/drivers/char/pcmcia/ipwireless/main.h
@@ -21,7 +21,6 @@
 #include <linux/sched.h>
 #include <linux/types.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -45,13 +44,9 @@
 	struct pcmcia_device *link;
 	int is_v2_card;
 
-	window_handle_t handle_attr_memory;
 	void __iomem *attr_memory;
-	win_req_t request_attr_memory;
 
-	window_handle_t handle_common_memory;
 	void __iomem *common_memory;
-	win_req_t request_common_memory;
 
 	/* Reference to attribute memory, containing CIS data */
 	void *attribute_memory;
diff --git a/drivers/char/pcmcia/ipwireless/tty.h b/drivers/char/pcmcia/ipwireless/tty.h
index 3e163d4..747b2d6 100644
--- a/drivers/char/pcmcia/ipwireless/tty.h
+++ b/drivers/char/pcmcia/ipwireless/tty.h
@@ -21,7 +21,6 @@
 #include <linux/types.h>
 #include <linux/sched.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 9ecd6be..be18100 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -70,7 +70,6 @@
 #include <linux/workqueue.h>
 #include <linux/hdlc.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -550,9 +549,6 @@
 
     /* Initialize the struct pcmcia_device structure */
 
-    link->conf.Attributes = 0;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-
     ret = mgslpc_config(link);
     if (ret)
 	    return ret;
@@ -565,20 +561,8 @@
 /* Card has been inserted.
  */
 
-static int mgslpc_ioprobe(struct pcmcia_device *p_dev,
-			  cistpl_cftable_entry_t *cfg,
-			  cistpl_cftable_entry_t *dflt,
-			  unsigned int vcc,
-			  void *priv_data)
+static int mgslpc_ioprobe(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (!cfg->io.nwin)
-		return -ENODEV;
-
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
-	p_dev->resource[0]->flags |= pcmcia_io_cfg_data_width(cfg->io.flags);
-	p_dev->io_lines = cfg->io.flags & CISTPL_IO_LINES_MASK;
-
 	return pcmcia_request_io(p_dev);
 }
 
@@ -590,32 +574,24 @@
     if (debug_level >= DEBUG_LEVEL_INFO)
 	    printk("mgslpc_config(0x%p)\n", link);
 
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
     ret = pcmcia_loop_config(link, mgslpc_ioprobe, NULL);
     if (ret != 0)
 	    goto failed;
 
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.ConfigIndex = 8;
-    link->conf.Present = PRESENT_OPTION;
+    link->config_index = 8;
+    link->config_regs = PRESENT_OPTION;
 
     ret = pcmcia_request_irq(link, mgslpc_isr);
     if (ret)
 	    goto failed;
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
     info->io_base = link->resource[0]->start;
     info->irq_level = link->irq;
-
-    dev_info(&link->dev, "index 0x%02x:",
-	    link->conf.ConfigIndex);
-    if (link->conf.Attributes & CONF_ENABLE_IRQ)
-	    printk(", irq %d", link->irq);
-    if (link->resource[0])
-	    printk(", io %pR", link->resource[0]);
-    printk("\n");
     return 0;
 
 failed:
@@ -2797,9 +2773,7 @@
 
 static struct pcmcia_driver mgslpc_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "synclink_cs",
-	},
+	.name		= "synclink_cs",
 	.probe		= mgslpc_probe,
 	.remove		= mgslpc_detach,
 	.id_table	= mgslpc_ids,
@@ -2835,8 +2809,6 @@
 {
 	int rc;
 
-	printk("Unloading %s: version %s\n", driver_name, driver_version);
-
 	while(mgslpc_device_list)
 		mgslpc_remove_device(mgslpc_device_list);
 
@@ -2859,8 +2831,6 @@
 	    BREAKPOINT();
     }
 
-    printk("%s %s\n", driver_name, driver_version);
-
     if ((rc = pcmcia_register_driver(&mgslpc_driver)) < 0)
 	    return rc;
 
@@ -4127,6 +4097,8 @@
 	if (cmd != SIOCWANDEV)
 		return hdlc_ioctl(dev, ifr, cmd);
 
+	memset(&new_line, 0, size);
+
 	switch(ifr->ifr_settings.type) {
 	case IF_GET_IFACE: /* return current sync_serial_settings */
 
diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c
index 05ad4a1..7c41335 100644
--- a/drivers/char/tpm/tpm.c
+++ b/drivers/char/tpm/tpm.c
@@ -47,6 +47,16 @@
 #define TPM_MAX_PROTECTED_ORDINAL 12
 #define TPM_PROTECTED_ORDINAL_MASK 0xFF
 
+/*
+ * Bug workaround - some TPM's don't flush the most
+ * recently changed pcr on suspend, so force the flush
+ * with an extend to the selected _unused_ non-volatile pcr.
+ */
+static int tpm_suspend_pcr;
+module_param_named(suspend_pcr, tpm_suspend_pcr, uint, 0644);
+MODULE_PARM_DESC(suspend_pcr,
+		 "PCR to use for dummy writes to faciltate flush on suspend.");
+
 static LIST_HEAD(tpm_chip_list);
 static DEFINE_SPINLOCK(driver_lock);
 static DECLARE_BITMAP(dev_mask, TPM_NUM_DEVICES);
@@ -1077,18 +1087,6 @@
 	.ordinal = TPM_ORD_SAVESTATE
 };
 
-/* Bug workaround - some TPM's don't flush the most
- * recently changed pcr on suspend, so force the flush
- * with an extend to the selected _unused_ non-volatile pcr.
- */
-static int tpm_suspend_pcr;
-static int __init tpm_suspend_setup(char *str)
-{
-	get_option(&str, &tpm_suspend_pcr);
-	return 1;
-}
-__setup("tpm_suspend_pcr=", tpm_suspend_setup);
-
 /*
  * We are about to suspend. Save the TPM state
  * so that it can be restored.
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index c810481..6c1b676 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -48,6 +48,9 @@
 	/* Used for exporting per-port information to debugfs */
 	struct dentry *debugfs_dir;
 
+	/* List of all the devices we're handling */
+	struct list_head portdevs;
+
 	/* Number of devices this driver is handling */
 	unsigned int index;
 
@@ -108,6 +111,9 @@
  * ports for that device (vdev->priv).
  */
 struct ports_device {
+	/* Next portdev in the list, head is in the pdrvdata struct */
+	struct list_head list;
+
 	/*
 	 * Workqueue handlers where we process deferred work after
 	 * notification
@@ -178,15 +184,21 @@
 	struct console cons;
 
 	/* Each port associates with a separate char device */
-	struct cdev cdev;
+	struct cdev *cdev;
 	struct device *dev;
 
+	/* Reference-counting to handle port hot-unplugs and file operations */
+	struct kref kref;
+
 	/* A waitqueue for poll() or blocking read operations */
 	wait_queue_head_t waitqueue;
 
 	/* The 'name' of the port that we expose via sysfs properties */
 	char *name;
 
+	/* We can notify apps of host connect / disconnect events via SIGIO */
+	struct fasync_struct *async_queue;
+
 	/* The 'id' to identify the port with the Host */
 	u32 id;
 
@@ -221,6 +233,41 @@
 	return port;
 }
 
+static struct port *find_port_by_devt_in_portdev(struct ports_device *portdev,
+						 dev_t dev)
+{
+	struct port *port;
+	unsigned long flags;
+
+	spin_lock_irqsave(&portdev->ports_lock, flags);
+	list_for_each_entry(port, &portdev->ports, list)
+		if (port->cdev->dev == dev)
+			goto out;
+	port = NULL;
+out:
+	spin_unlock_irqrestore(&portdev->ports_lock, flags);
+
+	return port;
+}
+
+static struct port *find_port_by_devt(dev_t dev)
+{
+	struct ports_device *portdev;
+	struct port *port;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdrvdata_lock, flags);
+	list_for_each_entry(portdev, &pdrvdata.portdevs, list) {
+		port = find_port_by_devt_in_portdev(portdev, dev);
+		if (port)
+			goto out;
+	}
+	port = NULL;
+out:
+	spin_unlock_irqrestore(&pdrvdata_lock, flags);
+	return port;
+}
+
 static struct port *find_port_by_id(struct ports_device *portdev, u32 id)
 {
 	struct port *port;
@@ -410,7 +457,10 @@
 static ssize_t send_control_msg(struct port *port, unsigned int event,
 				unsigned int value)
 {
-	return __send_control_msg(port->portdev, port->id, event, value);
+	/* Did the port get unplugged before userspace closed it? */
+	if (port->portdev)
+		return __send_control_msg(port->portdev, port->id, event, value);
+	return 0;
 }
 
 /* Callers must take the port->outvq_lock */
@@ -459,9 +509,12 @@
 
 	/*
 	 * Wait till the host acknowledges it pushed out the data we
-	 * sent.  This is done for ports in blocking mode or for data
-	 * from the hvc_console; the tty operations are performed with
-	 * spinlocks held so we can't sleep here.
+	 * sent.  This is done for data from the hvc_console; the tty
+	 * operations are performed with spinlocks held so we can't
+	 * sleep here.  An alternative would be to copy the data to a
+	 * buffer and relax the spinning requirement.  The downside is
+	 * we need to kmalloc a GFP_ATOMIC buffer each time the
+	 * console driver writes something out.
 	 */
 	while (!virtqueue_get_buf(out_vq, &len))
 		cpu_relax();
@@ -522,6 +575,10 @@
 /* The condition that must be true for polling to end */
 static bool will_read_block(struct port *port)
 {
+	if (!port->guest_connected) {
+		/* Port got hot-unplugged. Let's exit. */
+		return false;
+	}
 	return !port_has_data(port) && port->host_connected;
 }
 
@@ -572,6 +629,9 @@
 		if (ret < 0)
 			return ret;
 	}
+	/* Port got hot-unplugged. */
+	if (!port->guest_connected)
+		return -ENODEV;
 	/*
 	 * We could've received a disconnection message while we were
 	 * waiting for more data.
@@ -613,6 +673,9 @@
 		if (ret < 0)
 			return ret;
 	}
+	/* Port got hot-unplugged. */
+	if (!port->guest_connected)
+		return -ENODEV;
 
 	count = min((size_t)(32 * 1024), count);
 
@@ -626,6 +689,14 @@
 		goto free_buf;
 	}
 
+	/*
+	 * We now ask send_buf() to not spin for generic ports -- we
+	 * can re-use the same code path that non-blocking file
+	 * descriptors take for blocking file descriptors since the
+	 * wait is already done and we're certain the write will go
+	 * through to the host.
+	 */
+	nonblock = true;
 	ret = send_buf(port, buf, count, nonblock);
 
 	if (nonblock && ret > 0)
@@ -645,6 +716,10 @@
 	port = filp->private_data;
 	poll_wait(filp, &port->waitqueue, wait);
 
+	if (!port->guest_connected) {
+		/* Port got unplugged */
+		return POLLHUP;
+	}
 	ret = 0;
 	if (!will_read_block(port))
 		ret |= POLLIN | POLLRDNORM;
@@ -656,6 +731,8 @@
 	return ret;
 }
 
+static void remove_port(struct kref *kref);
+
 static int port_fops_release(struct inode *inode, struct file *filp)
 {
 	struct port *port;
@@ -676,6 +753,16 @@
 	reclaim_consumed_buffers(port);
 	spin_unlock_irq(&port->outvq_lock);
 
+	/*
+	 * Locks aren't necessary here as a port can't be opened after
+	 * unplug, and if a port isn't unplugged, a kref would already
+	 * exist for the port.  Plus, taking ports_lock here would
+	 * create a dependency on other locks taken by functions
+	 * inside remove_port if we're the last holder of the port,
+	 * creating many problems.
+	 */
+	kref_put(&port->kref, remove_port);
+
 	return 0;
 }
 
@@ -683,22 +770,31 @@
 {
 	struct cdev *cdev = inode->i_cdev;
 	struct port *port;
+	int ret;
 
-	port = container_of(cdev, struct port, cdev);
+	port = find_port_by_devt(cdev->dev);
 	filp->private_data = port;
 
+	/* Prevent against a port getting hot-unplugged at the same time */
+	spin_lock_irq(&port->portdev->ports_lock);
+	kref_get(&port->kref);
+	spin_unlock_irq(&port->portdev->ports_lock);
+
 	/*
 	 * Don't allow opening of console port devices -- that's done
 	 * via /dev/hvc
 	 */
-	if (is_console_port(port))
-		return -ENXIO;
+	if (is_console_port(port)) {
+		ret = -ENXIO;
+		goto out;
+	}
 
 	/* Allow only one process to open a particular port at a time */
 	spin_lock_irq(&port->inbuf_lock);
 	if (port->guest_connected) {
 		spin_unlock_irq(&port->inbuf_lock);
-		return -EMFILE;
+		ret = -EMFILE;
+		goto out;
 	}
 
 	port->guest_connected = true;
@@ -713,10 +809,23 @@
 	reclaim_consumed_buffers(port);
 	spin_unlock_irq(&port->outvq_lock);
 
+	nonseekable_open(inode, filp);
+
 	/* Notify host of port being opened */
 	send_control_msg(filp->private_data, VIRTIO_CONSOLE_PORT_OPEN, 1);
 
 	return 0;
+out:
+	kref_put(&port->kref, remove_port);
+	return ret;
+}
+
+static int port_fops_fasync(int fd, struct file *filp, int mode)
+{
+	struct port *port;
+
+	port = filp->private_data;
+	return fasync_helper(fd, filp, mode, &port->async_queue);
 }
 
 /*
@@ -732,6 +841,8 @@
 	.write = port_fops_write,
 	.poll  = port_fops_poll,
 	.release = port_fops_release,
+	.fasync = port_fops_fasync,
+	.llseek = no_llseek,
 };
 
 /*
@@ -990,6 +1101,12 @@
 	return nr_added_bufs;
 }
 
+static void send_sigio_to_port(struct port *port)
+{
+	if (port->async_queue && port->guest_connected)
+		kill_fasync(&port->async_queue, SIGIO, POLL_OUT);
+}
+
 static int add_port(struct ports_device *portdev, u32 id)
 {
 	char debugfs_name[16];
@@ -1004,6 +1121,7 @@
 		err = -ENOMEM;
 		goto fail;
 	}
+	kref_init(&port->kref);
 
 	port->portdev = portdev;
 	port->id = id;
@@ -1011,6 +1129,7 @@
 	port->name = NULL;
 	port->inbuf = NULL;
 	port->cons.hvc = NULL;
+	port->async_queue = NULL;
 
 	port->cons.ws.ws_row = port->cons.ws.ws_col = 0;
 
@@ -1021,14 +1140,20 @@
 	port->in_vq = portdev->in_vqs[port->id];
 	port->out_vq = portdev->out_vqs[port->id];
 
-	cdev_init(&port->cdev, &port_fops);
+	port->cdev = cdev_alloc();
+	if (!port->cdev) {
+		dev_err(&port->portdev->vdev->dev, "Error allocating cdev\n");
+		err = -ENOMEM;
+		goto free_port;
+	}
+	port->cdev->ops = &port_fops;
 
 	devt = MKDEV(portdev->chr_major, id);
-	err = cdev_add(&port->cdev, devt, 1);
+	err = cdev_add(port->cdev, devt, 1);
 	if (err < 0) {
 		dev_err(&port->portdev->vdev->dev,
 			"Error %d adding cdev for port %u\n", err, id);
-		goto free_port;
+		goto free_cdev;
 	}
 	port->dev = device_create(pdrvdata.class, &port->portdev->vdev->dev,
 				  devt, port, "vport%up%u",
@@ -1093,7 +1218,7 @@
 free_device:
 	device_destroy(pdrvdata.class, port->dev->devt);
 free_cdev:
-	cdev_del(&port->cdev);
+	cdev_del(port->cdev);
 free_port:
 	kfree(port);
 fail:
@@ -1102,21 +1227,45 @@
 	return err;
 }
 
-/* Remove all port-specific data. */
-static int remove_port(struct port *port)
+/* No users remain, remove all port-specific data. */
+static void remove_port(struct kref *kref)
+{
+	struct port *port;
+
+	port = container_of(kref, struct port, kref);
+
+	sysfs_remove_group(&port->dev->kobj, &port_attribute_group);
+	device_destroy(pdrvdata.class, port->dev->devt);
+	cdev_del(port->cdev);
+
+	kfree(port->name);
+
+	debugfs_remove(port->debugfs_file);
+
+	kfree(port);
+}
+
+/*
+ * Port got unplugged.  Remove port from portdev's list and drop the
+ * kref reference.  If no userspace has this port opened, it will
+ * result in immediate removal the port.
+ */
+static void unplug_port(struct port *port)
 {
 	struct port_buffer *buf;
 
+	spin_lock_irq(&port->portdev->ports_lock);
+	list_del(&port->list);
+	spin_unlock_irq(&port->portdev->ports_lock);
+
 	if (port->guest_connected) {
 		port->guest_connected = false;
 		port->host_connected = false;
 		wake_up_interruptible(&port->waitqueue);
-		send_control_msg(port, VIRTIO_CONSOLE_PORT_OPEN, 0);
-	}
 
-	spin_lock_irq(&port->portdev->ports_lock);
-	list_del(&port->list);
-	spin_unlock_irq(&port->portdev->ports_lock);
+		/* Let the app know the port is going down. */
+		send_sigio_to_port(port);
+	}
 
 	if (is_console_port(port)) {
 		spin_lock_irq(&pdrvdata_lock);
@@ -1135,9 +1284,6 @@
 		hvc_remove(port->cons.hvc);
 #endif
 	}
-	sysfs_remove_group(&port->dev->kobj, &port_attribute_group);
-	device_destroy(pdrvdata.class, port->dev->devt);
-	cdev_del(&port->cdev);
 
 	/* Remove unused data this port might have received. */
 	discard_port_data(port);
@@ -1148,12 +1294,19 @@
 	while ((buf = virtqueue_detach_unused_buf(port->in_vq)))
 		free_buf(buf);
 
-	kfree(port->name);
+	/*
+	 * We should just assume the device itself has gone off --
+	 * else a close on an open port later will try to send out a
+	 * control message.
+	 */
+	port->portdev = NULL;
 
-	debugfs_remove(port->debugfs_file);
-
-	kfree(port);
-	return 0;
+	/*
+	 * Locks around here are not necessary - a port can't be
+	 * opened after we removed the port struct from ports_list
+	 * above.
+	 */
+	kref_put(&port->kref, remove_port);
 }
 
 /* Any private messages that the Host and Guest want to share */
@@ -1192,7 +1345,7 @@
 		add_port(portdev, cpkt->id);
 		break;
 	case VIRTIO_CONSOLE_PORT_REMOVE:
-		remove_port(port);
+		unplug_port(port);
 		break;
 	case VIRTIO_CONSOLE_CONSOLE_PORT:
 		if (!cpkt->value)
@@ -1234,6 +1387,12 @@
 		spin_lock_irq(&port->outvq_lock);
 		reclaim_consumed_buffers(port);
 		spin_unlock_irq(&port->outvq_lock);
+
+		/*
+		 * If the guest is connected, it'll be interested in
+		 * knowing the host connection state changed.
+		 */
+		send_sigio_to_port(port);
 		break;
 	case VIRTIO_CONSOLE_PORT_NAME:
 		/*
@@ -1330,6 +1489,9 @@
 
 	wake_up_interruptible(&port->waitqueue);
 
+	/* Send a SIGIO indicating new data in case the process asked for it */
+	send_sigio_to_port(port);
+
 	if (is_console_port(port) && hvc_poll(port->cons.hvc))
 		hvc_kick();
 }
@@ -1566,6 +1728,10 @@
 		add_port(portdev, 0);
 	}
 
+	spin_lock_irq(&pdrvdata_lock);
+	list_add_tail(&portdev->list, &pdrvdata.portdevs);
+	spin_unlock_irq(&pdrvdata_lock);
+
 	__send_control_msg(portdev, VIRTIO_CONSOLE_BAD_ID,
 			   VIRTIO_CONSOLE_DEVICE_READY, 1);
 	return 0;
@@ -1589,23 +1755,41 @@
 {
 	struct ports_device *portdev;
 	struct port *port, *port2;
-	struct port_buffer *buf;
-	unsigned int len;
 
 	portdev = vdev->priv;
 
+	spin_lock_irq(&pdrvdata_lock);
+	list_del(&portdev->list);
+	spin_unlock_irq(&pdrvdata_lock);
+
+	/* Disable interrupts for vqs */
+	vdev->config->reset(vdev);
+	/* Finish up work that's lined up */
 	cancel_work_sync(&portdev->control_work);
 
 	list_for_each_entry_safe(port, port2, &portdev->ports, list)
-		remove_port(port);
+		unplug_port(port);
 
 	unregister_chrdev(portdev->chr_major, "virtio-portsdev");
 
-	while ((buf = virtqueue_get_buf(portdev->c_ivq, &len)))
-		free_buf(buf);
+	/*
+	 * When yanking out a device, we immediately lose the
+	 * (device-side) queues.  So there's no point in keeping the
+	 * guest side around till we drop our final reference.  This
+	 * also means that any ports which are in an open state will
+	 * have to just stop using the port, as the vqs are going
+	 * away.
+	 */
+	if (use_multiport(portdev)) {
+		struct port_buffer *buf;
+		unsigned int len;
 
-	while ((buf = virtqueue_detach_unused_buf(portdev->c_ivq)))
-		free_buf(buf);
+		while ((buf = virtqueue_get_buf(portdev->c_ivq, &len)))
+			free_buf(buf);
+
+		while ((buf = virtqueue_detach_unused_buf(portdev->c_ivq)))
+			free_buf(buf);
+	}
 
 	vdev->config->del_vqs(vdev);
 	kfree(portdev->in_vqs);
@@ -1652,6 +1836,7 @@
 			   PTR_ERR(pdrvdata.debugfs_dir));
 	}
 	INIT_LIST_HEAD(&pdrvdata.consoles);
+	INIT_LIST_HEAD(&pdrvdata.portdevs);
 
 	return register_virtio_driver(&virtio_console);
 }
diff --git a/drivers/dma/ioat/dma_v2.c b/drivers/dma/ioat/dma_v2.c
index 216f9d3..effd140 100644
--- a/drivers/dma/ioat/dma_v2.c
+++ b/drivers/dma/ioat/dma_v2.c
@@ -879,7 +879,7 @@
 	dma->device_issue_pending = ioat2_issue_pending;
 	dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
 	dma->device_free_chan_resources = ioat2_free_chan_resources;
-	dma->device_tx_status = ioat_tx_status;
+	dma->device_tx_status = ioat_dma_tx_status;
 
 	err = ioat_probe(device);
 	if (err)
diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index 70bb350..9dbb28b 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -39,7 +39,7 @@
 	  there're four debug levels (x=0,1,2,3 from low to high).
 	  Usually you should select 'N'.
 
- config EDAC_DECODE_MCE
+config EDAC_DECODE_MCE
 	tristate "Decode MCEs in human-readable form (only on AMD for now)"
 	depends on CPU_SUP_AMD && X86_MCE
 	default y
@@ -51,6 +51,16 @@
 	  which occur really early upon boot, before the module infrastructure
 	  has been initialized.
 
+config EDAC_MCE_INJ
+	tristate "Simple MCE injection interface over /sysfs"
+	depends on EDAC_DECODE_MCE
+	default n
+	help
+	  This is a simple interface to inject MCEs over /sysfs and test
+	  the MCE decoding code in EDAC.
+
+	  This is currently AMD-only.
+
 config EDAC_MM_EDAC
 	tristate "Main Memory EDAC (Error Detection And Correction) reporting"
 	help
@@ -66,13 +76,13 @@
 
 config EDAC_AMD64
 	tristate "AMD64 (Opteron, Athlon64) K8, F10h, F11h"
-	depends on EDAC_MM_EDAC && K8_NB && X86_64 && PCI && EDAC_DECODE_MCE
+	depends on EDAC_MM_EDAC && AMD_NB && X86_64 && PCI && EDAC_DECODE_MCE
 	help
 	  Support for error detection and correction on the AMD 64
 	  Families of Memory Controllers (K8, F10h and F11h)
 
 config EDAC_AMD64_ERROR_INJECTION
-	bool "Sysfs Error Injection facilities"
+	bool "Sysfs HW Error injection facilities"
 	depends on EDAC_AMD64
 	help
 	  Recent Opterons (Family 10h and later) provide for Memory Error
diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile
index ca6b1bb..32c7bc9 100644
--- a/drivers/edac/Makefile
+++ b/drivers/edac/Makefile
@@ -17,6 +17,9 @@
 edac_core-objs	+= edac_pci.o edac_pci_sysfs.o
 endif
 
+obj-$(CONFIG_EDAC_MCE_INJ)		+= mce_amd_inj.o
+
+edac_mce_amd-objs			:= mce_amd.o
 obj-$(CONFIG_EDAC_DECODE_MCE)		+= edac_mce_amd.o
 
 obj-$(CONFIG_EDAC_AMD76X)		+= amd76x_edac.o
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index e7d5d6b..8521401 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -1,5 +1,5 @@
 #include "amd64_edac.h"
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 
 static struct edac_pci_ctl_info *amd64_ctl_pci;
 
@@ -2073,11 +2073,18 @@
 		amd64_handle_ue(mci, info);
 }
 
-void amd64_decode_bus_error(int node_id, struct err_regs *regs)
+void amd64_decode_bus_error(int node_id, struct mce *m, u32 nbcfg)
 {
 	struct mem_ctl_info *mci = mci_lookup[node_id];
+	struct err_regs regs;
 
-	__amd64_decode_bus_error(mci, regs);
+	regs.nbsl  = (u32) m->status;
+	regs.nbsh  = (u32)(m->status >> 32);
+	regs.nbeal = (u32) m->addr;
+	regs.nbeah = (u32)(m->addr >> 32);
+	regs.nbcfg = nbcfg;
+
+	__amd64_decode_bus_error(mci, &regs);
 
 	/*
 	 * Check the UE bit of the NB status high register, if set generate some
@@ -2086,7 +2093,7 @@
 	 *
 	 * FIXME: this should go somewhere else, if at all.
 	 */
-	if (regs->nbsh & K8_NBSH_UC_ERR && !report_gart_errors)
+	if (regs.nbsh & K8_NBSH_UC_ERR && !report_gart_errors)
 		edac_mc_handle_ue_no_info(mci, "UE bit is set");
 
 }
@@ -2927,7 +2934,7 @@
 	 * to finish initialization of the MC instances.
 	 */
 	err = -ENODEV;
-	for (nb = 0; nb < num_k8_northbridges; nb++) {
+	for (nb = 0; nb < k8_northbridges.num; nb++) {
 		if (!pvt_lookup[nb])
 			continue;
 
diff --git a/drivers/edac/amd64_edac.h b/drivers/edac/amd64_edac.h
index 613b938..044aee4 100644
--- a/drivers/edac/amd64_edac.h
+++ b/drivers/edac/amd64_edac.h
@@ -72,7 +72,7 @@
 #include <linux/edac.h>
 #include <asm/msr.h>
 #include "edac_core.h"
-#include "edac_mce_amd.h"
+#include "mce_amd.h"
 
 #define amd64_printk(level, fmt, arg...) \
 	edac_printk(level, "amd64", fmt, ##arg)
@@ -482,11 +482,10 @@
 extern const char *to_msgs[2];
 extern const char *pp_msgs[4];
 extern const char *ii_msgs[4];
-extern const char *ext_msgs[32];
 extern const char *htlink_msgs[8];
 
 #ifdef CONFIG_EDAC_DEBUG
-#define NUM_DBG_ATTRS 9
+#define NUM_DBG_ATTRS 5
 #else
 #define NUM_DBG_ATTRS 0
 #endif
diff --git a/drivers/edac/amd64_edac_dbg.c b/drivers/edac/amd64_edac_dbg.c
index 59cf2cf..e356228 100644
--- a/drivers/edac/amd64_edac_dbg.c
+++ b/drivers/edac/amd64_edac_dbg.c
@@ -1,167 +1,16 @@
 #include "amd64_edac.h"
 
-/*
- * accept a hex value and store it into the virtual error register file, field:
- * nbeal and nbeah. Assume virtual error values have already been set for: NBSL,
- * NBSH and NBCFG. Then proceed to map the error values to a MC, CSROW and
- * CHANNEL
- */
-static ssize_t amd64_nbea_store(struct mem_ctl_info *mci, const char *data,
-				size_t count)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	unsigned long long value;
-	int ret = 0;
-
-	ret = strict_strtoull(data, 16, &value);
-	if (ret != -EINVAL) {
-		debugf0("received NBEA= 0x%llx\n", value);
-
-		/* place the value into the virtual error packet */
-		pvt->ctl_error_info.nbeal = (u32) value;
-		value >>= 32;
-		pvt->ctl_error_info.nbeah = (u32) value;
-
-		/* Process the Mapping request */
-		/* TODO: Add race prevention */
-		amd_decode_nb_mce(pvt->mc_node_id, &pvt->ctl_error_info, 1);
-
-		return count;
-	}
-	return ret;
+#define EDAC_DCT_ATTR_SHOW(reg)						\
+static ssize_t amd64_##reg##_show(struct mem_ctl_info *mci, char *data)	\
+{									\
+	struct amd64_pvt *pvt = mci->pvt_info;				\
+		return sprintf(data, "0x%016llx\n", (u64)pvt->reg);	\
 }
 
-/* display back what the last NBEA (MCA NB Address (MC4_ADDR)) was written */
-static ssize_t amd64_nbea_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	u64 value;
-
-	value = pvt->ctl_error_info.nbeah;
-	value <<= 32;
-	value |= pvt->ctl_error_info.nbeal;
-
-	return sprintf(data, "%llx\n", value);
-}
-
-/* store the NBSL (MCA NB Status Low (MC4_STATUS)) value user desires */
-static ssize_t amd64_nbsl_store(struct mem_ctl_info *mci, const char *data,
-				size_t count)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	unsigned long value;
-	int ret = 0;
-
-	ret = strict_strtoul(data, 16, &value);
-	if (ret != -EINVAL) {
-		debugf0("received NBSL= 0x%lx\n", value);
-
-		pvt->ctl_error_info.nbsl = (u32) value;
-
-		return count;
-	}
-	return ret;
-}
-
-/* display back what the last NBSL value written */
-static ssize_t amd64_nbsl_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	u32 value;
-
-	value = pvt->ctl_error_info.nbsl;
-
-	return sprintf(data, "%x\n", value);
-}
-
-/* store the NBSH (MCA NB Status High) value user desires */
-static ssize_t amd64_nbsh_store(struct mem_ctl_info *mci, const char *data,
-				size_t count)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	unsigned long value;
-	int ret = 0;
-
-	ret = strict_strtoul(data, 16, &value);
-	if (ret != -EINVAL) {
-		debugf0("received NBSH= 0x%lx\n", value);
-
-		pvt->ctl_error_info.nbsh = (u32) value;
-
-		return count;
-	}
-	return ret;
-}
-
-/* display back what the last NBSH value written */
-static ssize_t amd64_nbsh_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	u32 value;
-
-	value = pvt->ctl_error_info.nbsh;
-
-	return sprintf(data, "%x\n", value);
-}
-
-/* accept and store the NBCFG (MCA NB Configuration) value user desires */
-static ssize_t amd64_nbcfg_store(struct mem_ctl_info *mci,
-					const char *data, size_t count)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-	unsigned long value;
-	int ret = 0;
-
-	ret = strict_strtoul(data, 16, &value);
-	if (ret != -EINVAL) {
-		debugf0("received NBCFG= 0x%lx\n", value);
-
-		pvt->ctl_error_info.nbcfg = (u32) value;
-
-		return count;
-	}
-	return ret;
-}
-
-/* various show routines for the controls of a MCI */
-static ssize_t amd64_nbcfg_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-
-	return sprintf(data, "%x\n", pvt->ctl_error_info.nbcfg);
-}
-
-
-static ssize_t amd64_dhar_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-
-	return sprintf(data, "%x\n", pvt->dhar);
-}
-
-
-static ssize_t amd64_dbam_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-
-	return sprintf(data, "%x\n", pvt->dbam0);
-}
-
-
-static ssize_t amd64_topmem_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-
-	return sprintf(data, "%llx\n", pvt->top_mem);
-}
-
-
-static ssize_t amd64_topmem2_show(struct mem_ctl_info *mci, char *data)
-{
-	struct amd64_pvt *pvt = mci->pvt_info;
-
-	return sprintf(data, "%llx\n", pvt->top_mem2);
-}
+EDAC_DCT_ATTR_SHOW(dhar);
+EDAC_DCT_ATTR_SHOW(dbam0);
+EDAC_DCT_ATTR_SHOW(top_mem);
+EDAC_DCT_ATTR_SHOW(top_mem2);
 
 static ssize_t amd64_hole_show(struct mem_ctl_info *mci, char *data)
 {
@@ -182,38 +31,6 @@
 
 	{
 		.attr = {
-			.name = "nbea_ctl",
-			.mode = (S_IRUGO | S_IWUSR)
-		},
-		.show = amd64_nbea_show,
-		.store = amd64_nbea_store,
-	},
-	{
-		.attr = {
-			.name = "nbsl_ctl",
-			.mode = (S_IRUGO | S_IWUSR)
-		},
-		.show = amd64_nbsl_show,
-		.store = amd64_nbsl_store,
-	},
-	{
-		.attr = {
-			.name = "nbsh_ctl",
-			.mode = (S_IRUGO | S_IWUSR)
-		},
-		.show = amd64_nbsh_show,
-		.store = amd64_nbsh_store,
-	},
-	{
-		.attr = {
-			.name = "nbcfg_ctl",
-			.mode = (S_IRUGO | S_IWUSR)
-		},
-		.show = amd64_nbcfg_show,
-		.store = amd64_nbcfg_store,
-	},
-	{
-		.attr = {
 			.name = "dhar",
 			.mode = (S_IRUGO)
 		},
@@ -225,7 +42,7 @@
 			.name = "dbam",
 			.mode = (S_IRUGO)
 		},
-		.show = amd64_dbam_show,
+		.show = amd64_dbam0_show,
 		.store = NULL,
 	},
 	{
@@ -233,7 +50,7 @@
 			.name = "topmem",
 			.mode = (S_IRUGO)
 		},
-		.show = amd64_topmem_show,
+		.show = amd64_top_mem_show,
 		.store = NULL,
 	},
 	{
@@ -241,7 +58,7 @@
 			.name = "topmem2",
 			.mode = (S_IRUGO)
 		},
-		.show = amd64_topmem2_show,
+		.show = amd64_top_mem2_show,
 		.store = NULL,
 	},
 	{
diff --git a/drivers/edac/edac_device_sysfs.c b/drivers/edac/edac_device_sysfs.c
index 0709681..2941dca 100644
--- a/drivers/edac/edac_device_sysfs.c
+++ b/drivers/edac/edac_device_sysfs.c
@@ -13,6 +13,7 @@
 #include <linux/ctype.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/edac.h>
 
 #include "edac_core.h"
 #include "edac_module.h"
@@ -235,7 +236,7 @@
 	debugf1("%s()\n", __func__);
 
 	/* get the /sys/devices/system/edac reference */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class error\n", __func__);
 		err = -ENODEV;
@@ -255,7 +256,7 @@
 
 	if (!try_module_get(edac_dev->owner)) {
 		err = -ENODEV;
-		goto err_out;
+		goto err_mod_get;
 	}
 
 	/* register */
@@ -282,6 +283,9 @@
 err_kobj_reg:
 	module_put(edac_dev->owner);
 
+err_mod_get:
+	edac_put_sysfs_class();
+
 err_out:
 	return err;
 }
@@ -290,12 +294,11 @@
  * edac_device_unregister_sysfs_main_kobj:
  *	the '..../edac/<name>' kobject
  */
-void edac_device_unregister_sysfs_main_kobj(
-					struct edac_device_ctl_info *edac_dev)
+void edac_device_unregister_sysfs_main_kobj(struct edac_device_ctl_info *dev)
 {
 	debugf0("%s()\n", __func__);
 	debugf4("%s() name of kobject is: %s\n",
-		__func__, kobject_name(&edac_dev->kobj));
+		__func__, kobject_name(&dev->kobj));
 
 	/*
 	 * Unregister the edac device's kobject and
@@ -304,7 +307,8 @@
 	 *   a) module_put() this module
 	 *   b) 'kfree' the memory
 	 */
-	kobject_put(&edac_dev->kobj);
+	kobject_put(&dev->kobj);
+	edac_put_sysfs_class();
 }
 
 /* edac_dev -> instance information */
diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 8aad94d..a413586 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -11,6 +11,7 @@
 
 #include <linux/ctype.h>
 #include <linux/slab.h>
+#include <linux/edac.h>
 #include <linux/bug.h>
 
 #include "edac_core.h"
@@ -1011,13 +1012,13 @@
  */
 int edac_sysfs_setup_mc_kset(void)
 {
-	int err = 0;
+	int err = -EINVAL;
 	struct sysdev_class *edac_class;
 
 	debugf1("%s()\n", __func__);
 
 	/* get the /sys/devices/system/edac class reference */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class error=%d\n", __func__, err);
 		goto fail_out;
@@ -1028,15 +1029,16 @@
 	if (!mc_kset) {
 		err = -ENOMEM;
 		debugf1("%s() Failed to register '.../edac/mc'\n", __func__);
-		goto fail_out;
+		goto fail_kset;
 	}
 
 	debugf1("%s() Registered '.../edac/mc' kobject\n", __func__);
 
 	return 0;
 
+fail_kset:
+	edac_put_sysfs_class();
 
-	/* error unwind stack */
 fail_out:
 	return err;
 }
@@ -1049,5 +1051,6 @@
 void edac_sysfs_teardown_mc_kset(void)
 {
 	kset_unregister(mc_kset);
+	edac_put_sysfs_class();
 }
 
diff --git a/drivers/edac/edac_mce_amd.c b/drivers/edac/edac_mce_amd.c
deleted file mode 100644
index 9014df6..0000000
--- a/drivers/edac/edac_mce_amd.c
+++ /dev/null
@@ -1,452 +0,0 @@
-#include <linux/module.h>
-#include "edac_mce_amd.h"
-
-static bool report_gart_errors;
-static void (*nb_bus_decoder)(int node_id, struct err_regs *regs);
-
-void amd_report_gart_errors(bool v)
-{
-	report_gart_errors = v;
-}
-EXPORT_SYMBOL_GPL(amd_report_gart_errors);
-
-void amd_register_ecc_decoder(void (*f)(int, struct err_regs *))
-{
-	nb_bus_decoder = f;
-}
-EXPORT_SYMBOL_GPL(amd_register_ecc_decoder);
-
-void amd_unregister_ecc_decoder(void (*f)(int, struct err_regs *))
-{
-	if (nb_bus_decoder) {
-		WARN_ON(nb_bus_decoder != f);
-
-		nb_bus_decoder = NULL;
-	}
-}
-EXPORT_SYMBOL_GPL(amd_unregister_ecc_decoder);
-
-/*
- * string representation for the different MCA reported error types, see F3x48
- * or MSR0000_0411.
- */
-const char *tt_msgs[] = {        /* transaction type */
-	"instruction",
-	"data",
-	"generic",
-	"reserved"
-};
-EXPORT_SYMBOL_GPL(tt_msgs);
-
-const char *ll_msgs[] = {	/* cache level */
-	"L0",
-	"L1",
-	"L2",
-	"L3/generic"
-};
-EXPORT_SYMBOL_GPL(ll_msgs);
-
-const char *rrrr_msgs[] = {
-	"generic",
-	"generic read",
-	"generic write",
-	"data read",
-	"data write",
-	"inst fetch",
-	"prefetch",
-	"evict",
-	"snoop",
-	"reserved RRRR= 9",
-	"reserved RRRR= 10",
-	"reserved RRRR= 11",
-	"reserved RRRR= 12",
-	"reserved RRRR= 13",
-	"reserved RRRR= 14",
-	"reserved RRRR= 15"
-};
-EXPORT_SYMBOL_GPL(rrrr_msgs);
-
-const char *pp_msgs[] = {	/* participating processor */
-	"local node originated (SRC)",
-	"local node responded to request (RES)",
-	"local node observed as 3rd party (OBS)",
-	"generic"
-};
-EXPORT_SYMBOL_GPL(pp_msgs);
-
-const char *to_msgs[] = {
-	"no timeout",
-	"timed out"
-};
-EXPORT_SYMBOL_GPL(to_msgs);
-
-const char *ii_msgs[] = {	/* memory or i/o */
-	"mem access",
-	"reserved",
-	"i/o access",
-	"generic"
-};
-EXPORT_SYMBOL_GPL(ii_msgs);
-
-/*
- * Map the 4 or 5 (family-specific) bits of Extended Error code to the
- * string table.
- */
-const char *ext_msgs[] = {
-	"K8 ECC error",					/* 0_0000b */
-	"CRC error on link",				/* 0_0001b */
-	"Sync error packets on link",			/* 0_0010b */
-	"Master Abort during link operation",		/* 0_0011b */
-	"Target Abort during link operation",		/* 0_0100b */
-	"Invalid GART PTE entry during table walk",	/* 0_0101b */
-	"Unsupported atomic RMW command received",	/* 0_0110b */
-	"WDT error: NB transaction timeout",		/* 0_0111b */
-	"ECC/ChipKill ECC error",			/* 0_1000b */
-	"SVM DEV Error",				/* 0_1001b */
-	"Link Data error",				/* 0_1010b */
-	"Link/L3/Probe Filter Protocol error",		/* 0_1011b */
-	"NB Internal Arrays Parity error",		/* 0_1100b */
-	"DRAM Address/Control Parity error",		/* 0_1101b */
-	"Link Transmission error",			/* 0_1110b */
-	"GART/DEV Table Walk Data error"		/* 0_1111b */
-	"Res 0x100 error",				/* 1_0000b */
-	"Res 0x101 error",				/* 1_0001b */
-	"Res 0x102 error",				/* 1_0010b */
-	"Res 0x103 error",				/* 1_0011b */
-	"Res 0x104 error",				/* 1_0100b */
-	"Res 0x105 error",				/* 1_0101b */
-	"Res 0x106 error",				/* 1_0110b */
-	"Res 0x107 error",				/* 1_0111b */
-	"Res 0x108 error",				/* 1_1000b */
-	"Res 0x109 error",				/* 1_1001b */
-	"Res 0x10A error",				/* 1_1010b */
-	"Res 0x10B error",				/* 1_1011b */
-	"ECC error in L3 Cache Data",			/* 1_1100b */
-	"L3 Cache Tag error",				/* 1_1101b */
-	"L3 Cache LRU Parity error",			/* 1_1110b */
-	"Probe Filter error"				/* 1_1111b */
-};
-EXPORT_SYMBOL_GPL(ext_msgs);
-
-static void amd_decode_dc_mce(u64 mc0_status)
-{
-	u32 ec  = mc0_status & 0xffff;
-	u32 xec = (mc0_status >> 16) & 0xf;
-
-	pr_emerg("Data Cache Error");
-
-	if (xec == 1 && TLB_ERROR(ec))
-		pr_cont(": %s TLB multimatch.\n", LL_MSG(ec));
-	else if (xec == 0) {
-		if (mc0_status & (1ULL << 40))
-			pr_cont(" during Data Scrub.\n");
-		else if (TLB_ERROR(ec))
-			pr_cont(": %s TLB parity error.\n", LL_MSG(ec));
-		else if (MEM_ERROR(ec)) {
-			u8 ll   = ec & 0x3;
-			u8 tt   = (ec >> 2) & 0x3;
-			u8 rrrr = (ec >> 4) & 0xf;
-
-			/* see F10h BKDG (31116), Table 92. */
-			if (ll == 0x1) {
-				if (tt != 0x1)
-					goto wrong_dc_mce;
-
-				pr_cont(": Data/Tag %s error.\n", RRRR_MSG(ec));
-
-			} else if (ll == 0x2 && rrrr == 0x3)
-				pr_cont(" during L1 linefill from L2.\n");
-			else
-				goto wrong_dc_mce;
-		} else if (BUS_ERROR(ec) && boot_cpu_data.x86 == 0xf)
-			pr_cont(" during system linefill.\n");
-		else
-			goto wrong_dc_mce;
-	} else
-		goto wrong_dc_mce;
-
-	return;
-
-wrong_dc_mce:
-	pr_warning("Corrupted DC MCE info?\n");
-}
-
-static void amd_decode_ic_mce(u64 mc1_status)
-{
-	u32 ec  = mc1_status & 0xffff;
-	u32 xec = (mc1_status >> 16) & 0xf;
-
-	pr_emerg("Instruction Cache Error");
-
-	if (xec == 1 && TLB_ERROR(ec))
-		pr_cont(": %s TLB multimatch.\n", LL_MSG(ec));
-	else if (xec == 0) {
-		if (TLB_ERROR(ec))
-			pr_cont(": %s TLB Parity error.\n", LL_MSG(ec));
-		else if (BUS_ERROR(ec)) {
-			if (boot_cpu_data.x86 == 0xf &&
-			    (mc1_status & (1ULL << 58)))
-				pr_cont(" during system linefill.\n");
-			else
-				pr_cont(" during attempted NB data read.\n");
-		} else if (MEM_ERROR(ec)) {
-			u8 ll   = ec & 0x3;
-			u8 rrrr = (ec >> 4) & 0xf;
-
-			if (ll == 0x2)
-				pr_cont(" during a linefill from L2.\n");
-			else if (ll == 0x1) {
-
-				switch (rrrr) {
-				case 0x5:
-					pr_cont(": Parity error during "
-					       "data load.\n");
-					break;
-
-				case 0x7:
-					pr_cont(": Copyback Parity/Victim"
-						" error.\n");
-					break;
-
-				case 0x8:
-					pr_cont(": Tag Snoop error.\n");
-					break;
-
-				default:
-					goto wrong_ic_mce;
-					break;
-				}
-			}
-		} else
-			goto wrong_ic_mce;
-	} else
-		goto wrong_ic_mce;
-
-	return;
-
-wrong_ic_mce:
-	pr_warning("Corrupted IC MCE info?\n");
-}
-
-static void amd_decode_bu_mce(u64 mc2_status)
-{
-	u32 ec = mc2_status & 0xffff;
-	u32 xec = (mc2_status >> 16) & 0xf;
-
-	pr_emerg("Bus Unit Error");
-
-	if (xec == 0x1)
-		pr_cont(" in the write data buffers.\n");
-	else if (xec == 0x3)
-		pr_cont(" in the victim data buffers.\n");
-	else if (xec == 0x2 && MEM_ERROR(ec))
-		pr_cont(": %s error in the L2 cache tags.\n", RRRR_MSG(ec));
-	else if (xec == 0x0) {
-		if (TLB_ERROR(ec))
-			pr_cont(": %s error in a Page Descriptor Cache or "
-				"Guest TLB.\n", TT_MSG(ec));
-		else if (BUS_ERROR(ec))
-			pr_cont(": %s/ECC error in data read from NB: %s.\n",
-				RRRR_MSG(ec), PP_MSG(ec));
-		else if (MEM_ERROR(ec)) {
-			u8 rrrr = (ec >> 4) & 0xf;
-
-			if (rrrr >= 0x7)
-				pr_cont(": %s error during data copyback.\n",
-					RRRR_MSG(ec));
-			else if (rrrr <= 0x1)
-				pr_cont(": %s parity/ECC error during data "
-					"access from L2.\n", RRRR_MSG(ec));
-			else
-				goto wrong_bu_mce;
-		} else
-			goto wrong_bu_mce;
-	} else
-		goto wrong_bu_mce;
-
-	return;
-
-wrong_bu_mce:
-	pr_warning("Corrupted BU MCE info?\n");
-}
-
-static void amd_decode_ls_mce(u64 mc3_status)
-{
-	u32 ec  = mc3_status & 0xffff;
-	u32 xec = (mc3_status >> 16) & 0xf;
-
-	pr_emerg("Load Store Error");
-
-	if (xec == 0x0) {
-		u8 rrrr = (ec >> 4) & 0xf;
-
-		if (!BUS_ERROR(ec) || (rrrr != 0x3 && rrrr != 0x4))
-			goto wrong_ls_mce;
-
-		pr_cont(" during %s.\n", RRRR_MSG(ec));
-	}
-	return;
-
-wrong_ls_mce:
-	pr_warning("Corrupted LS MCE info?\n");
-}
-
-void amd_decode_nb_mce(int node_id, struct err_regs *regs, int handle_errors)
-{
-	u32 ec  = ERROR_CODE(regs->nbsl);
-
-	if (!handle_errors)
-		return;
-
-	/*
-	 * GART TLB error reporting is disabled by default. Bail out early.
-	 */
-	if (TLB_ERROR(ec) && !report_gart_errors)
-		return;
-
-	pr_emerg("Northbridge Error, node %d", node_id);
-
-	/*
-	 * F10h, revD can disable ErrCpu[3:0] so check that first and also the
-	 * value encoding has changed so interpret those differently
-	 */
-	if ((boot_cpu_data.x86 == 0x10) &&
-	    (boot_cpu_data.x86_model > 7)) {
-		if (regs->nbsh & K8_NBSH_ERR_CPU_VAL)
-			pr_cont(", core: %u\n", (u8)(regs->nbsh & 0xf));
-	} else {
-		u8 assoc_cpus = regs->nbsh & 0xf;
-
-		if (assoc_cpus > 0)
-			pr_cont(", core: %d", fls(assoc_cpus) - 1);
-
-		pr_cont("\n");
-	}
-
-	pr_emerg("%s.\n", EXT_ERR_MSG(regs->nbsl));
-
-	if (BUS_ERROR(ec) && nb_bus_decoder)
-		nb_bus_decoder(node_id, regs);
-}
-EXPORT_SYMBOL_GPL(amd_decode_nb_mce);
-
-static void amd_decode_fr_mce(u64 mc5_status)
-{
-	/* we have only one error signature so match all fields at once. */
-	if ((mc5_status & 0xffff) == 0x0f0f)
-		pr_emerg(" FR Error: CPU Watchdog timer expire.\n");
-	else
-		pr_warning("Corrupted FR MCE info?\n");
-}
-
-static inline void amd_decode_err_code(unsigned int ec)
-{
-	if (TLB_ERROR(ec)) {
-		pr_emerg("Transaction: %s, Cache Level %s\n",
-			 TT_MSG(ec), LL_MSG(ec));
-	} else if (MEM_ERROR(ec)) {
-		pr_emerg("Transaction: %s, Type: %s, Cache Level: %s",
-			 RRRR_MSG(ec), TT_MSG(ec), LL_MSG(ec));
-	} else if (BUS_ERROR(ec)) {
-		pr_emerg("Transaction type: %s(%s), %s, Cache Level: %s, "
-			 "Participating Processor: %s\n",
-			  RRRR_MSG(ec), II_MSG(ec), TO_MSG(ec), LL_MSG(ec),
-			  PP_MSG(ec));
-	} else
-		pr_warning("Huh? Unknown MCE error 0x%x\n", ec);
-}
-
-static int amd_decode_mce(struct notifier_block *nb, unsigned long val,
-			   void *data)
-{
-	struct mce *m = (struct mce *)data;
-	struct err_regs regs;
-	int node, ecc;
-
-	pr_emerg("MC%d_STATUS: ", m->bank);
-
-	pr_cont("%sorrected error, other errors lost: %s, "
-		 "CPU context corrupt: %s",
-		 ((m->status & MCI_STATUS_UC) ? "Unc"  : "C"),
-		 ((m->status & MCI_STATUS_OVER) ? "yes"  : "no"),
-		 ((m->status & MCI_STATUS_PCC) ? "yes" : "no"));
-
-	/* do the two bits[14:13] together */
-	ecc = (m->status >> 45) & 0x3;
-	if (ecc)
-		pr_cont(", %sECC Error", ((ecc == 2) ? "C" : "U"));
-
-	pr_cont("\n");
-
-	switch (m->bank) {
-	case 0:
-		amd_decode_dc_mce(m->status);
-		break;
-
-	case 1:
-		amd_decode_ic_mce(m->status);
-		break;
-
-	case 2:
-		amd_decode_bu_mce(m->status);
-		break;
-
-	case 3:
-		amd_decode_ls_mce(m->status);
-		break;
-
-	case 4:
-		regs.nbsl  = (u32) m->status;
-		regs.nbsh  = (u32)(m->status >> 32);
-		regs.nbeal = (u32) m->addr;
-		regs.nbeah = (u32)(m->addr >> 32);
-		node       = amd_get_nb_id(m->extcpu);
-
-		amd_decode_nb_mce(node, &regs, 1);
-		break;
-
-	case 5:
-		amd_decode_fr_mce(m->status);
-		break;
-
-	default:
-		break;
-	}
-
-	amd_decode_err_code(m->status & 0xffff);
-
-	return NOTIFY_STOP;
-}
-
-static struct notifier_block amd_mce_dec_nb = {
-	.notifier_call	= amd_decode_mce,
-};
-
-static int __init mce_amd_init(void)
-{
-	/*
-	 * We can decode MCEs for K8, F10h and F11h CPUs:
-	 */
-	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
-		return 0;
-
-	if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
-		return 0;
-
-	atomic_notifier_chain_register(&x86_mce_decoder_chain, &amd_mce_dec_nb);
-
-	return 0;
-}
-early_initcall(mce_amd_init);
-
-#ifdef MODULE
-static void __exit mce_amd_exit(void)
-{
-	atomic_notifier_chain_unregister(&x86_mce_decoder_chain, &amd_mce_dec_nb);
-}
-
-MODULE_DESCRIPTION("AMD MCE decoder");
-MODULE_ALIAS("edac-mce-amd");
-MODULE_LICENSE("GPL");
-module_exit(mce_amd_exit);
-#endif
diff --git a/drivers/edac/edac_module.c b/drivers/edac/edac_module.c
index 7e1374a..be4b075 100644
--- a/drivers/edac/edac_module.c
+++ b/drivers/edac/edac_module.c
@@ -27,15 +27,6 @@
 struct workqueue_struct *edac_workqueue;
 
 /*
- * sysfs object: /sys/devices/system/edac
- *	need to export to other files in this modules
- */
-static struct sysdev_class edac_class = {
-	.name = "edac",
-};
-static int edac_class_valid;
-
-/*
  * edac_op_state_to_string()
  */
 char *edac_op_state_to_string(int opstate)
@@ -55,60 +46,6 @@
 }
 
 /*
- * edac_get_edac_class()
- *
- *	return pointer to the edac class of 'edac'
- */
-struct sysdev_class *edac_get_edac_class(void)
-{
-	struct sysdev_class *classptr = NULL;
-
-	if (edac_class_valid)
-		classptr = &edac_class;
-
-	return classptr;
-}
-
-/*
- * edac_register_sysfs_edac_name()
- *
- *	register the 'edac' into /sys/devices/system
- *
- * return:
- *	0  success
- *	!0 error
- */
-static int edac_register_sysfs_edac_name(void)
-{
-	int err;
-
-	/* create the /sys/devices/system/edac directory */
-	err = sysdev_class_register(&edac_class);
-
-	if (err) {
-		debugf1("%s() error=%d\n", __func__, err);
-		return err;
-	}
-
-	edac_class_valid = 1;
-	return 0;
-}
-
-/*
- * sysdev_class_unregister()
- *
- *	unregister the 'edac' from /sys/devices/system
- */
-static void edac_unregister_sysfs_edac_name(void)
-{
-	/* only if currently registered, then unregister it */
-	if (edac_class_valid)
-		sysdev_class_unregister(&edac_class);
-
-	edac_class_valid = 0;
-}
-
-/*
  * edac_workqueue_setup
  *	initialize the edac work queue for polling operations
  */
@@ -154,21 +91,11 @@
 	edac_pci_clear_parity_errors();
 
 	/*
-	 * perform the registration of the /sys/devices/system/edac class object
-	 */
-	if (edac_register_sysfs_edac_name()) {
-		edac_printk(KERN_ERR, EDAC_MC,
-			"Error initializing 'edac' kobject\n");
-		err = -ENODEV;
-		goto error;
-	}
-
-	/*
 	 * now set up the mc_kset under the edac class object
 	 */
 	err = edac_sysfs_setup_mc_kset();
 	if (err)
-		goto sysfs_setup_fail;
+		goto error;
 
 	/* Setup/Initialize the workq for this core */
 	err = edac_workqueue_setup();
@@ -183,9 +110,6 @@
 workq_fail:
 	edac_sysfs_teardown_mc_kset();
 
-sysfs_setup_fail:
-	edac_unregister_sysfs_edac_name();
-
 error:
 	return err;
 }
@@ -201,7 +125,6 @@
 	/* tear down the various subsystems */
 	edac_workqueue_teardown();
 	edac_sysfs_teardown_mc_kset();
-	edac_unregister_sysfs_edac_name();
 }
 
 /*
diff --git a/drivers/edac/edac_module.h b/drivers/edac/edac_module.h
index 233d479..17aabb7 100644
--- a/drivers/edac/edac_module.h
+++ b/drivers/edac/edac_module.h
@@ -42,7 +42,6 @@
 				struct edac_device_ctl_info *edac_dev);
 extern int edac_device_create_sysfs(struct edac_device_ctl_info *edac_dev);
 extern void edac_device_remove_sysfs(struct edac_device_ctl_info *edac_dev);
-extern struct sysdev_class *edac_get_edac_class(void);
 
 /* edac core workqueue: single CPU mode */
 extern struct workqueue_struct *edac_workqueue;
diff --git a/drivers/edac/edac_pci_sysfs.c b/drivers/edac/edac_pci_sysfs.c
index c39697d..023b01c 100644
--- a/drivers/edac/edac_pci_sysfs.c
+++ b/drivers/edac/edac_pci_sysfs.c
@@ -7,7 +7,7 @@
  *
  */
 #include <linux/module.h>
-#include <linux/sysdev.h>
+#include <linux/edac.h>
 #include <linux/slab.h>
 #include <linux/ctype.h>
 
@@ -354,7 +354,7 @@
 	/* First time, so create the main kobject and its
 	 * controls and atributes
 	 */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class\n", __func__);
 		err = -ENODEV;
@@ -368,7 +368,7 @@
 	if (!try_module_get(THIS_MODULE)) {
 		debugf1("%s() try_module_get() failed\n", __func__);
 		err = -ENODEV;
-		goto decrement_count_fail;
+		goto mod_get_fail;
 	}
 
 	edac_pci_top_main_kobj = kzalloc(sizeof(struct kobject), GFP_KERNEL);
@@ -403,6 +403,9 @@
 kzalloc_fail:
 	module_put(THIS_MODULE);
 
+mod_get_fail:
+	edac_put_sysfs_class();
+
 decrement_count_fail:
 	/* if are on this error exit, nothing to tear down */
 	atomic_dec(&edac_pci_sysfs_refcount);
@@ -429,6 +432,7 @@
 			__func__);
 		kobject_put(edac_pci_top_main_kobj);
 	}
+	edac_put_sysfs_class();
 }
 
 /*
diff --git a/drivers/edac/edac_stub.c b/drivers/edac/edac_stub.c
index 20b428a..aab9707 100644
--- a/drivers/edac/edac_stub.c
+++ b/drivers/edac/edac_stub.c
@@ -3,10 +3,13 @@
  *
  * Author: Dave Jiang <djiang@mvista.com>
  *
- * 2007 (c) MontaVista Software, Inc. This file is licensed under
- * the terms of the GNU General Public License version 2. This program
- * is licensed "as is" without any warranty of any kind, whether express
- * or implied.
+ * 2007 (c) MontaVista Software, Inc.
+ * 2010 (c) Advanced Micro Devices Inc.
+ *	    Borislav Petkov <borislav.petkov@amd.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
  *
  */
 #include <linux/module.h>
@@ -23,6 +26,8 @@
 int edac_err_assert = 0;
 EXPORT_SYMBOL_GPL(edac_err_assert);
 
+static atomic_t edac_class_valid = ATOMIC_INIT(0);
+
 /*
  * called to determine if there is an EDAC driver interested in
  * knowing an event (such as NMI) occurred
@@ -44,3 +49,41 @@
 	edac_err_assert++;
 }
 EXPORT_SYMBOL_GPL(edac_atomic_assert_error);
+
+/*
+ * sysfs object: /sys/devices/system/edac
+ *	need to export to other files
+ */
+struct sysdev_class edac_class = {
+	.name = "edac",
+};
+EXPORT_SYMBOL_GPL(edac_class);
+
+/* return pointer to the 'edac' node in sysfs */
+struct sysdev_class *edac_get_sysfs_class(void)
+{
+	int err = 0;
+
+	if (atomic_read(&edac_class_valid))
+		goto out;
+
+	/* create the /sys/devices/system/edac directory */
+	err = sysdev_class_register(&edac_class);
+	if (err) {
+		printk(KERN_ERR "Error registering toplevel EDAC sysfs dir\n");
+		return NULL;
+	}
+
+out:
+	atomic_inc(&edac_class_valid);
+	return &edac_class;
+}
+EXPORT_SYMBOL_GPL(edac_get_sysfs_class);
+
+void edac_put_sysfs_class(void)
+{
+	/* last user unregisters it */
+	if (atomic_dec_and_test(&edac_class_valid))
+		sysdev_class_unregister(&edac_class);
+}
+EXPORT_SYMBOL_GPL(edac_put_sysfs_class);
diff --git a/drivers/edac/mce_amd.c b/drivers/edac/mce_amd.c
new file mode 100644
index 0000000..c018109
--- /dev/null
+++ b/drivers/edac/mce_amd.c
@@ -0,0 +1,680 @@
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include "mce_amd.h"
+
+static struct amd_decoder_ops *fam_ops;
+
+static u8 nb_err_cpumask = 0xf;
+
+static bool report_gart_errors;
+static void (*nb_bus_decoder)(int node_id, struct mce *m, u32 nbcfg);
+
+void amd_report_gart_errors(bool v)
+{
+	report_gart_errors = v;
+}
+EXPORT_SYMBOL_GPL(amd_report_gart_errors);
+
+void amd_register_ecc_decoder(void (*f)(int, struct mce *, u32))
+{
+	nb_bus_decoder = f;
+}
+EXPORT_SYMBOL_GPL(amd_register_ecc_decoder);
+
+void amd_unregister_ecc_decoder(void (*f)(int, struct mce *, u32))
+{
+	if (nb_bus_decoder) {
+		WARN_ON(nb_bus_decoder != f);
+
+		nb_bus_decoder = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(amd_unregister_ecc_decoder);
+
+/*
+ * string representation for the different MCA reported error types, see F3x48
+ * or MSR0000_0411.
+ */
+
+/* transaction type */
+const char *tt_msgs[] = { "INSN", "DATA", "GEN", "RESV" };
+EXPORT_SYMBOL_GPL(tt_msgs);
+
+/* cache level */
+const char *ll_msgs[] = { "RESV", "L1", "L2", "L3/GEN" };
+EXPORT_SYMBOL_GPL(ll_msgs);
+
+/* memory transaction type */
+const char *rrrr_msgs[] = {
+       "GEN", "RD", "WR", "DRD", "DWR", "IRD", "PRF", "EV", "SNP"
+};
+EXPORT_SYMBOL_GPL(rrrr_msgs);
+
+/* participating processor */
+const char *pp_msgs[] = { "SRC", "RES", "OBS", "GEN" };
+EXPORT_SYMBOL_GPL(pp_msgs);
+
+/* request timeout */
+const char *to_msgs[] = { "no timeout",	"timed out" };
+EXPORT_SYMBOL_GPL(to_msgs);
+
+/* memory or i/o */
+const char *ii_msgs[] = { "MEM", "RESV", "IO", "GEN" };
+EXPORT_SYMBOL_GPL(ii_msgs);
+
+static const char *f10h_nb_mce_desc[] = {
+	"HT link data error",
+	"Protocol error (link, L3, probe filter, etc.)",
+	"Parity error in NB-internal arrays",
+	"Link Retry due to IO link transmission error",
+	"L3 ECC data cache error",
+	"ECC error in L3 cache tag",
+	"L3 LRU parity bits error",
+	"ECC Error in the Probe Filter directory"
+};
+
+static bool f12h_dc_mce(u16 ec)
+{
+	bool ret = false;
+
+	if (MEM_ERROR(ec)) {
+		u8 ll = ec & 0x3;
+		ret = true;
+
+		if (ll == LL_L2)
+			pr_cont("during L1 linefill from L2.\n");
+		else if (ll == LL_L1)
+			pr_cont("Data/Tag %s error.\n", RRRR_MSG(ec));
+		else
+			ret = false;
+	}
+	return ret;
+}
+
+static bool f10h_dc_mce(u16 ec)
+{
+	u8 r4  = (ec >> 4) & 0xf;
+	u8 ll  = ec & 0x3;
+
+	if (r4 == R4_GEN && ll == LL_L1) {
+		pr_cont("during data scrub.\n");
+		return true;
+	}
+	return f12h_dc_mce(ec);
+}
+
+static bool k8_dc_mce(u16 ec)
+{
+	if (BUS_ERROR(ec)) {
+		pr_cont("during system linefill.\n");
+		return true;
+	}
+
+	return f10h_dc_mce(ec);
+}
+
+static bool f14h_dc_mce(u16 ec)
+{
+	u8 r4	 = (ec >> 4) & 0xf;
+	u8 ll	 = ec & 0x3;
+	u8 tt	 = (ec >> 2) & 0x3;
+	u8 ii	 = tt;
+	bool ret = true;
+
+	if (MEM_ERROR(ec)) {
+
+		if (tt != TT_DATA || ll != LL_L1)
+			return false;
+
+		switch (r4) {
+		case R4_DRD:
+		case R4_DWR:
+			pr_cont("Data/Tag parity error due to %s.\n",
+				(r4 == R4_DRD ? "load/hw prf" : "store"));
+			break;
+		case R4_EVICT:
+			pr_cont("Copyback parity error on a tag miss.\n");
+			break;
+		case R4_SNOOP:
+			pr_cont("Tag parity error during snoop.\n");
+			break;
+		default:
+			ret = false;
+		}
+	} else if (BUS_ERROR(ec)) {
+
+		if ((ii != II_MEM && ii != II_IO) || ll != LL_LG)
+			return false;
+
+		pr_cont("System read data error on a ");
+
+		switch (r4) {
+		case R4_RD:
+			pr_cont("TLB reload.\n");
+			break;
+		case R4_DWR:
+			pr_cont("store.\n");
+			break;
+		case R4_DRD:
+			pr_cont("load.\n");
+			break;
+		default:
+			ret = false;
+		}
+	} else {
+		ret = false;
+	}
+
+	return ret;
+}
+
+static void amd_decode_dc_mce(struct mce *m)
+{
+	u16 ec = m->status & 0xffff;
+	u8 xec = (m->status >> 16) & 0xf;
+
+	pr_emerg(HW_ERR "Data Cache Error: ");
+
+	/* TLB error signatures are the same across families */
+	if (TLB_ERROR(ec)) {
+		u8 tt = (ec >> 2) & 0x3;
+
+		if (tt == TT_DATA) {
+			pr_cont("%s TLB %s.\n", LL_MSG(ec),
+				(xec ? "multimatch" : "parity error"));
+			return;
+		}
+		else
+			goto wrong_dc_mce;
+	}
+
+	if (!fam_ops->dc_mce(ec))
+		goto wrong_dc_mce;
+
+	return;
+
+wrong_dc_mce:
+	pr_emerg(HW_ERR "Corrupted DC MCE info?\n");
+}
+
+static bool k8_ic_mce(u16 ec)
+{
+	u8 ll	 = ec & 0x3;
+	u8 r4	 = (ec >> 4) & 0xf;
+	bool ret = true;
+
+	if (!MEM_ERROR(ec))
+		return false;
+
+	if (ll == 0x2)
+		pr_cont("during a linefill from L2.\n");
+	else if (ll == 0x1) {
+		switch (r4) {
+		case R4_IRD:
+			pr_cont("Parity error during data load.\n");
+			break;
+
+		case R4_EVICT:
+			pr_cont("Copyback Parity/Victim error.\n");
+			break;
+
+		case R4_SNOOP:
+			pr_cont("Tag Snoop error.\n");
+			break;
+
+		default:
+			ret = false;
+			break;
+		}
+	} else
+		ret = false;
+
+	return ret;
+}
+
+static bool f14h_ic_mce(u16 ec)
+{
+	u8 ll    = ec & 0x3;
+	u8 tt    = (ec >> 2) & 0x3;
+	u8 r4  = (ec >> 4) & 0xf;
+	bool ret = true;
+
+	if (MEM_ERROR(ec)) {
+		if (tt != 0 || ll != 1)
+			ret = false;
+
+		if (r4 == R4_IRD)
+			pr_cont("Data/tag array parity error for a tag hit.\n");
+		else if (r4 == R4_SNOOP)
+			pr_cont("Tag error during snoop/victimization.\n");
+		else
+			ret = false;
+	}
+	return ret;
+}
+
+static void amd_decode_ic_mce(struct mce *m)
+{
+	u16 ec = m->status & 0xffff;
+	u8 xec = (m->status >> 16) & 0xf;
+
+	pr_emerg(HW_ERR "Instruction Cache Error: ");
+
+	if (TLB_ERROR(ec))
+		pr_cont("%s TLB %s.\n", LL_MSG(ec),
+			(xec ? "multimatch" : "parity error"));
+	else if (BUS_ERROR(ec)) {
+		bool k8 = (boot_cpu_data.x86 == 0xf && (m->status & BIT_64(58)));
+
+		pr_cont("during %s.\n", (k8 ? "system linefill" : "NB data read"));
+	} else if (fam_ops->ic_mce(ec))
+		;
+	else
+		pr_emerg(HW_ERR "Corrupted IC MCE info?\n");
+}
+
+static void amd_decode_bu_mce(struct mce *m)
+{
+	u32 ec = m->status & 0xffff;
+	u32 xec = (m->status >> 16) & 0xf;
+
+	pr_emerg(HW_ERR "Bus Unit Error");
+
+	if (xec == 0x1)
+		pr_cont(" in the write data buffers.\n");
+	else if (xec == 0x3)
+		pr_cont(" in the victim data buffers.\n");
+	else if (xec == 0x2 && MEM_ERROR(ec))
+		pr_cont(": %s error in the L2 cache tags.\n", RRRR_MSG(ec));
+	else if (xec == 0x0) {
+		if (TLB_ERROR(ec))
+			pr_cont(": %s error in a Page Descriptor Cache or "
+				"Guest TLB.\n", TT_MSG(ec));
+		else if (BUS_ERROR(ec))
+			pr_cont(": %s/ECC error in data read from NB: %s.\n",
+				RRRR_MSG(ec), PP_MSG(ec));
+		else if (MEM_ERROR(ec)) {
+			u8 rrrr = (ec >> 4) & 0xf;
+
+			if (rrrr >= 0x7)
+				pr_cont(": %s error during data copyback.\n",
+					RRRR_MSG(ec));
+			else if (rrrr <= 0x1)
+				pr_cont(": %s parity/ECC error during data "
+					"access from L2.\n", RRRR_MSG(ec));
+			else
+				goto wrong_bu_mce;
+		} else
+			goto wrong_bu_mce;
+	} else
+		goto wrong_bu_mce;
+
+	return;
+
+wrong_bu_mce:
+	pr_emerg(HW_ERR "Corrupted BU MCE info?\n");
+}
+
+static void amd_decode_ls_mce(struct mce *m)
+{
+	u16 ec = m->status & 0xffff;
+	u8 xec = (m->status >> 16) & 0xf;
+
+	if (boot_cpu_data.x86 == 0x14) {
+		pr_emerg("You shouldn't be seeing an LS MCE on this cpu family,"
+			 " please report on LKML.\n");
+		return;
+	}
+
+	pr_emerg(HW_ERR "Load Store Error");
+
+	if (xec == 0x0) {
+		u8 r4 = (ec >> 4) & 0xf;
+
+		if (!BUS_ERROR(ec) || (r4 != R4_DRD && r4 != R4_DWR))
+			goto wrong_ls_mce;
+
+		pr_cont(" during %s.\n", RRRR_MSG(ec));
+	} else
+		goto wrong_ls_mce;
+
+	return;
+
+wrong_ls_mce:
+	pr_emerg(HW_ERR "Corrupted LS MCE info?\n");
+}
+
+static bool k8_nb_mce(u16 ec, u8 xec)
+{
+	bool ret = true;
+
+	switch (xec) {
+	case 0x1:
+		pr_cont("CRC error detected on HT link.\n");
+		break;
+
+	case 0x5:
+		pr_cont("Invalid GART PTE entry during GART table walk.\n");
+		break;
+
+	case 0x6:
+		pr_cont("Unsupported atomic RMW received from an IO link.\n");
+		break;
+
+	case 0x0:
+	case 0x8:
+		if (boot_cpu_data.x86 == 0x11)
+			return false;
+
+		pr_cont("DRAM ECC error detected on the NB.\n");
+		break;
+
+	case 0xd:
+		pr_cont("Parity error on the DRAM addr/ctl signals.\n");
+		break;
+
+	default:
+		ret = false;
+		break;
+	}
+
+	return ret;
+}
+
+static bool f10h_nb_mce(u16 ec, u8 xec)
+{
+	bool ret = true;
+	u8 offset = 0;
+
+	if (k8_nb_mce(ec, xec))
+		return true;
+
+	switch(xec) {
+	case 0xa ... 0xc:
+		offset = 10;
+		break;
+
+	case 0xe:
+		offset = 11;
+		break;
+
+	case 0xf:
+		if (TLB_ERROR(ec))
+			pr_cont("GART Table Walk data error.\n");
+		else if (BUS_ERROR(ec))
+			pr_cont("DMA Exclusion Vector Table Walk error.\n");
+		else
+			ret = false;
+
+		goto out;
+		break;
+
+	case 0x1c ... 0x1f:
+		offset = 24;
+		break;
+
+	default:
+		ret = false;
+
+		goto out;
+		break;
+	}
+
+	pr_cont("%s.\n", f10h_nb_mce_desc[xec - offset]);
+
+out:
+	return ret;
+}
+
+static bool nb_noop_mce(u16 ec, u8 xec)
+{
+	return false;
+}
+
+void amd_decode_nb_mce(int node_id, struct mce *m, u32 nbcfg)
+{
+	u8 xec   = (m->status >> 16) & 0x1f;
+	u16 ec   = m->status & 0xffff;
+	u32 nbsh = (u32)(m->status >> 32);
+
+	pr_emerg(HW_ERR "Northbridge Error, node %d: ", node_id);
+
+	/*
+	 * F10h, revD can disable ErrCpu[3:0] so check that first and also the
+	 * value encoding has changed so interpret those differently
+	 */
+	if ((boot_cpu_data.x86 == 0x10) &&
+	    (boot_cpu_data.x86_model > 7)) {
+		if (nbsh & K8_NBSH_ERR_CPU_VAL)
+			pr_cont(", core: %u", (u8)(nbsh & nb_err_cpumask));
+	} else {
+		u8 assoc_cpus = nbsh & nb_err_cpumask;
+
+		if (assoc_cpus > 0)
+			pr_cont(", core: %d", fls(assoc_cpus) - 1);
+	}
+
+	switch (xec) {
+	case 0x2:
+		pr_cont("Sync error (sync packets on HT link detected).\n");
+		return;
+
+	case 0x3:
+		pr_cont("HT Master abort.\n");
+		return;
+
+	case 0x4:
+		pr_cont("HT Target abort.\n");
+		return;
+
+	case 0x7:
+		pr_cont("NB Watchdog timeout.\n");
+		return;
+
+	case 0x9:
+		pr_cont("SVM DMA Exclusion Vector error.\n");
+		return;
+
+	default:
+		break;
+	}
+
+	if (!fam_ops->nb_mce(ec, xec))
+		goto wrong_nb_mce;
+
+	if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10)
+		if ((xec == 0x8 || xec == 0x0) && nb_bus_decoder)
+			nb_bus_decoder(node_id, m, nbcfg);
+
+	return;
+
+wrong_nb_mce:
+	pr_emerg(HW_ERR "Corrupted NB MCE info?\n");
+}
+EXPORT_SYMBOL_GPL(amd_decode_nb_mce);
+
+static void amd_decode_fr_mce(struct mce *m)
+{
+	if (boot_cpu_data.x86 == 0xf ||
+	    boot_cpu_data.x86 == 0x11)
+		goto wrong_fr_mce;
+
+	/* we have only one error signature so match all fields at once. */
+	if ((m->status & 0xffff) == 0x0f0f) {
+		pr_emerg(HW_ERR "FR Error: CPU Watchdog timer expire.\n");
+		return;
+	}
+
+wrong_fr_mce:
+	pr_emerg(HW_ERR "Corrupted FR MCE info?\n");
+}
+
+static inline void amd_decode_err_code(u16 ec)
+{
+	if (TLB_ERROR(ec)) {
+		pr_emerg(HW_ERR "Transaction: %s, Cache Level: %s\n",
+			 TT_MSG(ec), LL_MSG(ec));
+	} else if (MEM_ERROR(ec)) {
+		pr_emerg(HW_ERR "Transaction: %s, Type: %s, Cache Level: %s\n",
+			 RRRR_MSG(ec), TT_MSG(ec), LL_MSG(ec));
+	} else if (BUS_ERROR(ec)) {
+		pr_emerg(HW_ERR "Transaction: %s (%s), %s, Cache Level: %s, "
+			 "Participating Processor: %s\n",
+			  RRRR_MSG(ec), II_MSG(ec), TO_MSG(ec), LL_MSG(ec),
+			  PP_MSG(ec));
+	} else
+		pr_emerg(HW_ERR "Huh? Unknown MCE error 0x%x\n", ec);
+}
+
+/*
+ * Filter out unwanted MCE signatures here.
+ */
+static bool amd_filter_mce(struct mce *m)
+{
+	u8 xec = (m->status >> 16) & 0x1f;
+
+	/*
+	 * NB GART TLB error reporting is disabled by default.
+	 */
+	if (m->bank == 4 && xec == 0x5 && !report_gart_errors)
+		return true;
+
+	return false;
+}
+
+int amd_decode_mce(struct notifier_block *nb, unsigned long val, void *data)
+{
+	struct mce *m = (struct mce *)data;
+	int node, ecc;
+
+	if (amd_filter_mce(m))
+		return NOTIFY_STOP;
+
+	pr_emerg(HW_ERR "MC%d_STATUS: ", m->bank);
+
+	pr_cont("%sorrected error, other errors lost: %s, "
+		 "CPU context corrupt: %s",
+		 ((m->status & MCI_STATUS_UC) ? "Unc"  : "C"),
+		 ((m->status & MCI_STATUS_OVER) ? "yes"  : "no"),
+		 ((m->status & MCI_STATUS_PCC) ? "yes" : "no"));
+
+	/* do the two bits[14:13] together */
+	ecc = (m->status >> 45) & 0x3;
+	if (ecc)
+		pr_cont(", %sECC Error", ((ecc == 2) ? "C" : "U"));
+
+	pr_cont("\n");
+
+	switch (m->bank) {
+	case 0:
+		amd_decode_dc_mce(m);
+		break;
+
+	case 1:
+		amd_decode_ic_mce(m);
+		break;
+
+	case 2:
+		amd_decode_bu_mce(m);
+		break;
+
+	case 3:
+		amd_decode_ls_mce(m);
+		break;
+
+	case 4:
+		node = amd_get_nb_id(m->extcpu);
+		amd_decode_nb_mce(node, m, 0);
+		break;
+
+	case 5:
+		amd_decode_fr_mce(m);
+		break;
+
+	default:
+		break;
+	}
+
+	amd_decode_err_code(m->status & 0xffff);
+
+	return NOTIFY_STOP;
+}
+EXPORT_SYMBOL_GPL(amd_decode_mce);
+
+static struct notifier_block amd_mce_dec_nb = {
+	.notifier_call	= amd_decode_mce,
+};
+
+static int __init mce_amd_init(void)
+{
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
+		return 0;
+
+	if ((boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x12) &&
+	    (boot_cpu_data.x86 != 0x14 || boot_cpu_data.x86_model > 0xf))
+		return 0;
+
+	fam_ops = kzalloc(sizeof(struct amd_decoder_ops), GFP_KERNEL);
+	if (!fam_ops)
+		return -ENOMEM;
+
+	switch (boot_cpu_data.x86) {
+	case 0xf:
+		fam_ops->dc_mce = k8_dc_mce;
+		fam_ops->ic_mce = k8_ic_mce;
+		fam_ops->nb_mce = k8_nb_mce;
+		break;
+
+	case 0x10:
+		fam_ops->dc_mce = f10h_dc_mce;
+		fam_ops->ic_mce = k8_ic_mce;
+		fam_ops->nb_mce = f10h_nb_mce;
+		break;
+
+	case 0x11:
+		fam_ops->dc_mce = k8_dc_mce;
+		fam_ops->ic_mce = k8_ic_mce;
+		fam_ops->nb_mce = f10h_nb_mce;
+		break;
+
+	case 0x12:
+		fam_ops->dc_mce = f12h_dc_mce;
+		fam_ops->ic_mce = k8_ic_mce;
+		fam_ops->nb_mce = nb_noop_mce;
+		break;
+
+	case 0x14:
+		nb_err_cpumask  = 0x3;
+		fam_ops->dc_mce = f14h_dc_mce;
+		fam_ops->ic_mce = f14h_ic_mce;
+		fam_ops->nb_mce = nb_noop_mce;
+		break;
+
+	default:
+		printk(KERN_WARNING "Huh? What family is that: %d?!\n",
+				    boot_cpu_data.x86);
+		kfree(fam_ops);
+		return -EINVAL;
+	}
+
+	pr_info("MCE: In-kernel MCE decoding enabled.\n");
+
+	atomic_notifier_chain_register(&x86_mce_decoder_chain, &amd_mce_dec_nb);
+
+	return 0;
+}
+early_initcall(mce_amd_init);
+
+#ifdef MODULE
+static void __exit mce_amd_exit(void)
+{
+	atomic_notifier_chain_unregister(&x86_mce_decoder_chain, &amd_mce_dec_nb);
+	kfree(fam_ops);
+}
+
+MODULE_DESCRIPTION("AMD MCE decoder");
+MODULE_ALIAS("edac-mce-amd");
+MODULE_LICENSE("GPL");
+module_exit(mce_amd_exit);
+#endif
diff --git a/drivers/edac/edac_mce_amd.h b/drivers/edac/mce_amd.h
similarity index 65%
rename from drivers/edac/edac_mce_amd.h
rename to drivers/edac/mce_amd.h
index df23ee0..35f6e0e 100644
--- a/drivers/edac/edac_mce_amd.h
+++ b/drivers/edac/mce_amd.h
@@ -1,11 +1,14 @@
 #ifndef _EDAC_MCE_AMD_H
 #define _EDAC_MCE_AMD_H
 
+#include <linux/notifier.h>
+
 #include <asm/mce.h>
 
+#define BIT_64(n)			(U64_C(1) << (n))
+
 #define ERROR_CODE(x)			((x) & 0xffff)
 #define EXT_ERROR_CODE(x)		(((x) >> 16) & 0x1f)
-#define EXT_ERR_MSG(x)			ext_msgs[EXT_ERROR_CODE(x)]
 
 #define LOW_SYNDROME(x)			(((x) >> 15) & 0xff)
 #define HIGH_SYNDROME(x)		(((x) >> 24) & 0xff)
@@ -20,13 +23,14 @@
 #define II_MSG(x)			ii_msgs[II(x)]
 #define LL(x)				(((x) >> 0) & 0x3)
 #define LL_MSG(x)			ll_msgs[LL(x)]
-#define RRRR(x)				(((x) >> 4) & 0xf)
-#define RRRR_MSG(x)			rrrr_msgs[RRRR(x)]
 #define TO(x)				(((x) >> 8) & 0x1)
 #define TO_MSG(x)			to_msgs[TO(x)]
 #define PP(x)				(((x) >> 9) & 0x3)
 #define PP_MSG(x)			pp_msgs[PP(x)]
 
+#define RRRR(x)				(((x) >> 4) & 0xf)
+#define RRRR_MSG(x)			((RRRR(x) < 9) ?  rrrr_msgs[RRRR(x)] : "Wrong R4!")
+
 #define K8_NBSH				0x4C
 
 #define K8_NBSH_VALID_BIT		BIT(31)
@@ -41,13 +45,45 @@
 #define K8_NBSH_UECC			BIT(13)
 #define K8_NBSH_ERR_SCRUBER		BIT(8)
 
+enum tt_ids {
+	TT_INSTR = 0,
+	TT_DATA,
+	TT_GEN,
+	TT_RESV,
+};
+
+enum ll_ids {
+	LL_RESV = 0,
+	LL_L1,
+	LL_L2,
+	LL_LG,
+};
+
+enum ii_ids {
+	II_MEM = 0,
+	II_RESV,
+	II_IO,
+	II_GEN,
+};
+
+enum rrrr_ids {
+	R4_GEN	= 0,
+	R4_RD,
+	R4_WR,
+	R4_DRD,
+	R4_DWR,
+	R4_IRD,
+	R4_PREF,
+	R4_EVICT,
+	R4_SNOOP,
+};
+
 extern const char *tt_msgs[];
 extern const char *ll_msgs[];
 extern const char *rrrr_msgs[];
 extern const char *pp_msgs[];
 extern const char *to_msgs[];
 extern const char *ii_msgs[];
-extern const char *ext_msgs[];
 
 /*
  * relevant NB regs
@@ -60,10 +96,19 @@
 	u32 nbeal;
 };
 
+/*
+ * per-family decoder ops
+ */
+struct amd_decoder_ops {
+	bool (*dc_mce)(u16);
+	bool (*ic_mce)(u16);
+	bool (*nb_mce)(u16, u8);
+};
 
 void amd_report_gart_errors(bool);
-void amd_register_ecc_decoder(void (*f)(int, struct err_regs *));
-void amd_unregister_ecc_decoder(void (*f)(int, struct err_regs *));
-void amd_decode_nb_mce(int, struct err_regs *, int);
+void amd_register_ecc_decoder(void (*f)(int, struct mce *, u32));
+void amd_unregister_ecc_decoder(void (*f)(int, struct mce *, u32));
+void amd_decode_nb_mce(int, struct mce *, u32);
+int amd_decode_mce(struct notifier_block *nb, unsigned long val, void *data);
 
 #endif /* _EDAC_MCE_AMD_H */
diff --git a/drivers/edac/mce_amd_inj.c b/drivers/edac/mce_amd_inj.c
new file mode 100644
index 0000000..8d0688f
--- /dev/null
+++ b/drivers/edac/mce_amd_inj.c
@@ -0,0 +1,171 @@
+/*
+ * A simple MCE injection facility for testing the MCE decoding code. This
+ * driver should be built as module so that it can be loaded on production
+ * kernels for testing purposes.
+ *
+ * This file may be distributed under the terms of the GNU General Public
+ * License version 2.
+ *
+ * Copyright (c) 2010:  Borislav Petkov <borislav.petkov@amd.com>
+ *			Advanced Micro Devices Inc.
+ */
+
+#include <linux/kobject.h>
+#include <linux/sysdev.h>
+#include <linux/edac.h>
+#include <asm/mce.h>
+
+#include "mce_amd.h"
+
+struct edac_mce_attr {
+	struct attribute attr;
+	ssize_t (*show) (struct kobject *kobj, struct edac_mce_attr *attr, char *buf);
+	ssize_t (*store)(struct kobject *kobj, struct edac_mce_attr *attr,
+			 const char *buf, size_t count);
+};
+
+#define EDAC_MCE_ATTR(_name, _mode, _show, _store)			\
+static struct edac_mce_attr mce_attr_##_name = __ATTR(_name, _mode, _show, _store)
+
+static struct kobject *mce_kobj;
+
+/*
+ * Collect all the MCi_XXX settings
+ */
+static struct mce i_mce;
+
+#define MCE_INJECT_STORE(reg)						\
+static ssize_t edac_inject_##reg##_store(struct kobject *kobj,		\
+					 struct edac_mce_attr *attr,	\
+					 const char *data, size_t count)\
+{									\
+	int ret = 0;							\
+	unsigned long value;						\
+									\
+	ret = strict_strtoul(data, 16, &value);				\
+	if (ret < 0)							\
+		printk(KERN_ERR "Error writing MCE " #reg " field.\n");	\
+									\
+	i_mce.reg = value;						\
+									\
+	return count;							\
+}
+
+MCE_INJECT_STORE(status);
+MCE_INJECT_STORE(misc);
+MCE_INJECT_STORE(addr);
+
+#define MCE_INJECT_SHOW(reg)						\
+static ssize_t edac_inject_##reg##_show(struct kobject *kobj,		\
+					struct edac_mce_attr *attr,	\
+					char *buf)			\
+{									\
+	return sprintf(buf, "0x%016llx\n", i_mce.reg);			\
+}
+
+MCE_INJECT_SHOW(status);
+MCE_INJECT_SHOW(misc);
+MCE_INJECT_SHOW(addr);
+
+EDAC_MCE_ATTR(status, 0644, edac_inject_status_show, edac_inject_status_store);
+EDAC_MCE_ATTR(misc, 0644, edac_inject_misc_show, edac_inject_misc_store);
+EDAC_MCE_ATTR(addr, 0644, edac_inject_addr_show, edac_inject_addr_store);
+
+/*
+ * This denotes into which bank we're injecting and triggers
+ * the injection, at the same time.
+ */
+static ssize_t edac_inject_bank_store(struct kobject *kobj,
+				      struct edac_mce_attr *attr,
+				      const char *data, size_t count)
+{
+	int ret = 0;
+	unsigned long value;
+
+	ret = strict_strtoul(data, 10, &value);
+	if (ret < 0) {
+		printk(KERN_ERR "Invalid bank value!\n");
+		return -EINVAL;
+	}
+
+	if (value > 5) {
+		printk(KERN_ERR "Non-existant MCE bank: %lu\n", value);
+		return -EINVAL;
+	}
+
+	i_mce.bank = value;
+
+	amd_decode_mce(NULL, 0, &i_mce);
+
+	return count;
+}
+
+static ssize_t edac_inject_bank_show(struct kobject *kobj,
+				     struct edac_mce_attr *attr, char *buf)
+{
+	return sprintf(buf, "%d\n", i_mce.bank);
+}
+
+EDAC_MCE_ATTR(bank, 0644, edac_inject_bank_show, edac_inject_bank_store);
+
+static struct edac_mce_attr *sysfs_attrs[] = { &mce_attr_status, &mce_attr_misc,
+					       &mce_attr_addr, &mce_attr_bank
+};
+
+static int __init edac_init_mce_inject(void)
+{
+	struct sysdev_class *edac_class = NULL;
+	int i, err = 0;
+
+	edac_class = edac_get_sysfs_class();
+	if (!edac_class)
+		return -EINVAL;
+
+	mce_kobj = kobject_create_and_add("mce", &edac_class->kset.kobj);
+	if (!mce_kobj) {
+		printk(KERN_ERR "Error creating a mce kset.\n");
+		err = -ENOMEM;
+		goto err_mce_kobj;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(sysfs_attrs); i++) {
+		err = sysfs_create_file(mce_kobj, &sysfs_attrs[i]->attr);
+		if (err) {
+			printk(KERN_ERR "Error creating %s in sysfs.\n",
+					sysfs_attrs[i]->attr.name);
+			goto err_sysfs_create;
+		}
+	}
+	return 0;
+
+err_sysfs_create:
+	while (i-- >= 0)
+		sysfs_remove_file(mce_kobj, &sysfs_attrs[i]->attr);
+
+	kobject_del(mce_kobj);
+
+err_mce_kobj:
+	edac_put_sysfs_class();
+
+	return err;
+}
+
+static void __exit edac_exit_mce_inject(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(sysfs_attrs); i++)
+		sysfs_remove_file(mce_kobj, &sysfs_attrs[i]->attr);
+
+	kobject_del(mce_kobj);
+
+	edac_put_sysfs_class();
+}
+
+module_init(edac_init_mce_inject);
+module_exit(edac_exit_mce_inject);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Borislav Petkov <borislav.petkov@amd.com>");
+MODULE_AUTHOR("AMD Inc.");
+MODULE_DESCRIPTION("MCE injection facility for testing MCE decoding");
diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index 1b05896..9dcb17d 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -2840,7 +2840,7 @@
 			       const struct pci_device_id *ent)
 {
 	struct fw_ohci *ohci;
-	u32 bus_options, max_receive, link_speed, version, link_enh;
+	u32 bus_options, max_receive, link_speed, version;
 	u64 guid;
 	int i, err, n_ir, n_it;
 	size_t size;
@@ -2894,23 +2894,6 @@
 	if (param_quirks)
 		ohci->quirks = param_quirks;
 
-	/* TI OHCI-Lynx and compatible: set recommended configuration bits. */
-	if (dev->vendor == PCI_VENDOR_ID_TI) {
-		pci_read_config_dword(dev, PCI_CFG_TI_LinkEnh, &link_enh);
-
-		/* adjust latency of ATx FIFO: use 1.7 KB threshold */
-		link_enh &= ~TI_LinkEnh_atx_thresh_mask;
-		link_enh |= TI_LinkEnh_atx_thresh_1_7K;
-
-		/* use priority arbitration for asynchronous responses */
-		link_enh |= TI_LinkEnh_enab_unfair;
-
-		/* required for aPhyEnhanceEnable to work */
-		link_enh |= TI_LinkEnh_enab_accel;
-
-		pci_write_config_dword(dev, PCI_CFG_TI_LinkEnh, link_enh);
-	}
-
 	ar_context_init(&ohci->ar_request_ctx, ohci,
 			OHCI1394_AsReqRcvContextControlSet);
 
diff --git a/drivers/firewire/ohci.h b/drivers/firewire/ohci.h
index 0e6c5a4..ef5e733 100644
--- a/drivers/firewire/ohci.h
+++ b/drivers/firewire/ohci.h
@@ -155,12 +155,4 @@
 
 #define OHCI1394_phy_tcode		0xe
 
-/* TI extensions */
-
-#define PCI_CFG_TI_LinkEnh		0xf4
-#define  TI_LinkEnh_enab_accel		0x00000002
-#define  TI_LinkEnh_enab_unfair		0x00000080
-#define  TI_LinkEnh_atx_thresh_mask	0x00003000
-#define  TI_LinkEnh_atx_thresh_1_7K	0x00001000
-
 #endif /* _FIREWIRE_OHCI_H */
diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig
index 280c9b5..88a3ae6 100644
--- a/drivers/firmware/Kconfig
+++ b/drivers/firmware/Kconfig
@@ -125,7 +125,7 @@
 config ISCSI_IBFT
 	tristate "iSCSI Boot Firmware Table Attributes module"
 	select ISCSI_BOOT_SYSFS
-	depends on ISCSI_IBFT_FIND && SCSI
+	depends on ISCSI_IBFT_FIND && SCSI && SCSI_LOWLEVEL
 	default	n
 	help
 	  This option enables support for detection and exposing of iSCSI
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index c74e4e8..2dd2c93 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -2231,6 +2231,9 @@
 	dev_priv->mchdev_lock = &mchdev_lock;
 	spin_unlock(&mchdev_lock);
 
+	/* XXX Prevent module unload due to memory corruption bugs. */
+	__module_get(THIS_MODULE);
+
 	return 0;
 
 out_workqueue_free:
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index 56ad9df..b61966c 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -238,8 +238,8 @@
 
 	drm_framebuffer_cleanup(&ifb->base);
 	if (ifb->obj) {
-		drm_gem_object_handle_unreference(ifb->obj);
 		drm_gem_object_unreference(ifb->obj);
+		ifb->obj = NULL;
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index d204771..dbd30b2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -352,7 +352,6 @@
 
 	if (nouveau_fb->nvbo) {
 		nouveau_bo_unmap(nouveau_fb->nvbo);
-		drm_gem_object_handle_unreference_unlocked(nouveau_fb->nvbo->gem);
 		drm_gem_object_unreference_unlocked(nouveau_fb->nvbo->gem);
 		nouveau_fb->nvbo = NULL;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_notifier.c b/drivers/gpu/drm/nouveau/nouveau_notifier.c
index 3c9964a..3ec181f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_notifier.c
+++ b/drivers/gpu/drm/nouveau/nouveau_notifier.c
@@ -79,7 +79,6 @@
 	mutex_lock(&dev->struct_mutex);
 	nouveau_bo_unpin(chan->notifier_bo);
 	mutex_unlock(&dev->struct_mutex);
-	drm_gem_object_handle_unreference_unlocked(chan->notifier_bo->gem);
 	drm_gem_object_unreference_unlocked(chan->notifier_bo->gem);
 	drm_mm_takedown(&chan->notifier_heap);
 }
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 79082d4..2f93d46 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1137,7 +1137,7 @@
 
 		WREG32(RCU_IND_INDEX, 0x203);
 		efuse_straps_3 = RREG32(RCU_IND_DATA);
-		efuse_box_bit_127_124 = (u8)(efuse_straps_3 & 0xF0000000) >> 28;
+		efuse_box_bit_127_124 = (u8)((efuse_straps_3 & 0xF0000000) >> 28);
 
 		switch(efuse_box_bit_127_124) {
 		case 0x0:
@@ -1407,6 +1407,7 @@
 	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
 	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	r600_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
 
@@ -1520,7 +1521,7 @@
 {
 	u32 tmp;
 
-	WREG32(CP_INT_CNTL, 0);
+	WREG32(CP_INT_CNTL, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
 	WREG32(GRBM_INT_CNTL, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index e151f16..e594223 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1030,6 +1030,7 @@
 		return r;
 	}
 	rdev->cp.ready = true;
+	rdev->mc.active_vram_size = rdev->mc.real_vram_size;
 	return 0;
 }
 
@@ -1047,6 +1048,7 @@
 void r100_cp_disable(struct radeon_device *rdev)
 {
 	/* Disable ring */
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	rdev->cp.ready = false;
 	WREG32(RADEON_CP_CSQ_MODE, 0);
 	WREG32(RADEON_CP_CSQ_CNTL, 0);
@@ -2295,6 +2297,7 @@
 	/* FIXME we don't use the second aperture yet when we could use it */
 	if (rdev->mc.visible_vram_size > rdev->mc.aper_size)
 		rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	config_aper_size = RREG32(RADEON_CONFIG_APER_SIZE);
 	if (rdev->flags & RADEON_IS_IGP) {
 		uint32_t tom;
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 7a04959..7b65e4ef 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1248,6 +1248,7 @@
 	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE);
 	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	r600_vram_gtt_location(rdev, &rdev->mc);
 
 	if (rdev->flags & RADEON_IS_IGP) {
@@ -1917,6 +1918,7 @@
  */
 void r600_cp_stop(struct radeon_device *rdev)
 {
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	WREG32(R_0086D8_CP_ME_CNTL, S_0086D8_CP_ME_HALT(1));
 }
 
@@ -2910,7 +2912,7 @@
 {
 	u32 tmp;
 
-	WREG32(CP_INT_CNTL, 0);
+	WREG32(CP_INT_CNTL, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
 	WREG32(GRBM_INT_CNTL, 0);
 	WREG32(DxMODE_INT_MASK, 0);
 	if (ASIC_IS_DCE3(rdev)) {
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index 9ceb2a1..3473c00 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -532,6 +532,7 @@
 	memcpy(ptr + rdev->r600_blit.ps_offset, r6xx_ps, r6xx_ps_size * 4);
 	radeon_bo_kunmap(rdev->r600_blit.shader_obj);
 	radeon_bo_unreserve(rdev->r600_blit.shader_obj);
+	rdev->mc.active_vram_size = rdev->mc.real_vram_size;
 	return 0;
 }
 
@@ -539,6 +540,7 @@
 {
 	int r;
 
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	if (rdev->r600_blit.shader_obj == NULL)
 		return;
 	/* If we can't reserve the bo, unref should be enough to destroy
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index a168d64..9ff38c9 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -344,6 +344,7 @@
 	 * about vram size near mc fb location */
 	u64			mc_vram_size;
 	u64			visible_vram_size;
+	u64			active_vram_size;
 	u64			gtt_size;
 	u64			gtt_start;
 	u64			gtt_end;
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 68932ba..8e43dda 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -1558,39 +1558,39 @@
 		switch (tv_info->ucTV_BootUpDefaultStandard) {
 		case ATOM_TV_NTSC:
 			tv_std = TV_STD_NTSC;
-			DRM_INFO("Default TV standard: NTSC\n");
+			DRM_DEBUG_KMS("Default TV standard: NTSC\n");
 			break;
 		case ATOM_TV_NTSCJ:
 			tv_std = TV_STD_NTSC_J;
-			DRM_INFO("Default TV standard: NTSC-J\n");
+			DRM_DEBUG_KMS("Default TV standard: NTSC-J\n");
 			break;
 		case ATOM_TV_PAL:
 			tv_std = TV_STD_PAL;
-			DRM_INFO("Default TV standard: PAL\n");
+			DRM_DEBUG_KMS("Default TV standard: PAL\n");
 			break;
 		case ATOM_TV_PALM:
 			tv_std = TV_STD_PAL_M;
-			DRM_INFO("Default TV standard: PAL-M\n");
+			DRM_DEBUG_KMS("Default TV standard: PAL-M\n");
 			break;
 		case ATOM_TV_PALN:
 			tv_std = TV_STD_PAL_N;
-			DRM_INFO("Default TV standard: PAL-N\n");
+			DRM_DEBUG_KMS("Default TV standard: PAL-N\n");
 			break;
 		case ATOM_TV_PALCN:
 			tv_std = TV_STD_PAL_CN;
-			DRM_INFO("Default TV standard: PAL-CN\n");
+			DRM_DEBUG_KMS("Default TV standard: PAL-CN\n");
 			break;
 		case ATOM_TV_PAL60:
 			tv_std = TV_STD_PAL_60;
-			DRM_INFO("Default TV standard: PAL-60\n");
+			DRM_DEBUG_KMS("Default TV standard: PAL-60\n");
 			break;
 		case ATOM_TV_SECAM:
 			tv_std = TV_STD_SECAM;
-			DRM_INFO("Default TV standard: SECAM\n");
+			DRM_DEBUG_KMS("Default TV standard: SECAM\n");
 			break;
 		default:
 			tv_std = TV_STD_NTSC;
-			DRM_INFO("Unknown TV standard; defaulting to NTSC\n");
+			DRM_DEBUG_KMS("Unknown TV standard; defaulting to NTSC\n");
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index a04b7a6..7b7ea26 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -913,47 +913,47 @@
 			switch (RBIOS8(tv_info + 7) & 0xf) {
 			case 1:
 				tv_std = TV_STD_NTSC;
-				DRM_INFO("Default TV standard: NTSC\n");
+				DRM_DEBUG_KMS("Default TV standard: NTSC\n");
 				break;
 			case 2:
 				tv_std = TV_STD_PAL;
-				DRM_INFO("Default TV standard: PAL\n");
+				DRM_DEBUG_KMS("Default TV standard: PAL\n");
 				break;
 			case 3:
 				tv_std = TV_STD_PAL_M;
-				DRM_INFO("Default TV standard: PAL-M\n");
+				DRM_DEBUG_KMS("Default TV standard: PAL-M\n");
 				break;
 			case 4:
 				tv_std = TV_STD_PAL_60;
-				DRM_INFO("Default TV standard: PAL-60\n");
+				DRM_DEBUG_KMS("Default TV standard: PAL-60\n");
 				break;
 			case 5:
 				tv_std = TV_STD_NTSC_J;
-				DRM_INFO("Default TV standard: NTSC-J\n");
+				DRM_DEBUG_KMS("Default TV standard: NTSC-J\n");
 				break;
 			case 6:
 				tv_std = TV_STD_SCART_PAL;
-				DRM_INFO("Default TV standard: SCART-PAL\n");
+				DRM_DEBUG_KMS("Default TV standard: SCART-PAL\n");
 				break;
 			default:
 				tv_std = TV_STD_NTSC;
-				DRM_INFO
+				DRM_DEBUG_KMS
 				    ("Unknown TV standard; defaulting to NTSC\n");
 				break;
 			}
 
 			switch ((RBIOS8(tv_info + 9) >> 2) & 0x3) {
 			case 0:
-				DRM_INFO("29.498928713 MHz TV ref clk\n");
+				DRM_DEBUG_KMS("29.498928713 MHz TV ref clk\n");
 				break;
 			case 1:
-				DRM_INFO("28.636360000 MHz TV ref clk\n");
+				DRM_DEBUG_KMS("28.636360000 MHz TV ref clk\n");
 				break;
 			case 2:
-				DRM_INFO("14.318180000 MHz TV ref clk\n");
+				DRM_DEBUG_KMS("14.318180000 MHz TV ref clk\n");
 				break;
 			case 3:
-				DRM_INFO("27.000000000 MHz TV ref clk\n");
+				DRM_DEBUG_KMS("27.000000000 MHz TV ref clk\n");
 				break;
 			default:
 				break;
@@ -1324,7 +1324,7 @@
 
 	if (tmds_info) {
 		ver = RBIOS8(tmds_info);
-		DRM_INFO("DFP table revision: %d\n", ver);
+		DRM_DEBUG_KMS("DFP table revision: %d\n", ver);
 		if (ver == 3) {
 			n = RBIOS8(tmds_info + 5) + 1;
 			if (n > 4)
@@ -1408,7 +1408,7 @@
 		offset = combios_get_table_offset(dev, COMBIOS_EXT_TMDS_INFO_TABLE);
 		if (offset) {
 			ver = RBIOS8(offset);
-			DRM_INFO("External TMDS Table revision: %d\n", ver);
+			DRM_DEBUG_KMS("External TMDS Table revision: %d\n", ver);
 			tmds->slave_addr = RBIOS8(offset + 4 + 2);
 			tmds->slave_addr >>= 1; /* 7 bit addressing */
 			gpio = RBIOS8(offset + 4 + 3);
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 5731fc9..3eef567 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -203,6 +203,7 @@
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 	struct radeon_device *rdev = crtc->dev->dev_private;
 	int xorigin = 0, yorigin = 0;
+	int w = radeon_crtc->cursor_width;
 
 	if (x < 0)
 		xorigin = -x + 1;
@@ -213,22 +214,7 @@
 	if (yorigin >= CURSOR_HEIGHT)
 		yorigin = CURSOR_HEIGHT - 1;
 
-	radeon_lock_cursor(crtc, true);
-	if (ASIC_IS_DCE4(rdev)) {
-		/* cursors are offset into the total surface */
-		x += crtc->x;
-		y += crtc->y;
-		DRM_DEBUG("x %d y %d c->x %d c->y %d\n", x, y, crtc->x, crtc->y);
-
-		/* XXX: check if evergreen has the same issues as avivo chips */
-		WREG32(EVERGREEN_CUR_POSITION + radeon_crtc->crtc_offset,
-		       ((xorigin ? 0 : x) << 16) |
-		       (yorigin ? 0 : y));
-		WREG32(EVERGREEN_CUR_HOT_SPOT + radeon_crtc->crtc_offset, (xorigin << 16) | yorigin);
-		WREG32(EVERGREEN_CUR_SIZE + radeon_crtc->crtc_offset,
-		       ((radeon_crtc->cursor_width - 1) << 16) | (radeon_crtc->cursor_height - 1));
-	} else if (ASIC_IS_AVIVO(rdev)) {
-		int w = radeon_crtc->cursor_width;
+	if (ASIC_IS_AVIVO(rdev)) {
 		int i = 0;
 		struct drm_crtc *crtc_p;
 
@@ -260,7 +246,17 @@
 			if (w <= 0)
 				w = 1;
 		}
+	}
 
+	radeon_lock_cursor(crtc, true);
+	if (ASIC_IS_DCE4(rdev)) {
+		WREG32(EVERGREEN_CUR_POSITION + radeon_crtc->crtc_offset,
+		       ((xorigin ? 0 : x) << 16) |
+		       (yorigin ? 0 : y));
+		WREG32(EVERGREEN_CUR_HOT_SPOT + radeon_crtc->crtc_offset, (xorigin << 16) | yorigin);
+		WREG32(EVERGREEN_CUR_SIZE + radeon_crtc->crtc_offset,
+		       ((w - 1) << 16) | (radeon_crtc->cursor_height - 1));
+	} else if (ASIC_IS_AVIVO(rdev)) {
 		WREG32(AVIVO_D1CUR_POSITION + radeon_crtc->crtc_offset,
 			     ((xorigin ? 0 : x) << 16) |
 			     (yorigin ? 0 : y));
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index 9cdf6a3..40b0c08 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -97,7 +97,6 @@
 		radeon_bo_unpin(rbo);
 		radeon_bo_unreserve(rbo);
 	}
-	drm_gem_object_handle_unreference(gobj);
 	drm_gem_object_unreference_unlocked(gobj);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 0afd1e6..b3b5306 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -69,7 +69,7 @@
 	u32 c = 0;
 
 	rbo->placement.fpfn = 0;
-	rbo->placement.lpfn = 0;
+	rbo->placement.lpfn = rbo->rdev->mc.active_vram_size >> PAGE_SHIFT;
 	rbo->placement.placement = rbo->placements;
 	rbo->placement.busy_placement = rbo->placements;
 	if (domain & RADEON_GEM_DOMAIN_VRAM)
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index 353998d..3481bc7 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -124,11 +124,8 @@
 	int r;
 
 	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0);
-	if (unlikely(r != 0)) {
-		if (r != -ERESTARTSYS)
-			dev_err(bo->rdev->dev, "%p reserve failed for wait\n", bo);
+	if (unlikely(r != 0))
 		return r;
-	}
 	spin_lock(&bo->tbo.lock);
 	if (mem_type)
 		*mem_type = bo->tbo.mem.mem_type;
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index cc05b23..51d5f7b 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -693,6 +693,7 @@
 	rdev->mc.real_vram_size = RREG32(RADEON_CONFIG_MEMSIZE);
 	rdev->mc.mc_vram_size = rdev->mc.real_vram_size;
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev);
 	base = RREG32_MC(R_000004_MC_FB_LOCATION);
 	base = G_000004_MC_FB_START(base) << 16;
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 3e3f757..4dc2a87 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -157,6 +157,7 @@
 	rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0);
 	rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0);
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	base = RREG32_MC(R_000100_MCCFG_FB_LOCATION);
 	base = G_000100_MC_FB_START(base) << 16;
 	rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev);
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index bfa59db..9490da7 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -267,6 +267,7 @@
  */
 void r700_cp_stop(struct radeon_device *rdev)
 {
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	WREG32(CP_ME_CNTL, (CP_ME_HALT | CP_PFP_HALT));
 }
 
@@ -992,6 +993,7 @@
 	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE);
 	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
+	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
 	r600_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index cb4cf7e..db809e0 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -442,6 +442,43 @@
 }
 
 /**
+ * Call bo::reserved and with the lru lock held.
+ * Will release GPU memory type usage on destruction.
+ * This is the place to put in driver specific hooks.
+ * Will release the bo::reserved lock and the
+ * lru lock on exit.
+ */
+
+static void ttm_bo_cleanup_memtype_use(struct ttm_buffer_object *bo)
+{
+	struct ttm_bo_global *glob = bo->glob;
+
+	if (bo->ttm) {
+
+		/**
+		 * Release the lru_lock, since we don't want to have
+		 * an atomic requirement on ttm_tt[unbind|destroy].
+		 */
+
+		spin_unlock(&glob->lru_lock);
+		ttm_tt_unbind(bo->ttm);
+		ttm_tt_destroy(bo->ttm);
+		bo->ttm = NULL;
+		spin_lock(&glob->lru_lock);
+	}
+
+	if (bo->mem.mm_node) {
+		drm_mm_put_block(bo->mem.mm_node);
+		bo->mem.mm_node = NULL;
+	}
+
+	atomic_set(&bo->reserved, 0);
+	wake_up_all(&bo->event_queue);
+	spin_unlock(&glob->lru_lock);
+}
+
+
+/**
  * If bo idle, remove from delayed- and lru lists, and unref.
  * If not idle, and already on delayed list, do nothing.
  * If not idle, and not on delayed list, put on delayed list,
@@ -456,6 +493,7 @@
 	int ret;
 
 	spin_lock(&bo->lock);
+retry:
 	(void) ttm_bo_wait(bo, false, false, !remove_all);
 
 	if (!bo->sync_obj) {
@@ -464,31 +502,52 @@
 		spin_unlock(&bo->lock);
 
 		spin_lock(&glob->lru_lock);
-		put_count = ttm_bo_del_from_lru(bo);
+		ret = ttm_bo_reserve_locked(bo, false, !remove_all, false, 0);
 
-		ret = ttm_bo_reserve_locked(bo, false, false, false, 0);
-		BUG_ON(ret);
-		if (bo->ttm)
-			ttm_tt_unbind(bo->ttm);
+		/**
+		 * Someone else has the object reserved. Bail and retry.
+		 */
+
+		if (unlikely(ret == -EBUSY)) {
+			spin_unlock(&glob->lru_lock);
+			spin_lock(&bo->lock);
+			goto requeue;
+		}
+
+		/**
+		 * We can re-check for sync object without taking
+		 * the bo::lock since setting the sync object requires
+		 * also bo::reserved. A busy object at this point may
+		 * be caused by another thread starting an accelerated
+		 * eviction.
+		 */
+
+		if (unlikely(bo->sync_obj)) {
+			atomic_set(&bo->reserved, 0);
+			wake_up_all(&bo->event_queue);
+			spin_unlock(&glob->lru_lock);
+			spin_lock(&bo->lock);
+			if (remove_all)
+				goto retry;
+			else
+				goto requeue;
+		}
+
+		put_count = ttm_bo_del_from_lru(bo);
 
 		if (!list_empty(&bo->ddestroy)) {
 			list_del_init(&bo->ddestroy);
 			++put_count;
 		}
-		if (bo->mem.mm_node) {
-			drm_mm_put_block(bo->mem.mm_node);
-			bo->mem.mm_node = NULL;
-		}
-		spin_unlock(&glob->lru_lock);
 
-		atomic_set(&bo->reserved, 0);
+		ttm_bo_cleanup_memtype_use(bo);
 
 		while (put_count--)
 			kref_put(&bo->list_kref, ttm_bo_ref_bug);
 
 		return 0;
 	}
-
+requeue:
 	spin_lock(&glob->lru_lock);
 	if (list_empty(&bo->ddestroy)) {
 		void *sync_obj = bo->sync_obj;
diff --git a/drivers/hid/hid-cando.c b/drivers/hid/hid-cando.c
index 4267a6f..5925bdc 100644
--- a/drivers/hid/hid-cando.c
+++ b/drivers/hid/hid-cando.c
@@ -237,6 +237,8 @@
 			USB_DEVICE_ID_CANDO_MULTI_TOUCH) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CANDO,
 			USB_DEVICE_ID_CANDO_MULTI_TOUCH_11_6) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_CANDO,
+		USB_DEVICE_ID_CANDO_MULTI_TOUCH_15_6) },
 	{ }
 };
 MODULE_DEVICE_TABLE(hid, cando_devices);
diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index 3f72924..a0dea3d 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1292,6 +1292,7 @@
 	{ HID_USB_DEVICE(USB_VENDOR_ID_BTC, USB_DEVICE_ID_BTC_EMPREX_REMOTE_2) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CANDO, USB_DEVICE_ID_CANDO_MULTI_TOUCH) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CANDO, USB_DEVICE_ID_CANDO_MULTI_TOUCH_11_6) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_CANDO, USB_DEVICE_ID_CANDO_MULTI_TOUCH_15_6) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHERRY, USB_DEVICE_ID_CHERRY_CYMOTION) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHERRY, USB_DEVICE_ID_CHERRY_CYMOTION_SOLAR) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHICONY, USB_DEVICE_ID_CHICONY_TACTICAL_PAD) },
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 765a4f5..c5ae5f1 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -134,6 +134,7 @@
 #define USB_VENDOR_ID_CANDO		0x2087
 #define USB_DEVICE_ID_CANDO_MULTI_TOUCH	0x0a01
 #define USB_DEVICE_ID_CANDO_MULTI_TOUCH_11_6 0x0b03
+#define USB_DEVICE_ID_CANDO_MULTI_TOUCH_15_6 0x0f01
 
 #define USB_VENDOR_ID_CH		0x068e
 #define USB_DEVICE_ID_CH_PRO_PEDALS	0x00f2
@@ -503,6 +504,7 @@
 
 #define USB_VENDOR_ID_TURBOX		0x062a
 #define USB_DEVICE_ID_TURBOX_KEYBOARD	0x0201
+#define USB_DEVICE_ID_TURBOX_TOUCHSCREEN_MOSART	0x7100
 
 #define USB_VENDOR_ID_TWINHAN		0x6253
 #define USB_DEVICE_ID_TWINHAN_IR_REMOTE	0x0100
diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c
index 47d70c5..a3866b5 100644
--- a/drivers/hid/hidraw.c
+++ b/drivers/hid/hidraw.c
@@ -109,6 +109,12 @@
 	int ret = 0;
 
 	mutex_lock(&minors_lock);
+
+	if (!hidraw_table[minor]) {
+		ret = -ENODEV;
+		goto out;
+	}
+
 	dev = hidraw_table[minor]->hid;
 
 	if (!dev->hid_output_raw_report) {
@@ -244,6 +250,10 @@
 
 	mutex_lock(&minors_lock);
 	dev = hidraw_table[minor];
+	if (!dev) {
+		ret = -ENODEV;
+		goto out;
+	}
 
 	switch (cmd) {
 		case HIDIOCGRDESCSIZE:
@@ -317,6 +327,7 @@
 
 		ret = -ENOTTY;
 	}
+out:
 	mutex_unlock(&minors_lock);
 	return ret;
 }
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 70da318..f0260c6 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -36,6 +36,7 @@
 	{ USB_VENDOR_ID_DWAV, USB_DEVICE_ID_EGALAX_TOUCHCONTROLLER, HID_QUIRK_MULTI_INPUT | HID_QUIRK_NOGET },
 	{ USB_VENDOR_ID_DWAV, USB_DEVICE_ID_DWAV_EGALAX_MULTITOUCH, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_MOJO, USB_DEVICE_ID_RETRO_ADAPTER, HID_QUIRK_MULTI_INPUT },
+	{ USB_VENDOR_ID_TURBOX, USB_DEVICE_ID_TURBOX_TOUCHSCREEN_MOSART, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_DRIVING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FLYING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FIGHTING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
diff --git a/drivers/i2c/busses/i2c-cpm.c b/drivers/i2c/busses/i2c-cpm.c
index f7bd261..f2de3be 100644
--- a/drivers/i2c/busses/i2c-cpm.c
+++ b/drivers/i2c/busses/i2c-cpm.c
@@ -677,6 +677,11 @@
 	dev_dbg(&ofdev->dev, "hw routines for %s registered.\n",
 		cpm->adap.name);
 
+	/*
+	 * register OF I2C devices
+	 */
+	of_i2c_register_devices(&cpm->adap);
+
 	return 0;
 out_shut:
 	cpm_i2c_shutdown(cpm);
diff --git a/drivers/i2c/busses/i2c-davinci.c b/drivers/i2c/busses/i2c-davinci.c
index b8feac5..5795c83 100644
--- a/drivers/i2c/busses/i2c-davinci.c
+++ b/drivers/i2c/busses/i2c-davinci.c
@@ -331,21 +331,16 @@
 	INIT_COMPLETION(dev->cmd_complete);
 	dev->cmd_err = 0;
 
-	/* Take I2C out of reset, configure it as master and set the
-	 * start bit */
-	flag = DAVINCI_I2C_MDR_IRS | DAVINCI_I2C_MDR_MST | DAVINCI_I2C_MDR_STT;
+	/* Take I2C out of reset and configure it as master */
+	flag = DAVINCI_I2C_MDR_IRS | DAVINCI_I2C_MDR_MST;
 
 	/* if the slave address is ten bit address, enable XA bit */
 	if (msg->flags & I2C_M_TEN)
 		flag |= DAVINCI_I2C_MDR_XA;
 	if (!(msg->flags & I2C_M_RD))
 		flag |= DAVINCI_I2C_MDR_TRX;
-	if (stop)
-		flag |= DAVINCI_I2C_MDR_STP;
-	if (msg->len == 0) {
+	if (msg->len == 0)
 		flag |= DAVINCI_I2C_MDR_RM;
-		flag &= ~DAVINCI_I2C_MDR_STP;
-	}
 
 	/* Enable receive or transmit interrupts */
 	w = davinci_i2c_read_reg(dev, DAVINCI_I2C_IMR_REG);
@@ -358,17 +353,28 @@
 	dev->terminate = 0;
 
 	/*
+	 * Write mode register first as needed for correct behaviour
+	 * on OMAP-L138, but don't set STT yet to avoid a race with XRDY
+	 * occuring before we have loaded DXR
+	 */
+	davinci_i2c_write_reg(dev, DAVINCI_I2C_MDR_REG, flag);
+
+	/*
 	 * First byte should be set here, not after interrupt,
 	 * because transmit-data-ready interrupt can come before
 	 * NACK-interrupt during sending of previous message and
 	 * ICDXR may have wrong data
+	 * It also saves us one interrupt, slightly faster
 	 */
 	if ((!(msg->flags & I2C_M_RD)) && dev->buf_len) {
 		davinci_i2c_write_reg(dev, DAVINCI_I2C_DXR_REG, *dev->buf++);
 		dev->buf_len--;
 	}
 
-	/* write the data into mode register; start transmitting */
+	/* Set STT to begin transmit now DXR is loaded */
+	flag |= DAVINCI_I2C_MDR_STT;
+	if (stop && msg->len != 0)
+		flag |= DAVINCI_I2C_MDR_STP;
 	davinci_i2c_write_reg(dev, DAVINCI_I2C_MDR_REG, flag);
 
 	r = wait_for_completion_interruptible_timeout(&dev->cmd_complete,
diff --git a/drivers/i2c/busses/i2c-ibm_iic.c b/drivers/i2c/busses/i2c-ibm_iic.c
index 43ca32f..89eedf4 100644
--- a/drivers/i2c/busses/i2c-ibm_iic.c
+++ b/drivers/i2c/busses/i2c-ibm_iic.c
@@ -761,6 +761,9 @@
 	dev_info(&ofdev->dev, "using %s mode\n",
 		 dev->fast_mode ? "fast (400 kHz)" : "standard (100 kHz)");
 
+	/* Now register all the child nodes */
+	of_i2c_register_devices(adap);
+
 	return 0;
 
 error_cleanup:
diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
index d1ff940..4c2a62b 100644
--- a/drivers/i2c/busses/i2c-imx.c
+++ b/drivers/i2c/busses/i2c-imx.c
@@ -159,15 +159,9 @@
 
 static int i2c_imx_trx_complete(struct imx_i2c_struct *i2c_imx)
 {
-	int result;
+	wait_event_timeout(i2c_imx->queue, i2c_imx->i2csr & I2SR_IIF, HZ / 10);
 
-	result = wait_event_interruptible_timeout(i2c_imx->queue,
-		i2c_imx->i2csr & I2SR_IIF, HZ / 10);
-
-	if (unlikely(result < 0)) {
-		dev_dbg(&i2c_imx->adapter.dev, "<%s> result < 0\n", __func__);
-		return result;
-	} else if (unlikely(!(i2c_imx->i2csr & I2SR_IIF))) {
+	if (unlikely(!(i2c_imx->i2csr & I2SR_IIF))) {
 		dev_dbg(&i2c_imx->adapter.dev, "<%s> Timeout\n", __func__);
 		return -ETIMEDOUT;
 	}
@@ -295,7 +289,7 @@
 		i2c_imx->i2csr = temp;
 		temp &= ~I2SR_IIF;
 		writeb(temp, i2c_imx->base + IMX_I2C_I2SR);
-		wake_up_interruptible(&i2c_imx->queue);
+		wake_up(&i2c_imx->queue);
 		return IRQ_HANDLED;
 	}
 
diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c
index a1c419a..b74e6dc 100644
--- a/drivers/i2c/busses/i2c-mpc.c
+++ b/drivers/i2c/busses/i2c-mpc.c
@@ -632,6 +632,7 @@
 		dev_err(i2c->dev, "failed to add adapter\n");
 		goto fail_add;
 	}
+	of_i2c_register_devices(&i2c->adap);
 
 	return result;
 
diff --git a/drivers/i2c/busses/i2c-pca-isa.c b/drivers/i2c/busses/i2c-pca-isa.c
index bbd7760..29933f8 100644
--- a/drivers/i2c/busses/i2c-pca-isa.c
+++ b/drivers/i2c/busses/i2c-pca-isa.c
@@ -71,8 +71,8 @@
 
 static int pca_isa_waitforcompletion(void *pd)
 {
-	long ret = ~0;
 	unsigned long timeout;
+	long ret;
 
 	if (irq > -1) {
 		ret = wait_event_timeout(pca_wait,
@@ -81,11 +81,15 @@
 	} else {
 		/* Do polling */
 		timeout = jiffies + pca_isa_ops.timeout;
-		while (((pca_isa_readbyte(pd, I2C_PCA_CON)
-				& I2C_PCA_CON_SI) == 0)
-				&& (ret = time_before(jiffies, timeout)))
+		do {
+			ret = time_before(jiffies, timeout);
+			if (pca_isa_readbyte(pd, I2C_PCA_CON)
+					& I2C_PCA_CON_SI)
+				break;
 			udelay(100);
+		} while (ret);
 	}
+
 	return ret > 0;
 }
 
diff --git a/drivers/i2c/busses/i2c-pca-platform.c b/drivers/i2c/busses/i2c-pca-platform.c
index ef5c784..5f6d7f8 100644
--- a/drivers/i2c/busses/i2c-pca-platform.c
+++ b/drivers/i2c/busses/i2c-pca-platform.c
@@ -80,8 +80,8 @@
 static int i2c_pca_pf_waitforcompletion(void *pd)
 {
 	struct i2c_pca_pf_data *i2c = pd;
-	long ret = ~0;
 	unsigned long timeout;
+	long ret;
 
 	if (i2c->irq) {
 		ret = wait_event_timeout(i2c->wait,
@@ -90,10 +90,13 @@
 	} else {
 		/* Do polling */
 		timeout = jiffies + i2c->adap.timeout;
-		while (((i2c->algo_data.read_byte(i2c, I2C_PCA_CON)
-				& I2C_PCA_CON_SI) == 0)
-				&& (ret = time_before(jiffies, timeout)))
+		do {
+			ret = time_before(jiffies, timeout);
+			if (i2c->algo_data.read_byte(i2c, I2C_PCA_CON)
+					& I2C_PCA_CON_SI)
+				break;
 			udelay(100);
+		} while (ret);
 	}
 
 	return ret > 0;
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 6649176..bea4c50 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -32,7 +32,6 @@
 #include <linux/init.h>
 #include <linux/idr.h>
 #include <linux/mutex.h>
-#include <linux/of_i2c.h>
 #include <linux/of_device.h>
 #include <linux/completion.h>
 #include <linux/hardirq.h>
@@ -197,11 +196,12 @@
 {
 	const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
-
-	if (pm)
-		return pm->suspend ? pm->suspend(dev) : 0;
+	if (pm) {
+		if (pm_runtime_suspended(dev))
+			return 0;
+		else
+			return pm->suspend ? pm->suspend(dev) : 0;
+	}
 
 	return i2c_legacy_suspend(dev, PMSG_SUSPEND);
 }
@@ -216,12 +216,6 @@
 	else
 		ret = i2c_legacy_resume(dev);
 
-	if (!ret) {
-		pm_runtime_disable(dev);
-		pm_runtime_set_active(dev);
-		pm_runtime_enable(dev);
-	}
-
 	return ret;
 }
 
@@ -229,11 +223,12 @@
 {
 	const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
-
-	if (pm)
-		return pm->freeze ? pm->freeze(dev) : 0;
+	if (pm) {
+		if (pm_runtime_suspended(dev))
+			return 0;
+		else
+			return pm->freeze ? pm->freeze(dev) : 0;
+	}
 
 	return i2c_legacy_suspend(dev, PMSG_FREEZE);
 }
@@ -242,11 +237,12 @@
 {
 	const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
-
-	if (pm)
-		return pm->thaw ? pm->thaw(dev) : 0;
+	if (pm) {
+		if (pm_runtime_suspended(dev))
+			return 0;
+		else
+			return pm->thaw ? pm->thaw(dev) : 0;
+	}
 
 	return i2c_legacy_resume(dev);
 }
@@ -255,11 +251,12 @@
 {
 	const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
-
-	if (pm)
-		return pm->poweroff ? pm->poweroff(dev) : 0;
+	if (pm) {
+		if (pm_runtime_suspended(dev))
+			return 0;
+		else
+			return pm->poweroff ? pm->poweroff(dev) : 0;
+	}
 
 	return i2c_legacy_suspend(dev, PMSG_HIBERNATE);
 }
@@ -876,9 +873,6 @@
 	if (adap->nr < __i2c_first_dynamic_bus_num)
 		i2c_scan_static_board_info(adap);
 
-	/* Register devices from the device tree */
-	of_i2c_register_devices(adap);
-
 	/* Notify drivers */
 	mutex_lock(&core_lock);
 	bus_for_each_drv(&i2c_bus_type, NULL, adap, __process_new_adapter);
diff --git a/drivers/ide/ide-cs.c b/drivers/ide/ide-cs.c
index 2a4cb9c..404843e 100644
--- a/drivers/ide/ide-cs.c
+++ b/drivers/ide/ide-cs.c
@@ -43,7 +43,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/cisreg.h>
@@ -72,17 +71,6 @@
 
 static void ide_detach(struct pcmcia_device *p_dev);
 
-
-
-
-/*======================================================================
-
-    ide_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int ide_probe(struct pcmcia_device *link)
 {
     ide_info_t *info;
@@ -97,23 +85,12 @@
     info->p_dev = link;
     link->priv = info;
 
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-    link->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO |
+	    CONF_AUTO_SET_VPP | CONF_AUTO_CHECK_VCC;
 
     return ide_config(link);
 } /* ide_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void ide_detach(struct pcmcia_device *link)
 {
     ide_info_t *info = link->priv;
@@ -187,79 +164,31 @@
     return NULL;
 }
 
-/*======================================================================
-
-    ide_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ide device available to the system.
-
-======================================================================*/
-
-struct pcmcia_config_check {
-	unsigned long ctl_base;
-	int skip_vcc;
-	int is_kme;
-};
-
-static int pcmcia_check_one_config(struct pcmcia_device *pdev,
-				   cistpl_cftable_entry_t *cfg,
-				   cistpl_cftable_entry_t *dflt,
-				   unsigned int vcc,
-				   void *priv_data)
+static int pcmcia_check_one_config(struct pcmcia_device *pdev, void *priv_data)
 {
-	struct pcmcia_config_check *stk = priv_data;
+	int *is_kme = priv_data;
 
-	/* Check for matching Vcc, unless we're desperate */
-	if (!stk->skip_vcc) {
-		if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-			if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000)
-				return -ENODEV;
-		} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-			if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] / 10000)
-				return -ENODEV;
-		}
+	if (!(pdev->resource[0]->flags & IO_DATA_PATH_WIDTH_8)) {
+		pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+		pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	}
+	pdev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	pdev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
 
-	if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		pdev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		pdev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		pdev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-
-		pdev->conf.ConfigIndex = cfg->index;
-		pdev->resource[0]->start = io->win[0].base;
-		if (!(io->flags & CISTPL_IO_16BIT)) {
-			pdev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-			pdev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-		}
-		if (io->nwin == 2) {
-			pdev->resource[0]->end = 8;
-			pdev->resource[1]->start = io->win[1].base;
-			pdev->resource[1]->end = (stk->is_kme) ? 2 : 1;
-			if (pcmcia_request_io(pdev) != 0)
-				return -ENODEV;
-			stk->ctl_base = pdev->resource[1]->start;
-		} else if ((io->nwin == 1) && (io->win[0].len >= 16)) {
-			pdev->resource[0]->end = io->win[0].len;
-			pdev->resource[1]->end = 0;
-			if (pcmcia_request_io(pdev) != 0)
-				return -ENODEV;
-			stk->ctl_base = pdev->resource[0]->start + 0x0e;
-		} else
+	if (pdev->resource[1]->end) {
+		pdev->resource[0]->end = 8;
+		pdev->resource[1]->end = (*is_kme) ? 2 : 1;
+	} else {
+		if (pdev->resource[0]->end < 16)
 			return -ENODEV;
-		/* If we've got this far, we're done */
-		return 0;
 	}
-	return -ENODEV;
+
+	return pcmcia_request_io(pdev);
 }
 
 static int ide_config(struct pcmcia_device *link)
 {
     ide_info_t *info = link->priv;
-    struct pcmcia_config_check *stk = NULL;
     int ret = 0, is_kme = 0;
     unsigned long io_base, ctl_base;
     struct ide_host *host;
@@ -270,23 +199,21 @@
 	      ((link->card_id == PRODID_KME_KXLC005_A) ||
 	       (link->card_id == PRODID_KME_KXLC005_B)));
 
-    stk = kzalloc(sizeof(*stk), GFP_KERNEL);
-    if (!stk)
-	    goto err_mem;
-    stk->is_kme = is_kme;
-    stk->skip_vcc = io_base = ctl_base = 0;
-
-    if (pcmcia_loop_config(link, pcmcia_check_one_config, stk)) {
-	    stk->skip_vcc = 1;
-	    if (pcmcia_loop_config(link, pcmcia_check_one_config, stk))
+    if (pcmcia_loop_config(link, pcmcia_check_one_config, &is_kme)) {
+	    link->config_flags &= ~CONF_AUTO_CHECK_VCC;
+	    if (pcmcia_loop_config(link, pcmcia_check_one_config, &is_kme))
 		    goto failed; /* No suitable config found */
     }
     io_base = link->resource[0]->start;
-    ctl_base = stk->ctl_base;
+    if (link->resource[1]->end)
+	    ctl_base = link->resource[1]->start;
+    else
+	    ctl_base = link->resource[0]->start + 0x0e;
 
     if (!link->irq)
 	    goto failed;
-    ret = pcmcia_request_configuration(link, &link->conf);
+
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -311,29 +238,15 @@
     info->host = host;
     dev_info(&link->dev, "ide-cs: hd%c: Vpp = %d.%d\n",
 	    'a' + host->ports[0]->index * 2,
-	    link->conf.Vpp / 10, link->conf.Vpp % 10);
+	    link->vpp / 10, link->vpp % 10);
 
-    kfree(stk);
     return 0;
 
-err_mem:
-    printk(KERN_NOTICE "ide-cs: ide_config failed memory allocation\n");
-    goto failed;
-
 failed:
-    kfree(stk);
     ide_release(link);
     return -ENODEV;
 } /* ide_config */
 
-/*======================================================================
-
-    After a card is removed, ide_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void ide_release(struct pcmcia_device *link)
 {
     ide_info_t *info = link->priv;
@@ -359,15 +272,6 @@
 } /* ide_release */
 
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.  A CARD_REMOVAL event
-    also sets some flags to discourage the ide drivers from
-    talking to the ports.
-
-======================================================================*/
-
 static struct pcmcia_device_id ide_ids[] = {
 	PCMCIA_DEVICE_FUNC_ID(4),
 	PCMCIA_DEVICE_MANF_CARD(0x0000, 0x0000),	/* Corsair */
@@ -440,9 +344,7 @@
 
 static struct pcmcia_driver ide_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "ide-cs",
-	},
+	.name		= "ide-cs",
 	.probe		= ide_probe,
 	.remove		= ide_detach,
 	.id_table       = ide_ids,
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 0906fc5..cb3ccf3 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -59,18 +59,11 @@
 #include <linux/hrtimer.h>	/* ktime_get_real() */
 #include <trace/events/power.h>
 #include <linux/sched.h>
+#include <asm/mwait.h>
 
 #define INTEL_IDLE_VERSION "0.4"
 #define PREFIX "intel_idle: "
 
-#define MWAIT_SUBSTATE_MASK	(0xf)
-#define MWAIT_CSTATE_MASK	(0xf)
-#define MWAIT_SUBSTATE_SIZE	(4)
-#define MWAIT_MAX_NUM_CSTATES	8
-#define CPUID_MWAIT_LEAF (5)
-#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1)
-#define CPUID5_ECX_INTERRUPT_BREAK	(0x2)
-
 static struct cpuidle_driver intel_idle_driver = {
 	.name = "intel_idle",
 	.owner = THIS_MODULE,
@@ -157,13 +150,13 @@
 	{ /* MWAIT C5 */ },
 	{ /* MWAIT C6 */
 		.name = "ATM-C6",
-		.desc = "MWAIT 0x40",
-		.driver_data = (void *) 0x40,
+		.desc = "MWAIT 0x52",
+		.driver_data = (void *) 0x52,
 		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
-		.exit_latency = 200,
+		.exit_latency = 140,
 		.power_usage = 150,
-		.target_residency = 800,
-		.enter = NULL },	/* disabled */
+		.target_residency = 560,
+		.enter = &intel_idle },
 };
 
 /**
diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index c908c5f..af9ee31 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -28,7 +28,7 @@
 	int minor;
 	struct input_handle handle;
 	wait_queue_head_t wait;
-	struct evdev_client *grab;
+	struct evdev_client __rcu *grab;
 	struct list_head client_list;
 	spinlock_t client_lock; /* protects client_list */
 	struct mutex mutex;
@@ -669,6 +669,9 @@
 
 		if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCGABS(0))) {
 
+			if (!dev->absinfo)
+				return -EINVAL;
+
 			t = _IOC_NR(cmd) & ABS_MAX;
 			abs = dev->absinfo[t];
 
@@ -680,10 +683,13 @@
 		}
 	}
 
-	if (_IOC_DIR(cmd) == _IOC_READ) {
+	if (_IOC_DIR(cmd) == _IOC_WRITE) {
 
 		if ((_IOC_NR(cmd) & ~ABS_MAX) == _IOC_NR(EVIOCSABS(0))) {
 
+			if (!dev->absinfo)
+				return -EINVAL;
+
 			t = _IOC_NR(cmd) & ABS_MAX;
 
 			if (copy_from_user(&abs, p, min_t(size_t,
diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c
index d85bd8a..22239e9 100644
--- a/drivers/input/joydev.c
+++ b/drivers/input/joydev.c
@@ -483,6 +483,9 @@
 
 	memcpy(joydev->abspam, abspam, len);
 
+	for (i = 0; i < joydev->nabs; i++)
+		joydev->absmap[joydev->abspam[i]] = i;
+
  out:
 	kfree(abspam);
 	return retval;
diff --git a/drivers/input/misc/hp_sdc_rtc.c b/drivers/input/misc/hp_sdc_rtc.c
index c190664..7e2c12a 100644
--- a/drivers/input/misc/hp_sdc_rtc.c
+++ b/drivers/input/misc/hp_sdc_rtc.c
@@ -104,7 +104,7 @@
 	t.endidx =		91;
 	t.seq =			tseq;
 	t.act.semaphore =	&tsem;
-	init_MUTEX_LOCKED(&tsem);
+	sema_init(&tsem, 0);
 	
 	if (hp_sdc_enqueue_transaction(&t)) return -1;
 	
@@ -698,7 +698,7 @@
 		return -ENODEV;
 #endif
 
-	init_MUTEX(&i8042tregs);
+	sema_init(&i8042tregs, 1);
 
 	if ((ret = hp_sdc_request_timer_irq(&hp_sdc_rtc_isr)))
 		return ret;
diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c
index 0d4266a..3606985 100644
--- a/drivers/input/misc/uinput.c
+++ b/drivers/input/misc/uinput.c
@@ -404,6 +404,13 @@
 		retval = uinput_validate_absbits(dev);
 		if (retval < 0)
 			goto exit;
+		if (test_bit(ABS_MT_SLOT, dev->absbit)) {
+			int nslot = input_abs_get_max(dev, ABS_MT_SLOT) + 1;
+			input_mt_create_slots(dev, nslot);
+			input_set_events_per_packet(dev, 6 * nslot);
+		} else if (test_bit(ABS_MT_POSITION_X, dev->absbit)) {
+			input_set_events_per_packet(dev, 60);
+		}
 	}
 
 	udev->state = UIST_SETUP_COMPLETE;
diff --git a/drivers/input/serio/hil_mlc.c b/drivers/input/serio/hil_mlc.c
index c92f4ed..e5624d8 100644
--- a/drivers/input/serio/hil_mlc.c
+++ b/drivers/input/serio/hil_mlc.c
@@ -915,15 +915,15 @@
 	mlc->ostarted = 0;
 
 	rwlock_init(&mlc->lock);
-	init_MUTEX(&mlc->osem);
+	sema_init(&mlc->osem, 1);
 
-	init_MUTEX(&mlc->isem);
+	sema_init(&mlc->isem, 1);
 	mlc->icount = -1;
 	mlc->imatch = 0;
 
 	mlc->opercnt = 0;
 
-	init_MUTEX_LOCKED(&(mlc->csem));
+	sema_init(&(mlc->csem), 0);
 
 	hil_mlc_clear_di_scratch(mlc);
 	hil_mlc_clear_di_map(mlc, 0);
diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c
index bcc2d30..8c0b51c 100644
--- a/drivers/input/serio/hp_sdc.c
+++ b/drivers/input/serio/hp_sdc.c
@@ -905,7 +905,7 @@
 	ts_sync[1]	= 0x0f;
 	ts_sync[2] = ts_sync[3]	= ts_sync[4] = ts_sync[5] = 0;
 	t_sync.act.semaphore = &s_sync;
-	init_MUTEX_LOCKED(&s_sync);
+	sema_init(&s_sync, 0);
 	hp_sdc_enqueue_transaction(&t_sync);
 	down(&s_sync); /* Wait for t_sync to complete */
 
@@ -1039,7 +1039,7 @@
 		return hp_sdc.dev_err;
 	}
 
-	init_MUTEX_LOCKED(&tq_init_sem);
+	sema_init(&tq_init_sem, 0);
 
 	tq_init.actidx		= 0;
 	tq_init.idx		= 1;
diff --git a/drivers/input/tablet/wacom_sys.c b/drivers/input/tablet/wacom_sys.c
index 42ba369..b35876e 100644
--- a/drivers/input/tablet/wacom_sys.c
+++ b/drivers/input/tablet/wacom_sys.c
@@ -103,27 +103,26 @@
 static int wacom_open(struct input_dev *dev)
 {
 	struct wacom *wacom = input_get_drvdata(dev);
+	int retval = 0;
+
+	if (usb_autopm_get_interface(wacom->intf) < 0)
+		return -EIO;
 
 	mutex_lock(&wacom->lock);
 
-	wacom->irq->dev = wacom->usbdev;
-
-	if (usb_autopm_get_interface(wacom->intf) < 0) {
-		mutex_unlock(&wacom->lock);
-		return -EIO;
-	}
-
 	if (usb_submit_urb(wacom->irq, GFP_KERNEL)) {
-		usb_autopm_put_interface(wacom->intf);
-		mutex_unlock(&wacom->lock);
-		return -EIO;
+		retval = -EIO;
+		goto out;
 	}
 
 	wacom->open = true;
 	wacom->intf->needs_remote_wakeup = 1;
 
+out:
 	mutex_unlock(&wacom->lock);
-	return 0;
+	if (retval)
+		usb_autopm_put_interface(wacom->intf);
+	return retval;
 }
 
 static void wacom_close(struct input_dev *dev)
@@ -135,6 +134,8 @@
 	wacom->open = false;
 	wacom->intf->needs_remote_wakeup = 0;
 	mutex_unlock(&wacom->lock);
+
+	usb_autopm_put_interface(wacom->intf);
 }
 
 static int wacom_parse_hid(struct usb_interface *intf, struct hid_descriptor *hid_desc,
diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c
index 6e29bad..47fd7a0 100644
--- a/drivers/input/tablet/wacom_wac.c
+++ b/drivers/input/tablet/wacom_wac.c
@@ -442,8 +442,10 @@
 	/* general pen packet */
 	if ((data[1] & 0xb8) == 0xa0) {
 		t = (data[6] << 2) | ((data[7] >> 6) & 3);
-		if (features->type >= INTUOS4S && features->type <= INTUOS4L)
+		if ((features->type >= INTUOS4S && features->type <= INTUOS4L) ||
+		    features->type == WACOM_21UX2) {
 			t = (t << 1) | (data[1] & 1);
+		}
 		input_report_abs(input, ABS_PRESSURE, t);
 		input_report_abs(input, ABS_TILT_X,
 				((data[7] << 1) & 0x7e) | (data[8] >> 7));
diff --git a/drivers/isdn/act2000/act2000.h b/drivers/isdn/act2000/act2000.h
index d4c5051..88c9423 100644
--- a/drivers/isdn/act2000/act2000.h
+++ b/drivers/isdn/act2000/act2000.h
@@ -141,9 +141,9 @@
 	__u8            rcvhdr[8];
 } irq_data_isa;
 
-typedef union irq_data {
+typedef union act2000_irq_data {
 	irq_data_isa isa;
-} irq_data;
+} act2000_irq_data;
 
 /*
  * Per card driver data
@@ -176,7 +176,7 @@
 	char   *status_buf_read;
 	char   *status_buf_write;
 	char   *status_buf_end;
-	irq_data idat;			/* Data used for IRQ handler        */
+	act2000_irq_data idat;		/* Data used for IRQ handler        */
 	isdn_if interface;		/* Interface to upper layer         */
 	char regname[35];		/* Name used for request_region     */
 } act2000_card;
diff --git a/drivers/isdn/hardware/avm/avm_cs.c b/drivers/isdn/hardware/avm/avm_cs.c
index 09b1795..91f06a3 100644
--- a/drivers/isdn/hardware/avm/avm_cs.c
+++ b/drivers/isdn/hardware/avm/avm_cs.c
@@ -20,7 +20,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -39,87 +38,32 @@
 
 /*====================================================================*/
 
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card insertion
-   and ejection events.  They are invoked from the skeleton event
-   handler.
-*/
-
 static int avmcs_config(struct pcmcia_device *link);
 static void avmcs_release(struct pcmcia_device *link);
-
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void avmcs_detach(struct pcmcia_device *p_dev);
 
-/*======================================================================
-
-    avmcs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-    
-======================================================================*/
-
 static int avmcs_probe(struct pcmcia_device *p_dev)
 {
-
-    /* The io structure describes IO port mapping */
-    p_dev->resource[0]->end = 16;
-    p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-
     /* General socket configuration */
-    p_dev->conf.Attributes = CONF_ENABLE_IRQ;
-    p_dev->conf.IntType = INT_MEMORY_AND_IO;
-    p_dev->conf.ConfigIndex = 1;
-    p_dev->conf.Present = PRESENT_OPTION;
+    p_dev->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+    p_dev->config_index = 1;
+    p_dev->config_regs = PRESENT_OPTION;
 
     return avmcs_config(p_dev);
 } /* avmcs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
 
 static void avmcs_detach(struct pcmcia_device *link)
 {
 	avmcs_release(link);
 } /* avmcs_detach */
 
-/*======================================================================
-
-    avmcs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-    
-======================================================================*/
-
-static int avmcs_configcheck(struct pcmcia_device *p_dev,
-			     cistpl_cftable_entry_t *cf,
-			     cistpl_cftable_entry_t *dflt,
-			     unsigned int vcc,
-			     void *priv_data)
+static int avmcs_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cf->io.nwin <= 0)
-		return -ENODEV;
+	p_dev->resource[0]->end = 16;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
 
-	p_dev->resource[0]->start = cf->io.win[0].base;
-	p_dev->resource[0]->end = cf->io.win[0].len;
 	return pcmcia_request_io(p_dev);
 }
 
@@ -150,7 +94,7 @@
 	/*
          * configure the PCMCIA socket
 	  */
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0) {
 	    pcmcia_disable_device(link);
 	    break;
@@ -197,13 +141,6 @@
 
 } /* avmcs_config */
 
-/*======================================================================
-
-    After a card is removed, avmcs_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-    
-======================================================================*/
 
 static void avmcs_release(struct pcmcia_device *link)
 {
@@ -222,9 +159,7 @@
 
 static struct pcmcia_driver avmcs_driver = {
 	.owner	= THIS_MODULE,
-	.drv	= {
-		.name	= "avm_cs",
-	},
+	.name		= "avm_cs",
 	.probe = avmcs_probe,
 	.remove	= avmcs_detach,
 	.id_table = avmcs_ids,
diff --git a/drivers/isdn/hisax/avma1_cs.c b/drivers/isdn/hisax/avma1_cs.c
index 94263c2..ac4dd78 100644
--- a/drivers/isdn/hisax/avma1_cs.c
+++ b/drivers/isdn/hisax/avma1_cs.c
@@ -20,7 +20,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include "hisax_cfg.h"
@@ -40,67 +39,22 @@
 
 /*====================================================================*/
 
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card insertion
-   and ejection events.  They are invoked from the skeleton event
-   handler.
-*/
-
 static int avma1cs_config(struct pcmcia_device *link) __devinit ;
 static void avma1cs_release(struct pcmcia_device *link);
-
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void avma1cs_detach(struct pcmcia_device *p_dev) __devexit ;
 
-
-/*======================================================================
-
-    avma1cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-    
-======================================================================*/
-
 static int __devinit avma1cs_probe(struct pcmcia_device *p_dev)
 {
     dev_dbg(&p_dev->dev, "avma1cs_attach()\n");
 
-    /* The io structure describes IO port mapping */
-    p_dev->resource[0]->end = 16;
-    p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-    p_dev->resource[1]->end = 16;
-    p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_16;
-
     /* General socket configuration */
-    p_dev->conf.Attributes = CONF_ENABLE_IRQ;
-    p_dev->conf.IntType = INT_MEMORY_AND_IO;
-    p_dev->conf.ConfigIndex = 1;
-    p_dev->conf.Present = PRESENT_OPTION;
+    p_dev->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+    p_dev->config_index = 1;
+    p_dev->config_regs = PRESENT_OPTION;
 
     return avma1cs_config(p_dev);
 } /* avma1cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void __devexit avma1cs_detach(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "avma1cs_detach(0x%p)\n", link);
@@ -108,26 +62,13 @@
 	kfree(link->priv);
 } /* avma1cs_detach */
 
-/*======================================================================
-
-    avma1cs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-    
-======================================================================*/
-
-static int avma1cs_configcheck(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cf,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int avma1cs_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cf->io.nwin <= 0)
-		return -ENODEV;
-
-	p_dev->resource[0]->start = cf->io.win[0].base;
-	p_dev->resource[0]->end = cf->io.win[0].len;
+	p_dev->resource[0]->end = 16;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
 	p_dev->io_lines = 5;
+
 	return pcmcia_request_io(p_dev);
 }
 
@@ -161,7 +102,7 @@
 	/*
 	 * configure the PCMCIA socket
 	 */
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0) {
 	    pcmcia_disable_device(link);
 	    break;
@@ -175,9 +116,6 @@
 	return -ENODEV;
     }
 
-    printk(KERN_NOTICE "avma1_cs: checking at i/o %#x, irq %d\n",
-		(unsigned int) link->resource[0]->start, link->irq);
-
     icard.para[0] = link->irq;
     icard.para[1] = link->resource[0]->start;
     icard.protocol = isdnprot;
@@ -196,14 +134,6 @@
     return 0;
 } /* avma1cs_config */
 
-/*======================================================================
-
-    After a card is removed, avma1cs_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-    
-======================================================================*/
-
 static void avma1cs_release(struct pcmcia_device *link)
 {
 	unsigned long minor = (unsigned long) link->priv;
@@ -216,7 +146,6 @@
 	pcmcia_disable_device(link);
 } /* avma1cs_release */
 
-
 static struct pcmcia_device_id avma1cs_ids[] = {
 	PCMCIA_DEVICE_PROD_ID12("AVM", "ISDN A", 0x95d42008, 0xadc9d4bb),
 	PCMCIA_DEVICE_PROD_ID12("ISDN", "CARD", 0x8d9761c8, 0x01c5aa7b),
@@ -226,19 +155,15 @@
 
 static struct pcmcia_driver avma1cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "avma1_cs",
-	},
+	.name		= "avma1_cs",
 	.probe		= avma1cs_probe,
 	.remove		= __devexit_p(avma1cs_detach),
 	.id_table	= avma1cs_ids,
 };
 
-/*====================================================================*/
-
 static int __init init_avma1_cs(void)
 {
-	return(pcmcia_register_driver(&avma1cs_driver));
+	return pcmcia_register_driver(&avma1cs_driver);
 }
 
 static void __exit exit_avma1_cs(void)
diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c
index 6f9afcd..b133378 100644
--- a/drivers/isdn/hisax/config.c
+++ b/drivers/isdn/hisax/config.c
@@ -801,6 +801,16 @@
 	ll_unload(csta);
 }
 
+static irqreturn_t card_irq(int intno, void *dev_id)
+{
+	struct IsdnCardState *cs = dev_id;
+	irqreturn_t ret = cs->irq_func(intno, cs);
+
+	if (ret == IRQ_HANDLED)
+		cs->irq_cnt++;
+	return ret;
+}
+
 static int init_card(struct IsdnCardState *cs)
 {
 	int 	irq_cnt, cnt = 3, ret;
@@ -809,10 +819,10 @@
 		ret = cs->cardmsg(cs, CARD_INIT, NULL);
 		return(ret);
 	}
-	irq_cnt = kstat_irqs(cs->irq);
+	irq_cnt = cs->irq_cnt = 0;
 	printk(KERN_INFO "%s: IRQ %d count %d\n", CardType[cs->typ],
 	       cs->irq, irq_cnt);
-	if (request_irq(cs->irq, cs->irq_func, cs->irq_flags, "HiSax", cs)) {
+	if (request_irq(cs->irq, card_irq, cs->irq_flags, "HiSax", cs)) {
 		printk(KERN_WARNING "HiSax: couldn't get interrupt %d\n",
 		       cs->irq);
 		return 1;
@@ -822,8 +832,8 @@
 		/* Timeout 10ms */
 		msleep(10);
 		printk(KERN_INFO "%s: IRQ %d count %d\n",
-		       CardType[cs->typ], cs->irq, kstat_irqs(cs->irq));
-		if (kstat_irqs(cs->irq) == irq_cnt) {
+		       CardType[cs->typ], cs->irq, cs->irq_cnt);
+		if (cs->irq_cnt == irq_cnt) {
 			printk(KERN_WARNING
 			       "%s: IRQ(%d) getting no interrupts during init %d\n",
 			       CardType[cs->typ], cs->irq, 4 - cnt);
diff --git a/drivers/isdn/hisax/elsa_cs.c b/drivers/isdn/hisax/elsa_cs.c
index b3c08aa..496d477 100644
--- a/drivers/isdn/hisax/elsa_cs.c
+++ b/drivers/isdn/hisax/elsa_cs.c
@@ -46,7 +46,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -64,26 +63,8 @@
 static int protocol = 2;        /* EURO-ISDN Default */
 module_param(protocol, int, 0);
 
-/*====================================================================*/
-
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card insertion
-   and ejection events.  They are invoked from the elsa_cs event
-   handler.
-*/
-
 static int elsa_cs_config(struct pcmcia_device *link) __devinit ;
 static void elsa_cs_release(struct pcmcia_device *link);
-
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void elsa_cs_detach(struct pcmcia_device *p_dev) __devexit;
 
 typedef struct local_info_t {
@@ -92,18 +73,6 @@
     int			cardnr;
 } local_info_t;
 
-/*======================================================================
-
-    elsa_cs_attach() creates an "instance" of the driver, allocatingx
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int __devinit elsa_cs_probe(struct pcmcia_device *link)
 {
     local_info_t *local;
@@ -119,31 +88,9 @@
 
     local->cardnr = -1;
 
-    /*
-      General socket configuration defaults can go here.  In this
-      client, we assume very little, and rely on the CIS for almost
-      everything.  In most clients, many details (i.e., number, sizes,
-      and attributes of IO windows) are fixed by the nature of the
-      device, and can be hard-wired here.
-    */
-    link->resource[0]->end = 8;
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-
     return elsa_cs_config(link);
 } /* elsa_cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void __devexit elsa_cs_detach(struct pcmcia_device *link)
 {
 	local_info_t *info = link->priv;
@@ -156,27 +103,17 @@
 	kfree(info);
 } /* elsa_cs_detach */
 
-/*======================================================================
-
-    elsa_cs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
-static int elsa_cs_configcheck(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cf,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int elsa_cs_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int j;
 
 	p_dev->io_lines = 3;
+	p_dev->resource[0]->end = 8;
+	p_dev->resource[0]->flags &= IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 
-	if ((cf->io.nwin > 0) && cf->io.win[0].base) {
+	if ((p_dev->resource[0]->end) && p_dev->resource[0]->start) {
 		printk(KERN_INFO "(elsa_cs: looks like the 96 model)\n");
-		p_dev->resource[0]->start = cf->io.win[0].base;
 		if (!pcmcia_request_io(p_dev))
 			return 0;
 	} else {
@@ -199,6 +136,8 @@
     dev_dbg(&link->dev, "elsa_config(0x%p)\n", link);
     dev = link->priv;
 
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
     i = pcmcia_loop_config(link, elsa_cs_configcheck, NULL);
     if (i != 0)
 	goto failed;
@@ -206,21 +145,10 @@
     if (!link->irq)
 	goto failed;
 
-    i = pcmcia_request_configuration(link, &link->conf);
+    i = pcmcia_enable_device(link);
     if (i != 0)
 	goto failed;
 
-    /* Finally, report what we've done */
-    dev_info(&link->dev, "index 0x%02x: ",
-	    link->conf.ConfigIndex);
-    if (link->conf.Attributes & CONF_ENABLE_IRQ)
-	printk(", irq %d", link->irq);
-    if (link->resource[0])
-	printk(" & %pR", link->resource[0]);
-    if (link->resource[1])
-	printk(" & %pR", link->resource[1]);
-    printk("\n");
-
     icard.para[0] = link->irq;
     icard.para[1] = link->resource[0]->start;
     icard.protocol = protocol;
@@ -240,14 +168,6 @@
     return -ENODEV;
 } /* elsa_cs_config */
 
-/*======================================================================
-
-    After a card is removed, elsa_cs_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void elsa_cs_release(struct pcmcia_device *link)
 {
     local_info_t *local = link->priv;
@@ -291,9 +211,7 @@
 
 static struct pcmcia_driver elsa_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "elsa_cs",
-	},
+	.name		= "elsa_cs",
 	.probe		= elsa_cs_probe,
 	.remove		= __devexit_p(elsa_cs_detach),
 	.id_table	= elsa_ids,
diff --git a/drivers/isdn/hisax/hisax.h b/drivers/isdn/hisax/hisax.h
index 832a878..32ab392 100644
--- a/drivers/isdn/hisax/hisax.h
+++ b/drivers/isdn/hisax/hisax.h
@@ -959,6 +959,7 @@
 	u_long		event;
 	struct work_struct tqueue;
 	struct timer_list dbusytimer;
+	unsigned int	irq_cnt;
 #ifdef ERROR_STATISTIC
 	int		err_crc;
 	int		err_tx;
diff --git a/drivers/isdn/hisax/sedlbauer_cs.c b/drivers/isdn/hisax/sedlbauer_cs.c
index a024192..360204b 100644
--- a/drivers/isdn/hisax/sedlbauer_cs.c
+++ b/drivers/isdn/hisax/sedlbauer_cs.c
@@ -46,7 +46,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -64,26 +63,9 @@
 static int protocol = 2;        /* EURO-ISDN Default */
 module_param(protocol, int, 0);
 
-/*====================================================================*/
-
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card
-   insertion and ejection events.  They are invoked from the sedlbauer
-   event handler. 
-*/
-
 static int sedlbauer_config(struct pcmcia_device *link) __devinit ;
 static void sedlbauer_release(struct pcmcia_device *link);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void sedlbauer_detach(struct pcmcia_device *p_dev) __devexit;
 
 typedef struct local_info_t {
@@ -92,18 +74,6 @@
     int			cardnr;
 } local_info_t;
 
-/*======================================================================
-
-    sedlbauer_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-    
-======================================================================*/
-
 static int __devinit sedlbauer_probe(struct pcmcia_device *link)
 {
     local_info_t *local;
@@ -118,35 +88,9 @@
     local->p_dev = link;
     link->priv = local;
 
-    /*
-      General socket configuration defaults can go here.  In this
-      client, we assume very little, and rely on the CIS for almost
-      everything.  In most clients, many details (i.e., number, sizes,
-      and attributes of IO windows) are fixed by the nature of the
-      device, and can be hard-wired here.
-    */
-
-    /* from old sedl_cs 
-    */
-    /* The io structure describes IO port mapping */
-    link->resource[0]->end = 8;
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-
-    link->conf.Attributes = 0;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-
     return sedlbauer_config(link);
 } /* sedlbauer_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void __devexit sedlbauer_detach(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "sedlbauer_detach(0x%p)\n", link);
@@ -158,70 +102,15 @@
 	kfree(link->priv);
 } /* sedlbauer_detach */
 
-/*======================================================================
-
-    sedlbauer_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-    
-======================================================================*/
-static int sedlbauer_config_check(struct pcmcia_device *p_dev,
-				  cistpl_cftable_entry_t *cfg,
-				  cistpl_cftable_entry_t *dflt,
-				  unsigned int vcc,
-				  void *priv_data)
+static int sedlbauer_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-		if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM]/10000)
-			return -ENODEV;
-	} else if (dflt->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-		if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM]/10000)
-			return -ENODEV;
-	}
-
-	if (cfg->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM]/10000;
-	else if (dflt->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM]/10000;
-
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-					pcmcia_io_cfg_data_width(io->flags);
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		p_dev->io_lines = 3;
-		if (pcmcia_request_io(p_dev) != 0)
-			return -ENODEV;
-	}
-
-	return 0;
+	p_dev->io_lines = 3;
+	return pcmcia_request_io(p_dev);
 }
 
-
-
 static int __devinit sedlbauer_config(struct pcmcia_device *link)
 {
     int ret;
@@ -229,44 +118,17 @@
 
     dev_dbg(&link->dev, "sedlbauer_config(0x%p)\n", link);
 
-    /*
-      In this loop, we scan the CIS for configuration table entries,
-      each of which describes a valid card configuration, including
-      voltage, IO window, memory window, and interrupt settings.
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_CHECK_VCC |
+	    CONF_AUTO_SET_VPP | CONF_AUTO_AUDIO | CONF_AUTO_SET_IO;
 
-      We make no assumptions about the card to be configured: we use
-      just the information available in the CIS.  In an ideal world,
-      this would work for any PCMCIA card, but it requires a complete
-      and accurate CIS.  In practice, a driver usually "knows" most of
-      these things without consulting the CIS, and most client drivers
-      will only use the CIS to fill in implementation-defined details.
-    */
     ret = pcmcia_loop_config(link, sedlbauer_config_check, NULL);
     if (ret)
 	    goto failed;
 
-    /*
-       This actually configures the PCMCIA socket -- setting up
-       the I/O windows and the interrupt mapping, and putting the
-       card and host interface into "Memory and IO" mode.
-    */
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
-    /* Finally, report what we've done */
-    dev_info(&link->dev, "index 0x%02x:",
-	   link->conf.ConfigIndex);
-    if (link->conf.Vpp)
-	printk(", Vpp %d.%d", link->conf.Vpp/10, link->conf.Vpp%10);
-    if (link->conf.Attributes & CONF_ENABLE_IRQ)
-	printk(", irq %d", link->irq);
-    if (link->resource[0])
-	printk(" & %pR", link->resource[0]);
-    if (link->resource[1])
-	printk(" & %pR", link->resource[1]);
-    printk("\n");
-
     icard.para[0] = link->irq;
     icard.para[1] = link->resource[0]->start;
     icard.protocol = protocol;
@@ -290,14 +152,6 @@
 
 } /* sedlbauer_config */
 
-/*======================================================================
-
-    After a card is removed, sedlbauer_release() will unregister the
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-    
-======================================================================*/
-
 static void sedlbauer_release(struct pcmcia_device *link)
 {
     local_info_t *local = link->priv;
@@ -346,9 +200,7 @@
 
 static struct pcmcia_driver sedlbauer_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "sedlbauer_cs",
-	},
+	.name		= "sedlbauer_cs",
 	.probe		= sedlbauer_probe,
 	.remove		= __devexit_p(sedlbauer_detach),
 	.id_table	= sedlbauer_ids,
diff --git a/drivers/isdn/hisax/teles_cs.c b/drivers/isdn/hisax/teles_cs.c
index 7296102..282a446 100644
--- a/drivers/isdn/hisax/teles_cs.c
+++ b/drivers/isdn/hisax/teles_cs.c
@@ -27,7 +27,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -45,26 +44,8 @@
 static int protocol = 2;        /* EURO-ISDN Default */
 module_param(protocol, int, 0);
 
-/*====================================================================*/
-
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card insertion
-   and ejection events.  They are invoked from the teles_cs event
-   handler.
-*/
-
 static int teles_cs_config(struct pcmcia_device *link) __devinit ;
 static void teles_cs_release(struct pcmcia_device *link);
-
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void teles_detach(struct pcmcia_device *p_dev) __devexit ;
 
 typedef struct local_info_t {
@@ -73,18 +54,6 @@
     int			cardnr;
 } local_info_t;
 
-/*======================================================================
-
-    teles_attach() creates an "instance" of the driver, allocatingx
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int __devinit teles_probe(struct pcmcia_device *link)
 {
     local_info_t *local;
@@ -99,31 +68,11 @@
     local->p_dev = link;
     link->priv = local;
 
-    /*
-      General socket configuration defaults can go here.  In this
-      client, we assume very little, and rely on the CIS for almost
-      everything.  In most clients, many details (i.e., number, sizes,
-      and attributes of IO windows) are fixed by the nature of the
-      device, and can be hard-wired here.
-    */
-    link->resource[0]->end = 96;
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
     return teles_cs_config(link);
 } /* teles_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void __devexit teles_detach(struct pcmcia_device *link)
 {
 	local_info_t *info = link->priv;
@@ -136,27 +85,17 @@
 	kfree(info);
 } /* teles_detach */
 
-/*======================================================================
-
-    teles_cs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
-static int teles_cs_configcheck(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cf,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int teles_cs_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int j;
 
 	p_dev->io_lines = 5;
+	p_dev->resource[0]->end = 96;
+	p_dev->resource[0]->flags &= IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 
-	if ((cf->io.nwin > 0) && cf->io.win[0].base) {
+	if ((p_dev->resource[0]->end) && p_dev->resource[0]->start) {
 		printk(KERN_INFO "(teles_cs: looks like the 96 model)\n");
-		p_dev->resource[0]->start = cf->io.win[0].base;
 		if (!pcmcia_request_io(p_dev))
 			return 0;
 	} else {
@@ -186,21 +125,10 @@
     if (!link->irq)
         goto cs_failed;
 
-    i = pcmcia_request_configuration(link, &link->conf);
+    i = pcmcia_enable_device(link);
     if (i != 0)
       goto cs_failed;
 
-    /* Finally, report what we've done */
-    dev_info(&link->dev, "index 0x%02x:",
-	    link->conf.ConfigIndex);
-    if (link->conf.Attributes & CONF_ENABLE_IRQ)
-	    printk(", irq %d", link->irq);
-    if (link->resource[0])
-	printk(" & %pR", link->resource[0]);
-    if (link->resource[1])
-	printk(" & %pR", link->resource[1]);
-    printk("\n");
-
     icard.para[0] = link->irq;
     icard.para[1] = link->resource[0]->start;
     icard.protocol = protocol;
@@ -222,14 +150,6 @@
     return -ENODEV;
 } /* teles_cs_config */
 
-/*======================================================================
-
-    After a card is removed, teles_cs_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void teles_cs_release(struct pcmcia_device *link)
 {
     local_info_t *local = link->priv;
@@ -273,9 +193,7 @@
 
 static struct pcmcia_driver teles_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "teles_cs",
-	},
+	.name		= "teles_cs",
 	.probe		= teles_probe,
 	.remove		= __devexit_p(teles_detach),
 	.id_table       = teles_ids,
diff --git a/drivers/isdn/sc/interrupt.c b/drivers/isdn/sc/interrupt.c
index 485be8b..f0225bc 100644
--- a/drivers/isdn/sc/interrupt.c
+++ b/drivers/isdn/sc/interrupt.c
@@ -112,11 +112,19 @@
 			}
 			else if(callid>=0x0000 && callid<=0x7FFF)
 			{
+				int len;
+
 				pr_debug("%s: Got Incoming Call\n",
 						sc_adapter[card]->devicename);
-				strcpy(setup.phone,&(rcvmsg.msg_data.byte_array[4]));
-				strcpy(setup.eazmsn,
-					sc_adapter[card]->channel[rcvmsg.phy_link_no-1].dn);
+				len = strlcpy(setup.phone, &(rcvmsg.msg_data.byte_array[4]),
+						sizeof(setup.phone));
+				if (len >= sizeof(setup.phone))
+					continue;
+				len = strlcpy(setup.eazmsn,
+						sc_adapter[card]->channel[rcvmsg.phy_link_no - 1].dn,
+						sizeof(setup.eazmsn));
+				if (len >= sizeof(setup.eazmsn))
+					continue;
 				setup.si1 = 7;
 				setup.si2 = 0;
 				setup.plan = 0;
@@ -176,7 +184,9 @@
 		 * Handle a GetMyNumber Rsp
 		 */
 		if (IS_CE_MESSAGE(rcvmsg,Call,0,GetMyNumber)){
-			strcpy(sc_adapter[card]->channel[rcvmsg.phy_link_no-1].dn,rcvmsg.msg_data.byte_array);
+			strlcpy(sc_adapter[card]->channel[rcvmsg.phy_link_no - 1].dn,
+				rcvmsg.msg_data.byte_array,
+				sizeof(rcvmsg.msg_data.byte_array));
 			continue;
 		}
 			
diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c
index 1c4ee6e..bf64e49 100644
--- a/drivers/macintosh/adb.c
+++ b/drivers/macintosh/adb.c
@@ -83,7 +83,7 @@
 BLOCKING_NOTIFIER_HEAD(adb_client_list);
 static int adb_got_sleep;
 static int adb_inited;
-static DECLARE_MUTEX(adb_probe_mutex);
+static DEFINE_SEMAPHORE(adb_probe_mutex);
 static int sleepy_trackpad;
 static int autopoll_devs;
 int __adb_probe_sync;
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index ed4900a..e4fb58d 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1000,10 +1000,11 @@
 				page = bitmap->sb_page;
 				offset = sizeof(bitmap_super_t);
 				if (!file)
-					read_sb_page(bitmap->mddev,
-						     bitmap->mddev->bitmap_info.offset,
-						     page,
-						     index, count);
+					page = read_sb_page(
+						bitmap->mddev,
+						bitmap->mddev->bitmap_info.offset,
+						page,
+						index, count);
 			} else if (file) {
 				page = read_page(file, index, bitmap, count);
 				offset = 0;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index ad83a4d..0b830bb 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1839,7 +1839,9 @@
 
 		/* take from bio_init */
 		bio->bi_next = NULL;
+		bio->bi_flags &= ~(BIO_POOL_MASK-1);
 		bio->bi_flags |= 1 << BIO_UPTODATE;
+		bio->bi_comp_cpu = -1;
 		bio->bi_rw = READ;
 		bio->bi_vcnt = 0;
 		bio->bi_idx = 0;
@@ -1912,7 +1914,7 @@
 			    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
 				break;
 			BUG_ON(sync_blocks < (PAGE_SIZE>>9));
-			if (len > (sync_blocks<<9))
+			if ((len >> 9) > sync_blocks)
 				len = sync_blocks<<9;
 		}
 
diff --git a/drivers/media/IR/ir-keytable.c b/drivers/media/IR/ir-keytable.c
index 7e82a9d..7961d59 100644
--- a/drivers/media/IR/ir-keytable.c
+++ b/drivers/media/IR/ir-keytable.c
@@ -319,7 +319,7 @@
 	 * a keyup event might follow immediately after the keydown.
 	 */
 	spin_lock_irqsave(&ir->keylock, flags);
-	if (time_is_after_eq_jiffies(ir->keyup_jiffies))
+	if (time_is_before_eq_jiffies(ir->keyup_jiffies))
 		ir_keyup(ir);
 	spin_unlock_irqrestore(&ir->keylock, flags);
 }
@@ -510,6 +510,13 @@
 		   (ir_dev->props && ir_dev->props->driver_type == RC_DRIVER_IR_RAW) ?
 			" in raw mode" : "");
 
+	/*
+	 * Default delay of 250ms is too short for some protocols, expecially
+	 * since the timeout is currently set to 250ms. Increase it to 500ms,
+	 * to avoid wrong repetition of the keycodes.
+	 */
+	input_dev->rep[REP_DELAY] = 500;
+
 	return 0;
 
 out_event:
diff --git a/drivers/media/IR/ir-lirc-codec.c b/drivers/media/IR/ir-lirc-codec.c
index 77b5946..e63f757 100644
--- a/drivers/media/IR/ir-lirc-codec.c
+++ b/drivers/media/IR/ir-lirc-codec.c
@@ -267,7 +267,7 @@
 			features |= LIRC_CAN_SET_SEND_CARRIER;
 
 		if (ir_dev->props->s_tx_duty_cycle)
-			features |= LIRC_CAN_SET_REC_DUTY_CYCLE;
+			features |= LIRC_CAN_SET_SEND_DUTY_CYCLE;
 	}
 
 	if (ir_dev->props->s_rx_carrier_range)
diff --git a/drivers/media/IR/ir-raw-event.c b/drivers/media/IR/ir-raw-event.c
index 43094e7..8e0e1b1 100644
--- a/drivers/media/IR/ir-raw-event.c
+++ b/drivers/media/IR/ir-raw-event.c
@@ -279,9 +279,11 @@
 			"rc%u",  (unsigned int)ir->devno);
 
 	if (IS_ERR(ir->raw->thread)) {
+		int ret = PTR_ERR(ir->raw->thread);
+
 		kfree(ir->raw);
 		ir->raw = NULL;
-		return PTR_ERR(ir->raw->thread);
+		return ret;
 	}
 
 	mutex_lock(&ir_raw_handler_lock);
diff --git a/drivers/media/IR/ir-sysfs.c b/drivers/media/IR/ir-sysfs.c
index 96dafc4..46d4246 100644
--- a/drivers/media/IR/ir-sysfs.c
+++ b/drivers/media/IR/ir-sysfs.c
@@ -67,13 +67,14 @@
 	char *tmp = buf;
 	int i;
 
-	if (ir_dev->props->driver_type == RC_DRIVER_SCANCODE) {
+	if (ir_dev->props && ir_dev->props->driver_type == RC_DRIVER_SCANCODE) {
 		enabled = ir_dev->rc_tab.ir_type;
 		allowed = ir_dev->props->allowed_protos;
-	} else {
+	} else if (ir_dev->raw) {
 		enabled = ir_dev->raw->enabled_protocols;
 		allowed = ir_raw_get_allowed_protocols();
-	}
+	} else
+		return sprintf(tmp, "[builtin]\n");
 
 	IR_dprintk(1, "allowed - 0x%llx, enabled - 0x%llx\n",
 		   (long long)allowed,
@@ -121,10 +122,14 @@
 	int rc, i, count = 0;
 	unsigned long flags;
 
-	if (ir_dev->props->driver_type == RC_DRIVER_SCANCODE)
+	if (ir_dev->props && ir_dev->props->driver_type == RC_DRIVER_SCANCODE)
 		type = ir_dev->rc_tab.ir_type;
-	else
+	else if (ir_dev->raw)
 		type = ir_dev->raw->enabled_protocols;
+	else {
+		IR_dprintk(1, "Protocol switching not supported\n");
+		return -EINVAL;
+	}
 
 	while ((tmp = strsep((char **) &data, " \n")) != NULL) {
 		if (!*tmp)
@@ -185,7 +190,7 @@
 		}
 	}
 
-	if (ir_dev->props->driver_type == RC_DRIVER_SCANCODE) {
+	if (ir_dev->props && ir_dev->props->driver_type == RC_DRIVER_SCANCODE) {
 		spin_lock_irqsave(&ir_dev->rc_tab.lock, flags);
 		ir_dev->rc_tab.ir_type = type;
 		spin_unlock_irqrestore(&ir_dev->rc_tab.lock, flags);
diff --git a/drivers/media/IR/keymaps/rc-rc6-mce.c b/drivers/media/IR/keymaps/rc-rc6-mce.c
index 64264f7..39557ad 100644
--- a/drivers/media/IR/keymaps/rc-rc6-mce.c
+++ b/drivers/media/IR/keymaps/rc-rc6-mce.c
@@ -19,6 +19,7 @@
 
 	{ 0x800f0416, KEY_PLAY },
 	{ 0x800f0418, KEY_PAUSE },
+	{ 0x800f046e, KEY_PLAYPAUSE },
 	{ 0x800f0419, KEY_STOP },
 	{ 0x800f0417, KEY_RECORD },
 
@@ -37,6 +38,8 @@
 	{ 0x800f0411, KEY_VOLUMEDOWN },
 	{ 0x800f0412, KEY_CHANNELUP },
 	{ 0x800f0413, KEY_CHANNELDOWN },
+	{ 0x800f043a, KEY_BRIGHTNESSUP },
+	{ 0x800f0480, KEY_BRIGHTNESSDOWN },
 
 	{ 0x800f0401, KEY_NUMERIC_1 },
 	{ 0x800f0402, KEY_NUMERIC_2 },
diff --git a/drivers/media/IR/mceusb.c b/drivers/media/IR/mceusb.c
index ac6bb2c..bc620e1 100644
--- a/drivers/media/IR/mceusb.c
+++ b/drivers/media/IR/mceusb.c
@@ -120,6 +120,10 @@
 	{ USB_DEVICE(VENDOR_PHILIPS, 0x0613) },
 	/* Philips eHome Infrared Transceiver */
 	{ USB_DEVICE(VENDOR_PHILIPS, 0x0815) },
+	/* Philips/Spinel plus IR transceiver for ASUS */
+	{ USB_DEVICE(VENDOR_PHILIPS, 0x206c) },
+	/* Philips/Spinel plus IR transceiver for ASUS */
+	{ USB_DEVICE(VENDOR_PHILIPS, 0x2088) },
 	/* Realtek MCE IR Receiver */
 	{ USB_DEVICE(VENDOR_REALTEK, 0x0161) },
 	/* SMK/Toshiba G83C0004D410 */
diff --git a/drivers/media/dvb/dvb-usb/dib0700_core.c b/drivers/media/dvb/dvb-usb/dib0700_core.c
index fe81834..48397f1 100644
--- a/drivers/media/dvb/dvb-usb/dib0700_core.c
+++ b/drivers/media/dvb/dvb-usb/dib0700_core.c
@@ -673,9 +673,6 @@
 			else
 				dev->props.rc.core.bulk_mode = false;
 
-			/* Need a higher delay, to avoid wrong repeat */
-			dev->rc_input_dev->rep[REP_DELAY] = 500;
-
 			dib0700_rc_setup(dev);
 
 			return 0;
diff --git a/drivers/media/dvb/dvb-usb/dib0700_devices.c b/drivers/media/dvb/dvb-usb/dib0700_devices.c
index f634d2e..e06acd1 100644
--- a/drivers/media/dvb/dvb-usb/dib0700_devices.c
+++ b/drivers/media/dvb/dvb-usb/dib0700_devices.c
@@ -940,6 +940,58 @@
 	return adap->fe == NULL ? -ENODEV : 0;
 }
 
+/* STK7770P */
+static struct dib7000p_config dib7770p_dib7000p_config = {
+	.output_mpeg2_in_188_bytes = 1,
+
+	.agc_config_count = 1,
+	.agc = &dib7070_agc_config,
+	.bw  = &dib7070_bw_config_12_mhz,
+	.tuner_is_baseband = 1,
+	.spur_protect = 1,
+
+	.gpio_dir = DIB7000P_GPIO_DEFAULT_DIRECTIONS,
+	.gpio_val = DIB7000P_GPIO_DEFAULT_VALUES,
+	.gpio_pwm_pos = DIB7000P_GPIO_DEFAULT_PWM_POS,
+
+	.hostbus_diversity = 1,
+	.enable_current_mirror = 1,
+	.disable_sample_and_hold = 0,
+};
+
+static int stk7770p_frontend_attach(struct dvb_usb_adapter *adap)
+{
+	struct usb_device_descriptor *p = &adap->dev->udev->descriptor;
+	if (p->idVendor  == cpu_to_le16(USB_VID_PINNACLE) &&
+	    p->idProduct == cpu_to_le16(USB_PID_PINNACLE_PCTV72E))
+		dib0700_set_gpio(adap->dev, GPIO6, GPIO_OUT, 0);
+	else
+		dib0700_set_gpio(adap->dev, GPIO6, GPIO_OUT, 1);
+	msleep(10);
+	dib0700_set_gpio(adap->dev, GPIO9, GPIO_OUT, 1);
+	dib0700_set_gpio(adap->dev, GPIO4, GPIO_OUT, 1);
+	dib0700_set_gpio(adap->dev, GPIO7, GPIO_OUT, 1);
+	dib0700_set_gpio(adap->dev, GPIO10, GPIO_OUT, 0);
+
+	dib0700_ctrl_clock(adap->dev, 72, 1);
+
+	msleep(10);
+	dib0700_set_gpio(adap->dev, GPIO10, GPIO_OUT, 1);
+	msleep(10);
+	dib0700_set_gpio(adap->dev, GPIO0, GPIO_OUT, 1);
+
+	if (dib7000p_i2c_enumeration(&adap->dev->i2c_adap, 1, 18,
+				     &dib7770p_dib7000p_config) != 0) {
+		err("%s: dib7000p_i2c_enumeration failed.  Cannot continue\n",
+		    __func__);
+		return -ENODEV;
+	}
+
+	adap->fe = dvb_attach(dib7000p_attach, &adap->dev->i2c_adap, 0x80,
+		&dib7770p_dib7000p_config);
+	return adap->fe == NULL ? -ENODEV : 0;
+}
+
 /* DIB807x generic */
 static struct dibx000_agc_config dib807x_agc_config[2] = {
 	{
@@ -1781,7 +1833,7 @@
 /* 60 */{ USB_DEVICE(USB_VID_TERRATEC,	USB_PID_TERRATEC_CINERGY_T_XXS_2) },
 	{ USB_DEVICE(USB_VID_DIBCOM,    USB_PID_DIBCOM_STK807XPVR) },
 	{ USB_DEVICE(USB_VID_DIBCOM,    USB_PID_DIBCOM_STK807XP) },
-	{ USB_DEVICE(USB_VID_PIXELVIEW, USB_PID_PIXELVIEW_SBTVD) },
+	{ USB_DEVICE_VER(USB_VID_PIXELVIEW, USB_PID_PIXELVIEW_SBTVD, 0x000, 0x3f00) },
 	{ USB_DEVICE(USB_VID_EVOLUTEPC, USB_PID_TVWAY_PLUS) },
 /* 65 */{ USB_DEVICE(USB_VID_PINNACLE,	USB_PID_PINNACLE_PCTV73ESE) },
 	{ USB_DEVICE(USB_VID_PINNACLE,	USB_PID_PINNACLE_PCTV282E) },
@@ -2406,7 +2458,7 @@
 				.pid_filter_count = 32,
 				.pid_filter       = stk70x0p_pid_filter,
 				.pid_filter_ctrl  = stk70x0p_pid_filter_ctrl,
-				.frontend_attach  = stk7070p_frontend_attach,
+				.frontend_attach  = stk7770p_frontend_attach,
 				.tuner_attach     = dib7770p_tuner_attach,
 
 				DIB0700_DEFAULT_STREAMING_CONFIG(0x02),
diff --git a/drivers/media/dvb/dvb-usb/opera1.c b/drivers/media/dvb/dvb-usb/opera1.c
index 6b22ec6..f896337 100644
--- a/drivers/media/dvb/dvb-usb/opera1.c
+++ b/drivers/media/dvb/dvb-usb/opera1.c
@@ -483,9 +483,7 @@
 		}
 	}
 	kfree(p);
-	if (fw) {
-		release_firmware(fw);
-	}
+	release_firmware(fw);
 	return ret;
 }
 
diff --git a/drivers/media/dvb/frontends/dib7000p.c b/drivers/media/dvb/frontends/dib7000p.c
index 2e28b97..3aed0d4 100644
--- a/drivers/media/dvb/frontends/dib7000p.c
+++ b/drivers/media/dvb/frontends/dib7000p.c
@@ -260,6 +260,9 @@
 
 //	dprintk( "908: %x, 909: %x\n", reg_908, reg_909);
 
+	reg_909 |= (state->cfg.disable_sample_and_hold & 1) << 4;
+	reg_908 |= (state->cfg.enable_current_mirror & 1) << 7;
+
 	dib7000p_write_word(state, 908, reg_908);
 	dib7000p_write_word(state, 909, reg_909);
 }
@@ -778,7 +781,10 @@
 		default:
 		case GUARD_INTERVAL_1_32: value *= 1; break;
 	}
-	state->div_sync_wait = (value * 3) / 2 + 32; // add 50% SFN margin + compensate for one DVSY-fifo TODO
+	if (state->cfg.diversity_delay == 0)
+		state->div_sync_wait = (value * 3) / 2 + 48; // add 50% SFN margin + compensate for one DVSY-fifo
+	else
+		state->div_sync_wait = (value * 3) / 2 + state->cfg.diversity_delay; // add 50% SFN margin + compensate for one DVSY-fifo
 
 	/* deactive the possibility of diversity reception if extended interleaver */
 	state->div_force_off = !1 && ch->u.ofdm.transmission_mode != TRANSMISSION_MODE_8K;
diff --git a/drivers/media/dvb/frontends/dib7000p.h b/drivers/media/dvb/frontends/dib7000p.h
index 805dd13..da17345 100644
--- a/drivers/media/dvb/frontends/dib7000p.h
+++ b/drivers/media/dvb/frontends/dib7000p.h
@@ -33,6 +33,11 @@
 	int (*agc_control) (struct dvb_frontend *, u8 before);
 
 	u8 output_mode;
+	u8 disable_sample_and_hold : 1;
+
+	u8 enable_current_mirror : 1;
+	u8 diversity_delay;
+
 };
 
 #define DEFAULT_DIB7000P_I2C_ADDRESS 18
diff --git a/drivers/media/dvb/siano/smscoreapi.c b/drivers/media/dvb/siano/smscoreapi.c
index d93468c..ff3b0fa 100644
--- a/drivers/media/dvb/siano/smscoreapi.c
+++ b/drivers/media/dvb/siano/smscoreapi.c
@@ -1098,33 +1098,26 @@
  *
  * @return pointer to descriptor on success, NULL on error.
  */
-struct smscore_buffer_t *smscore_getbuffer(struct smscore_device_t *coredev)
+
+struct smscore_buffer_t *get_entry(struct smscore_device_t *coredev)
 {
 	struct smscore_buffer_t *cb = NULL;
 	unsigned long flags;
 
-	DEFINE_WAIT(wait);
-
 	spin_lock_irqsave(&coredev->bufferslock, flags);
-
-	/* This function must return a valid buffer, since the buffer list is
-	 * finite, we check that there is an available buffer, if not, we wait
-	 * until such buffer become available.
-	 */
-
-	prepare_to_wait(&coredev->buffer_mng_waitq, &wait, TASK_INTERRUPTIBLE);
-	if (list_empty(&coredev->buffers)) {
-		spin_unlock_irqrestore(&coredev->bufferslock, flags);
-		schedule();
-		spin_lock_irqsave(&coredev->bufferslock, flags);
+	if (!list_empty(&coredev->buffers)) {
+		cb = (struct smscore_buffer_t *) coredev->buffers.next;
+		list_del(&cb->entry);
 	}
-
-	finish_wait(&coredev->buffer_mng_waitq, &wait);
-
-	cb = (struct smscore_buffer_t *) coredev->buffers.next;
-	list_del(&cb->entry);
-
 	spin_unlock_irqrestore(&coredev->bufferslock, flags);
+	return cb;
+}
+
+struct smscore_buffer_t *smscore_getbuffer(struct smscore_device_t *coredev)
+{
+	struct smscore_buffer_t *cb = NULL;
+
+	wait_event(coredev->buffer_mng_waitq, (cb = get_entry(coredev)));
 
 	return cb;
 }
diff --git a/drivers/media/radio/si470x/radio-si470x-i2c.c b/drivers/media/radio/si470x/radio-si470x-i2c.c
index 67a4ec8..4ce541a 100644
--- a/drivers/media/radio/si470x/radio-si470x-i2c.c
+++ b/drivers/media/radio/si470x/radio-si470x-i2c.c
@@ -395,7 +395,7 @@
 	radio->registers[POWERCFG] = POWERCFG_ENABLE;
 	if (si470x_set_register(radio, POWERCFG) < 0) {
 		retval = -EIO;
-		goto err_all;
+		goto err_video;
 	}
 	msleep(110);
 
diff --git a/drivers/media/video/cx231xx/Makefile b/drivers/media/video/cx231xx/Makefile
index 755dd0c..6f2b573 100644
--- a/drivers/media/video/cx231xx/Makefile
+++ b/drivers/media/video/cx231xx/Makefile
@@ -11,4 +11,5 @@
 EXTRA_CFLAGS += -Idrivers/media/common/tuners
 EXTRA_CFLAGS += -Idrivers/media/dvb/dvb-core
 EXTRA_CFLAGS += -Idrivers/media/dvb/frontends
+EXTRA_CFLAGS += -Idrivers/media/dvb/dvb-usb
 
diff --git a/drivers/media/video/cx231xx/cx231xx-cards.c b/drivers/media/video/cx231xx/cx231xx-cards.c
index 6bdc0ef..f2a4900 100644
--- a/drivers/media/video/cx231xx/cx231xx-cards.c
+++ b/drivers/media/video/cx231xx/cx231xx-cards.c
@@ -32,6 +32,7 @@
 #include <media/v4l2-chip-ident.h>
 
 #include <media/cx25840.h>
+#include "dvb-usb-ids.h"
 #include "xc5000.h"
 
 #include "cx231xx.h"
@@ -175,6 +176,8 @@
 	 .driver_info = CX231XX_BOARD_CNXT_RDE_250},
 	{USB_DEVICE(0x0572, 0x58A1),
 	 .driver_info = CX231XX_BOARD_CNXT_RDU_250},
+	{USB_DEVICE_VER(USB_VID_PIXELVIEW, USB_PID_PIXELVIEW_SBTVD, 0x4000,0x4fff),
+	 .driver_info = CX231XX_BOARD_UNKNOWN},
 	{},
 };
 
@@ -226,14 +229,16 @@
 		     dev->board.name, dev->model);
 
 	/* set the direction for GPIO pins */
-	cx231xx_set_gpio_direction(dev, dev->board.tuner_gpio->bit, 1);
-	cx231xx_set_gpio_value(dev, dev->board.tuner_gpio->bit, 1);
-	cx231xx_set_gpio_direction(dev, dev->board.tuner_sif_gpio, 1);
+	if (dev->board.tuner_gpio) {
+		cx231xx_set_gpio_direction(dev, dev->board.tuner_gpio->bit, 1);
+		cx231xx_set_gpio_value(dev, dev->board.tuner_gpio->bit, 1);
+		cx231xx_set_gpio_direction(dev, dev->board.tuner_sif_gpio, 1);
 
-	/* request some modules if any required */
+		/* request some modules if any required */
 
-	/* reset the Tuner */
-	cx231xx_gpio_set(dev, dev->board.tuner_gpio);
+		/* reset the Tuner */
+		cx231xx_gpio_set(dev, dev->board.tuner_gpio);
+	}
 
 	/* set the mode to Analog mode initially */
 	cx231xx_set_mode(dev, CX231XX_ANALOG_MODE);
diff --git a/drivers/media/video/cx25840/cx25840-core.c b/drivers/media/video/cx25840/cx25840-core.c
index 86ca8c2..f5a3e74 100644
--- a/drivers/media/video/cx25840/cx25840-core.c
+++ b/drivers/media/video/cx25840/cx25840-core.c
@@ -1996,7 +1996,7 @@
 
 		state->volume = v4l2_ctrl_new_std(&state->hdl,
 			&cx25840_audio_ctrl_ops, V4L2_CID_AUDIO_VOLUME,
-			0, 65335, 65535 / 100, default_volume);
+			0, 65535, 65535 / 100, default_volume);
 		state->mute = v4l2_ctrl_new_std(&state->hdl,
 			&cx25840_audio_ctrl_ops, V4L2_CID_AUDIO_MUTE,
 			0, 1, 1, 0);
diff --git a/drivers/media/video/cx88/Kconfig b/drivers/media/video/cx88/Kconfig
index 99dbae1..0fa85cb 100644
--- a/drivers/media/video/cx88/Kconfig
+++ b/drivers/media/video/cx88/Kconfig
@@ -17,7 +17,7 @@
 
 config VIDEO_CX88_ALSA
 	tristate "Conexant 2388x DMA audio support"
-	depends on VIDEO_CX88 && SND && EXPERIMENTAL
+	depends on VIDEO_CX88 && SND
 	select SND_PCM
 	---help---
 	  This is a video4linux driver for direct (DMA) audio on
diff --git a/drivers/media/video/gspca/gspca.c b/drivers/media/video/gspca/gspca.c
index b984610..78abc1c 100644
--- a/drivers/media/video/gspca/gspca.c
+++ b/drivers/media/video/gspca/gspca.c
@@ -223,6 +223,7 @@
 		usb_rcvintpipe(dev, ep->bEndpointAddress),
 		buffer, buffer_len,
 		int_irq, (void *)gspca_dev, interval);
+	urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 	gspca_dev->int_urb = urb;
 	ret = usb_submit_urb(urb, GFP_KERNEL);
 	if (ret < 0) {
diff --git a/drivers/media/video/gspca/sn9c20x.c b/drivers/media/video/gspca/sn9c20x.c
index 83a718f..9052d57 100644
--- a/drivers/media/video/gspca/sn9c20x.c
+++ b/drivers/media/video/gspca/sn9c20x.c
@@ -2357,8 +2357,7 @@
 			    (data[33] << 10);
 		avg_lum >>= 9;
 		atomic_set(&sd->avg_lum, avg_lum);
-		gspca_frame_add(gspca_dev, LAST_PACKET,
-				data, len);
+		gspca_frame_add(gspca_dev, LAST_PACKET, NULL, 0);
 		return;
 	}
 	if (gspca_dev->last_packet_type == LAST_PACKET) {
diff --git a/drivers/media/video/ivtv/ivtvfb.c b/drivers/media/video/ivtv/ivtvfb.c
index be03a71..f0316d0 100644
--- a/drivers/media/video/ivtv/ivtvfb.c
+++ b/drivers/media/video/ivtv/ivtvfb.c
@@ -466,6 +466,8 @@
 			struct fb_vblank vblank;
 			u32 trace;
 
+			memset(&vblank, 0, sizeof(struct fb_vblank));
+
 			vblank.flags = FB_VBLANK_HAVE_COUNT |FB_VBLANK_HAVE_VCOUNT |
 					FB_VBLANK_HAVE_VSYNC;
 			trace = read_reg(IVTV_REG_DEC_LINE_FIELD) >> 16;
diff --git a/drivers/media/video/mem2mem_testdev.c b/drivers/media/video/mem2mem_testdev.c
index 4525335..a7210d9 100644
--- a/drivers/media/video/mem2mem_testdev.c
+++ b/drivers/media/video/mem2mem_testdev.c
@@ -239,7 +239,7 @@
 		return -EFAULT;
 	}
 
-	if (in_buf->vb.size < out_buf->vb.size) {
+	if (in_buf->vb.size > out_buf->vb.size) {
 		v4l2_err(&dev->v4l2_dev, "Output buffer is too small\n");
 		return -EINVAL;
 	}
@@ -1014,6 +1014,7 @@
 	v4l2_m2m_release(dev->m2m_dev);
 	del_timer_sync(&dev->timer);
 	video_unregister_device(dev->vfd);
+	video_device_release(dev->vfd);
 	v4l2_device_unregister(&dev->v4l2_dev);
 	kfree(dev);
 
diff --git a/drivers/media/video/mt9m111.c b/drivers/media/video/mt9m111.c
index 758a4db..c71af4e 100644
--- a/drivers/media/video/mt9m111.c
+++ b/drivers/media/video/mt9m111.c
@@ -447,6 +447,9 @@
 	dev_dbg(&client->dev, "%s left=%d, top=%d, width=%d, height=%d\n",
 		__func__, rect.left, rect.top, rect.width, rect.height);
 
+	if (a->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
+		return -EINVAL;
+
 	ret = mt9m111_make_rect(client, &rect);
 	if (!ret)
 		mt9m111->rect = rect;
@@ -466,12 +469,14 @@
 
 static int mt9m111_cropcap(struct v4l2_subdev *sd, struct v4l2_cropcap *a)
 {
+	if (a->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
+		return -EINVAL;
+
 	a->bounds.left			= MT9M111_MIN_DARK_COLS;
 	a->bounds.top			= MT9M111_MIN_DARK_ROWS;
 	a->bounds.width			= MT9M111_MAX_WIDTH;
 	a->bounds.height		= MT9M111_MAX_HEIGHT;
 	a->defrect			= a->bounds;
-	a->type				= V4L2_BUF_TYPE_VIDEO_CAPTURE;
 	a->pixelaspect.numerator	= 1;
 	a->pixelaspect.denominator	= 1;
 
@@ -487,6 +492,7 @@
 	mf->width	= mt9m111->rect.width;
 	mf->height	= mt9m111->rect.height;
 	mf->code	= mt9m111->fmt->code;
+	mf->colorspace	= mt9m111->fmt->colorspace;
 	mf->field	= V4L2_FIELD_NONE;
 
 	return 0;
diff --git a/drivers/media/video/mt9v022.c b/drivers/media/video/mt9v022.c
index e7cd23c..b48473c 100644
--- a/drivers/media/video/mt9v022.c
+++ b/drivers/media/video/mt9v022.c
@@ -402,9 +402,6 @@
 		if (mt9v022->model != V4L2_IDENT_MT9V022IX7ATC)
 			return -EINVAL;
 		break;
-	case 0:
-		/* No format change, only geometry */
-		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/media/video/mx2_camera.c b/drivers/media/video/mx2_camera.c
index 66ff174..b6ea672 100644
--- a/drivers/media/video/mx2_camera.c
+++ b/drivers/media/video/mx2_camera.c
@@ -378,6 +378,9 @@
 
 	spin_lock_irqsave(&pcdev->lock, flags);
 
+	if (*fb_active == NULL)
+		goto out;
+
 	vb = &(*fb_active)->vb;
 	dev_dbg(pcdev->dev, "%s (vb=0x%p) 0x%08lx %d\n", __func__,
 		vb, vb->baddr, vb->bsize);
@@ -402,6 +405,7 @@
 
 	*fb_active = buf;
 
+out:
 	spin_unlock_irqrestore(&pcdev->lock, flags);
 }
 
diff --git a/drivers/media/video/pvrusb2/pvrusb2-ctrl.c b/drivers/media/video/pvrusb2/pvrusb2-ctrl.c
index 1b992b8..55ea914 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-ctrl.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-ctrl.c
@@ -513,7 +513,7 @@
 			if (ret >= 0) {
 				ret = pvr2_ctrl_range_check(cptr,*valptr);
 			}
-			if (maskptr) *maskptr = ~0;
+			*maskptr = ~0;
 		} else if (cptr->info->type == pvr2_ctl_bool) {
 			ret = parse_token(ptr,len,valptr,boolNames,
 					  ARRAY_SIZE(boolNames));
@@ -522,7 +522,7 @@
 			} else if (ret == 0) {
 				*valptr = (*valptr & 1) ? !0 : 0;
 			}
-			if (maskptr) *maskptr = 1;
+			*maskptr = 1;
 		} else if (cptr->info->type == pvr2_ctl_enum) {
 			ret = parse_token(
 				ptr,len,valptr,
@@ -531,7 +531,7 @@
 			if (ret >= 0) {
 				ret = pvr2_ctrl_range_check(cptr,*valptr);
 			}
-			if (maskptr) *maskptr = ~0;
+			*maskptr = ~0;
 		} else if (cptr->info->type == pvr2_ctl_bitmask) {
 			ret = parse_tlist(
 				ptr,len,maskptr,valptr,
diff --git a/drivers/media/video/s5p-fimc/fimc-core.c b/drivers/media/video/s5p-fimc/fimc-core.c
index b151c7b..6961c55 100644
--- a/drivers/media/video/s5p-fimc/fimc-core.c
+++ b/drivers/media/video/s5p-fimc/fimc-core.c
@@ -393,6 +393,37 @@
 	dbg("ctx->out_order_1p= %d", ctx->out_order_1p);
 }
 
+static void fimc_prepare_dma_offset(struct fimc_ctx *ctx, struct fimc_frame *f)
+{
+	struct samsung_fimc_variant *variant = ctx->fimc_dev->variant;
+
+	f->dma_offset.y_h = f->offs_h;
+	if (!variant->pix_hoff)
+		f->dma_offset.y_h *= (f->fmt->depth >> 3);
+
+	f->dma_offset.y_v = f->offs_v;
+
+	f->dma_offset.cb_h = f->offs_h;
+	f->dma_offset.cb_v = f->offs_v;
+
+	f->dma_offset.cr_h = f->offs_h;
+	f->dma_offset.cr_v = f->offs_v;
+
+	if (!variant->pix_hoff) {
+		if (f->fmt->planes_cnt == 3) {
+			f->dma_offset.cb_h >>= 1;
+			f->dma_offset.cr_h >>= 1;
+		}
+		if (f->fmt->color == S5P_FIMC_YCBCR420) {
+			f->dma_offset.cb_v >>= 1;
+			f->dma_offset.cr_v >>= 1;
+		}
+	}
+
+	dbg("in_offset: color= %d, y_h= %d, y_v= %d",
+	    f->fmt->color, f->dma_offset.y_h, f->dma_offset.y_v);
+}
+
 /**
  * fimc_prepare_config - check dimensions, operation and color mode
  *			 and pre-calculate offset and the scaling coefficients.
@@ -406,7 +437,6 @@
 {
 	struct fimc_frame *s_frame, *d_frame;
 	struct fimc_vid_buffer *buf = NULL;
-	struct samsung_fimc_variant *variant = ctx->fimc_dev->variant;
 	int ret = 0;
 
 	s_frame = &ctx->s_frame;
@@ -419,61 +449,16 @@
 			swap(d_frame->width, d_frame->height);
 		}
 
-		/* Prepare the output offset ratios for scaler. */
-		d_frame->dma_offset.y_h = d_frame->offs_h;
-		if (!variant->pix_hoff)
-			d_frame->dma_offset.y_h *= (d_frame->fmt->depth >> 3);
+		/* Prepare the DMA offset ratios for scaler. */
+		fimc_prepare_dma_offset(ctx, &ctx->s_frame);
+		fimc_prepare_dma_offset(ctx, &ctx->d_frame);
 
-		d_frame->dma_offset.y_v = d_frame->offs_v;
-
-		d_frame->dma_offset.cb_h = d_frame->offs_h;
-		d_frame->dma_offset.cb_v = d_frame->offs_v;
-
-		d_frame->dma_offset.cr_h = d_frame->offs_h;
-		d_frame->dma_offset.cr_v = d_frame->offs_v;
-
-		if (!variant->pix_hoff && d_frame->fmt->planes_cnt == 3) {
-			d_frame->dma_offset.cb_h >>= 1;
-			d_frame->dma_offset.cb_v >>= 1;
-			d_frame->dma_offset.cr_h >>= 1;
-			d_frame->dma_offset.cr_v >>= 1;
-		}
-
-		dbg("out offset: color= %d, y_h= %d, y_v= %d",
-			d_frame->fmt->color,
-			d_frame->dma_offset.y_h, d_frame->dma_offset.y_v);
-
-		/* Prepare the input offset ratios for scaler. */
-		s_frame->dma_offset.y_h = s_frame->offs_h;
-		if (!variant->pix_hoff)
-			s_frame->dma_offset.y_h *= (s_frame->fmt->depth >> 3);
-		s_frame->dma_offset.y_v = s_frame->offs_v;
-
-		s_frame->dma_offset.cb_h = s_frame->offs_h;
-		s_frame->dma_offset.cb_v = s_frame->offs_v;
-
-		s_frame->dma_offset.cr_h = s_frame->offs_h;
-		s_frame->dma_offset.cr_v = s_frame->offs_v;
-
-		if (!variant->pix_hoff && s_frame->fmt->planes_cnt == 3) {
-			s_frame->dma_offset.cb_h >>= 1;
-			s_frame->dma_offset.cb_v >>= 1;
-			s_frame->dma_offset.cr_h >>= 1;
-			s_frame->dma_offset.cr_v >>= 1;
-		}
-
-		dbg("in offset: color= %d, y_h= %d, y_v= %d",
-			s_frame->fmt->color, s_frame->dma_offset.y_h,
-			s_frame->dma_offset.y_v);
-
-		fimc_set_yuv_order(ctx);
-
-		/* Check against the scaler ratio. */
 		if (s_frame->height > (SCALER_MAX_VRATIO * d_frame->height) ||
 		    s_frame->width > (SCALER_MAX_HRATIO * d_frame->width)) {
 			err("out of scaler range");
 			return -EINVAL;
 		}
+		fimc_set_yuv_order(ctx);
 	}
 
 	/* Input DMA mode is not allowed when the scaler is disabled. */
@@ -822,7 +807,8 @@
 	} else {
 		v4l2_err(&ctx->fimc_dev->m2m.v4l2_dev,
 			 "Wrong buffer/video queue type (%d)\n", f->type);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto s_fmt_out;
 	}
 
 	pix = &f->fmt.pix;
@@ -1414,8 +1400,10 @@
 	}
 
 	fimc->work_queue = create_workqueue(dev_name(&fimc->pdev->dev));
-	if (!fimc->work_queue)
+	if (!fimc->work_queue) {
+		ret = -ENOMEM;
 		goto err_irq;
+	}
 
 	ret = fimc_register_m2m_device(fimc);
 	if (ret)
@@ -1492,6 +1480,7 @@
 };
 
 static struct samsung_fimc_variant fimc01_variant_s5pv210 = {
+	.pix_hoff	= 1,
 	.has_inp_rot	= 1,
 	.has_out_rot	= 1,
 	.min_inp_pixsize = 16,
@@ -1506,6 +1495,7 @@
 };
 
 static struct samsung_fimc_variant fimc2_variant_s5pv210 = {
+	.pix_hoff	 = 1,
 	.min_inp_pixsize = 16,
 	.min_out_pixsize = 32,
 
diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c
index ec697fc..bb8d83d 100644
--- a/drivers/media/video/saa7134/saa7134-cards.c
+++ b/drivers/media/video/saa7134/saa7134-cards.c
@@ -4323,13 +4323,13 @@
 	},
 	[SAA7134_BOARD_BEHOLD_COLUMBUS_TVFM] = {
 		/*       Beholder Intl. Ltd. 2008      */
-		/*Dmitry Belimov <d.belimov@gmail.com> */
-		.name           = "Beholder BeholdTV Columbus TVFM",
+		/* Dmitry Belimov <d.belimov@gmail.com> */
+		.name           = "Beholder BeholdTV Columbus TV/FM",
 		.audio_clock    = 0x00187de7,
 		.tuner_type     = TUNER_ALPS_TSBE5_PAL,
-		.radio_type     = UNSET,
-		.tuner_addr     = ADDR_UNSET,
-		.radio_addr     = ADDR_UNSET,
+		.radio_type     = TUNER_TEA5767,
+		.tuner_addr     = 0xc2 >> 1,
+		.radio_addr     = 0xc0 >> 1,
 		.tda9887_conf   = TDA9887_PRESENT,
 		.gpiomask       = 0x000A8004,
 		.inputs         = {{
diff --git a/drivers/media/video/saa7164/saa7164-buffer.c b/drivers/media/video/saa7164/saa7164-buffer.c
index 5713f3a..ddd25d3 100644
--- a/drivers/media/video/saa7164/saa7164-buffer.c
+++ b/drivers/media/video/saa7164/saa7164-buffer.c
@@ -136,10 +136,11 @@
 int saa7164_buffer_dealloc(struct saa7164_tsport *port,
 	struct saa7164_buffer *buf)
 {
-	struct saa7164_dev *dev = port->dev;
+	struct saa7164_dev *dev;
 
-	if ((buf == 0) || (port == 0))
+	if (!buf || !port)
 		return SAA_ERR_BAD_PARAMETER;
+	dev = port->dev;
 
 	dprintk(DBGLVL_BUF, "%s() deallocating buffer @ 0x%p\n", __func__, buf);
 
diff --git a/drivers/media/video/uvc/uvc_driver.c b/drivers/media/video/uvc/uvc_driver.c
index 8bdd940..2ac85d8 100644
--- a/drivers/media/video/uvc/uvc_driver.c
+++ b/drivers/media/video/uvc/uvc_driver.c
@@ -486,6 +486,12 @@
 			    max(frame->dwFrameInterval[0],
 				frame->dwDefaultFrameInterval));
 
+		if (dev->quirks & UVC_QUIRK_RESTRICT_FRAME_RATE) {
+			frame->bFrameIntervalType = 1;
+			frame->dwFrameInterval[0] =
+				frame->dwDefaultFrameInterval;
+		}
+
 		uvc_trace(UVC_TRACE_DESCR, "- %ux%u (%u.%u fps)\n",
 			frame->wWidth, frame->wHeight,
 			10000000/frame->dwDefaultFrameInterval,
@@ -2026,6 +2032,15 @@
 	  .bInterfaceClass	= USB_CLASS_VENDOR_SPEC,
 	  .bInterfaceSubClass	= 1,
 	  .bInterfaceProtocol	= 0 },
+	/* Chicony CNF7129 (Asus EEE 100HE) */
+	{ .match_flags		= USB_DEVICE_ID_MATCH_DEVICE
+				| USB_DEVICE_ID_MATCH_INT_INFO,
+	  .idVendor		= 0x04f2,
+	  .idProduct		= 0xb071,
+	  .bInterfaceClass	= USB_CLASS_VIDEO,
+	  .bInterfaceSubClass	= 1,
+	  .bInterfaceProtocol	= 0,
+	  .driver_info		= UVC_QUIRK_RESTRICT_FRAME_RATE },
 	/* Alcor Micro AU3820 (Future Boy PC USB Webcam) */
 	{ .match_flags		= USB_DEVICE_ID_MATCH_DEVICE
 				| USB_DEVICE_ID_MATCH_INT_INFO,
@@ -2091,6 +2106,15 @@
 	  .bInterfaceProtocol	= 0,
 	  .driver_info		= UVC_QUIRK_PROBE_MINMAX
 				| UVC_QUIRK_PROBE_DEF },
+	/* IMC Networks (Medion Akoya) */
+	{ .match_flags		= USB_DEVICE_ID_MATCH_DEVICE
+				| USB_DEVICE_ID_MATCH_INT_INFO,
+	  .idVendor		= 0x13d3,
+	  .idProduct		= 0x5103,
+	  .bInterfaceClass	= USB_CLASS_VIDEO,
+	  .bInterfaceSubClass	= 1,
+	  .bInterfaceProtocol	= 0,
+	  .driver_info		= UVC_QUIRK_STREAM_NO_FID },
 	/* Syntek (HP Spartan) */
 	{ .match_flags		= USB_DEVICE_ID_MATCH_DEVICE
 				| USB_DEVICE_ID_MATCH_INT_INFO,
diff --git a/drivers/media/video/uvc/uvcvideo.h b/drivers/media/video/uvc/uvcvideo.h
index bdacf3b..892e0e5 100644
--- a/drivers/media/video/uvc/uvcvideo.h
+++ b/drivers/media/video/uvc/uvcvideo.h
@@ -182,6 +182,7 @@
 #define UVC_QUIRK_IGNORE_SELECTOR_UNIT	0x00000020
 #define UVC_QUIRK_FIX_BANDWIDTH		0x00000080
 #define UVC_QUIRK_PROBE_DEF		0x00000100
+#define UVC_QUIRK_RESTRICT_FRAME_RATE	0x00000200
 
 /* Format flags */
 #define UVC_FMT_FLAG_COMPRESSED		0x00000001
diff --git a/drivers/media/video/v4l2-compat-ioctl32.c b/drivers/media/video/v4l2-compat-ioctl32.c
index 073f013..86294ed3 100644
--- a/drivers/media/video/v4l2-compat-ioctl32.c
+++ b/drivers/media/video/v4l2-compat-ioctl32.c
@@ -193,17 +193,24 @@
 struct video_code32 {
 	char		loadwhat[16];	/* name or tag of file being passed */
 	compat_int_t	datasize;
-	unsigned char	*data;
+	compat_uptr_t	data;
 };
 
-static int get_microcode32(struct video_code *kp, struct video_code32 __user *up)
+static struct video_code __user *get_microcode32(struct video_code32 *kp)
 {
-	if (!access_ok(VERIFY_READ, up, sizeof(struct video_code32)) ||
-		copy_from_user(kp->loadwhat, up->loadwhat, sizeof(up->loadwhat)) ||
-		get_user(kp->datasize, &up->datasize) ||
-		copy_from_user(kp->data, up->data, up->datasize))
-			return -EFAULT;
-	return 0;
+	struct video_code __user *up;
+
+	up = compat_alloc_user_space(sizeof(*up));
+
+	/*
+	 * NOTE! We don't actually care if these fail. If the
+	 * user address is invalid, the native ioctl will do
+	 * the error handling for us
+	 */
+	(void) copy_to_user(up->loadwhat, kp->loadwhat, sizeof(up->loadwhat));
+	(void) put_user(kp->datasize, &up->datasize);
+	(void) put_user(compat_ptr(kp->data), &up->data);
+	return up;
 }
 
 #define VIDIOCGTUNER32		_IOWR('v', 4, struct video_tuner32)
@@ -739,7 +746,7 @@
 		struct video_tuner vt;
 		struct video_buffer vb;
 		struct video_window vw;
-		struct video_code vc;
+		struct video_code32 vc;
 		struct video_audio va;
 #endif
 		struct v4l2_format v2f;
@@ -818,8 +825,11 @@
 		break;
 
 	case VIDIOCSMICROCODE:
-		err = get_microcode32(&karg.vc, up);
-		compatible_arg = 0;
+		/* Copy the 32-bit "video_code32" to kernel space */
+		if (copy_from_user(&karg.vc, up, sizeof(karg.vc)))
+			return -EFAULT;
+		/* Convert the 32-bit version to a 64-bit version in user space */
+		up = get_microcode32(&karg.vc);
 		break;
 
 	case VIDIOCSFREQ:
diff --git a/drivers/media/video/videobuf-dma-contig.c b/drivers/media/video/videobuf-dma-contig.c
index 372b87e..6ff9e4b 100644
--- a/drivers/media/video/videobuf-dma-contig.c
+++ b/drivers/media/video/videobuf-dma-contig.c
@@ -393,8 +393,10 @@
 	}
 
 	/* read() method */
-	dma_free_coherent(q->dev, mem->size, mem->vaddr, mem->dma_handle);
-	mem->vaddr = NULL;
+	if (mem->vaddr) {
+		dma_free_coherent(q->dev, mem->size, mem->vaddr, mem->dma_handle);
+		mem->vaddr = NULL;
+	}
 }
 EXPORT_SYMBOL_GPL(videobuf_dma_contig_free);
 
diff --git a/drivers/media/video/videobuf-dma-sg.c b/drivers/media/video/videobuf-dma-sg.c
index 06f9a9c..2ad0bc2 100644
--- a/drivers/media/video/videobuf-dma-sg.c
+++ b/drivers/media/video/videobuf-dma-sg.c
@@ -94,7 +94,7 @@
  * must free the memory.
  */
 static struct scatterlist *videobuf_pages_to_sg(struct page **pages,
-						int nr_pages, int offset)
+					int nr_pages, int offset, size_t size)
 {
 	struct scatterlist *sglist;
 	int i;
@@ -110,12 +110,14 @@
 		/* DMA to highmem pages might not work */
 		goto highmem;
 	sg_set_page(&sglist[0], pages[0], PAGE_SIZE - offset, offset);
+	size -= PAGE_SIZE - offset;
 	for (i = 1; i < nr_pages; i++) {
 		if (NULL == pages[i])
 			goto nopage;
 		if (PageHighMem(pages[i]))
 			goto highmem;
-		sg_set_page(&sglist[i], pages[i], PAGE_SIZE, 0);
+		sg_set_page(&sglist[i], pages[i], min(PAGE_SIZE, size), 0);
+		size -= min(PAGE_SIZE, size);
 	}
 	return sglist;
 
@@ -170,7 +172,8 @@
 
 	first = (data          & PAGE_MASK) >> PAGE_SHIFT;
 	last  = ((data+size-1) & PAGE_MASK) >> PAGE_SHIFT;
-	dma->offset   = data & ~PAGE_MASK;
+	dma->offset = data & ~PAGE_MASK;
+	dma->size = size;
 	dma->nr_pages = last-first+1;
 	dma->pages = kmalloc(dma->nr_pages * sizeof(struct page *), GFP_KERNEL);
 	if (NULL == dma->pages)
@@ -252,7 +255,7 @@
 
 	if (dma->pages) {
 		dma->sglist = videobuf_pages_to_sg(dma->pages, dma->nr_pages,
-						   dma->offset);
+						   dma->offset, dma->size);
 	}
 	if (dma->vaddr) {
 		dma->sglist = videobuf_vmalloc_to_sg(dma->vaddr,
diff --git a/drivers/mfd/ab8500-spi.c b/drivers/mfd/ab8500-spi.c
index e1c8b62..01b6d58 100644
--- a/drivers/mfd/ab8500-spi.c
+++ b/drivers/mfd/ab8500-spi.c
@@ -83,6 +83,11 @@
 	struct ab8500 *ab8500;
 	int ret;
 
+	spi->bits_per_word = 24;
+	ret = spi_setup(spi);
+	if (ret < 0)
+		return ret;
+
 	ab8500 = kzalloc(sizeof *ab8500, GFP_KERNEL);
 	if (!ab8500)
 		return -ENOMEM;
diff --git a/drivers/mfd/twl4030-irq.c b/drivers/mfd/twl4030-irq.c
index 097f24d..b9fda70 100644
--- a/drivers/mfd/twl4030-irq.c
+++ b/drivers/mfd/twl4030-irq.c
@@ -78,7 +78,7 @@
 	u8	irq_lines;		/* number of supported irq lines */
 
 	/* SIR ignored -- set interrupt, for testing only */
-	struct irq_data {
+	struct sih_irq_data {
 		u8	isr_offset;
 		u8	imr_offset;
 	} mask[2];
@@ -810,7 +810,7 @@
 	twl4030_irq_chip = dummy_irq_chip;
 	twl4030_irq_chip.name = "twl4030";
 
-	twl4030_sih_irq_chip.ack = dummy_irq_chip.ack;
+	twl4030_sih_irq_chip.irq_ack = dummy_irq_chip.irq_ack;
 
 	for (i = irq_base; i < irq_end; i++) {
 		set_irq_chip_and_handler(i, &twl4030_irq_chip,
diff --git a/drivers/misc/bh1780gli.c b/drivers/misc/bh1780gli.c
index 714c6b4..d5f3a3f 100644
--- a/drivers/misc/bh1780gli.c
+++ b/drivers/misc/bh1780gli.c
@@ -190,7 +190,6 @@
 
 	ddata = i2c_get_clientdata(client);
 	sysfs_remove_group(&client->dev.kobj, &bh1780_attr_group);
-	i2c_set_clientdata(client, NULL);
 	kfree(ddata);
 
 	return 0;
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 5db49b1..09eee6d 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -1631,6 +1631,19 @@
 	if (host->bus_ops && !host->bus_dead) {
 		if (host->bus_ops->suspend)
 			err = host->bus_ops->suspend(host);
+		if (err == -ENOSYS || !host->bus_ops->resume) {
+			/*
+			 * We simply "remove" the card in this case.
+			 * It will be redetected on resume.
+			 */
+			if (host->bus_ops->remove)
+				host->bus_ops->remove(host);
+			mmc_claim_host(host);
+			mmc_detach_bus(host);
+			mmc_release_host(host);
+			host->pm_flags = 0;
+			err = 0;
+		}
 	}
 	mmc_bus_put(host);
 
diff --git a/drivers/mmc/host/sdricoh_cs.c b/drivers/mmc/host/sdricoh_cs.c
index 7aa65bb..f472c27 100644
--- a/drivers/mmc/host/sdricoh_cs.c
+++ b/drivers/mmc/host/sdricoh_cs.c
@@ -30,7 +30,6 @@
 #include <linux/ioport.h>
 #include <linux/scatterlist.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <linux/io.h>
@@ -536,9 +535,7 @@
 #endif
 
 static struct pcmcia_driver sdricoh_driver = {
-	.drv = {
-		.name = DRIVER_NAME,
-		},
+	.name = DRIVER_NAME,
 	.probe = sdricoh_pcmcia_probe,
 	.remove = sdricoh_pcmcia_detach,
 	.id_table = pcmcia_ids,
diff --git a/drivers/mtd/maps/pcmciamtd.c b/drivers/mtd/maps/pcmciamtd.c
index e9ca5ba..57a1acf 100644
--- a/drivers/mtd/maps/pcmciamtd.c
+++ b/drivers/mtd/maps/pcmciamtd.c
@@ -16,7 +16,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -101,7 +100,7 @@
 static caddr_t remap_window(struct map_info *map, unsigned long to)
 {
 	struct pcmciamtd_dev *dev = (struct pcmciamtd_dev *)map->map_priv_1;
-	window_handle_t win = (window_handle_t)map->map_priv_2;
+	struct resource *win = (struct resource *) map->map_priv_2;
 	unsigned int offset;
 	int ret;
 
@@ -316,30 +315,19 @@
 {
 	struct pcmciamtd_dev *dev = (struct pcmciamtd_dev *)map->map_priv_1;
 	struct pcmcia_device *link = dev->p_dev;
-	modconf_t mod;
-	int ret;
-
-	mod.Attributes = CONF_VPP1_CHANGE_VALID | CONF_VPP2_CHANGE_VALID;
-	mod.Vcc = 0;
-	mod.Vpp1 = mod.Vpp2 = on ? dev->vpp : 0;
 
 	DEBUG(2, "dev = %p on = %d vpp = %d\n", dev, on, dev->vpp);
-	ret = pcmcia_modify_configuration(link, &mod);
+	pcmcia_fixup_vpp(link, on ? dev->vpp : 0);
 }
 
 
-/* After a card is removed, pcmciamtd_release() will unregister the
- * device, and release the PCMCIA configuration.  If the device is
- * still open, this will be postponed until it is closed.
- */
-
 static void pcmciamtd_release(struct pcmcia_device *link)
 {
 	struct pcmciamtd_dev *dev = link->priv;
 
 	DEBUG(3, "link = 0x%p", link);
 
-	if (link->win) {
+	if (link->resource[2]->end) {
 		if(dev->win_base) {
 			iounmap(dev->win_base);
 			dev->win_base = NULL;
@@ -482,18 +470,12 @@
 }
 
 
-/* pcmciamtd_config() is scheduled to run after a CARD_INSERTION event
- * is received, to configure the PCMCIA socket, and to make the
- * MTD device available to the system.
- */
-
 static int pcmciamtd_config(struct pcmcia_device *link)
 {
 	struct pcmciamtd_dev *dev = link->priv;
 	struct mtd_info *mtd = NULL;
-	win_req_t req;
 	int ret;
-	int i;
+	int i, j = 0;
 	static char *probes[] = { "jedec_probe", "cfi_probe" };
 	int new_name = 0;
 
@@ -520,28 +502,34 @@
 	 * smaller windows until we succeed
 	 */
 
-	req.Attributes =  WIN_MEMORY_TYPE_CM | WIN_ENABLE;
-	req.Attributes |= (dev->pcmcia_map.bankwidth == 1) ? WIN_DATA_WIDTH_8 : WIN_DATA_WIDTH_16;
-	req.Base = 0;
-	req.AccessSpeed = mem_speed;
-	link->win = (window_handle_t)link;
-	req.Size = (force_size) ? force_size << 20 : MAX_PCMCIA_ADDR;
+	link->resource[2]->flags |=  WIN_MEMORY_TYPE_CM | WIN_ENABLE;
+	link->resource[2]->flags |= (dev->pcmcia_map.bankwidth == 1) ?
+					WIN_DATA_WIDTH_8 : WIN_DATA_WIDTH_16;
+	link->resource[2]->start = 0;
+	link->resource[2]->end = (force_size) ? force_size << 20 :
+					MAX_PCMCIA_ADDR;
 	dev->win_size = 0;
 
 	do {
 		int ret;
-		DEBUG(2, "requesting window with size = %dKiB memspeed = %d",
-		      req.Size >> 10, req.AccessSpeed);
-		ret = pcmcia_request_window(link, &req, &link->win);
+		DEBUG(2, "requesting window with size = %luKiB memspeed = %d",
+			(unsigned long) resource_size(link->resource[2]) >> 10,
+			mem_speed);
+		ret = pcmcia_request_window(link, link->resource[2], mem_speed);
 		DEBUG(2, "ret = %d dev->win_size = %d", ret, dev->win_size);
 		if(ret) {
-			req.Size >>= 1;
+			j++;
+			link->resource[2]->start = 0;
+			link->resource[2]->end = (force_size) ?
+					force_size << 20 : MAX_PCMCIA_ADDR;
+			link->resource[2]->end >>= j;
 		} else {
-			DEBUG(2, "Got window of size %dKiB", req.Size >> 10);
-			dev->win_size = req.Size;
+			DEBUG(2, "Got window of size %luKiB", (unsigned long)
+				resource_size(link->resource[2]) >> 10);
+			dev->win_size = resource_size(link->resource[2]);
 			break;
 		}
-	} while(req.Size >= 0x1000);
+	} while (link->resource[2]->end >= 0x1000);
 
 	DEBUG(2, "dev->win_size = %d", dev->win_size);
 
@@ -553,33 +541,31 @@
 	DEBUG(1, "Allocated a window of %dKiB", dev->win_size >> 10);
 
 	/* Get write protect status */
-	DEBUG(2, "window handle = 0x%8.8lx", (unsigned long)link->win);
-	dev->win_base = ioremap(req.Base, req.Size);
+	dev->win_base = ioremap(link->resource[2]->start,
+				resource_size(link->resource[2]));
 	if(!dev->win_base) {
-		dev_err(&dev->p_dev->dev, "ioremap(%lu, %u) failed\n",
-			req.Base, req.Size);
+		dev_err(&dev->p_dev->dev, "ioremap(%pR) failed\n",
+			link->resource[2]);
 		pcmciamtd_release(link);
 		return -ENODEV;
 	}
-	DEBUG(1, "mapped window dev = %p req.base = 0x%lx base = %p size = 0x%x",
-	      dev, req.Base, dev->win_base, req.Size);
+	DEBUG(1, "mapped window dev = %p @ %pR, base = %p",
+	      dev, link->resource[2], dev->win_base);
 
 	dev->offset = 0;
 	dev->pcmcia_map.map_priv_1 = (unsigned long)dev;
-	dev->pcmcia_map.map_priv_2 = (unsigned long)link->win;
+	dev->pcmcia_map.map_priv_2 = (unsigned long)link->resource[2];
 
 	dev->vpp = (vpp) ? vpp : link->socket->socket.Vpp;
-	link->conf.Attributes = 0;
 	if(setvpp == 2) {
-		link->conf.Vpp = dev->vpp;
+		link->vpp = dev->vpp;
 	} else {
-		link->conf.Vpp = 0;
+		link->vpp = 0;
 	}
 
-	link->conf.IntType = INT_MEMORY;
-	link->conf.ConfigIndex = 0;
+	link->config_index = 0;
 	DEBUG(2, "Setting Configuration");
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret != 0) {
 		if (dev->win_base) {
 			iounmap(dev->win_base);
@@ -680,12 +666,6 @@
 }
 
 
-/* This deletes a driver "instance".  The device is de-registered
- * with Card Services.  If it has been released, all local data
- * structures are freed.  Otherwise, the structures will be freed
- * when the device is released.
- */
-
 static void pcmciamtd_detach(struct pcmcia_device *link)
 {
 	struct pcmciamtd_dev *dev = link->priv;
@@ -703,11 +683,6 @@
 }
 
 
-/* pcmciamtd_attach() creates an "instance" of the driver, allocating
- * local data structures for one device.  The device is registered
- * with Card Services.
- */
-
 static int pcmciamtd_probe(struct pcmcia_device *link)
 {
 	struct pcmciamtd_dev *dev;
@@ -720,9 +695,6 @@
 	dev->p_dev = link;
 	link->priv = dev;
 
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY;
-
 	return pcmciamtd_config(link);
 }
 
@@ -757,9 +729,7 @@
 MODULE_DEVICE_TABLE(pcmcia, pcmciamtd_ids);
 
 static struct pcmcia_driver pcmciamtd_driver = {
-	.drv		= {
-		.name	= "pcmciamtd"
-	},
+	.name		= "pcmciamtd",
 	.probe		= pcmciamtd_probe,
 	.remove		= pcmciamtd_detach,
 	.owner		= THIS_MODULE,
@@ -771,8 +741,6 @@
 
 static int __init init_pcmciamtd(void)
 {
-	info(DRIVER_DESC);
-
 	if(bankwidth && bankwidth != 1 && bankwidth != 2) {
 		info("bad bankwidth (%d), using default", bankwidth);
 		bankwidth = 2;
diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index b2828e8..214b03a 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -30,6 +30,8 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/completion.h>
 
 #include <asm/mach/flash.h>
 #include <mach/mxc_nand.h>
@@ -151,7 +153,7 @@
 	int			irq;
 	int			eccsize;
 
-	wait_queue_head_t	irq_waitq;
+	struct completion	op_completion;
 
 	uint8_t			*data_buf;
 	unsigned int		buf_start;
@@ -164,6 +166,7 @@
 	void			(*send_read_id)(struct mxc_nand_host *);
 	uint16_t		(*get_dev_status)(struct mxc_nand_host *);
 	int			(*check_int)(struct mxc_nand_host *);
+	void			(*irq_control)(struct mxc_nand_host *, int);
 };
 
 /* OOB placement block for use with hardware ecc generation */
@@ -216,9 +219,12 @@
 {
 	struct mxc_nand_host *host = dev_id;
 
-	disable_irq_nosync(irq);
+	if (!host->check_int(host))
+		return IRQ_NONE;
 
-	wake_up(&host->irq_waitq);
+	host->irq_control(host, 0);
+
+	complete(&host->op_completion);
 
 	return IRQ_HANDLED;
 }
@@ -245,11 +251,54 @@
 	if (!(tmp & NFC_V1_V2_CONFIG2_INT))
 		return 0;
 
-	writew(tmp & ~NFC_V1_V2_CONFIG2_INT, NFC_V1_V2_CONFIG2);
+	if (!cpu_is_mx21())
+		writew(tmp & ~NFC_V1_V2_CONFIG2_INT, NFC_V1_V2_CONFIG2);
 
 	return 1;
 }
 
+/*
+ * It has been observed that the i.MX21 cannot read the CONFIG2:INT bit
+ * if interrupts are masked (CONFIG1:INT_MSK is set). To handle this, the
+ * driver can enable/disable the irq line rather than simply masking the
+ * interrupts.
+ */
+static void irq_control_mx21(struct mxc_nand_host *host, int activate)
+{
+	if (activate)
+		enable_irq(host->irq);
+	else
+		disable_irq_nosync(host->irq);
+}
+
+static void irq_control_v1_v2(struct mxc_nand_host *host, int activate)
+{
+	uint16_t tmp;
+
+	tmp = readw(NFC_V1_V2_CONFIG1);
+
+	if (activate)
+		tmp &= ~NFC_V1_V2_CONFIG1_INT_MSK;
+	else
+		tmp |= NFC_V1_V2_CONFIG1_INT_MSK;
+
+	writew(tmp, NFC_V1_V2_CONFIG1);
+}
+
+static void irq_control_v3(struct mxc_nand_host *host, int activate)
+{
+	uint32_t tmp;
+
+	tmp = readl(NFC_V3_CONFIG2);
+
+	if (activate)
+		tmp &= ~NFC_V3_CONFIG2_INT_MSK;
+	else
+		tmp |= NFC_V3_CONFIG2_INT_MSK;
+
+	writel(tmp, NFC_V3_CONFIG2);
+}
+
 /* This function polls the NANDFC to wait for the basic operation to
  * complete by checking the INT bit of config2 register.
  */
@@ -259,10 +308,9 @@
 
 	if (useirq) {
 		if (!host->check_int(host)) {
-
-			enable_irq(host->irq);
-
-			wait_event(host->irq_waitq, host->check_int(host));
+			INIT_COMPLETION(host->op_completion);
+			host->irq_control(host, 1);
+			wait_for_completion(&host->op_completion);
 		}
 	} else {
 		while (max_retries-- > 0) {
@@ -799,6 +847,7 @@
 		NFC_V3_CONFIG2_2CMD_PHASES |
 		NFC_V3_CONFIG2_SPAS(mtd->oobsize >> 1) |
 		NFC_V3_CONFIG2_ST_CMD(0x70) |
+		NFC_V3_CONFIG2_INT_MSK |
 		NFC_V3_CONFIG2_NUM_ADDR_PHASE0;
 
 	if (chip->ecc.mode == NAND_ECC_HW)
@@ -1024,6 +1073,10 @@
 		host->send_read_id = send_read_id_v1_v2;
 		host->get_dev_status = get_dev_status_v1_v2;
 		host->check_int = check_int_v1_v2;
+		if (cpu_is_mx21())
+			host->irq_control = irq_control_mx21;
+		else
+			host->irq_control = irq_control_v1_v2;
 	}
 
 	if (nfc_is_v21()) {
@@ -1062,6 +1115,7 @@
 		host->send_read_id = send_read_id_v3;
 		host->check_int = check_int_v3;
 		host->get_dev_status = get_dev_status_v3;
+		host->irq_control = irq_control_v3;
 		oob_smallpage = &nandv2_hw_eccoob_smallpage;
 		oob_largepage = &nandv2_hw_eccoob_largepage;
 	} else
@@ -1093,14 +1147,34 @@
 		this->options |= NAND_USE_FLASH_BBT;
 	}
 
-	init_waitqueue_head(&host->irq_waitq);
+	init_completion(&host->op_completion);
 
 	host->irq = platform_get_irq(pdev, 0);
 
+	/*
+	 * mask the interrupt. For i.MX21 explicitely call
+	 * irq_control_v1_v2 to use the mask bit. We can't call
+	 * disable_irq_nosync() for an interrupt we do not own yet.
+	 */
+	if (cpu_is_mx21())
+		irq_control_v1_v2(host, 0);
+	else
+		host->irq_control(host, 0);
+
 	err = request_irq(host->irq, mxc_nfc_irq, IRQF_DISABLED, DRIVER_NAME, host);
 	if (err)
 		goto eirq;
 
+	host->irq_control(host, 0);
+
+	/*
+	 * Now that the interrupt is disabled make sure the interrupt
+	 * mask bit is cleared on i.MX21. Otherwise we can't read
+	 * the interrupt status bit on this machine.
+	 */
+	if (cpu_is_mx21())
+		irq_control_v1_v2(host, 1);
+
 	/* first scan to find the device and get the page size */
 	if (nand_scan_ident(mtd, 1, NULL)) {
 		err = -ENXIO;
diff --git a/drivers/net/3c527.c b/drivers/net/3c527.c
index 70705d1..eca55c5 100644
--- a/drivers/net/3c527.c
+++ b/drivers/net/3c527.c
@@ -522,7 +522,7 @@
 	lp->tx_len 		= lp->exec_box->data[9];   /* Transmit list count */
 	lp->rx_len 		= lp->exec_box->data[11];  /* Receive list count */
 
-	init_MUTEX_LOCKED(&lp->cmd_mutex);
+	sema_init(&lp->cmd_mutex, 0);
 	init_completion(&lp->execution_cmd);
 	init_completion(&lp->xceiver_cmd);
 
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 2cc81a5..5db667c 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2428,7 +2428,7 @@
 
 config MV643XX_ETH
 	tristate "Marvell Discovery (643XX) and Orion ethernet support"
-	depends on MV64X60 || PPC32 || PLAT_ORION
+	depends on (MV64X60 || PPC32 || PLAT_ORION) && INET
 	select INET_LRO
 	select PHYLIB
 	help
@@ -2803,7 +2803,7 @@
 
 config PASEMI_MAC
 	tristate "PA Semi 1/10Gbit MAC"
-	depends on PPC_PASEMI && PCI
+	depends on PPC_PASEMI && PCI && INET
 	select PHYLIB
 	select INET_LRO
 	help
diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index 1e620e2..efeffdf 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -2170,8 +2170,6 @@
 	dev->irq = sdev->irq;
 	SET_ETHTOOL_OPS(dev, &b44_ethtool_ops);
 
-	netif_carrier_off(dev);
-
 	err = ssb_bus_powerup(sdev->bus, 0);
 	if (err) {
 		dev_err(sdev->dev,
@@ -2213,6 +2211,8 @@
 		goto err_out_powerdown;
 	}
 
+	netif_carrier_off(dev);
+
 	ssb_set_drvdata(sdev, dev);
 
 	/* Chip reset provides power to the b44 MAC & PCI cores, which
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 3b16f62..e953c6a 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5164,6 +5164,15 @@
 		res = dev_alloc_name(bond_dev, "bond%d");
 		if (res < 0)
 			goto out;
+	} else {
+		/*
+		 * If we're given a name to register
+		 * we need to ensure that its not already
+		 * registered
+		 */
+		res = -EEXIST;
+		if (__dev_get_by_name(net, name) != NULL)
+			goto out;
 	}
 
 	res = register_netdevice(bond_dev);
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index a333b42..6372610 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -533,8 +533,15 @@
 	int length = cqe->num_bytes_transfered - 4;	/*remove CRC */
 
 	skb_put(skb, length);
-	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->protocol = eth_type_trans(skb, dev);
+
+	/* The packet was not an IPV4 packet so a complemented checksum was
+	   calculated. The value is found in the Internet Checksum field. */
+	if (cqe->status & EHEA_CQE_BLIND_CKSUM) {
+		skb->ip_summed = CHECKSUM_COMPLETE;
+		skb->csum = csum_unfold(~cqe->inet_checksum_value);
+	} else
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
 }
 
 static inline struct sk_buff *get_skb_by_index(struct sk_buff **skb_array,
diff --git a/drivers/net/ehea/ehea_qmr.h b/drivers/net/ehea/ehea_qmr.h
index f608a6c..3810473 100644
--- a/drivers/net/ehea/ehea_qmr.h
+++ b/drivers/net/ehea/ehea_qmr.h
@@ -150,6 +150,7 @@
 #define EHEA_CQE_TYPE_RQ           0x60
 #define EHEA_CQE_STAT_ERR_MASK     0x700F
 #define EHEA_CQE_STAT_FAT_ERR_MASK 0xF
+#define EHEA_CQE_BLIND_CKSUM       0x8000
 #define EHEA_CQE_STAT_ERR_TCP      0x4000
 #define EHEA_CQE_STAT_ERR_IP       0x2000
 #define EHEA_CQE_STAT_ERR_CRC      0x1000
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 768b840..cce32d4 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -678,24 +678,37 @@
 {
 	struct fec_enet_private *fep = netdev_priv(dev);
 	struct phy_device *phy_dev = NULL;
-	int ret;
+	char mdio_bus_id[MII_BUS_ID_SIZE];
+	char phy_name[MII_BUS_ID_SIZE + 3];
+	int phy_id;
 
 	fep->phy_dev = NULL;
 
-	/* find the first phy */
-	phy_dev = phy_find_first(fep->mii_bus);
-	if (!phy_dev) {
-		printk(KERN_ERR "%s: no PHY found\n", dev->name);
-		return -ENODEV;
+	/* check for attached phy */
+	for (phy_id = 0; (phy_id < PHY_MAX_ADDR); phy_id++) {
+		if ((fep->mii_bus->phy_mask & (1 << phy_id)))
+			continue;
+		if (fep->mii_bus->phy_map[phy_id] == NULL)
+			continue;
+		if (fep->mii_bus->phy_map[phy_id]->phy_id == 0)
+			continue;
+		strncpy(mdio_bus_id, fep->mii_bus->id, MII_BUS_ID_SIZE);
+		break;
 	}
 
-	/* attach the mac to the phy */
-	ret = phy_connect_direct(dev, phy_dev,
-			     &fec_enet_adjust_link, 0,
-			     PHY_INTERFACE_MODE_MII);
-	if (ret) {
-		printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name);
-		return ret;
+	if (phy_id >= PHY_MAX_ADDR) {
+		printk(KERN_INFO "%s: no PHY, assuming direct connection "
+			"to switch\n", dev->name);
+		strncpy(mdio_bus_id, "0", MII_BUS_ID_SIZE);
+		phy_id = 0;
+	}
+
+	snprintf(phy_name, MII_BUS_ID_SIZE, PHY_ID_FMT, mdio_bus_id, phy_id);
+	phy_dev = phy_connect(dev, phy_name, &fec_enet_adjust_link, 0,
+		PHY_INTERFACE_MODE_MII);
+	if (IS_ERR(phy_dev)) {
+		printk(KERN_ERR "%s: could not attach to PHY\n", dev->name);
+		return PTR_ERR(phy_dev);
 	}
 
 	/* mask with MAC supported features */
@@ -738,7 +751,7 @@
 	fep->mii_bus->read = fec_enet_mdio_read;
 	fep->mii_bus->write = fec_enet_mdio_write;
 	fep->mii_bus->reset = fec_enet_mdio_reset;
-	snprintf(fep->mii_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
+	snprintf(fep->mii_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id + 1);
 	fep->mii_bus->priv = fep;
 	fep->mii_bus->parent = &pdev->dev;
 
@@ -1311,6 +1324,9 @@
 	if (ret)
 		goto failed_mii_init;
 
+	/* Carrier starts down, phylib will bring it up */
+	netif_carrier_off(ndev);
+
 	ret = register_netdev(ndev);
 	if (ret)
 		goto failed_register;
diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c
index 4b52c76..3e5d0b6 100644
--- a/drivers/net/hamradio/6pack.c
+++ b/drivers/net/hamradio/6pack.c
@@ -608,7 +608,7 @@
 
 	spin_lock_init(&sp->lock);
 	atomic_set(&sp->refcnt, 1);
-	init_MUTEX_LOCKED(&sp->dead_sem);
+	sema_init(&sp->dead_sem, 0);
 
 	/* !!! length of the buffers. MTU is IP MTU, not PACLEN!  */
 
diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c
index 66e88bd..4c62839 100644
--- a/drivers/net/hamradio/mkiss.c
+++ b/drivers/net/hamradio/mkiss.c
@@ -747,7 +747,7 @@
 
 	spin_lock_init(&ax->buflock);
 	atomic_set(&ax->refcnt, 1);
-	init_MUTEX_LOCKED(&ax->dead_sem);
+	sema_init(&ax->dead_sem, 0);
 
 	ax->tty = tty;
 	tty->disc_data = ax;
diff --git a/drivers/net/irda/sir_dev.c b/drivers/net/irda/sir_dev.c
index 1b051da..51d7444 100644
--- a/drivers/net/irda/sir_dev.c
+++ b/drivers/net/irda/sir_dev.c
@@ -909,7 +909,7 @@
 	dev->tx_skb = NULL;
 
 	spin_lock_init(&dev->tx_lock);
-	init_MUTEX(&dev->fsm.sem);
+	sema_init(&dev->fsm.sem, 1);
 
 	dev->drv = drv;
 	dev->netdev = ndev;
diff --git a/drivers/net/pcmcia/3c574_cs.c b/drivers/net/pcmcia/3c574_cs.c
index c683f77..ff824e1 100644
--- a/drivers/net/pcmcia/3c574_cs.c
+++ b/drivers/net/pcmcia/3c574_cs.c
@@ -87,7 +87,6 @@
 #include <linux/bitops.h>
 #include <linux/mii.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -280,25 +279,15 @@
 	spin_lock_init(&lp->window_lock);
 	link->resource[0]->end = 32;
 	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-	link->conf.ConfigIndex = 1;
+	link->config_flags |= CONF_ENABLE_IRQ;
+	link->config_index = 1;
 
 	dev->netdev_ops = &el3_netdev_ops;
 	SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
 	dev->watchdog_timeo = TX_TIMEOUT;
 
 	return tc574_config(link);
-} /* tc574_attach */
-
-/*
-
-	This deletes a driver "instance".  The device is de-registered
-	with Card Services.  If it has been released, all local data
-	structures are freed.  Otherwise, the structures will be freed
-	when the device is released.
-
-*/
+}
 
 static void tc574_detach(struct pcmcia_device *link)
 {
@@ -313,12 +302,6 @@
 	free_netdev(dev);
 } /* tc574_detach */
 
-/*
-	tc574_config() is scheduled to run after a CARD_INSERTION event
-	is received, to configure the PCMCIA socket, and to make the
-	ethernet device available to the system.
-*/
-
 static const char *ram_split[] = {"5:3", "3:1", "1:1", "3:5"};
 
 static int tc574_config(struct pcmcia_device *link)
@@ -352,7 +335,7 @@
 	if (ret)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -465,12 +448,6 @@
 
 } /* tc574_config */
 
-/*
-	After a card is removed, tc574_release() will unregister the net
-	device, and release the PCMCIA configuration.  If the device is
-	still open, this will be postponed until it is closed.
-*/
-
 static void tc574_release(struct pcmcia_device *link)
 {
 	pcmcia_disable_device(link);
@@ -1198,9 +1175,7 @@
 
 static struct pcmcia_driver tc574_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "3c574_cs",
-	},
+	.name		= "3c574_cs",
 	.probe		= tc574_probe,
 	.remove		= tc574_detach,
 	.id_table       = tc574_ids,
diff --git a/drivers/net/pcmcia/3c589_cs.c b/drivers/net/pcmcia/3c589_cs.c
index 61f9cf2..a07e222 100644
--- a/drivers/net/pcmcia/3c589_cs.c
+++ b/drivers/net/pcmcia/3c589_cs.c
@@ -41,7 +41,6 @@
 #include <linux/bitops.h>
 #include <linux/jiffies.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -176,14 +175,6 @@
 
 static void tc589_detach(struct pcmcia_device *p_dev);
 
-/*======================================================================
-
-    tc589_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static const struct net_device_ops el3_netdev_ops = {
 	.ndo_open		= el3_open,
 	.ndo_stop		= el3_close,
@@ -216,9 +207,8 @@
     link->resource[0]->end = 16;
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
 
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.ConfigIndex = 1;
+    link->config_flags |= CONF_ENABLE_IRQ;
+    link->config_index = 1;
 
     dev->netdev_ops = &el3_netdev_ops;
     dev->watchdog_timeo = TX_TIMEOUT;
@@ -226,16 +216,7 @@
     SET_ETHTOOL_OPS(dev, &netdev_ethtool_ops);
 
     return tc589_config(link);
-} /* tc589_attach */
-
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
+}
 
 static void tc589_detach(struct pcmcia_device *link)
 {
@@ -250,14 +231,6 @@
     free_netdev(dev);
 } /* tc589_detach */
 
-/*======================================================================
-
-    tc589_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-
-======================================================================*/
-
 static int tc589_config(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
@@ -294,7 +267,7 @@
     if (ret)
 	    goto failed;
 
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -352,14 +325,6 @@
     return -ENODEV;
 } /* tc589_config */
 
-/*======================================================================
-
-    After a card is removed, tc589_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void tc589_release(struct pcmcia_device *link)
 {
 	pcmcia_disable_device(link);
@@ -955,9 +920,7 @@
 
 static struct pcmcia_driver tc589_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "3c589_cs",
-	},
+	.name		= "3c589_cs",
 	.probe		= tc589_probe,
 	.remove		= tc589_detach,
 	.id_table	= tc589_ids,
diff --git a/drivers/net/pcmcia/axnet_cs.c b/drivers/net/pcmcia/axnet_cs.c
index 5f05ffb..9e8b28b 100644
--- a/drivers/net/pcmcia/axnet_cs.c
+++ b/drivers/net/pcmcia/axnet_cs.c
@@ -39,7 +39,6 @@
 #include <linux/mii.h>
 #include "../8390.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -140,14 +139,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/*======================================================================
-
-    axnet_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int axnet_probe(struct pcmcia_device *link)
 {
     axnet_dev_t *info;
@@ -166,8 +157,7 @@
     info = PRIV(dev);
     info->p_dev = link;
     link->priv = dev;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ;
 
     dev->netdev_ops = &axnet_netdev_ops;
 
@@ -177,15 +167,6 @@
     return axnet_config(link);
 } /* axnet_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void axnet_detach(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
@@ -231,7 +212,7 @@
     };
 
     /* Not much of a test, but the alternatives are messy */
-    if (link->conf.ConfigBase != 0x03c0)
+    if (link->config_base != 0x03c0)
 	return 0;
 
     axnet_reset_8390(dev);
@@ -248,14 +229,6 @@
     return 1;
 } /* get_prom */
 
-/*======================================================================
-
-    axnet_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-
-======================================================================*/
-
 static int try_io_port(struct pcmcia_device *link)
 {
     int j, ret;
@@ -286,35 +259,16 @@
     }
 }
 
-static int axnet_configcheck(struct pcmcia_device *p_dev,
-			     cistpl_cftable_entry_t *cfg,
-			     cistpl_cftable_entry_t *dflt,
-			     unsigned int vcc,
-			     void *priv_data)
+static int axnet_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
-	int i;
-	cistpl_io_t *io = &cfg->io;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	if (cfg->index == 0 || cfg->io.nwin == 0)
+	p_dev->config_index = 0x05;
+	if (p_dev->resource[0]->end + p_dev->resource[1]->end < 32)
 		return -ENODEV;
 
-	p_dev->conf.ConfigIndex = 0x05;
-	/* For multifunction cards, by convention, we configure the
-	   network function with window 0, and serial with window 1 */
-	if (io->nwin > 1) {
-		i = (io->win[1].len > io->win[0].len);
-		p_dev->resource[1]->start = io->win[1-i].base;
-		p_dev->resource[1]->end = io->win[1-i].len;
-	} else {
-		i = p_dev->resource[1]->end = 0;
-	}
-	p_dev->resource[0]->start = io->win[i].base;
-	p_dev->resource[0]->end = io->win[i].len;
-	p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-	if (p_dev->resource[0]->end + p_dev->resource[1]->end >= 32)
-		return try_io_port(p_dev);
-
-	return -ENODEV;
+	return try_io_port(p_dev);
 }
 
 static int axnet_config(struct pcmcia_device *link)
@@ -326,20 +280,19 @@
     dev_dbg(&link->dev, "axnet_config(0x%p)\n", link);
 
     /* don't trust the CIS on this; Linksys got it wrong */
-    link->conf.Present = 0x63;
+    link->config_regs = 0x63;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
     ret = pcmcia_loop_config(link, axnet_configcheck, NULL);
     if (ret != 0)
 	goto failed;
 
     if (!link->irq)
 	    goto failed;
+
+    if (resource_size(link->resource[1]) == 8)
+	link->config_flags |= CONF_ENABLE_SPKR;
     
-    if (resource_size(link->resource[1]) == 8) {
-	link->conf.Attributes |= CONF_ENABLE_SPKR;
-	link->conf.Status = CCSR_AUDIO_ENA;
-    }
-    
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -414,14 +367,6 @@
     return -ENODEV;
 } /* axnet_config */
 
-/*======================================================================
-
-    After a card is removed, axnet_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void axnet_release(struct pcmcia_device *link)
 {
 	pcmcia_disable_device(link);
@@ -783,9 +728,7 @@
 
 static struct pcmcia_driver axnet_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "axnet_cs",
-	},
+	.name		= "axnet_cs",
 	.probe		= axnet_probe,
 	.remove		= axnet_detach,
 	.id_table       = axnet_ids,
diff --git a/drivers/net/pcmcia/com20020_cs.c b/drivers/net/pcmcia/com20020_cs.c
index 3c400cf..b706a72 100644
--- a/drivers/net/pcmcia/com20020_cs.c
+++ b/drivers/net/pcmcia/com20020_cs.c
@@ -43,7 +43,6 @@
 #include <linux/arcdevice.h>
 #include <linux/com20020.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -123,14 +122,6 @@
     struct net_device       *dev;
 } com20020_dev_t;
 
-/*======================================================================
-
-    com20020_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int com20020_probe(struct pcmcia_device *p_dev)
 {
     com20020_dev_t *info;
@@ -160,8 +151,7 @@
 
     p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
     p_dev->resource[0]->end = 16;
-    p_dev->conf.Attributes = CONF_ENABLE_IRQ;
-    p_dev->conf.IntType = INT_MEMORY_AND_IO;
+    p_dev->config_flags |= CONF_ENABLE_IRQ;
 
     info->dev = dev;
     p_dev->priv = info;
@@ -174,15 +164,6 @@
     return -ENOMEM;
 } /* com20020_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void com20020_detach(struct pcmcia_device *link)
 {
     struct com20020_dev_t *info = link->priv;
@@ -221,14 +202,6 @@
 
 } /* com20020_detach */
 
-/*======================================================================
-
-    com20020_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
 static int com20020_config(struct pcmcia_device *link)
 {
     struct arcnet_local *lp;
@@ -282,7 +255,7 @@
 
     dev->irq = link->irq;
 
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -316,14 +289,6 @@
     return -ENODEV;
 } /* com20020_config */
 
-/*======================================================================
-
-    After a card is removed, com20020_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void com20020_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "com20020_release\n");
@@ -366,9 +331,7 @@
 
 static struct pcmcia_driver com20020_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "com20020_cs",
-	},
+	.name		= "com20020_cs",
 	.probe		= com20020_probe,
 	.remove		= com20020_detach,
 	.id_table	= com20020_ids,
diff --git a/drivers/net/pcmcia/fmvj18x_cs.c b/drivers/net/pcmcia/fmvj18x_cs.c
index 98fffb0..1c32759 100644
--- a/drivers/net/pcmcia/fmvj18x_cs.c
+++ b/drivers/net/pcmcia/fmvj18x_cs.c
@@ -49,7 +49,6 @@
 #include <linux/ioport.h>
 #include <linux/crc32.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -252,8 +251,7 @@
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 
     /* General socket configuration */
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ;
 
     dev->netdev_ops = &fjn_netdev_ops;
     dev->watchdog_timeo = TX_TIMEOUT;
@@ -313,7 +311,7 @@
 	ret = pcmcia_request_io(link);
 	if (ret == 0) {
 	    /* calculate ConfigIndex value */
-	    link->conf.ConfigIndex = 
+	    link->config_index =
 		((link->resource[0]->start & 0x0f0) >> 3) | 0x22;
 	    return ret;
 	}
@@ -321,11 +319,7 @@
     return ret;	/* RequestIO failed */
 }
 
-static int fmvj18x_ioprobe(struct pcmcia_device *p_dev,
-			   cistpl_cftable_entry_t *cfg,
-			   cistpl_cftable_entry_t *dflt,
-			   unsigned int vcc,
-			   void *priv_data)
+static int fmvj18x_ioprobe(struct pcmcia_device *p_dev, void *priv_data)
 {
 	return 0; /* strange, but that's what the code did already before... */
 }
@@ -362,28 +356,28 @@
 		link->card_id == PRODID_TDK_NP9610 ||
 		link->card_id == PRODID_TDK_MN3200) {
 		/* MultiFunction Card */
-		link->conf.ConfigBase = 0x800;
-		link->conf.ConfigIndex = 0x47;
+		link->config_base = 0x800;
+		link->config_index = 0x47;
 		link->resource[1]->end = 8;
 	    }
 	    break;
 	case MANFID_NEC:
 	    cardtype = NEC; /* MultiFunction Card */
-	    link->conf.ConfigBase = 0x800;
-	    link->conf.ConfigIndex = 0x47;
+	    link->config_base = 0x800;
+	    link->config_index = 0x47;
 	    link->resource[1]->end = 8;
 	    break;
 	case MANFID_KME:
 	    cardtype = KME; /* MultiFunction Card */
-	    link->conf.ConfigBase = 0x800;
-	    link->conf.ConfigIndex = 0x47;
+	    link->config_base = 0x800;
+	    link->config_index = 0x47;
 	    link->resource[1]->end = 8;
 	    break;
 	case MANFID_CONTEC:
 	    cardtype = CONTEC;
 	    break;
 	case MANFID_FUJITSU:
-	    if (link->conf.ConfigBase == 0x0fe0)
+	    if (link->config_base == 0x0fe0)
 		cardtype = MBH10302;
 	    else if (link->card_id == PRODID_FUJITSU_MBH10302) 
                 /* RATOC REX-5588/9822/4886's PRODID are 0004(=MBH10302),
@@ -403,10 +397,10 @@
 	case MANFID_FUJITSU:
 	    if (link->card_id == PRODID_FUJITSU_MBH10304) {
 		cardtype = XXX10304;    /* MBH10304 with buggy CIS */
-	        link->conf.ConfigIndex = 0x20;
+		link->config_index = 0x20;
 	    } else {
 		cardtype = MBH10302;    /* NextCom NC5310, etc. */
-		link->conf.ConfigIndex = 1;
+		link->config_index = 1;
 	    }
 	    break;
 	case MANFID_UNGERMANN:
@@ -414,7 +408,7 @@
 	    break;
 	default:
 	    cardtype = MBH10302;
-	    link->conf.ConfigIndex = 1;
+	    link->config_index = 1;
 	}
     }
 
@@ -432,7 +426,7 @@
     ret = pcmcia_request_irq(link, fjn_interrupt);
     if (ret)
 	    goto failed;
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -544,20 +538,18 @@
 
 static int fmvj18x_get_hwinfo(struct pcmcia_device *link, u_char *node_id)
 {
-    win_req_t req;
     u_char __iomem *base;
     int i, j;
 
     /* Allocate a small memory window */
-    req.Attributes = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
-    req.Base = 0; req.Size = 0;
-    req.AccessSpeed = 0;
-    i = pcmcia_request_window(link, &req, &link->win);
+    link->resource[2]->flags |= WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
+    link->resource[2]->start = 0; link->resource[2]->end = 0;
+    i = pcmcia_request_window(link, link->resource[2], 0);
     if (i != 0)
 	return -1;
 
-    base = ioremap(req.Base, req.Size);
-    pcmcia_map_mem_page(link, link->win, 0);
+    base = ioremap(link->resource[2]->start, resource_size(link->resource[2]));
+    pcmcia_map_mem_page(link, link->resource[2], 0);
 
     /*
      *  MBH10304 CISTPL_FUNCE_LAN_NODE_ID format
@@ -582,7 +574,7 @@
     }
 
     iounmap(base);
-    j = pcmcia_release_window(link, link->win);
+    j = pcmcia_release_window(link, link->resource[2]);
     return (i != 0x200) ? 0 : -1;
 
 } /* fmvj18x_get_hwinfo */
@@ -590,27 +582,26 @@
 
 static int fmvj18x_setup_mfc(struct pcmcia_device *link)
 {
-    win_req_t req;
     int i;
     struct net_device *dev = link->priv;
     unsigned int ioaddr;
     local_info_t *lp = netdev_priv(dev);
 
     /* Allocate a small memory window */
-    req.Attributes = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
-    req.Base = 0; req.Size = 0;
-    req.AccessSpeed = 0;
-    i = pcmcia_request_window(link, &req, &link->win);
+    link->resource[3]->flags = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
+    link->resource[3]->start = link->resource[3]->end = 0;
+    i = pcmcia_request_window(link, link->resource[3], 0);
     if (i != 0)
 	return -1;
 
-    lp->base = ioremap(req.Base, req.Size);
+    lp->base = ioremap(link->resource[3]->start,
+		       resource_size(link->resource[3]));
     if (lp->base == NULL) {
 	printk(KERN_NOTICE "fmvj18x_cs: ioremap failed\n");
 	return -1;
     }
 
-    i = pcmcia_map_mem_page(link, link->win, 0);
+    i = pcmcia_map_mem_page(link, link->resource[3], 0);
     if (i != 0) {
 	iounmap(lp->base);
 	lp->base = NULL;
@@ -638,7 +629,6 @@
     struct net_device *dev = link->priv;
     local_info_t *lp = netdev_priv(dev);
     u_char __iomem *tmp;
-    int j;
 
     dev_dbg(&link->dev, "fmvj18x_release\n");
 
@@ -646,7 +636,6 @@
 	tmp = lp->base;
 	lp->base = NULL;    /* set NULL before iounmap */
 	iounmap(tmp);
-	j = pcmcia_release_window(link, link->win);
     }
 
     pcmcia_disable_device(link);
@@ -708,9 +697,7 @@
 
 static struct pcmcia_driver fmvj18x_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "fmvj18x_cs",
-	},
+	.name		= "fmvj18x_cs",
 	.probe		= fmvj18x_probe,
 	.remove		= fmvj18x_detach,
 	.id_table       = fmvj18x_ids,
diff --git a/drivers/net/pcmcia/ibmtr_cs.c b/drivers/net/pcmcia/ibmtr_cs.c
index b0d06a3..bf7dff9 100644
--- a/drivers/net/pcmcia/ibmtr_cs.c
+++ b/drivers/net/pcmcia/ibmtr_cs.c
@@ -57,7 +57,6 @@
 #include <linux/trdevice.h>
 #include <linux/ibmtr.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -102,9 +101,8 @@
 
 typedef struct ibmtr_dev_t {
 	struct pcmcia_device	*p_dev;
-    struct net_device	*dev;
-    window_handle_t     sram_win_handle;
-    struct tok_info	*ti;
+	struct net_device	*dev;
+	struct tok_info		*ti;
 } ibmtr_dev_t;
 
 static void netdev_get_drvinfo(struct net_device *dev,
@@ -123,14 +121,6 @@
 	return tok_interrupt(irq, dev);
 };
 
-/*======================================================================
-
-    ibmtr_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int __devinit ibmtr_attach(struct pcmcia_device *link)
 {
     ibmtr_dev_t *info;
@@ -153,9 +143,8 @@
 
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
     link->resource[0]->end = 4;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.Present = PRESENT_OPTION;
+    link->config_flags |= CONF_ENABLE_IRQ;
+    link->config_regs = PRESENT_OPTION;
 
     info->dev = dev;
 
@@ -164,15 +153,6 @@
     return ibmtr_config(link);
 } /* ibmtr_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void ibmtr_detach(struct pcmcia_device *link)
 {
     struct ibmtr_dev_t *info = link->priv;
@@ -197,26 +177,17 @@
     kfree(info);
 } /* ibmtr_detach */
 
-/*======================================================================
-
-    ibmtr_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    token-ring device available to the system.
-
-======================================================================*/
-
 static int __devinit ibmtr_config(struct pcmcia_device *link)
 {
     ibmtr_dev_t *info = link->priv;
     struct net_device *dev = info->dev;
     struct tok_info *ti = netdev_priv(dev);
-    win_req_t req;
     int i, ret;
 
     dev_dbg(&link->dev, "ibmtr_config\n");
 
-    link->conf.ConfigIndex = 0x61;
     link->io_lines = 16;
+    link->config_index = 0x61;
 
     /* Determine if this is PRIMARY or ALTERNATE. */
 
@@ -240,39 +211,39 @@
     ti->global_int_enable=GLOBAL_INT_ENABLE+((dev->irq==9) ? 2 : dev->irq);
 
     /* Allocate the MMIO memory window */
-    req.Attributes = WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
-    req.Attributes |= WIN_USE_WAIT;
-    req.Base = 0; 
-    req.Size = 0x2000;
-    req.AccessSpeed = 250;
-    ret = pcmcia_request_window(link, &req, &link->win);
+    link->resource[2]->flags |= WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
+    link->resource[2]->flags |= WIN_USE_WAIT;
+    link->resource[2]->start = 0;
+    link->resource[2]->end = 0x2000;
+    ret = pcmcia_request_window(link, link->resource[2], 250);
     if (ret)
 	    goto failed;
 
-    ret = pcmcia_map_mem_page(link, link->win, mmiobase);
+    ret = pcmcia_map_mem_page(link, link->resource[2], mmiobase);
     if (ret)
 	    goto failed;
-    ti->mmio = ioremap(req.Base, req.Size);
+    ti->mmio = ioremap(link->resource[2]->start,
+		    resource_size(link->resource[2]));
 
     /* Allocate the SRAM memory window */
-    req.Attributes = WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
-    req.Attributes |= WIN_USE_WAIT;
-    req.Base = 0;
-    req.Size = sramsize * 1024;
-    req.AccessSpeed = 250;
-    ret = pcmcia_request_window(link, &req, &info->sram_win_handle);
+    link->resource[3]->flags = WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
+    link->resource[3]->flags |= WIN_USE_WAIT;
+    link->resource[3]->start = 0;
+    link->resource[3]->end = sramsize * 1024;
+    ret = pcmcia_request_window(link, link->resource[3], 250);
     if (ret)
 	    goto failed;
 
-    ret = pcmcia_map_mem_page(link, info->sram_win_handle, srambase);
+    ret = pcmcia_map_mem_page(link, link->resource[3], srambase);
     if (ret)
 	    goto failed;
 
     ti->sram_base = srambase >> 12;
-    ti->sram_virt = ioremap(req.Base, req.Size);
-    ti->sram_phys = req.Base;
+    ti->sram_virt = ioremap(link->resource[3]->start,
+		    resource_size(link->resource[3]));
+    ti->sram_phys = link->resource[3]->start;
 
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -301,14 +272,6 @@
     return -ENODEV;
 } /* ibmtr_config */
 
-/*======================================================================
-
-    After a card is removed, ibmtr_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void ibmtr_release(struct pcmcia_device *link)
 {
 	ibmtr_dev_t *info = link->priv;
@@ -316,7 +279,7 @@
 
 	dev_dbg(&link->dev, "ibmtr_release\n");
 
-	if (link->win) {
+	if (link->resource[2]->end) {
 		struct tok_info *ti = netdev_priv(dev);
 		iounmap(ti->mmio);
 	}
@@ -398,9 +361,7 @@
 
 static struct pcmcia_driver ibmtr_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "ibmtr_cs",
-	},
+	.name		= "ibmtr_cs",
 	.probe		= ibmtr_attach,
 	.remove		= ibmtr_detach,
 	.id_table       = ibmtr_ids,
diff --git a/drivers/net/pcmcia/nmclan_cs.c b/drivers/net/pcmcia/nmclan_cs.c
index 68f2dee..1eca4f5 100644
--- a/drivers/net/pcmcia/nmclan_cs.c
+++ b/drivers/net/pcmcia/nmclan_cs.c
@@ -146,7 +146,6 @@
 #include <linux/ioport.h>
 #include <linux/bitops.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
@@ -435,13 +434,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/* ----------------------------------------------------------------------------
-nmclan_attach
-	Creates an "instance" of the driver, allocating local data
-	structures for one device.  The device is registered with Card
-	Services.
----------------------------------------------------------------------------- */
-
 static int nmclan_probe(struct pcmcia_device *link)
 {
     mace_private *lp;
@@ -460,10 +452,9 @@
     spin_lock_init(&lp->bank_lock);
     link->resource[0]->end = 32;
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.ConfigIndex = 1;
-    link->conf.Present = PRESENT_OPTION;
+    link->config_flags |= CONF_ENABLE_IRQ;
+    link->config_index = 1;
+    link->config_regs = PRESENT_OPTION;
 
     lp->tx_free_frames=AM2150_MAX_TX_FRAMES;
 
@@ -474,14 +465,6 @@
     return nmclan_config(link);
 } /* nmclan_attach */
 
-/* ----------------------------------------------------------------------------
-nmclan_detach
-	This deletes a driver "instance".  The device is de-registered
-	with Card Services.  If it has been released, all local data
-	structures are freed.  Otherwise, the structures will be freed
-	when the device is released.
----------------------------------------------------------------------------- */
-
 static void nmclan_detach(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
@@ -625,13 +608,6 @@
   return 0;
 } /* mace_init */
 
-/* ----------------------------------------------------------------------------
-nmclan_config
-	This routine is scheduled to run after a CARD_INSERTION event
-	is received, to configure the PCMCIA socket, and to make the
-	ethernet device available to the system.
----------------------------------------------------------------------------- */
-
 static int nmclan_config(struct pcmcia_device *link)
 {
   struct net_device *dev = link->priv;
@@ -650,7 +626,7 @@
   ret = pcmcia_request_exclusive_irq(link, mace_interrupt);
   if (ret)
 	  goto failed;
-  ret = pcmcia_request_configuration(link, &link->conf);
+  ret = pcmcia_enable_device(link);
   if (ret)
 	  goto failed;
 
@@ -712,12 +688,6 @@
 	return -ENODEV;
 } /* nmclan_config */
 
-/* ----------------------------------------------------------------------------
-nmclan_release
-	After a card is removed, nmclan_release() will unregister the
-	net device, and release the PCMCIA configuration.  If the device
-	is still open, this will be postponed until it is closed.
----------------------------------------------------------------------------- */
 static void nmclan_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "nmclan_release\n");
@@ -1535,9 +1505,7 @@
 
 static struct pcmcia_driver nmclan_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "nmclan_cs",
-	},
+	.name		= "nmclan_cs",
 	.probe		= nmclan_probe,
 	.remove		= nmclan_detach,
 	.id_table       = nmclan_ids,
diff --git a/drivers/net/pcmcia/pcnet_cs.c b/drivers/net/pcmcia/pcnet_cs.c
index f9b509a..5d7d1d3 100644
--- a/drivers/net/pcmcia/pcnet_cs.c
+++ b/drivers/net/pcmcia/pcnet_cs.c
@@ -42,7 +42,6 @@
 #include <linux/mii.h>
 #include "../8390.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -238,14 +237,6 @@
 #endif
 };
 
-/*======================================================================
-
-    pcnet_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int pcnet_probe(struct pcmcia_device *link)
 {
     pcnet_dev_t *info;
@@ -260,23 +251,13 @@
     info->p_dev = link;
     link->priv = dev;
 
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
     dev->netdev_ops = &pcnet_netdev_ops;
 
     return pcnet_config(link);
 } /* pcnet_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void pcnet_detach(struct pcmcia_device *link)
 {
 	struct net_device *dev = link->priv;
@@ -300,22 +281,22 @@
 static hw_info_t *get_hwinfo(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
-    win_req_t req;
     u_char __iomem *base, *virt;
     int i, j;
 
     /* Allocate a small memory window */
-    req.Attributes = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
-    req.Base = 0; req.Size = 0;
-    req.AccessSpeed = 0;
-    i = pcmcia_request_window(link, &req, &link->win);
+    link->resource[2]->flags |= WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
+    link->resource[2]->start = 0; link->resource[2]->end = 0;
+    i = pcmcia_request_window(link, link->resource[2], 0);
     if (i != 0)
 	return NULL;
 
-    virt = ioremap(req.Base, req.Size);
+    virt = ioremap(link->resource[2]->start,
+	    resource_size(link->resource[2]));
     for (i = 0; i < NR_INFO; i++) {
-	pcmcia_map_mem_page(link, link->win, hw_info[i].offset & ~(req.Size-1));
-	base = &virt[hw_info[i].offset & (req.Size-1)];
+	pcmcia_map_mem_page(link, link->resource[2],
+		hw_info[i].offset & ~(resource_size(link->resource[2])-1));
+	base = &virt[hw_info[i].offset & (resource_size(link->resource[2])-1)];
 	if ((readb(base+0) == hw_info[i].a0) &&
 	    (readb(base+2) == hw_info[i].a1) &&
 	    (readb(base+4) == hw_info[i].a2)) {
@@ -326,7 +307,7 @@
     }
 
     iounmap(virt);
-    j = pcmcia_release_window(link, link->win);
+    j = pcmcia_release_window(link, link->resource[2]);
     return (i < NR_INFO) ? hw_info+i : NULL;
 } /* get_hwinfo */
 
@@ -421,7 +402,7 @@
     int i, j;
 
     /* Not much of a test, but the alternatives are messy */
-    if (link->conf.ConfigBase != 0x03c0)
+    if (link->config_base != 0x03c0)
 	return NULL;
 
     outb_p(0x01, ioaddr + EN0_DCFG);	/* Set word-wide access. */
@@ -463,14 +444,6 @@
     return &default_info;
 } /* get_hwired */
 
-/*======================================================================
-
-    pcnet_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-
-======================================================================*/
-
 static int try_io_port(struct pcmcia_device *link)
 {
     int j, ret;
@@ -502,43 +475,22 @@
     }
 }
 
-static int pcnet_confcheck(struct pcmcia_device *p_dev,
-			   cistpl_cftable_entry_t *cfg,
-			   cistpl_cftable_entry_t *dflt,
-			   unsigned int vcc,
-			   void *priv_data)
+static int pcnet_confcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int *priv = priv_data;
 	int try = (*priv & 0x1);
-	int i;
-	cistpl_io_t *io = &cfg->io;
 
-	if (cfg->index == 0 || cfg->io.nwin == 0)
+	*priv &= (p_dev->resource[2]->end >= 0x4000) ? 0x10 : ~0x10;
+
+	if (p_dev->config_index == 0)
 		return -EINVAL;
 
-	/* For multifunction cards, by convention, we configure the
-	   network function with window 0, and serial with window 1 */
-	if (io->nwin > 1) {
-		i = (io->win[1].len > io->win[0].len);
-		p_dev->resource[1]->start = io->win[1-i].base;
-		p_dev->resource[1]->end = io->win[1-i].len;
-	} else {
-		i = p_dev->resource[1]->end = 0;
-	}
+	if (p_dev->resource[0]->end + p_dev->resource[1]->end < 32)
+		return -EINVAL;
 
-	*priv &= ((cfg->mem.nwin == 1) &&
-		  (cfg->mem.win[0].len >= 0x4000)) ? 0x10 : ~0x10;
-
-	p_dev->resource[0]->start = io->win[i].base;
-	p_dev->resource[0]->end = io->win[i].len;
-	if (!try)
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-	else
+	if (try)
 		p_dev->io_lines = 16;
-	if (p_dev->resource[0]->end + p_dev->resource[1]->end >= 32)
-		return try_io_port(p_dev);
-
-	return -EINVAL;
+	return try_io_port(p_dev);
 }
 
 static hw_info_t *pcnet_try_config(struct pcmcia_device *link,
@@ -560,15 +512,14 @@
 	if (!link->irq)
 		return NULL;
 
-	if (resource_size(link->resource[1]) == 8) {
-		link->conf.Attributes |= CONF_ENABLE_SPKR;
-		link->conf.Status = CCSR_AUDIO_ENA;
-	}
+	if (resource_size(link->resource[1]) == 8)
+		link->config_flags |= CONF_ENABLE_SPKR;
+
 	if ((link->manf_id == MANFID_IBM) &&
 	    (link->card_id == PRODID_IBM_HOME_AND_AWAY))
-		link->conf.ConfigIndex |= 0x10;
+		link->config_index |= 0x10;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		return NULL;
 
@@ -583,7 +534,7 @@
 	} else
 		dev->if_port = 0;
 
-	if ((link->conf.ConfigBase == 0x03c0) &&
+	if ((link->config_base == 0x03c0) &&
 	    (link->manf_id == 0x149) && (link->card_id == 0xc1ab)) {
 		dev_info(&link->dev,
 			"this is an AX88190 card - use axnet_cs instead.\n");
@@ -689,14 +640,6 @@
     return -ENODEV;
 } /* pcnet_config */
 
-/*======================================================================
-
-    After a card is removed, pcnet_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void pcnet_release(struct pcmcia_device *link)
 {
 	pcnet_dev_t *info = PRIV(link->priv);
@@ -709,15 +652,6 @@
 	pcmcia_disable_device(link);
 }
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.  A CARD_REMOVAL event
-    also sets some flags to discourage the net drivers from trying
-    to talk to the card any more.
-
-======================================================================*/
-
 static int pcnet_suspend(struct pcmcia_device *link)
 {
 	struct net_device *dev = link->priv;
@@ -1486,7 +1420,6 @@
 {
     struct net_device *dev = link->priv;
     pcnet_dev_t *info = PRIV(dev);
-    win_req_t req;
     int i, window_size, offset, ret;
 
     window_size = (stop_pg - start_pg) << 8;
@@ -1497,22 +1430,22 @@
     window_size = roundup_pow_of_two(window_size);
 
     /* Allocate a memory window */
-    req.Attributes = WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
-    req.Attributes |= WIN_USE_WAIT;
-    req.Base = 0; req.Size = window_size;
-    req.AccessSpeed = mem_speed;
-    ret = pcmcia_request_window(link, &req, &link->win);
+    link->resource[3]->flags |= WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM|WIN_ENABLE;
+    link->resource[3]->flags |= WIN_USE_WAIT;
+    link->resource[3]->start = 0; link->resource[3]->end = window_size;
+    ret = pcmcia_request_window(link, link->resource[3], mem_speed);
     if (ret)
 	    goto failed;
 
     offset = (start_pg << 8) + cm_offset;
     offset -= offset % window_size;
-    ret = pcmcia_map_mem_page(link, link->win, offset);
+    ret = pcmcia_map_mem_page(link, link->resource[3], offset);
     if (ret)
 	    goto failed;
 
     /* Try scribbling on the buffer */
-    info->base = ioremap(req.Base, window_size);
+    info->base = ioremap(link->resource[3]->start,
+			resource_size(link->resource[3]));
     for (i = 0; i < (TX_PAGES<<8); i += 2)
 	__raw_writew((i>>1), info->base+offset+i);
     udelay(100);
@@ -1521,19 +1454,20 @@
     pcnet_reset_8390(dev);
     if (i != (TX_PAGES<<8)) {
 	iounmap(info->base);
-	pcmcia_release_window(link, link->win);
-	info->base = NULL; link->win = 0;
+	pcmcia_release_window(link, link->resource[3]);
+	info->base = NULL;
 	goto failed;
     }
 
     ei_status.mem = info->base + offset;
-    ei_status.priv = req.Size;
+    ei_status.priv = resource_size(link->resource[3]);
     dev->mem_start = (u_long)ei_status.mem;
-    dev->mem_end = dev->mem_start + req.Size;
+    dev->mem_end = dev->mem_start + resource_size(link->resource[3]);
 
     ei_status.tx_start_page = start_pg;
     ei_status.rx_start_page = start_pg + TX_PAGES;
-    ei_status.stop_page = start_pg + ((req.Size - offset) >> 8);
+    ei_status.stop_page = start_pg + (
+	    (resource_size(link->resource[3]) - offset) >> 8);
 
     /* set up block i/o functions */
     ei_status.get_8390_hdr = &shmem_get_8390_hdr;
@@ -1772,9 +1706,7 @@
 MODULE_FIRMWARE("cis/tamarack.cis");
 
 static struct pcmcia_driver pcnet_driver = {
-	.drv		= {
-		.name	= "pcnet_cs",
-	},
+	.name		= "pcnet_cs",
 	.probe		= pcnet_probe,
 	.remove		= pcnet_detach,
 	.owner		= THIS_MODULE,
diff --git a/drivers/net/pcmcia/smc91c92_cs.c b/drivers/net/pcmcia/smc91c92_cs.c
index 377367d..0af2fc8 100644
--- a/drivers/net/pcmcia/smc91c92_cs.c
+++ b/drivers/net/pcmcia/smc91c92_cs.c
@@ -44,7 +44,6 @@
 #include <linux/jiffies.h>
 #include <linux/firmware.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -300,14 +299,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/*======================================================================
-
-  smc91c92_attach() creates an "instance" of the driver, allocating
-  local data structures for one device.  The device is registered
-  with Card Services.
-
-======================================================================*/
-
 static int smc91c92_probe(struct pcmcia_device *link)
 {
     struct smc_private *smc;
@@ -324,10 +315,6 @@
     link->priv = dev;
 
     spin_lock_init(&smc->lock);
-    link->resource[0]->end = 16;
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
 
     /* The SMC91c92-specific entries in the device structure. */
     dev->netdev_ops = &smc_netdev_ops;
@@ -343,15 +330,6 @@
     return smc91c92_config(link);
 } /* smc91c92_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void smc91c92_detach(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
@@ -412,26 +390,28 @@
     mdelay(200);
 
     /* Now read and write the COR... */
-    tmp = readb(smc->base + link->conf.ConfigBase + CISREG_COR);
+    tmp = readb(smc->base + link->config_base + CISREG_COR);
     udelay(5);
-    writeb(tmp, smc->base + link->conf.ConfigBase + CISREG_COR);
+    writeb(tmp, smc->base + link->config_base + CISREG_COR);
 
     return 0;
 }
 
-static int mhz_mfc_config_check(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cf,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int mhz_mfc_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int k;
-	p_dev->resource[1]->start = cf->io.win[0].base;
+	p_dev->io_lines = 16;
+	p_dev->resource[1]->start = p_dev->resource[0]->start;
+	p_dev->resource[1]->end = 8;
+	p_dev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[0]->end = 16;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	for (k = 0; k < 0x400; k += 0x10) {
 		if (k & 0x80)
 			continue;
 		p_dev->resource[0]->start = k ^ 0x300;
-		p_dev->io_lines = 16;
 		if (!pcmcia_request_io(p_dev))
 			return 0;
 	}
@@ -442,14 +422,11 @@
 {
     struct net_device *dev = link->priv;
     struct smc_private *smc = netdev_priv(dev);
-    win_req_t req;
     unsigned int offset;
     int i;
 
-    link->conf.Attributes |= CONF_ENABLE_SPKR;
-    link->conf.Status = CCSR_AUDIO_ENA;
-    link->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
-    link->resource[1]->end = 8;
+    link->config_flags |= CONF_ENABLE_SPKR | CONF_ENABLE_IRQ |
+	    CONF_AUTO_SET_IO;
 
     /* The Megahertz combo cards have modem-like CIS entries, so
        we have to explicitly try a bunch of port combinations. */
@@ -459,16 +436,16 @@
     dev->base_addr = link->resource[0]->start;
 
     /* Allocate a memory window, for accessing the ISR */
-    req.Attributes = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
-    req.Base = req.Size = 0;
-    req.AccessSpeed = 0;
-    i = pcmcia_request_window(link, &req, &link->win);
+    link->resource[2]->flags = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
+    link->resource[2]->start = link->resource[2]->end = 0;
+    i = pcmcia_request_window(link, link->resource[2], 0);
     if (i != 0)
 	    return -ENODEV;
 
-    smc->base = ioremap(req.Base, req.Size);
-    offset = (smc->manfid == MANFID_MOTOROLA) ? link->conf.ConfigBase : 0;
-    i = pcmcia_map_mem_page(link, link->win, offset);
+    smc->base = ioremap(link->resource[2]->start,
+		    resource_size(link->resource[2]));
+    offset = (smc->manfid == MANFID_MOTOROLA) ? link->config_base : 0;
+    i = pcmcia_map_mem_page(link, link->resource[2], offset);
     if ((i == 0) &&
 	(smc->manfid == MANFID_MEGAHERTZ) &&
 	(smc->cardid == PRODID_MEGAHERTZ_EM3288))
@@ -591,14 +568,12 @@
 
 /*====================================================================*/
 
-static int smc_configcheck(struct pcmcia_device *p_dev,
-			   cistpl_cftable_entry_t *cf,
-			   cistpl_cftable_entry_t *dflt,
-			   unsigned int vcc,
-			   void *priv_data)
+static int smc_configcheck(struct pcmcia_device *p_dev, void *priv_data)
 {
-	p_dev->resource[0]->start = cf->io.win[0].base;
-	p_dev->io_lines = cf->io.flags & CISTPL_IO_LINES_MASK;
+	p_dev->resource[0]->end = 16;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
+
 	return pcmcia_request_io(p_dev);
 }
 
@@ -607,7 +582,8 @@
     struct net_device *dev = link->priv;
     int i;
 
-    link->resource[0]->end = 16;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
     i = pcmcia_loop_config(link, smc_configcheck, NULL);
     if (!i)
 	    dev->base_addr = link->resource[0]->start;
@@ -640,15 +616,14 @@
     static const unsigned int com[4] = { 0x3f8, 0x2f8, 0x3e8, 0x2e8 };
     int i, j;
 
-    link->conf.Attributes |= CONF_ENABLE_SPKR;
-    link->conf.Status = CCSR_AUDIO_ENA;
+    link->config_flags |= CONF_ENABLE_SPKR | CONF_ENABLE_IRQ;
     link->resource[0]->end = 64;
     link->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
     link->resource[1]->end = 8;
 
     /* Enable Hard Decode, LAN, Modem */
-    link->conf.ConfigIndex = 0x23;
     link->io_lines = 16;
+    link->config_index = 0x23;
 
     for (i = j = 0; j < 4; j++) {
 	link->resource[1]->start = com[j];
@@ -658,7 +633,7 @@
     }
     if (i != 0) {
 	/* Fallback: turn off hard decode */
-	link->conf.ConfigIndex = 0x03;
+	link->config_index = 0x03;
 	link->resource[1]->end = 0;
 	i = pcmcia_request_io(link);
     }
@@ -817,27 +792,16 @@
     }
 
     if (width) {
-	    modconf_t mod = {
-		    .Attributes = CONF_IO_CHANGE_WIDTH,
-	    };
 	    printk(KERN_INFO "smc91c92_cs: using 8-bit IO window.\n");
 
 	    smc91c92_suspend(link);
-	    pcmcia_modify_configuration(link, &mod);
+	    pcmcia_fixup_iowidth(link);
 	    smc91c92_resume(link);
 	    return check_sig(link);
     }
     return -ENODEV;
 }
 
-/*======================================================================
-
-    smc91c92_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-
-======================================================================*/
-
 static int smc91c92_config(struct pcmcia_device *link)
 {
     struct net_device *dev = link->priv;
@@ -869,7 +833,7 @@
     i = pcmcia_request_irq(link, smc_interrupt);
     if (i)
 	    goto config_failed;
-    i = pcmcia_request_configuration(link, &link->conf);
+    i = pcmcia_enable_device(link);
     if (i)
 	    goto config_failed;
 
@@ -988,18 +952,10 @@
     return -ENODEV;
 } /* smc91c92_config */
 
-/*======================================================================
-
-    After a card is removed, smc91c92_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void smc91c92_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "smc91c92_release\n");
-	if (link->win) {
+	if (link->resource[2]->end) {
 		struct net_device *dev = link->priv;
 		struct smc_private *smc = netdev_priv(dev);
 		iounmap(smc->base);
@@ -2101,9 +2057,7 @@
 
 static struct pcmcia_driver smc91c92_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "smc91c92_cs",
-	},
+	.name		= "smc91c92_cs",
 	.probe		= smc91c92_probe,
 	.remove		= smc91c92_detach,
 	.id_table       = smc91c92_ids,
diff --git a/drivers/net/pcmcia/xirc2ps_cs.c b/drivers/net/pcmcia/xirc2ps_cs.c
index f5819526..1fece61 100644
--- a/drivers/net/pcmcia/xirc2ps_cs.c
+++ b/drivers/net/pcmcia/xirc2ps_cs.c
@@ -82,7 +82,6 @@
 #include <linux/bitops.h>
 #include <linux/mii.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -267,33 +266,11 @@
 static void mii_wr(unsigned int ioaddr, u_char phyaddr, u_char phyreg,
 		   unsigned data, int len);
 
-/*
- * The event() function is this driver's Card Services event handler.
- * It will be called by Card Services when an appropriate card status
- * event is received.  The config() and release() entry points are
- * used to configure or release a socket, in response to card insertion
- * and ejection events.  They are invoked from the event handler.
- */
-
 static int has_ce2_string(struct pcmcia_device * link);
 static int xirc2ps_config(struct pcmcia_device * link);
 static void xirc2ps_release(struct pcmcia_device * link);
-
-/****************
- * The attach() and detach() entry points are used to create and destroy
- * "instances" of the driver, where each instance represents everything
- * needed to manage one actual PCMCIA card.
- */
-
 static void xirc2ps_detach(struct pcmcia_device *p_dev);
 
-/****************
- * You'll also need to prototype all the functions that will actually
- * be used to talk to your device.  See 'pcmem_cs' for a good example
- * of a fully self-sufficient driver; the other drivers rely more or
- * less on other parts of the kernel.
- */
-
 static irqreturn_t xirc2ps_interrupt(int irq, void *dev_id);
 
 typedef struct local_info_t {
@@ -501,16 +478,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/****************
- * xirc2ps_attach() creates an "instance" of the driver, allocating
- * local data structures for one device.  The device is registered
- * with Card Services.
- *
- * The dev_link structure is initialized, but we don't actually
- * configure the card at this point -- we wait until we receive a
- * card insertion event.
- */
-
 static int
 xirc2ps_probe(struct pcmcia_device *link)
 {
@@ -529,9 +496,7 @@
     link->priv = dev;
 
     /* General socket configuration */
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.ConfigIndex = 1;
+    link->config_index = 1;
 
     /* Fill in card specific entries */
     dev->netdev_ops = &netdev_ops;
@@ -542,13 +507,6 @@
     return xirc2ps_config(link);
 } /* xirc2ps_attach */
 
-/****************
- *  This deletes a driver "instance".  The device is de-registered
- *  with Card Services.  If it has been released, all local data
- *  structures are freed.  Otherwise, the structures will be freed
- *  when the device is released.
- */
-
 static void
 xirc2ps_detach(struct pcmcia_device *link)
 {
@@ -667,44 +625,53 @@
 }
 
 static int
-xirc2ps_config_modem(struct pcmcia_device *p_dev,
-		     cistpl_cftable_entry_t *cf,
-		     cistpl_cftable_entry_t *dflt,
-		     unsigned int vcc,
-		     void *priv_data)
+xirc2ps_config_modem(struct pcmcia_device *p_dev, void *priv_data)
 {
 	unsigned int ioaddr;
 
-	if (cf->io.nwin > 0  &&  (cf->io.win[0].base & 0xf) == 8) {
-		for (ioaddr = 0x300; ioaddr < 0x400; ioaddr += 0x10) {
-			p_dev->resource[1]->start = cf->io.win[0].base;
-			p_dev->resource[0]->start = ioaddr;
-			if (!pcmcia_request_io(p_dev))
-				return 0;
-		}
+	if ((p_dev->resource[0]->start & 0xf) == 8)
+		return -ENODEV;
+
+	p_dev->resource[0]->end = 16;
+	p_dev->resource[1]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
+	p_dev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->io_lines = 10;
+
+	p_dev->resource[1]->start = p_dev->resource[0]->start;
+	for (ioaddr = 0x300; ioaddr < 0x400; ioaddr += 0x10) {
+		p_dev->resource[0]->start = ioaddr;
+		if (!pcmcia_request_io(p_dev))
+			return 0;
 	}
 	return -ENODEV;
 }
 
 static int
-xirc2ps_config_check(struct pcmcia_device *p_dev,
-		     cistpl_cftable_entry_t *cf,
-		     cistpl_cftable_entry_t *dflt,
-		     unsigned int vcc,
-		     void *priv_data)
+xirc2ps_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int *pass = priv_data;
+	resource_size_t tmp = p_dev->resource[1]->start;
 
-	if (cf->io.nwin > 0 && (cf->io.win[0].base & 0xf) == 8) {
-		p_dev->resource[1]->start = cf->io.win[0].base;
-		p_dev->resource[0]->start = p_dev->resource[1]->start
-			+ (*pass ? (cf->index & 0x20 ? -24:8)
-			   : (cf->index & 0x20 ?   8:-24));
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -ENODEV;
+	tmp += (*pass ? (p_dev->config_index & 0x20 ? -24 : 8)
+		: (p_dev->config_index & 0x20 ?   8 : -24));
 
+	if ((p_dev->resource[0]->start & 0xf) == 8)
+		return -ENODEV;
+
+	p_dev->resource[0]->end = 18;
+	p_dev->resource[1]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
+	p_dev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->io_lines = 10;
+
+	p_dev->resource[1]->start = p_dev->resource[0]->start;
+	p_dev->resource[0]->start = tmp;
+	return pcmcia_request_io(p_dev);
 }
 
 
@@ -727,11 +694,6 @@
 };
 
 
-/****************
- * xirc2ps_config() is scheduled to run after a CARD_INSERTION event
- * is received, to configure the PCMCIA socket, and to make the
- * ethernet device available to the system.
- */
 static int
 xirc2ps_config(struct pcmcia_device * link)
 {
@@ -807,32 +769,24 @@
 	goto failure;
     }
 
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
-    link->io_lines = 10;
     if (local->modem) {
 	int pass;
+	link->config_flags |= CONF_AUTO_SET_IO;
 
-	if (do_sound) {
-	    link->conf.Attributes |= CONF_ENABLE_SPKR;
-	    link->conf.Status |= CCSR_AUDIO_ENA;
-	}
-	link->resource[1]->end = 8;
-	link->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
 	if (local->dingo) {
 	    /* Take the Modem IO port from the CIS and scan for a free
 	     * Ethernet port */
-	    link->resource[0]->end = 16; /* no Mako stuff anymore */
 	    if (!pcmcia_loop_config(link, xirc2ps_config_modem, NULL))
 		    goto port_found;
 	} else {
-	    link->resource[0]->end = 18;
 	    /* We do 2 passes here: The first one uses the regular mapping and
 	     * the second tries again, thereby considering that the 32 ports are
 	     * mirrored every 32 bytes. Actually we use a mirrored port for
 	     * the Mako if (on the first pass) the COR bit 5 is set.
 	     */
 	    for (pass=0; pass < 2; pass++)
-		    if (!pcmcia_loop_config(link, xirc2ps_config_check, &pass))
+		    if (!pcmcia_loop_config(link, xirc2ps_config_check,
+						    &pass))
 			    goto port_found;
 	    /* if special option:
 	     * try to configure as Ethernet only.
@@ -840,7 +794,9 @@
 	}
 	printk(KNOT_XIRC "no ports available\n");
     } else {
+	link->io_lines = 10;
 	link->resource[0]->end = 16;
+	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
 	for (ioaddr = 0x300; ioaddr < 0x400; ioaddr += 0x10) {
 	    link->resource[0]->start = ioaddr;
 	    if (!(err = pcmcia_request_io(link)))
@@ -861,16 +817,14 @@
     if ((err=pcmcia_request_irq(link, xirc2ps_interrupt)))
 	goto config_error;
 
-    /****************
-     * This actually configures the PCMCIA socket -- setting up
-     * the I/O windows and the interrupt mapping.
-     */
-    if ((err=pcmcia_request_configuration(link, &link->conf)))
+    link->config_flags |= CONF_ENABLE_IRQ;
+    if (do_sound)
+	    link->config_flags |= CONF_ENABLE_SPKR;
+
+    if ((err = pcmcia_enable_device(link)))
 	goto config_error;
 
     if (local->dingo) {
-	win_req_t req;
-
 	/* Reset the modem's BAR to the correct value
 	 * This is necessary because in the RequestConfiguration call,
 	 * the base address of the ethernet port (BasePort1) is written
@@ -890,14 +844,14 @@
 	 * is at 0x0800. So we allocate a window into the attribute
 	 * memory and write direct to the CIS registers
 	 */
-	req.Attributes = WIN_DATA_WIDTH_8|WIN_MEMORY_TYPE_AM|WIN_ENABLE;
-	req.Base = req.Size = 0;
-	req.AccessSpeed = 0;
-	if ((err = pcmcia_request_window(link, &req, &link->win)))
+	link->resource[2]->flags = WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_AM |
+					WIN_ENABLE;
+	link->resource[2]->start = link->resource[2]->end = 0;
+	if ((err = pcmcia_request_window(link, link->resource[2], 0)))
 	    goto config_error;
 
-	local->dingo_ccr = ioremap(req.Base,0x1000) + 0x0800;
-	if ((err = pcmcia_map_mem_page(link, link->win, 0)))
+	local->dingo_ccr = ioremap(link->resource[2]->start, 0x1000) + 0x0800;
+	if ((err = pcmcia_map_mem_page(link, link->resource[2], 0)))
 	    goto config_error;
 
 	/* Setup the CCRs; there are no infos in the CIS about the Ethernet
@@ -978,17 +932,12 @@
     return -ENODEV;
 } /* xirc2ps_config */
 
-/****************
- * After a card is removed, xirc2ps_release() will unregister the net
- * device, and release the PCMCIA configuration.  If the device is
- * still open, this will be postponed until it is closed.
- */
 static void
 xirc2ps_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "release\n");
 
-	if (link->win) {
+	if (link->resource[2]->end) {
 		struct net_device *dev = link->priv;
 		local_info_t *local = netdev_priv(dev);
 		if (local->dingo)
@@ -1830,9 +1779,7 @@
 
 static struct pcmcia_driver xirc2ps_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "xirc2ps_cs",
-	},
+	.name		= "xirc2ps_cs",
 	.probe		= xirc2ps_probe,
 	.remove		= xirc2ps_detach,
 	.id_table       = xirc2ps_ids,
diff --git a/drivers/net/ppp_async.c b/drivers/net/ppp_async.c
index af50a53..78d70a6 100644
--- a/drivers/net/ppp_async.c
+++ b/drivers/net/ppp_async.c
@@ -184,7 +184,7 @@
 	tasklet_init(&ap->tsk, ppp_async_process, (unsigned long) ap);
 
 	atomic_set(&ap->refcnt, 1);
-	init_MUTEX_LOCKED(&ap->dead_sem);
+	sema_init(&ap->dead_sem, 0);
 
 	ap->chan.private = ap;
 	ap->chan.ops = &async_ops;
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index a0da4a1..992db2f 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -1212,7 +1212,8 @@
 	if ((RTL_R8(ChipCmd) & CmdRxEnb) == 0)
 		return;
 
-	counters = pci_alloc_consistent(tp->pci_dev, sizeof(*counters), &paddr);
+	counters = dma_alloc_coherent(&tp->pci_dev->dev, sizeof(*counters),
+				      &paddr, GFP_KERNEL);
 	if (!counters)
 		return;
 
@@ -1233,7 +1234,8 @@
 	RTL_W32(CounterAddrLow, 0);
 	RTL_W32(CounterAddrHigh, 0);
 
-	pci_free_consistent(tp->pci_dev, sizeof(*counters), counters, paddr);
+	dma_free_coherent(&tp->pci_dev->dev, sizeof(*counters), counters,
+			  paddr);
 }
 
 static void rtl8169_get_ethtool_stats(struct net_device *dev,
@@ -3292,15 +3294,15 @@
 
 	/*
 	 * Rx and Tx desscriptors needs 256 bytes alignment.
-	 * pci_alloc_consistent provides more.
+	 * dma_alloc_coherent provides more.
 	 */
-	tp->TxDescArray = pci_alloc_consistent(pdev, R8169_TX_RING_BYTES,
-					       &tp->TxPhyAddr);
+	tp->TxDescArray = dma_alloc_coherent(&pdev->dev, R8169_TX_RING_BYTES,
+					     &tp->TxPhyAddr, GFP_KERNEL);
 	if (!tp->TxDescArray)
 		goto err_pm_runtime_put;
 
-	tp->RxDescArray = pci_alloc_consistent(pdev, R8169_RX_RING_BYTES,
-					       &tp->RxPhyAddr);
+	tp->RxDescArray = dma_alloc_coherent(&pdev->dev, R8169_RX_RING_BYTES,
+					     &tp->RxPhyAddr, GFP_KERNEL);
 	if (!tp->RxDescArray)
 		goto err_free_tx_0;
 
@@ -3334,12 +3336,12 @@
 err_release_ring_2:
 	rtl8169_rx_clear(tp);
 err_free_rx_1:
-	pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
-			    tp->RxPhyAddr);
+	dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,
+			  tp->RxPhyAddr);
 	tp->RxDescArray = NULL;
 err_free_tx_0:
-	pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
-			    tp->TxPhyAddr);
+	dma_free_coherent(&pdev->dev, R8169_TX_RING_BYTES, tp->TxDescArray,
+			  tp->TxPhyAddr);
 	tp->TxDescArray = NULL;
 err_pm_runtime_put:
 	pm_runtime_put_noidle(&pdev->dev);
@@ -3975,7 +3977,7 @@
 {
 	struct pci_dev *pdev = tp->pci_dev;
 
-	pci_unmap_single(pdev, le64_to_cpu(desc->addr), tp->rx_buf_sz,
+	dma_unmap_single(&pdev->dev, le64_to_cpu(desc->addr), tp->rx_buf_sz,
 			 PCI_DMA_FROMDEVICE);
 	dev_kfree_skb(*sk_buff);
 	*sk_buff = NULL;
@@ -4000,7 +4002,7 @@
 static struct sk_buff *rtl8169_alloc_rx_skb(struct pci_dev *pdev,
 					    struct net_device *dev,
 					    struct RxDesc *desc, int rx_buf_sz,
-					    unsigned int align)
+					    unsigned int align, gfp_t gfp)
 {
 	struct sk_buff *skb;
 	dma_addr_t mapping;
@@ -4008,13 +4010,13 @@
 
 	pad = align ? align : NET_IP_ALIGN;
 
-	skb = netdev_alloc_skb(dev, rx_buf_sz + pad);
+	skb = __netdev_alloc_skb(dev, rx_buf_sz + pad, gfp);
 	if (!skb)
 		goto err_out;
 
 	skb_reserve(skb, align ? ((pad - 1) & (unsigned long)skb->data) : pad);
 
-	mapping = pci_map_single(pdev, skb->data, rx_buf_sz,
+	mapping = dma_map_single(&pdev->dev, skb->data, rx_buf_sz,
 				 PCI_DMA_FROMDEVICE);
 
 	rtl8169_map_to_asic(desc, mapping, rx_buf_sz);
@@ -4039,7 +4041,7 @@
 }
 
 static u32 rtl8169_rx_fill(struct rtl8169_private *tp, struct net_device *dev,
-			   u32 start, u32 end)
+			   u32 start, u32 end, gfp_t gfp)
 {
 	u32 cur;
 
@@ -4054,7 +4056,7 @@
 
 		skb = rtl8169_alloc_rx_skb(tp->pci_dev, dev,
 					   tp->RxDescArray + i,
-					   tp->rx_buf_sz, tp->align);
+					   tp->rx_buf_sz, tp->align, gfp);
 		if (!skb)
 			break;
 
@@ -4082,7 +4084,7 @@
 	memset(tp->tx_skb, 0x0, NUM_TX_DESC * sizeof(struct ring_info));
 	memset(tp->Rx_skbuff, 0x0, NUM_RX_DESC * sizeof(struct sk_buff *));
 
-	if (rtl8169_rx_fill(tp, dev, 0, NUM_RX_DESC) != NUM_RX_DESC)
+	if (rtl8169_rx_fill(tp, dev, 0, NUM_RX_DESC, GFP_KERNEL) != NUM_RX_DESC)
 		goto err_out;
 
 	rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1);
@@ -4099,7 +4101,8 @@
 {
 	unsigned int len = tx_skb->len;
 
-	pci_unmap_single(pdev, le64_to_cpu(desc->addr), len, PCI_DMA_TODEVICE);
+	dma_unmap_single(&pdev->dev, le64_to_cpu(desc->addr), len,
+			 PCI_DMA_TODEVICE);
 	desc->opts1 = 0x00;
 	desc->opts2 = 0x00;
 	desc->addr = 0x00;
@@ -4243,7 +4246,8 @@
 		txd = tp->TxDescArray + entry;
 		len = frag->size;
 		addr = ((void *) page_address(frag->page)) + frag->page_offset;
-		mapping = pci_map_single(tp->pci_dev, addr, len, PCI_DMA_TODEVICE);
+		mapping = dma_map_single(&tp->pci_dev->dev, addr, len,
+					 PCI_DMA_TODEVICE);
 
 		/* anti gcc 2.95.3 bugware (sic) */
 		status = opts1 | len | (RingEnd * !((entry + 1) % NUM_TX_DESC));
@@ -4313,7 +4317,8 @@
 		tp->tx_skb[entry].skb = skb;
 	}
 
-	mapping = pci_map_single(tp->pci_dev, skb->data, len, PCI_DMA_TODEVICE);
+	mapping = dma_map_single(&tp->pci_dev->dev, skb->data, len,
+				 PCI_DMA_TODEVICE);
 
 	tp->tx_skb[entry].len = len;
 	txd->addr = cpu_to_le64(mapping);
@@ -4477,8 +4482,8 @@
 	if (!skb)
 		goto out;
 
-	pci_dma_sync_single_for_cpu(tp->pci_dev, addr, pkt_size,
-				    PCI_DMA_FROMDEVICE);
+	dma_sync_single_for_cpu(&tp->pci_dev->dev, addr, pkt_size,
+				PCI_DMA_FROMDEVICE);
 	skb_copy_from_linear_data(*sk_buff, skb->data, pkt_size);
 	*sk_buff = skb;
 	done = true;
@@ -4549,11 +4554,11 @@
 			rtl8169_rx_csum(skb, desc);
 
 			if (rtl8169_try_rx_copy(&skb, tp, pkt_size, addr)) {
-				pci_dma_sync_single_for_device(pdev, addr,
+				dma_sync_single_for_device(&pdev->dev, addr,
 					pkt_size, PCI_DMA_FROMDEVICE);
 				rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
 			} else {
-				pci_unmap_single(pdev, addr, tp->rx_buf_sz,
+				dma_unmap_single(&pdev->dev, addr, tp->rx_buf_sz,
 						 PCI_DMA_FROMDEVICE);
 				tp->Rx_skbuff[entry] = NULL;
 			}
@@ -4583,7 +4588,7 @@
 	count = cur_rx - tp->cur_rx;
 	tp->cur_rx = cur_rx;
 
-	delta = rtl8169_rx_fill(tp, dev, tp->dirty_rx, tp->cur_rx);
+	delta = rtl8169_rx_fill(tp, dev, tp->dirty_rx, tp->cur_rx, GFP_ATOMIC);
 	if (!delta && count)
 		netif_info(tp, intr, dev, "no Rx buffer allocated\n");
 	tp->dirty_rx += delta;
@@ -4769,10 +4774,10 @@
 
 	free_irq(dev->irq, dev);
 
-	pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
-			    tp->RxPhyAddr);
-	pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
-			    tp->TxPhyAddr);
+	dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,
+			  tp->RxPhyAddr);
+	dma_free_coherent(&pdev->dev, R8169_TX_RING_BYTES, tp->TxDescArray,
+			  tp->TxPhyAddr);
 	tp->TxDescArray = NULL;
 	tp->RxDescArray = NULL;
 
diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index 40e5c46..465ae7e 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -43,6 +43,7 @@
 #include <linux/seq_file.h>
 #include <linux/mii.h>
 #include <linux/slab.h>
+#include <linux/dmi.h>
 #include <asm/irq.h>
 
 #include "skge.h"
@@ -3868,6 +3869,8 @@
 	netif_info(skge, probe, skge->netdev, "addr %pM\n", dev->dev_addr);
 }
 
+static int only_32bit_dma;
+
 static int __devinit skge_probe(struct pci_dev *pdev,
 				const struct pci_device_id *ent)
 {
@@ -3889,7 +3892,7 @@
 
 	pci_set_master(pdev);
 
-	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
+	if (!only_32bit_dma && !pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
 		using_dac = 1;
 		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
 	} else if (!(err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)))) {
@@ -4147,8 +4150,21 @@
 	.shutdown =	skge_shutdown,
 };
 
+static struct dmi_system_id skge_32bit_dma_boards[] = {
+	{
+		.ident = "Gigabyte nForce boards",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "Gigabyte Technology Co"),
+			DMI_MATCH(DMI_BOARD_NAME, "nForce"),
+		},
+	},
+	{}
+};
+
 static int __init skge_init_module(void)
 {
+	if (dmi_check_system(skge_32bit_dma_boards))
+		only_32bit_dma = 1;
 	skge_debug_init();
 	return pci_register_driver(&skge_driver);
 }
diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index bc3af78..1ec4b9e 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -4666,7 +4666,7 @@
 				       desc_idx, *post_ptr);
 		drop_it_no_recycle:
 			/* Other statistics kept track of by card. */
-			tp->net_stats.rx_dropped++;
+			tp->rx_dropped++;
 			goto next_pkt;
 		}
 
@@ -4726,7 +4726,7 @@
 		if (len > (tp->dev->mtu + ETH_HLEN) &&
 		    skb->protocol != htons(ETH_P_8021Q)) {
 			dev_kfree_skb(skb);
-			goto next_pkt;
+			goto drop_it_no_recycle;
 		}
 
 		if (desc->type_flags & RXD_FLAG_VLAN &&
@@ -9240,6 +9240,8 @@
 	stats->rx_missed_errors = old_stats->rx_missed_errors +
 		get_stat64(&hw_stats->rx_discards);
 
+	stats->rx_dropped = tp->rx_dropped;
+
 	return stats;
 }
 
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 4937bd1..be7ff13 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -2759,7 +2759,7 @@
 
 
 	/* begin "everything else" cacheline(s) section */
-	struct rtnl_link_stats64	net_stats;
+	unsigned long			rx_dropped;
 	struct rtnl_link_stats64	net_stats_prev;
 	struct tg3_ethtool_stats	estats;
 	struct tg3_ethtool_stats	estats_prev;
diff --git a/drivers/net/wan/cosa.c b/drivers/net/wan/cosa.c
index 04c6cd4..10bafd5 100644
--- a/drivers/net/wan/cosa.c
+++ b/drivers/net/wan/cosa.c
@@ -575,7 +575,7 @@
 
 		/* Initialize the chardev data structures */
 		mutex_init(&chan->rlock);
-		init_MUTEX(&chan->wsem);
+		sema_init(&chan->wsem, 1);
 
 		/* Register the network interface */
 		if (!(chan->netdev = alloc_hdlcdev(chan))) {
diff --git a/drivers/net/wimax/i2400m/rx.c b/drivers/net/wimax/i2400m/rx.c
index 8cc9e31..1737d14 100644
--- a/drivers/net/wimax/i2400m/rx.c
+++ b/drivers/net/wimax/i2400m/rx.c
@@ -1244,16 +1244,16 @@
 	int i, result;
 	struct device *dev = i2400m_dev(i2400m);
 	const struct i2400m_msg_hdr *msg_hdr;
-	size_t pl_itr, pl_size, skb_len;
+	size_t pl_itr, pl_size;
 	unsigned long flags;
-	unsigned num_pls, single_last;
+	unsigned num_pls, single_last, skb_len;
 
 	skb_len = skb->len;
-	d_fnstart(4, dev, "(i2400m %p skb %p [size %zu])\n",
+	d_fnstart(4, dev, "(i2400m %p skb %p [size %u])\n",
 		  i2400m, skb, skb_len);
 	result = -EIO;
 	msg_hdr = (void *) skb->data;
-	result = i2400m_rx_msg_hdr_check(i2400m, msg_hdr, skb->len);
+	result = i2400m_rx_msg_hdr_check(i2400m, msg_hdr, skb_len);
 	if (result < 0)
 		goto error_msg_hdr_check;
 	result = -EIO;
@@ -1261,10 +1261,10 @@
 	pl_itr = sizeof(*msg_hdr) +	/* Check payload descriptor(s) */
 		num_pls * sizeof(msg_hdr->pld[0]);
 	pl_itr = ALIGN(pl_itr, I2400M_PL_ALIGN);
-	if (pl_itr > skb->len) {	/* got all the payload descriptors? */
+	if (pl_itr > skb_len) {	/* got all the payload descriptors? */
 		dev_err(dev, "RX: HW BUG? message too short (%u bytes) for "
 			"%u payload descriptors (%zu each, total %zu)\n",
-			skb->len, num_pls, sizeof(msg_hdr->pld[0]), pl_itr);
+			skb_len, num_pls, sizeof(msg_hdr->pld[0]), pl_itr);
 		goto error_pl_descr_short;
 	}
 	/* Walk each payload payload--check we really got it */
@@ -1272,7 +1272,7 @@
 		/* work around old gcc warnings */
 		pl_size = i2400m_pld_size(&msg_hdr->pld[i]);
 		result = i2400m_rx_pl_descr_check(i2400m, &msg_hdr->pld[i],
-						  pl_itr, skb->len);
+						  pl_itr, skb_len);
 		if (result < 0)
 			goto error_pl_descr_check;
 		single_last = num_pls == 1 || i == num_pls - 1;
@@ -1290,16 +1290,16 @@
 	if (i < i2400m->rx_pl_min)
 		i2400m->rx_pl_min = i;
 	i2400m->rx_num++;
-	i2400m->rx_size_acc += skb->len;
-	if (skb->len < i2400m->rx_size_min)
-		i2400m->rx_size_min = skb->len;
-	if (skb->len > i2400m->rx_size_max)
-		i2400m->rx_size_max = skb->len;
+	i2400m->rx_size_acc += skb_len;
+	if (skb_len < i2400m->rx_size_min)
+		i2400m->rx_size_min = skb_len;
+	if (skb_len > i2400m->rx_size_max)
+		i2400m->rx_size_max = skb_len;
 	spin_unlock_irqrestore(&i2400m->rx_lock, flags);
 error_pl_descr_check:
 error_pl_descr_short:
 error_msg_hdr_check:
-	d_fnend(4, dev, "(i2400m %p skb %p [size %zu]) = %d\n",
+	d_fnend(4, dev, "(i2400m %p skb %p [size %u]) = %d\n",
 		i2400m, skb, skb_len, result);
 	return result;
 }
diff --git a/drivers/net/wireless/airo_cs.c b/drivers/net/wireless/airo_cs.c
index 9a121a5..df2484d 100644
--- a/drivers/net/wireless/airo_cs.c
+++ b/drivers/net/wireless/airo_cs.c
@@ -32,7 +32,6 @@
 #include <linux/timer.h>
 #include <linux/netdevice.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -54,58 +53,21 @@
 
 /*====================================================================*/
 
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card
-   insertion and ejection events.  They are invoked from the airo_cs
-   event handler.
-*/
-
 static int airo_config(struct pcmcia_device *link);
 static void airo_release(struct pcmcia_device *link);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void airo_detach(struct pcmcia_device *p_dev);
 
 typedef struct local_info_t {
 	struct net_device *eth_dev;
 } local_info_t;
 
-/*======================================================================
-
-  airo_attach() creates an "instance" of the driver, allocating
-  local data structures for one device.  The device is registered
-  with Card Services.
-
-  The dev_link structure is initialized, but we don't actually
-  configure the card at this point -- we wait until we receive a
-  card insertion event.
-
-  ======================================================================*/
-
 static int airo_probe(struct pcmcia_device *p_dev)
 {
 	local_info_t *local;
 
 	dev_dbg(&p_dev->dev, "airo_attach()\n");
 
-	/*
-	  General socket configuration defaults can go here.  In this
-	  client, we assume very little, and rely on the CIS for almost
-	  everything.  In most clients, many details (i.e., number, sizes,
-	  and attributes of IO windows) are fixed by the nature of the
-	  device, and can be hard-wired here.
-	*/
-	p_dev->conf.Attributes = 0;
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
-
 	/* Allocate space for private device-specific data */
 	local = kzalloc(sizeof(local_info_t), GFP_KERNEL);
 	if (!local) {
@@ -117,15 +79,6 @@
 	return airo_config(p_dev);
 } /* airo_attach */
 
-/*======================================================================
-
-  This deletes a driver "instance".  The device is de-registered
-  with Card Services.  If it has been released, all local data
-  structures are freed.  Otherwise, the structures will be freed
-  when the device is released.
-
-  ======================================================================*/
-
 static void airo_detach(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "airo_detach\n");
@@ -140,60 +93,12 @@
 	kfree(link->priv);
 } /* airo_detach */
 
-/*======================================================================
-
-  airo_config() is scheduled to run after a CARD_INSERTION event
-  is received, to configure the PCMCIA socket, and to make the
-  device available to the system.
-
-  ======================================================================*/
-
-static int airo_cs_config_check(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int airo_cs_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM]/10000;
-	else if (dflt->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM]/10000;
-
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-					pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-	}
-
-	/* This reserves IO space but doesn't actually enable it */
-	if (pcmcia_request_io(p_dev) != 0)
-		return -ENODEV;
-
-	/* If we got this far, we're cool! */
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
 
@@ -206,20 +111,9 @@
 
 	dev_dbg(&link->dev, "airo_config\n");
 
-	/*
-	 * In this loop, we scan the CIS for configuration table
-	 * entries, each of which describes a valid card
-	 * configuration, including voltage, IO window, memory window,
-	 * and interrupt settings.
-	 *
-	 * We make no assumptions about the card to be configured: we
-	 * use just the information available in the CIS.  In an ideal
-	 * world, this would work for any PCMCIA card, but it requires
-	 * a complete and accurate CIS.  In practice, a driver usually
-	 * "knows" most of these things without consulting the CIS,
-	 * and most client drivers will only use the CIS to fill in
-	 * implementation-defined details.
-	 */
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_VPP |
+		CONF_AUTO_AUDIO | CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, airo_cs_config_check, NULL);
 	if (ret)
 		goto failed;
@@ -227,12 +121,7 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	  This actually configures the PCMCIA socket -- setting up
-	  the I/O windows and the interrupt mapping, and putting the
-	  card and host interface into "Memory and IO" mode.
-	*/
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 	((local_info_t *)link->priv)->eth_dev =
@@ -241,17 +130,6 @@
 	if (!((local_info_t *)link->priv)->eth_dev)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x: ",
-	       link->conf.ConfigIndex);
-	if (link->conf.Vpp)
-		printk(", Vpp %d.%d", link->conf.Vpp/10, link->conf.Vpp%10);
-	printk(", irq %d", link->irq);
-	if (link->resource[0])
-		printk(" & %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
 	return 0;
 
  failed:
@@ -259,14 +137,6 @@
 	return -ENODEV;
 } /* airo_config */
 
-/*======================================================================
-
-  After a card is removed, airo_release() will unregister the
-  device, and release the PCMCIA configuration.  If the device is
-  still open, this will be postponed until it is closed.
-
-  ======================================================================*/
-
 static void airo_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "airo_release\n");
@@ -305,9 +175,7 @@
 
 static struct pcmcia_driver airo_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "airo_cs",
-	},
+	.name		= "airo_cs",
 	.probe		= airo_probe,
 	.remove		= airo_detach,
 	.id_table       = airo_ids,
@@ -315,12 +183,12 @@
 	.resume		= airo_resume,
 };
 
-static int airo_cs_init(void)
+static int __init airo_cs_init(void)
 {
 	return pcmcia_register_driver(&airo_driver);
 }
 
-static void airo_cs_cleanup(void)
+static void __exit airo_cs_cleanup(void)
 {
 	pcmcia_unregister_driver(&airo_driver);
 }
diff --git a/drivers/net/wireless/ath/ath9k/ani.c b/drivers/net/wireless/ath/ath9k/ani.c
index cc648b6..a3d95cc 100644
--- a/drivers/net/wireless/ath/ath9k/ani.c
+++ b/drivers/net/wireless/ath/ath9k/ani.c
@@ -543,7 +543,7 @@
 	if (conf_is_ht40(conf))
 		return clockrate * 2;
 
-	return clockrate * 2;
+	return clockrate;
 }
 
 static int32_t ath9k_hw_ani_get_listen_time(struct ath_hw *ah)
diff --git a/drivers/net/wireless/atmel_cs.c b/drivers/net/wireless/atmel_cs.c
index 3b63216..c96e19d 100644
--- a/drivers/net/wireless/atmel_cs.c
+++ b/drivers/net/wireless/atmel_cs.c
@@ -42,7 +42,6 @@
 #include <linux/moduleparam.h>
 #include <linux/device.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -64,58 +63,21 @@
 
 /*====================================================================*/
 
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card
-   insertion and ejection events.  They are invoked from the atmel_cs
-   event handler.
-*/
-
 static int atmel_config(struct pcmcia_device *link);
 static void atmel_release(struct pcmcia_device *link);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static void atmel_detach(struct pcmcia_device *p_dev);
 
 typedef struct local_info_t {
 	struct net_device *eth_dev;
 } local_info_t;
 
-/*======================================================================
-
-  atmel_attach() creates an "instance" of the driver, allocating
-  local data structures for one device.  The device is registered
-  with Card Services.
-
-  The dev_link structure is initialized, but we don't actually
-  configure the card at this point -- we wait until we receive a
-  card insertion event.
-
-  ======================================================================*/
-
 static int atmel_probe(struct pcmcia_device *p_dev)
 {
 	local_info_t *local;
 
 	dev_dbg(&p_dev->dev, "atmel_attach()\n");
 
-	/*
-	  General socket configuration defaults can go here.  In this
-	  client, we assume very little, and rely on the CIS for almost
-	  everything.  In most clients, many details (i.e., number, sizes,
-	  and attributes of IO windows) are fixed by the nature of the
-	  device, and can be hard-wired here.
-	*/
-	p_dev->conf.Attributes = 0;
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
-
 	/* Allocate space for private device-specific data */
 	local = kzalloc(sizeof(local_info_t), GFP_KERNEL);
 	if (!local) {
@@ -127,15 +89,6 @@
 	return atmel_config(p_dev);
 } /* atmel_attach */
 
-/*======================================================================
-
-  This deletes a driver "instance".  The device is de-registered
-  with Card Services.  If it has been released, all local data
-  structures are freed.  Otherwise, the structures will be freed
-  when the device is released.
-
-  ======================================================================*/
-
 static void atmel_detach(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "atmel_detach\n");
@@ -145,14 +98,6 @@
 	kfree(link->priv);
 }
 
-/*======================================================================
-
-  atmel_config() is scheduled to run after a CARD_INSERTION event
-  is received, to configure the PCMCIA socket, and to make the
-  device available to the system.
-
-  ======================================================================*/
-
 /* Call-back function to interrogate PCMCIA-specific information
    about the current existance of the card */
 static int card_present(void *arg)
@@ -165,47 +110,11 @@
 	return 0;
 }
 
-static int atmel_config_check(struct pcmcia_device *p_dev,
-			      cistpl_cftable_entry_t *cfg,
-			      cistpl_cftable_entry_t *dflt,
-			      unsigned int vcc,
-			      void *priv_data)
+static int atmel_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM]/10000;
-	else if (dflt->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM]/10000;
-
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-					pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-	}
-
-	/* This reserves IO space but doesn't actually enable it */
 	return pcmcia_request_io(p_dev);
 }
 
@@ -220,18 +129,9 @@
 
 	dev_dbg(&link->dev, "atmel_config\n");
 
-	/*
-	  In this loop, we scan the CIS for configuration table entries,
-	  each of which describes a valid card configuration, including
-	  voltage, IO window, memory window, and interrupt settings.
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_VPP |
+		CONF_AUTO_AUDIO | CONF_AUTO_SET_IO;
 
-	  We make no assumptions about the card to be configured: we use
-	  just the information available in the CIS.  In an ideal world,
-	  this would work for any PCMCIA card, but it requires a complete
-	  and accurate CIS.  In practice, a driver usually "knows" most of
-	  these things without consulting the CIS, and most client drivers
-	  will only use the CIS to fill in implementation-defined details.
-	*/
 	if (pcmcia_loop_config(link, atmel_config_check, NULL))
 		goto failed;
 
@@ -240,12 +140,7 @@
 		goto failed;
 	}
 
-	/*
-	  This actually configures the PCMCIA socket -- setting up
-	  the I/O windows and the interrupt mapping, and putting the
-	  card and host interface into "Memory and IO" mode.
-	*/
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -267,14 +162,6 @@
 	return -ENODEV;
 }
 
-/*======================================================================
-
-  After a card is removed, atmel_release() will unregister the
-  device, and release the PCMCIA configuration.  If the device is
-  still open, this will be postponed until it is closed.
-
-  ======================================================================*/
-
 static void atmel_release(struct pcmcia_device *link)
 {
 	struct net_device *dev = ((local_info_t*)link->priv)->eth_dev;
@@ -353,9 +240,7 @@
 
 static struct pcmcia_driver atmel_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "atmel_cs",
-        },
+	.name		= "atmel_cs",
 	.probe          = atmel_probe,
 	.remove		= atmel_detach,
 	.id_table	= atmel_ids,
@@ -363,12 +248,12 @@
 	.resume		= atmel_resume,
 };
 
-static int atmel_cs_init(void)
+static int __init atmel_cs_init(void)
 {
         return pcmcia_register_driver(&atmel_driver);
 }
 
-static void atmel_cs_cleanup(void)
+static void __exit atmel_cs_cleanup(void)
 {
         pcmcia_unregister_driver(&atmel_driver);
 }
diff --git a/drivers/net/wireless/b43/pcmcia.c b/drivers/net/wireless/b43/pcmcia.c
index dfbc41d..7dcba5f 100644
--- a/drivers/net/wireless/b43/pcmcia.c
+++ b/drivers/net/wireless/b43/pcmcia.c
@@ -26,7 +26,6 @@
 #include <linux/ssb/ssb.h>
 #include <linux/slab.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -63,7 +62,6 @@
 static int __devinit b43_pcmcia_probe(struct pcmcia_device *dev)
 {
 	struct ssb_bus *ssb;
-	win_req_t win;
 	int err = -ENOMEM;
 	int res = 0;
 
@@ -73,30 +71,28 @@
 
 	err = -ENODEV;
 
-	dev->conf.Attributes = CONF_ENABLE_IRQ;
-	dev->conf.IntType = INT_MEMORY_AND_IO;
+	dev->config_flags |= CONF_ENABLE_IRQ;
 
-	win.Attributes =  WIN_ENABLE | WIN_DATA_WIDTH_16 |
+	dev->resource[2]->flags |=  WIN_ENABLE | WIN_DATA_WIDTH_16 |
 			 WIN_USE_WAIT;
-	win.Base = 0;
-	win.Size = SSB_CORE_SIZE;
-	win.AccessSpeed = 250;
-	res = pcmcia_request_window(dev, &win, &dev->win);
+	dev->resource[2]->start = 0;
+	dev->resource[2]->end = SSB_CORE_SIZE;
+	res = pcmcia_request_window(dev, dev->resource[2], 250);
 	if (res != 0)
 		goto err_kfree_ssb;
 
-	res = pcmcia_map_mem_page(dev, dev->win, 0);
+	res = pcmcia_map_mem_page(dev, dev->resource[2], 0);
 	if (res != 0)
 		goto err_disable;
 
 	if (!dev->irq)
 		goto err_disable;
 
-	res = pcmcia_request_configuration(dev, &dev->conf);
+	res = pcmcia_enable_device(dev);
 	if (res != 0)
 		goto err_disable;
 
-	err = ssb_bus_pcmciabus_register(ssb, dev, win.Base);
+	err = ssb_bus_pcmciabus_register(ssb, dev, dev->resource[2]->start);
 	if (err)
 		goto err_disable;
 	dev->priv = ssb;
@@ -125,9 +121,7 @@
 
 static struct pcmcia_driver b43_pcmcia_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-				.name = "b43-pcmcia",
-			},
+	.name		= "b43-pcmcia",
 	.id_table	= b43_pcmcia_tbl,
 	.probe		= b43_pcmcia_probe,
 	.remove		= __devexit_p(b43_pcmcia_remove),
diff --git a/drivers/net/wireless/hostap/hostap_cs.c b/drivers/net/wireless/hostap/hostap_cs.c
index ba54d1b..bd8a413 100644
--- a/drivers/net/wireless/hostap/hostap_cs.c
+++ b/drivers/net/wireless/hostap/hostap_cs.c
@@ -12,7 +12,6 @@
 #include <linux/wireless.h>
 #include <net/iw_handler.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -437,7 +436,6 @@
 	int ret;
 
 	PDEBUG(DEBUG_HW, "%s: setting Vcc=33 (constant)\n", dev_info);
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
 
 	ret = prism2_config(p_dev);
 	if (ret) {
@@ -468,74 +466,11 @@
 }
 
 
-/* run after a CARD_INSERTION event is received to configure the PCMCIA
- * socket and make the device available to the system */
-
-static int prism2_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cfg,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int prism2_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	PDEBUG(DEBUG_EXTRA, "Checking CFTABLE_ENTRY 0x%02X "
-	       "(default 0x%02X)\n", cfg->index, dflt->index);
-
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] /
-		    10000 && !ignore_cis_vcc) {
-			PDEBUG(DEBUG_EXTRA, "  Vcc mismatch - skipping"
-			       " this entry\n");
-			return -ENODEV;
-		}
-	} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] /
-		    10000 && !ignore_cis_vcc) {
-			PDEBUG(DEBUG_EXTRA, "  Vcc (default) mismatch "
-			       "- skipping this entry\n");
-			return -ENODEV;
-		}
-	}
-
-	if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp = dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	PDEBUG(DEBUG_EXTRA, "IO window settings: cfg->io.nwin=%d "
-	       "dflt->io.nwin=%d\n",
-	       cfg->io.nwin, dflt->io.nwin);
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-					pcmcia_io_cfg_data_width(io->flags);
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-	}
-
-	/* This reserves IO space but doesn't actually enable it */
 	return pcmcia_request_io(p_dev);
 }
 
@@ -557,6 +492,10 @@
 	}
 
 	/* Look for an appropriate configuration table entry in the CIS */
+	link->config_flags |= CONF_AUTO_SET_VPP | CONF_AUTO_AUDIO |
+		CONF_AUTO_CHECK_VCC | CONF_AUTO_SET_IO | CONF_ENABLE_IRQ;
+	if (ignore_cis_vcc)
+		link->config_flags &= ~CONF_AUTO_CHECK_VCC;
 	ret = pcmcia_loop_config(link, prism2_config_check, NULL);
 	if (ret) {
 		if (!ignore_cis_vcc)
@@ -588,12 +527,7 @@
 	if (ret)
 		goto failed_unlock;
 
-	/*
-	 * This actually configures the PCMCIA socket -- setting up
-	 * the I/O windows and the interrupt mapping, and putting the
-	 * card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed_unlock;
 
@@ -602,20 +536,6 @@
 
 	spin_unlock_irqrestore(&local->irq_init_lock, flags);
 
-	/* Finally, report what we've done */
-	printk(KERN_INFO "%s: index 0x%02x: ",
-	       dev_info, link->conf.ConfigIndex);
-	if (link->conf.Vpp)
-		printk(", Vpp %d.%d", link->conf.Vpp / 10,
-		       link->conf.Vpp % 10);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %d", link->irq);
-	if (link->resource[0])
-		printk(" & %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	local->shutdown = 0;
 
 	sandisk_enable_wireless(dev);
@@ -627,7 +547,7 @@
 	return ret;
 
  failed_unlock:
-	 spin_unlock_irqrestore(&local->irq_init_lock, flags);
+	spin_unlock_irqrestore(&local->irq_init_lock, flags);
  failed:
 	kfree(hw_priv);
 	prism2_release((u_long)link);
@@ -779,9 +699,7 @@
 
 
 static struct pcmcia_driver hostap_driver = {
-	.drv		= {
-		.name	= "hostap_cs",
-	},
+	.name		= "hostap_cs",
 	.probe		= hostap_cs_probe,
 	.remove		= prism2_detach,
 	.owner		= THIS_MODULE,
diff --git a/drivers/net/wireless/libertas/if_cs.c b/drivers/net/wireless/libertas/if_cs.c
index 9c29839..ff1280f 100644
--- a/drivers/net/wireless/libertas/if_cs.c
+++ b/drivers/net/wireless/libertas/if_cs.c
@@ -28,7 +28,6 @@
 #include <linux/firmware.h>
 #include <linux/netdevice.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -761,15 +760,6 @@
 }
 
 
-/********************************************************************/
-/* Card Services                                                    */
-/********************************************************************/
-
-/*
- * After a card is removed, if_cs_release() will unregister the
- * device, and release the PCMCIA configuration.  If the device is
- * still open, this will be postponed until it is closed.
- */
 static void if_cs_release(struct pcmcia_device *p_dev)
 {
 	struct if_cs_card *card = p_dev->priv;
@@ -785,31 +775,12 @@
 }
 
 
-/*
- * This creates an "instance" of the driver, allocating local data
- * structures for one device.  The device is registered with Card
- * Services.
- *
- * The dev_link structure is initialized, but we don't actually
- * configure the card at this point -- we wait until we receive a card
- * insertion event.
- */
-
-static int if_cs_ioprobe(struct pcmcia_device *p_dev,
-			 cistpl_cftable_entry_t *cfg,
-			 cistpl_cftable_entry_t *dflt,
-			 unsigned int vcc,
-			 void *priv_data)
+static int if_cs_ioprobe(struct pcmcia_device *p_dev, void *priv_data)
 {
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
 	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
 
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	if (cfg->io.nwin != 1) {
+	if (p_dev->resource[1]->end) {
 		lbs_pr_err("wrong CIS (check number of IO windows)\n");
 		return -ENODEV;
 	}
@@ -835,15 +806,13 @@
 	card->p_dev = p_dev;
 	p_dev->priv = card;
 
-	p_dev->conf.Attributes = 0;
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
+	p_dev->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
 	if (pcmcia_loop_config(p_dev, if_cs_ioprobe, NULL)) {
 		lbs_pr_err("error in pcmcia_loop_config\n");
 		goto out1;
 	}
 
-
 	/*
 	 * Allocate an interrupt line.  Note that this does not assign
 	 * a handler to the interrupt, unless the 'Handler' member of
@@ -861,14 +830,9 @@
 		goto out1;
 	}
 
-	/*
-	 * This actually configures the PCMCIA socket -- setting up
-	 * the I/O windows and the interrupt mapping, and putting the
-	 * card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(p_dev, &p_dev->conf);
+	ret = pcmcia_enable_device(p_dev);
 	if (ret) {
-		lbs_pr_err("error in pcmcia_request_configuration\n");
+		lbs_pr_err("error in pcmcia_enable_device\n");
 		goto out2;
 	}
 
@@ -962,12 +926,6 @@
 }
 
 
-/*
- * This deletes a driver "instance".  The device is de-registered with
- * Card Services.  If it has been released, all local data structures
- * are freed.  Otherwise, the structures will be freed when the device
- * is released.
- */
 static void if_cs_detach(struct pcmcia_device *p_dev)
 {
 	struct if_cs_card *card = p_dev->priv;
@@ -1000,9 +958,7 @@
 
 static struct pcmcia_driver lbs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= DRV_NAME,
-	},
+	.name		= DRV_NAME,
 	.probe		= if_cs_probe,
 	.remove		= if_cs_detach,
 	.id_table       = if_cs_ids,
diff --git a/drivers/net/wireless/orinoco/orinoco_cs.c b/drivers/net/wireless/orinoco/orinoco_cs.c
index ef46a2d..71b3d68 100644
--- a/drivers/net/wireless/orinoco/orinoco_cs.c
+++ b/drivers/net/wireless/orinoco/orinoco_cs.c
@@ -17,7 +17,6 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/delay.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -93,14 +92,6 @@
 /* PCMCIA stuff     						    */
 /********************************************************************/
 
-/*
- * This creates an "instance" of the driver, allocating local data
- * structures for one device.  The device is registered with Card
- * Services.
- *
- * The dev_link structure is initialized, but we don't actually
- * configure the card at this point -- we wait until we receive a card
- * insertion event.  */
 static int
 orinoco_cs_probe(struct pcmcia_device *link)
 {
@@ -117,23 +108,9 @@
 	card->p_dev = link;
 	link->priv = priv;
 
-	/* General socket configuration defaults can go here.  In this
-	 * client, we assume very little, and rely on the CIS for
-	 * almost everything.  In most clients, many details (i.e.,
-	 * number, sizes, and attributes of IO windows) are fixed by
-	 * the nature of the device, and can be hard-wired here. */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	return orinoco_cs_config(link);
 }				/* orinoco_cs_attach */
 
-/*
- * This deletes a driver "instance".  The device is de-registered with
- * Card Services.  If it has been released, all local data structures
- * are freed.  Otherwise, the structures will be freed when the device
- * is released.
- */
 static void orinoco_cs_detach(struct pcmcia_device *link)
 {
 	struct orinoco_private *priv = link->priv;
@@ -145,76 +122,12 @@
 	free_orinocodev(priv);
 }				/* orinoco_cs_detach */
 
-/*
- * orinoco_cs_config() is scheduled to run after a CARD_INSERTION
- * event is received, to configure the PCMCIA socket, and to make the
- * device available to the system.
- */
-
-static int orinoco_cs_config_check(struct pcmcia_device *p_dev,
-				   cistpl_cftable_entry_t *cfg,
-				   cistpl_cftable_entry_t *dflt,
-				   unsigned int vcc,
-				   void *priv_data)
+static int orinoco_cs_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		goto next_entry;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Use power settings for Vcc and Vpp if present */
-	/* Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000) {
-			DEBUG(2, "%s: Vcc mismatch (vcc = %d, CIS = %d)\n",
-			      __func__, vcc,
-			      cfg->vcc.param[CISTPL_POWER_VNOM] / 10000);
-			if (!ignore_cis_vcc)
-				goto next_entry;
-		}
-	} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] / 10000) {
-			DEBUG(2, "%s: Vcc mismatch (vcc = %d, CIS = %d)\n",
-			      __func__, vcc,
-			      dflt->vcc.param[CISTPL_POWER_VNOM] / 10000);
-			if (!ignore_cis_vcc)
-				goto next_entry;
-		}
-	}
-
-	if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-
-		/* This reserves IO space but doesn't actually enable it */
-		if (pcmcia_request_io(p_dev) != 0)
-			goto next_entry;
-	}
-	return 0;
-
-next_entry:
-	pcmcia_disable_device(p_dev);
-	return -ENODEV;
+	return pcmcia_request_io(p_dev);
 };
 
 static int
@@ -225,20 +138,10 @@
 	int ret;
 	void __iomem *mem;
 
-	/*
-	 * In this loop, we scan the CIS for configuration table
-	 * entries, each of which describes a valid card
-	 * configuration, including voltage, IO window, memory window,
-	 * and interrupt settings.
-	 *
-	 * We make no assumptions about the card to be configured: we
-	 * use just the information available in the CIS.  In an ideal
-	 * world, this would work for any PCMCIA card, but it requires
-	 * a complete and accurate CIS.  In practice, a driver usually
-	 * "knows" most of these things without consulting the CIS,
-	 * and most client drivers will only use the CIS to fill in
-	 * implementation-defined details.
-	 */
+	link->config_flags |= CONF_AUTO_SET_VPP | CONF_AUTO_CHECK_VCC |
+		CONF_AUTO_SET_IO | CONF_ENABLE_IRQ;
+	if (ignore_cis_vcc)
+		link->config_flags &= ~CONF_AUTO_CHECK_VCC;
 	ret = pcmcia_loop_config(link, orinoco_cs_config_check, NULL);
 	if (ret) {
 		if (!ignore_cis_vcc)
@@ -262,12 +165,7 @@
 
 	hermes_struct_init(hw, mem, HERMES_16BIT_REGSPACING);
 
-	/*
-	 * This actually configures the PCMCIA socket -- setting up
-	 * the I/O windows and the interrupt mapping, and putting the
-	 * card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -291,11 +189,6 @@
 	return -ENODEV;
 }				/* orinoco_cs_config */
 
-/*
- * After a card is removed, orinoco_cs_release() will unregister the
- * device, and release the PCMCIA configuration.  If the device is
- * still open, this will be postponed until it is closed.
- */
 static void
 orinoco_cs_release(struct pcmcia_device *link)
 {
@@ -344,12 +237,6 @@
 /* Module initialization					    */
 /********************************************************************/
 
-/* Can't be declared "const" or the whole __initdata section will
- * become const */
-static char version[] __initdata = DRIVER_NAME " " DRIVER_VERSION
-	" (David Gibson <hermes@gibson.dropbear.id.au>, "
-	"Pavel Roskin <proski@gnu.org>, et al)";
-
 static struct pcmcia_device_id orinoco_cs_ids[] = {
 	PCMCIA_DEVICE_MANF_CARD(0x0101, 0x0777), /* 3Com AirConnect PCI 777A */
 	PCMCIA_DEVICE_MANF_CARD(0x0156, 0x0002), /* Lucent Orinoco and old Intersil */
@@ -441,9 +328,7 @@
 
 static struct pcmcia_driver orinoco_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= DRIVER_NAME,
-	},
+	.name		= DRIVER_NAME,
 	.probe		= orinoco_cs_probe,
 	.remove		= orinoco_cs_detach,
 	.id_table       = orinoco_cs_ids,
@@ -454,8 +339,6 @@
 static int __init
 init_orinoco_cs(void)
 {
-	printk(KERN_DEBUG "%s\n", version);
-
 	return pcmcia_register_driver(&orinoco_driver);
 }
 
diff --git a/drivers/net/wireless/orinoco/spectrum_cs.c b/drivers/net/wireless/orinoco/spectrum_cs.c
index 873877e..fb859a5 100644
--- a/drivers/net/wireless/orinoco/spectrum_cs.c
+++ b/drivers/net/wireless/orinoco/spectrum_cs.c
@@ -25,7 +25,6 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/delay.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -154,14 +153,6 @@
 /* PCMCIA stuff     						    */
 /********************************************************************/
 
-/*
- * This creates an "instance" of the driver, allocating local data
- * structures for one device.  The device is registered with Card
- * Services.
- *
- * The dev_link structure is initialized, but we don't actually
- * configure the card at this point -- we wait until we receive a card
- * insertion event.  */
 static int
 spectrum_cs_probe(struct pcmcia_device *link)
 {
@@ -179,23 +170,9 @@
 	card->p_dev = link;
 	link->priv = priv;
 
-	/* General socket configuration defaults can go here.  In this
-	 * client, we assume very little, and rely on the CIS for
-	 * almost everything.  In most clients, many details (i.e.,
-	 * number, sizes, and attributes of IO windows) are fixed by
-	 * the nature of the device, and can be hard-wired here. */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	return spectrum_cs_config(link);
 }				/* spectrum_cs_attach */
 
-/*
- * This deletes a driver "instance".  The device is de-registered with
- * Card Services.  If it has been released, all local data structures
- * are freed.  Otherwise, the structures will be freed when the device
- * is released.
- */
 static void spectrum_cs_detach(struct pcmcia_device *link)
 {
 	struct orinoco_private *priv = link->priv;
@@ -207,76 +184,13 @@
 	free_orinocodev(priv);
 }				/* spectrum_cs_detach */
 
-/*
- * spectrum_cs_config() is scheduled to run after a CARD_INSERTION
- * event is received, to configure the PCMCIA socket, and to make the
- * device available to the system.
- */
-
 static int spectrum_cs_config_check(struct pcmcia_device *p_dev,
-				    cistpl_cftable_entry_t *cfg,
-				    cistpl_cftable_entry_t *dflt,
-				    unsigned int vcc,
 				    void *priv_data)
 {
-	if (cfg->index == 0)
-		goto next_entry;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Use power settings for Vcc and Vpp if present */
-	/* Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000) {
-			DEBUG(2, "%s: Vcc mismatch (vcc = %d, CIS = %d)\n",
-			      __func__, vcc,
-			      cfg->vcc.param[CISTPL_POWER_VNOM] / 10000);
-			if (!ignore_cis_vcc)
-				goto next_entry;
-		}
-	} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
-		if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] / 10000) {
-			DEBUG(2, "%s: Vcc mismatch (vcc = %d, CIS = %d)\n",
-			      __func__, vcc,
-			      dflt->vcc.param[CISTPL_POWER_VNOM] / 10000);
-			if (!ignore_cis_vcc)
-				goto next_entry;
-		}
-	}
-
-	if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-	else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-
-		/* This reserves IO space but doesn't actually enable it */
-		if (pcmcia_request_io(p_dev) != 0)
-			goto next_entry;
-	}
-	return 0;
-
-next_entry:
-	pcmcia_disable_device(p_dev);
-	return -ENODEV;
+	return pcmcia_request_io(p_dev);
 };
 
 static int
@@ -287,20 +201,10 @@
 	int ret;
 	void __iomem *mem;
 
-	/*
-	 * In this loop, we scan the CIS for configuration table
-	 * entries, each of which describes a valid card
-	 * configuration, including voltage, IO window, memory window,
-	 * and interrupt settings.
-	 *
-	 * We make no assumptions about the card to be configured: we
-	 * use just the information available in the CIS.  In an ideal
-	 * world, this would work for any PCMCIA card, but it requires
-	 * a complete and accurate CIS.  In practice, a driver usually
-	 * "knows" most of these things without consulting the CIS,
-	 * and most client drivers will only use the CIS to fill in
-	 * implementation-defined details.
-	 */
+	link->config_flags |= CONF_AUTO_SET_VPP | CONF_AUTO_CHECK_VCC |
+		CONF_AUTO_SET_IO | CONF_ENABLE_IRQ;
+	if (ignore_cis_vcc)
+		link->config_flags &= ~CONF_AUTO_CHECK_VCC;
 	ret = pcmcia_loop_config(link, spectrum_cs_config_check, NULL);
 	if (ret) {
 		if (!ignore_cis_vcc)
@@ -325,12 +229,7 @@
 	hermes_struct_init(hw, mem, HERMES_16BIT_REGSPACING);
 	hw->eeprom_pda = true;
 
-	/*
-	 * This actually configures the PCMCIA socket -- setting up
-	 * the I/O windows and the interrupt mapping, and putting the
-	 * card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -358,11 +257,6 @@
 	return -ENODEV;
 }				/* spectrum_cs_config */
 
-/*
- * After a card is removed, spectrum_cs_release() will unregister the
- * device, and release the PCMCIA configuration.  If the device is
- * still open, this will be postponed until it is closed.
- */
 static void
 spectrum_cs_release(struct pcmcia_device *link)
 {
@@ -407,12 +301,6 @@
 /* Module initialization					    */
 /********************************************************************/
 
-/* Can't be declared "const" or the whole __initdata section will
- * become const */
-static char version[] __initdata = DRIVER_NAME " " DRIVER_VERSION
-	" (Pavel Roskin <proski@gnu.org>,"
-	" David Gibson <hermes@gibson.dropbear.id.au>, et al)";
-
 static struct pcmcia_device_id spectrum_cs_ids[] = {
 	PCMCIA_DEVICE_MANF_CARD(0x026c, 0x0001), /* Symbol Spectrum24 LA4137 */
 	PCMCIA_DEVICE_MANF_CARD(0x0104, 0x0001), /* Socket Communications CF */
@@ -423,9 +311,7 @@
 
 static struct pcmcia_driver orinoco_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= DRIVER_NAME,
-	},
+	.name		= DRIVER_NAME,
 	.probe		= spectrum_cs_probe,
 	.remove		= spectrum_cs_detach,
 	.suspend	= spectrum_cs_suspend,
@@ -436,8 +322,6 @@
 static int __init
 init_spectrum_cs(void)
 {
-	printk(KERN_DEBUG "%s\n", version);
-
 	return pcmcia_register_driver(&orinoco_driver);
 }
 
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
index 88560d0..af5b17c 100644
--- a/drivers/net/wireless/ray_cs.c
+++ b/drivers/net/wireless/ray_cs.c
@@ -46,7 +46,6 @@
 #include <linux/ethtool.h>
 #include <linux/ieee80211.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -169,13 +168,6 @@
  */
 static char *phy_addr = NULL;
 
-
-/* A struct pcmcia_device structure has fields for most things that are needed
-   to keep track of a socket, but there will usually be some device
-   specific information that also needs to be kept track of.  The
-   'priv' pointer in a struct pcmcia_device structure can be used to point to
-   a device-specific private data structure, like this.
-*/
 static unsigned int ray_mem_speed = 500;
 
 /* WARNING: THIS DRIVER IS NOT CAPABLE OF HANDLING MULTIPLE DEVICES! */
@@ -290,14 +282,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/*=============================================================================
-    ray_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-=============================================================================*/
 static int ray_probe(struct pcmcia_device *p_dev)
 {
 	ray_dev_t *local;
@@ -318,9 +302,8 @@
 	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
 
 	/* General socket configuration */
-	p_dev->conf.Attributes = CONF_ENABLE_IRQ;
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
-	p_dev->conf.ConfigIndex = 1;
+	p_dev->config_flags |= CONF_ENABLE_IRQ;
+	p_dev->config_index = 1;
 
 	p_dev->priv = dev;
 
@@ -353,12 +336,6 @@
 	return -ENOMEM;
 } /* ray_attach */
 
-/*=============================================================================
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-=============================================================================*/
 static void ray_detach(struct pcmcia_device *link)
 {
 	struct net_device *dev;
@@ -381,17 +358,11 @@
 	dev_dbg(&link->dev, "ray_cs ray_detach ending\n");
 } /* ray_detach */
 
-/*=============================================================================
-    ray_config() is run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-=============================================================================*/
 #define MAX_TUPLE_SIZE 128
 static int ray_config(struct pcmcia_device *link)
 {
 	int ret = 0;
 	int i;
-	win_req_t req;
 	struct net_device *dev = (struct net_device *)link->priv;
 	ray_dev_t *local = netdev_priv(dev);
 
@@ -412,54 +383,50 @@
 		goto failed;
 	dev->irq = link->irq;
 
-	/* This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
 /*** Set up 32k window for shared memory (transmit and control) ************/
-	req.Attributes =
-	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
-	req.Base = 0;
-	req.Size = 0x8000;
-	req.AccessSpeed = ray_mem_speed;
-	ret = pcmcia_request_window(link, &req, &link->win);
+	link->resource[2]->flags |= WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
+	link->resource[2]->start = 0;
+	link->resource[2]->end = 0x8000;
+	ret = pcmcia_request_window(link, link->resource[2], ray_mem_speed);
 	if (ret)
 		goto failed;
-	ret = pcmcia_map_mem_page(link, link->win, 0);
+	ret = pcmcia_map_mem_page(link, link->resource[2], 0);
 	if (ret)
 		goto failed;
-	local->sram = ioremap(req.Base, req.Size);
+	local->sram = ioremap(link->resource[2]->start,
+			resource_size(link->resource[2]));
 
 /*** Set up 16k window for shared memory (receive buffer) ***************/
-	req.Attributes =
+	link->resource[3]->flags |=
 	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_CM | WIN_ENABLE | WIN_USE_WAIT;
-	req.Base = 0;
-	req.Size = 0x4000;
-	req.AccessSpeed = ray_mem_speed;
-	ret = pcmcia_request_window(link, &req, &local->rmem_handle);
+	link->resource[3]->start = 0;
+	link->resource[3]->end = 0x4000;
+	ret = pcmcia_request_window(link, link->resource[3], ray_mem_speed);
 	if (ret)
 		goto failed;
-	ret = pcmcia_map_mem_page(link, local->rmem_handle, 0x8000);
+	ret = pcmcia_map_mem_page(link, link->resource[3], 0x8000);
 	if (ret)
 		goto failed;
-	local->rmem = ioremap(req.Base, req.Size);
+	local->rmem = ioremap(link->resource[3]->start,
+			resource_size(link->resource[3]));
 
 /*** Set up window for attribute memory ***********************************/
-	req.Attributes =
+	link->resource[4]->flags |=
 	    WIN_DATA_WIDTH_8 | WIN_MEMORY_TYPE_AM | WIN_ENABLE | WIN_USE_WAIT;
-	req.Base = 0;
-	req.Size = 0x1000;
-	req.AccessSpeed = ray_mem_speed;
-	ret = pcmcia_request_window(link, &req, &local->amem_handle);
+	link->resource[4]->start = 0;
+	link->resource[4]->end = 0x1000;
+	ret = pcmcia_request_window(link, link->resource[4], ray_mem_speed);
 	if (ret)
 		goto failed;
-	ret = pcmcia_map_mem_page(link, local->amem_handle, 0);
+	ret = pcmcia_map_mem_page(link, link->resource[4], 0);
 	if (ret)
 		goto failed;
-	local->amem = ioremap(req.Base, req.Size);
+	local->amem = ioremap(link->resource[4]->start,
+			resource_size(link->resource[4]));
 
 	dev_dbg(&link->dev, "ray_config sram=%p\n", local->sram);
 	dev_dbg(&link->dev, "ray_config rmem=%p\n", local->rmem);
@@ -775,11 +742,7 @@
 	local->card_status = CARD_DOING_ACQ;
 }
 
-/*============================================================================
-    After a card is removed, ray_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-=============================================================================*/
+
 static void ray_release(struct pcmcia_device *link)
 {
 	struct net_device *dev = link->priv;
@@ -2847,9 +2810,7 @@
 
 static struct pcmcia_driver ray_driver = {
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "ray_cs",
-		},
+	.name = "ray_cs",
 	.probe = ray_probe,
 	.remove = ray_detach,
 	.id_table = ray_ids,
diff --git a/drivers/net/wireless/ray_cs.h b/drivers/net/wireless/ray_cs.h
index 9f01ddb..e79848f 100644
--- a/drivers/net/wireless/ray_cs.h
+++ b/drivers/net/wireless/ray_cs.h
@@ -25,8 +25,6 @@
 typedef struct ray_dev_t {
     int card_status;
     int authentication_state;
-    window_handle_t amem_handle;   /* handle to window for attribute memory  */
-    window_handle_t rmem_handle;   /* handle to window for rx buffer on card */
     void __iomem *sram;            /* pointer to beginning of shared RAM     */
     void __iomem *amem;            /* pointer to attribute mem window        */
     void __iomem *rmem;            /* pointer to receive buffer window       */
diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
index a1cc2d4..ca3f896 100644
--- a/drivers/net/wireless/wl3501_cs.c
+++ b/drivers/net/wireless/wl3501_cs.c
@@ -48,7 +48,6 @@
 
 #include <net/iw_handler.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -78,13 +77,6 @@
 #define WL3501_RESUME	0
 #define WL3501_SUSPEND	1
 
-/*
- * The event() function is this driver's Card Services event handler.  It will
- * be called by Card Services when an appropriate card status event is
- * received. The config() and release() entry points are used to configure or
- * release a socket, in response to card insertion and ejection events.  They
- * are invoked from the wl24 event handler.
- */
 static int wl3501_config(struct pcmcia_device *link);
 static void wl3501_release(struct pcmcia_device *link);
 
@@ -1869,15 +1861,6 @@
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-/**
- * wl3501_attach - creates an "instance" of the driver
- *
- * Creates an "instance" of the driver, allocating local data structures for
- * one device.  The device is registered with Card Services.
- *
- * The dev_link structure is initialized, but we don't actually configure the
- * card at this point -- we wait until we receive a card insertion event.
- */
 static int wl3501_probe(struct pcmcia_device *p_dev)
 {
 	struct net_device *dev;
@@ -1888,9 +1871,8 @@
 	p_dev->resource[0]->flags	= IO_DATA_PATH_WIDTH_8;
 
 	/* General socket configuration */
-	p_dev->conf.Attributes	= CONF_ENABLE_IRQ;
-	p_dev->conf.IntType	= INT_MEMORY_AND_IO;
-	p_dev->conf.ConfigIndex	= 1;
+	p_dev->config_flags	= CONF_ENABLE_IRQ;
+	p_dev->config_index	= 1;
 
 	dev = alloc_etherdev(sizeof(struct wl3501_card));
 	if (!dev)
@@ -1914,14 +1896,6 @@
 	return -ENOMEM;
 }
 
-/**
- * wl3501_config - configure the PCMCIA socket and make eth device available
- * @link - FILL_IN
- *
- * wl3501_config() is scheduled to run after a CARD_INSERTION event is
- * received, to configure the PCMCIA socket, and to make the ethernet device
- * available to the system.
- */
 static int wl3501_config(struct pcmcia_device *link)
 {
 	struct net_device *dev = link->priv;
@@ -1952,10 +1926,7 @@
 	if (ret)
 		goto failed;
 
-	/* This actually configures the PCMCIA socket -- setting up the I/O
-	 * windows and the interrupt mapping.  */
-
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -2010,14 +1981,6 @@
 	return -ENODEV;
 }
 
-/**
- * wl3501_release - unregister the net, release PCMCIA configuration
- * @arg - link
- *
- * After a card is removed, wl3501_release() will unregister the net device,
- * and release the PCMCIA configuration.  If the device is still open, this
- * will be postponed until it is closed.
- */
 static void wl3501_release(struct pcmcia_device *link)
 {
 	pcmcia_disable_device(link);
@@ -2056,9 +2019,7 @@
 
 static struct pcmcia_driver wl3501_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "wl3501_cs",
-	},
+	.name		= "wl3501_cs",
 	.probe		= wl3501_probe,
 	.remove		= wl3501_detach,
 	.id_table	= wl3501_ids,
diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c
index b336cd9..f9bda64 100644
--- a/drivers/oprofile/oprof.c
+++ b/drivers/oprofile/oprof.c
@@ -225,26 +225,17 @@
 	mutex_unlock(&start_mutex);
 }
 
-int oprofile_set_backtrace(unsigned long val)
+int oprofile_set_ulong(unsigned long *addr, unsigned long val)
 {
-	int err = 0;
+	int err = -EBUSY;
 
 	mutex_lock(&start_mutex);
-
-	if (oprofile_started) {
-		err = -EBUSY;
-		goto out;
+	if (!oprofile_started) {
+		*addr = val;
+		err = 0;
 	}
-
-	if (!oprofile_ops.backtrace) {
-		err = -EINVAL;
-		goto out;
-	}
-
-	oprofile_backtrace_depth = val;
-
-out:
 	mutex_unlock(&start_mutex);
+
 	return err;
 }
 
@@ -257,16 +248,9 @@
 		printk(KERN_INFO "oprofile: using timer interrupt.\n");
 		err = oprofile_timer_init(&oprofile_ops);
 		if (err)
-			goto out_arch;
+			return err;
 	}
-	err = oprofilefs_register();
-	if (err)
-		goto out_arch;
-	return 0;
-
-out_arch:
-	oprofile_arch_exit();
-	return err;
+	return oprofilefs_register();
 }
 
 
diff --git a/drivers/oprofile/oprof.h b/drivers/oprofile/oprof.h
index 47e12cb..177b73d 100644
--- a/drivers/oprofile/oprof.h
+++ b/drivers/oprofile/oprof.h
@@ -37,7 +37,7 @@
 int oprofile_timer_init(struct oprofile_operations *ops);
 void oprofile_timer_exit(void);
 
-int oprofile_set_backtrace(unsigned long depth);
+int oprofile_set_ulong(unsigned long *addr, unsigned long val);
 int oprofile_set_timeout(unsigned long time);
 
 #endif /* OPROF_H */
diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c
index bbd7516..ccf099e 100644
--- a/drivers/oprofile/oprofile_files.c
+++ b/drivers/oprofile/oprofile_files.c
@@ -79,14 +79,17 @@
 	if (*offset)
 		return -EINVAL;
 
+	if (!oprofile_ops.backtrace)
+		return -EINVAL;
+
 	retval = oprofilefs_ulong_from_user(&val, buf, count);
 	if (retval)
 		return retval;
 
-	retval = oprofile_set_backtrace(val);
-
+	retval = oprofile_set_ulong(&oprofile_backtrace_depth, val);
 	if (retval)
 		return retval;
+
 	return count;
 }
 
diff --git a/drivers/oprofile/oprofile_perf.c b/drivers/oprofile/oprofile_perf.c
new file mode 100644
index 0000000..9046f7b
--- /dev/null
+++ b/drivers/oprofile/oprofile_perf.c
@@ -0,0 +1,328 @@
+/*
+ * Copyright 2010 ARM Ltd.
+ *
+ * Perf-events backend for OProfile.
+ */
+#include <linux/perf_event.h>
+#include <linux/platform_device.h>
+#include <linux/oprofile.h>
+#include <linux/slab.h>
+
+/*
+ * Per performance monitor configuration as set via oprofilefs.
+ */
+struct op_counter_config {
+	unsigned long count;
+	unsigned long enabled;
+	unsigned long event;
+	unsigned long unit_mask;
+	unsigned long kernel;
+	unsigned long user;
+	struct perf_event_attr attr;
+};
+
+static int oprofile_perf_enabled;
+static DEFINE_MUTEX(oprofile_perf_mutex);
+
+static struct op_counter_config *counter_config;
+static struct perf_event **perf_events[nr_cpumask_bits];
+static int num_counters;
+
+/*
+ * Overflow callback for oprofile.
+ */
+static void op_overflow_handler(struct perf_event *event, int unused,
+			struct perf_sample_data *data, struct pt_regs *regs)
+{
+	int id;
+	u32 cpu = smp_processor_id();
+
+	for (id = 0; id < num_counters; ++id)
+		if (perf_events[cpu][id] == event)
+			break;
+
+	if (id != num_counters)
+		oprofile_add_sample(regs, id);
+	else
+		pr_warning("oprofile: ignoring spurious overflow "
+				"on cpu %u\n", cpu);
+}
+
+/*
+ * Called by oprofile_perf_setup to create perf attributes to mirror the oprofile
+ * settings in counter_config. Attributes are created as `pinned' events and
+ * so are permanently scheduled on the PMU.
+ */
+static void op_perf_setup(void)
+{
+	int i;
+	u32 size = sizeof(struct perf_event_attr);
+	struct perf_event_attr *attr;
+
+	for (i = 0; i < num_counters; ++i) {
+		attr = &counter_config[i].attr;
+		memset(attr, 0, size);
+		attr->type		= PERF_TYPE_RAW;
+		attr->size		= size;
+		attr->config		= counter_config[i].event;
+		attr->sample_period	= counter_config[i].count;
+		attr->pinned		= 1;
+	}
+}
+
+static int op_create_counter(int cpu, int event)
+{
+	struct perf_event *pevent;
+
+	if (!counter_config[event].enabled || perf_events[cpu][event])
+		return 0;
+
+	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
+						  cpu, NULL,
+						  op_overflow_handler);
+
+	if (IS_ERR(pevent))
+		return PTR_ERR(pevent);
+
+	if (pevent->state != PERF_EVENT_STATE_ACTIVE) {
+		perf_event_release_kernel(pevent);
+		pr_warning("oprofile: failed to enable event %d "
+				"on CPU %d\n", event, cpu);
+		return -EBUSY;
+	}
+
+	perf_events[cpu][event] = pevent;
+
+	return 0;
+}
+
+static void op_destroy_counter(int cpu, int event)
+{
+	struct perf_event *pevent = perf_events[cpu][event];
+
+	if (pevent) {
+		perf_event_release_kernel(pevent);
+		perf_events[cpu][event] = NULL;
+	}
+}
+
+/*
+ * Called by oprofile_perf_start to create active perf events based on the
+ * perviously configured attributes.
+ */
+static int op_perf_start(void)
+{
+	int cpu, event, ret = 0;
+
+	for_each_online_cpu(cpu) {
+		for (event = 0; event < num_counters; ++event) {
+			ret = op_create_counter(cpu, event);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+/*
+ * Called by oprofile_perf_stop at the end of a profiling run.
+ */
+static void op_perf_stop(void)
+{
+	int cpu, event;
+
+	for_each_online_cpu(cpu)
+		for (event = 0; event < num_counters; ++event)
+			op_destroy_counter(cpu, event);
+}
+
+static int oprofile_perf_create_files(struct super_block *sb, struct dentry *root)
+{
+	unsigned int i;
+
+	for (i = 0; i < num_counters; i++) {
+		struct dentry *dir;
+		char buf[4];
+
+		snprintf(buf, sizeof buf, "%d", i);
+		dir = oprofilefs_mkdir(sb, root, buf);
+		oprofilefs_create_ulong(sb, dir, "enabled", &counter_config[i].enabled);
+		oprofilefs_create_ulong(sb, dir, "event", &counter_config[i].event);
+		oprofilefs_create_ulong(sb, dir, "count", &counter_config[i].count);
+		oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask);
+		oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel);
+		oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user);
+	}
+
+	return 0;
+}
+
+static int oprofile_perf_setup(void)
+{
+	spin_lock(&oprofilefs_lock);
+	op_perf_setup();
+	spin_unlock(&oprofilefs_lock);
+	return 0;
+}
+
+static int oprofile_perf_start(void)
+{
+	int ret = -EBUSY;
+
+	mutex_lock(&oprofile_perf_mutex);
+	if (!oprofile_perf_enabled) {
+		ret = 0;
+		op_perf_start();
+		oprofile_perf_enabled = 1;
+	}
+	mutex_unlock(&oprofile_perf_mutex);
+	return ret;
+}
+
+static void oprofile_perf_stop(void)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled)
+		op_perf_stop();
+	oprofile_perf_enabled = 0;
+	mutex_unlock(&oprofile_perf_mutex);
+}
+
+#ifdef CONFIG_PM
+
+static int oprofile_perf_suspend(struct platform_device *dev, pm_message_t state)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled)
+		op_perf_stop();
+	mutex_unlock(&oprofile_perf_mutex);
+	return 0;
+}
+
+static int oprofile_perf_resume(struct platform_device *dev)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled && op_perf_start())
+		oprofile_perf_enabled = 0;
+	mutex_unlock(&oprofile_perf_mutex);
+	return 0;
+}
+
+static struct platform_driver oprofile_driver = {
+	.driver		= {
+		.name		= "oprofile-perf",
+	},
+	.resume		= oprofile_perf_resume,
+	.suspend	= oprofile_perf_suspend,
+};
+
+static struct platform_device *oprofile_pdev;
+
+static int __init init_driverfs(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&oprofile_driver);
+	if (ret)
+		return ret;
+
+	oprofile_pdev =	platform_device_register_simple(
+				oprofile_driver.driver.name, 0, NULL, 0);
+	if (IS_ERR(oprofile_pdev)) {
+		ret = PTR_ERR(oprofile_pdev);
+		platform_driver_unregister(&oprofile_driver);
+	}
+
+	return ret;
+}
+
+static void exit_driverfs(void)
+{
+	platform_device_unregister(oprofile_pdev);
+	platform_driver_unregister(&oprofile_driver);
+}
+
+#else
+
+static inline int  init_driverfs(void) { return 0; }
+static inline void exit_driverfs(void) { }
+
+#endif /* CONFIG_PM */
+
+void oprofile_perf_exit(void)
+{
+	int cpu, id;
+	struct perf_event *event;
+
+	for_each_possible_cpu(cpu) {
+		for (id = 0; id < num_counters; ++id) {
+			event = perf_events[cpu][id];
+			if (event)
+				perf_event_release_kernel(event);
+		}
+
+		kfree(perf_events[cpu]);
+	}
+
+	kfree(counter_config);
+	exit_driverfs();
+}
+
+int __init oprofile_perf_init(struct oprofile_operations *ops)
+{
+	int cpu, ret = 0;
+
+	ret = init_driverfs();
+	if (ret)
+		return ret;
+
+	memset(&perf_events, 0, sizeof(perf_events));
+
+	num_counters = perf_num_counters();
+	if (num_counters <= 0) {
+		pr_info("oprofile: no performance counters\n");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	counter_config = kcalloc(num_counters,
+			sizeof(struct op_counter_config), GFP_KERNEL);
+
+	if (!counter_config) {
+		pr_info("oprofile: failed to allocate %d "
+				"counters\n", num_counters);
+		ret = -ENOMEM;
+		num_counters = 0;
+		goto out;
+	}
+
+	for_each_possible_cpu(cpu) {
+		perf_events[cpu] = kcalloc(num_counters,
+				sizeof(struct perf_event *), GFP_KERNEL);
+		if (!perf_events[cpu]) {
+			pr_info("oprofile: failed to allocate %d perf events "
+					"for cpu %d\n", num_counters, cpu);
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+
+	ops->create_files	= oprofile_perf_create_files;
+	ops->setup		= oprofile_perf_setup;
+	ops->start		= oprofile_perf_start;
+	ops->stop		= oprofile_perf_stop;
+	ops->shutdown		= oprofile_perf_stop;
+	ops->cpu_type		= op_name_from_perf_id();
+
+	if (!ops->cpu_type)
+		ret = -ENODEV;
+	else
+		pr_info("oprofile: using %s\n", ops->cpu_type);
+
+out:
+	if (ret)
+		oprofile_perf_exit();
+
+	return ret;
+}
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index 2766a6d..1944621 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -91,16 +91,20 @@
 
 static ssize_t ulong_write_file(struct file *file, char const __user *buf, size_t count, loff_t *offset)
 {
-	unsigned long *value = file->private_data;
+	unsigned long value;
 	int retval;
 
 	if (*offset)
 		return -EINVAL;
 
-	retval = oprofilefs_ulong_from_user(value, buf, count);
-
+	retval = oprofilefs_ulong_from_user(&value, buf, count);
 	if (retval)
 		return retval;
+
+	retval = oprofile_set_ulong(file->private_data, value);
+	if (retval)
+		return retval;
+
 	return count;
 }
 
@@ -126,50 +130,41 @@
 };
 
 
-static struct dentry *__oprofilefs_create_file(struct super_block *sb,
+static int __oprofilefs_create_file(struct super_block *sb,
 	struct dentry *root, char const *name, const struct file_operations *fops,
-	int perm)
+	int perm, void *priv)
 {
 	struct dentry *dentry;
 	struct inode *inode;
 
 	dentry = d_alloc_name(root, name);
 	if (!dentry)
-		return NULL;
+		return -ENOMEM;
 	inode = oprofilefs_get_inode(sb, S_IFREG | perm);
 	if (!inode) {
 		dput(dentry);
-		return NULL;
+		return -ENOMEM;
 	}
 	inode->i_fop = fops;
 	d_add(dentry, inode);
-	return dentry;
+	dentry->d_inode->i_private = priv;
+	return 0;
 }
 
 
 int oprofilefs_create_ulong(struct super_block *sb, struct dentry *root,
 	char const *name, unsigned long *val)
 {
-	struct dentry *d = __oprofilefs_create_file(sb, root, name,
-						     &ulong_fops, 0644);
-	if (!d)
-		return -EFAULT;
-
-	d->d_inode->i_private = val;
-	return 0;
+	return __oprofilefs_create_file(sb, root, name,
+					&ulong_fops, 0644, val);
 }
 
 
 int oprofilefs_create_ro_ulong(struct super_block *sb, struct dentry *root,
 	char const *name, unsigned long *val)
 {
-	struct dentry *d = __oprofilefs_create_file(sb, root, name,
-						     &ulong_ro_fops, 0444);
-	if (!d)
-		return -EFAULT;
-
-	d->d_inode->i_private = val;
-	return 0;
+	return __oprofilefs_create_file(sb, root, name,
+					&ulong_ro_fops, 0444, val);
 }
 
 
@@ -189,31 +184,22 @@
 int oprofilefs_create_ro_atomic(struct super_block *sb, struct dentry *root,
 	char const *name, atomic_t *val)
 {
-	struct dentry *d = __oprofilefs_create_file(sb, root, name,
-						     &atomic_ro_fops, 0444);
-	if (!d)
-		return -EFAULT;
-
-	d->d_inode->i_private = val;
-	return 0;
+	return __oprofilefs_create_file(sb, root, name,
+					&atomic_ro_fops, 0444, val);
 }
 
 
 int oprofilefs_create_file(struct super_block *sb, struct dentry *root,
 	char const *name, const struct file_operations *fops)
 {
-	if (!__oprofilefs_create_file(sb, root, name, fops, 0644))
-		return -EFAULT;
-	return 0;
+	return __oprofilefs_create_file(sb, root, name, fops, 0644, NULL);
 }
 
 
 int oprofilefs_create_file_perm(struct super_block *sb, struct dentry *root,
 	char const *name, const struct file_operations *fops, int perm)
 {
-	if (!__oprofilefs_create_file(sb, root, name, fops, perm))
-		return -EFAULT;
-	return 0;
+	return __oprofilefs_create_file(sb, root, name, fops, perm, NULL);
 }
 
 
diff --git a/drivers/parport/parport_cs.c b/drivers/parport/parport_cs.c
index 23e50f4..787ebde 100644
--- a/drivers/parport/parport_cs.c
+++ b/drivers/parport/parport_cs.c
@@ -48,7 +48,6 @@
 #include <linux/parport.h>
 #include <linux/parport_pc.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/cisreg.h>
@@ -81,14 +80,6 @@
 static int parport_config(struct pcmcia_device *link);
 static void parport_cs_release(struct pcmcia_device *);
 
-/*======================================================================
-
-    parport_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int parport_probe(struct pcmcia_device *link)
 {
     parport_info_t *info;
@@ -101,23 +92,11 @@
     link->priv = info;
     info->p_dev = link;
 
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-    link->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
     return parport_config(link);
 } /* parport_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void parport_detach(struct pcmcia_device *link)
 {
     dev_dbg(&link->dev, "parport_detach\n");
@@ -127,36 +106,14 @@
     kfree(link->priv);
 } /* parport_detach */
 
-/*======================================================================
-
-    parport_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    parport device available to the system.
-
-======================================================================*/
-
-static int parport_config_check(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int parport_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		if (epp_mode)
-			p_dev->conf.ConfigIndex |= FORCE_EPP_MODE;
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin == 2) {
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		if (pcmcia_request_io(p_dev) != 0)
-			return -ENODEV;
-		return 0;
-	}
-	return -ENODEV;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int parport_config(struct pcmcia_device *link)
@@ -167,13 +124,16 @@
 
     dev_dbg(&link->dev, "parport_config\n");
 
+    if (epp_mode)
+	    link->config_index |= FORCE_EPP_MODE;
+
     ret = pcmcia_loop_config(link, parport_config_check, NULL);
     if (ret)
 	    goto failed;
 
     if (!link->irq)
 	    goto failed;
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -202,14 +162,6 @@
     return -ENODEV;
 } /* parport_config */
 
-/*======================================================================
-
-    After a card is removed, parport_cs_release() will unregister the
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-    
-======================================================================*/
-
 static void parport_cs_release(struct pcmcia_device *link)
 {
 	parport_info_t *info = link->priv;
@@ -236,9 +188,7 @@
 
 static struct pcmcia_driver parport_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "parport_cs",
-	},
+	.name		= "parport_cs",
 	.probe		= parport_probe,
 	.remove		= parport_detach,
 	.id_table	= parport_ids,
diff --git a/drivers/parport/share.c b/drivers/parport/share.c
index dffa5d4..a2d9d1e 100644
--- a/drivers/parport/share.c
+++ b/drivers/parport/share.c
@@ -306,7 +306,7 @@
 	spin_lock_init(&tmp->pardevice_lock);
 	tmp->ieee1284.mode = IEEE1284_MODE_COMPAT;
 	tmp->ieee1284.phase = IEEE1284_PH_FWD_IDLE;
-	init_MUTEX_LOCKED (&tmp->ieee1284.irq); /* actually a semaphore at 0 */
+	sema_init(&tmp->ieee1284.irq, 0);
 	tmp->spintime = parport_default_spintime;
 	atomic_set (&tmp->ref_count, 1);
 	INIT_LIST_HEAD(&tmp->full_list);
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 0a19708..0157708 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -36,6 +36,7 @@
 #include <linux/tboot.h>
 #include <linux/dmi.h>
 #include <linux/slab.h>
+#include <asm/iommu_table.h>
 
 #define PREFIX "DMAR: "
 
@@ -687,7 +688,7 @@
 	return 0;
 }
 
-void __init detect_intel_iommu(void)
+int __init detect_intel_iommu(void)
 {
 	int ret;
 
@@ -723,6 +724,8 @@
 	}
 	early_acpi_os_unmap_memory(dmar_tbl, dmar_tbl_size);
 	dmar_tbl = NULL;
+
+	return ret ? 1 : -ENODEV;
 }
 
 
@@ -1221,9 +1224,9 @@
 	}
 }
 
-void dmar_msi_unmask(unsigned int irq)
+void dmar_msi_unmask(struct irq_data *data)
 {
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 	unsigned long flag;
 
 	/* unmask it */
@@ -1234,10 +1237,10 @@
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 }
 
-void dmar_msi_mask(unsigned int irq)
+void dmar_msi_mask(struct irq_data *data)
 {
 	unsigned long flag;
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 
 	/* mask it */
 	spin_lock_irqsave(&iommu->register_lock, flag);
@@ -1455,3 +1458,4 @@
 		return 0;
 	return dmar->flags & 0x1;
 }
+IOMMU_INIT_POST(detect_intel_iommu);
diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c
index 98abf8b..834842a 100644
--- a/drivers/pci/htirq.c
+++ b/drivers/pci/htirq.c
@@ -57,28 +57,22 @@
 	*msg = cfg->msg;
 }
 
-void mask_ht_irq(unsigned int irq)
+void mask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	cfg = get_irq_data(irq);
-
-	msg = cfg->msg;
 	msg.address_lo |= 1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
-void unmask_ht_irq(unsigned int irq)
+void unmask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	cfg = get_irq_data(irq);
-
-	msg = cfg->msg;
 	msg.address_lo &= ~1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
 /**
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index fd1d286..ec87cd6 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -46,109 +46,24 @@
 }
 early_param("intremap", setup_intremap);
 
-struct irq_2_iommu {
-	struct intel_iommu *iommu;
-	u16 irte_index;
-	u16 sub_handle;
-	u8  irte_mask;
-};
-
-#ifdef CONFIG_GENERIC_HARDIRQS
-static struct irq_2_iommu *get_one_free_irq_2_iommu(int node)
-{
-	struct irq_2_iommu *iommu;
-
-	iommu = kzalloc_node(sizeof(*iommu), GFP_ATOMIC, node);
-	printk(KERN_DEBUG "alloc irq_2_iommu on node %d\n", node);
-
-	return iommu;
-}
-
-static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-
-	if (WARN_ON_ONCE(!desc))
-		return NULL;
-
-	return desc->irq_2_iommu;
-}
-
-static struct irq_2_iommu *irq_2_iommu_alloc(unsigned int irq)
-{
-	struct irq_desc *desc;
-	struct irq_2_iommu *irq_iommu;
-
-	desc = irq_to_desc(irq);
-	if (!desc) {
-		printk(KERN_INFO "can not get irq_desc for %d\n", irq);
-		return NULL;
-	}
-
-	irq_iommu = desc->irq_2_iommu;
-
-	if (!irq_iommu)
-		desc->irq_2_iommu = get_one_free_irq_2_iommu(irq_node(irq));
-
-	return desc->irq_2_iommu;
-}
-
-#else /* !CONFIG_SPARSE_IRQ */
-
-static struct irq_2_iommu irq_2_iommuX[NR_IRQS];
-
-static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
-{
-	if (irq < nr_irqs)
-		return &irq_2_iommuX[irq];
-
-	return NULL;
-}
-static struct irq_2_iommu *irq_2_iommu_alloc(unsigned int irq)
-{
-	return irq_2_iommu(irq);
-}
-#endif
-
 static DEFINE_SPINLOCK(irq_2_ir_lock);
 
-static struct irq_2_iommu *valid_irq_2_iommu(unsigned int irq)
+static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
 {
-	struct irq_2_iommu *irq_iommu;
-
-	irq_iommu = irq_2_iommu(irq);
-
-	if (!irq_iommu)
-		return NULL;
-
-	if (!irq_iommu->iommu)
-		return NULL;
-
-	return irq_iommu;
-}
-
-int irq_remapped(int irq)
-{
-	return valid_irq_2_iommu(irq) != NULL;
+	struct irq_cfg *cfg = get_irq_chip_data(irq);
+	return cfg ? &cfg->irq_2_iommu : NULL;
 }
 
 int get_irte(int irq, struct irte *entry)
 {
-	int index;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int index;
 
-	if (!entry)
+	if (!entry || !irq_iommu)
 		return -1;
 
 	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
 
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 	*entry = *(irq_iommu->iommu->ir_table->base + index);
@@ -160,21 +75,15 @@
 int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 {
 	struct ir_table *table = iommu->ir_table;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	u16 index, start_index;
 	unsigned int mask = 0;
 	unsigned long flags;
 	int i;
 
-	if (!count)
+	if (!count || !irq_iommu)
 		return -1;
 
-#ifndef CONFIG_SPARSE_IRQ
-	/* protect irq_2_iommu_alloc later */
-	if (irq >= nr_irqs)
-		return -1;
-#endif
-
 	/*
 	 * start the IRTE search from index 0.
 	 */
@@ -214,13 +123,6 @@
 	for (i = index; i < index + count; i++)
 		table->base[i].present = 1;
 
-	irq_iommu = irq_2_iommu_alloc(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		printk(KERN_ERR "can't allocate irq_2_iommu\n");
-		return -1;
-	}
-
 	irq_iommu->iommu = iommu;
 	irq_iommu->irte_index =  index;
 	irq_iommu->sub_handle = 0;
@@ -244,17 +146,14 @@
 
 int map_irq_to_irte_handle(int irq, u16 *sub_handle)
 {
-	int index;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int index;
+
+	if (!irq_iommu)
+		return -1;
 
 	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
 	*sub_handle = irq_iommu->sub_handle;
 	index = irq_iommu->irte_index;
 	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
@@ -263,18 +162,13 @@
 
 int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 {
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
 
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-
-	irq_iommu = irq_2_iommu_alloc(irq);
-
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		printk(KERN_ERR "can't allocate irq_2_iommu\n");
+	if (!irq_iommu)
 		return -1;
-	}
+
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 
 	irq_iommu->iommu = iommu;
 	irq_iommu->irte_index = index;
@@ -286,43 +180,18 @@
 	return 0;
 }
 
-int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
-{
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	irq_iommu->iommu = NULL;
-	irq_iommu->irte_index = 0;
-	irq_iommu->sub_handle = 0;
-	irq_2_iommu(irq)->irte_mask = 0;
-
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return 0;
-}
-
 int modify_irte(int irq, struct irte *irte_modified)
 {
-	int rc;
-	int index;
-	struct irte *irte;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	struct intel_iommu *iommu;
-	struct irq_2_iommu *irq_iommu;
 	unsigned long flags;
+	struct irte *irte;
+	int rc, index;
+
+	if (!irq_iommu)
+		return -1;
 
 	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
 
 	iommu = irq_iommu->iommu;
 
@@ -339,31 +208,6 @@
 	return rc;
 }
 
-int flush_irte(int irq)
-{
-	int rc;
-	int index;
-	struct intel_iommu *iommu;
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	iommu = irq_iommu->iommu;
-
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
-
-	rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return rc;
-}
-
 struct intel_iommu *map_hpet_to_ir(u8 hpet_id)
 {
 	int i;
@@ -420,16 +264,14 @@
 
 int free_irte(int irq)
 {
-	int rc = 0;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int rc;
+
+	if (!irq_iommu)
+		return -1;
 
 	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
 
 	rc = clear_entries(irq_iommu);
 
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 69b7be3..5fcf5ae 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -170,33 +170,31 @@
 	desc->masked = __msix_mask_irq(desc, flag);
 }
 
-static void msi_set_mask_bit(unsigned irq, u32 flag)
+static void msi_set_mask_bit(struct irq_data *data, u32 flag)
 {
-	struct msi_desc *desc = get_irq_msi(irq);
+	struct msi_desc *desc = irq_data_get_msi(data);
 
 	if (desc->msi_attrib.is_msix) {
 		msix_mask_irq(desc, flag);
 		readl(desc->mask_base);		/* Flush write to device */
 	} else {
-		unsigned offset = irq - desc->dev->irq;
+		unsigned offset = data->irq - desc->dev->irq;
 		msi_mask_irq(desc, 1 << offset, flag << offset);
 	}
 }
 
-void mask_msi_irq(unsigned int irq)
+void mask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 1);
+	msi_set_mask_bit(data, 1);
 }
 
-void unmask_msi_irq(unsigned int irq)
+void unmask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 0);
+	msi_set_mask_bit(data, 0);
 }
 
-void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	BUG_ON(entry->dev->current_state != PCI_D0);
 
 	if (entry->msi_attrib.is_msix) {
@@ -227,15 +225,13 @@
 
 void read_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	read_msi_msg_desc(desc, msg);
+	__read_msi_msg(entry, msg);
 }
 
-void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	/* Assert that the cache is valid, assuming that
 	 * valid messages are not all-zeroes. */
 	BUG_ON(!(entry->msg.address_hi | entry->msg.address_lo |
@@ -246,15 +242,13 @@
 
 void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	get_cached_msi_msg_desc(desc, msg);
+	__get_cached_msi_msg(entry, msg);
 }
 
-void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	if (entry->dev->current_state != PCI_D0) {
 		/* Don't touch the hardware now */
 	} else if (entry->msi_attrib.is_msix) {
@@ -292,9 +286,9 @@
 
 void write_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	write_msi_msg_desc(desc, msg);
+	__write_msi_msg(entry, msg);
 }
 
 static void free_msi_irqs(struct pci_dev *dev)
diff --git a/drivers/pcmcia/au1000_generic.c b/drivers/pcmcia/au1000_generic.c
index 88c4c40..95dd7c6 100644
--- a/drivers/pcmcia/au1000_generic.c
+++ b/drivers/pcmcia/au1000_generic.c
@@ -441,14 +441,12 @@
 
 
 out_err:
-	flush_scheduled_work();
 	ops->hw_shutdown(skt);
 	while (i-- > 0) {
 		skt = PCMCIA_SOCKET(i);
 
 		del_timer_sync(&skt->poll_timer);
 		pcmcia_unregister_socket(&skt->socket);
-		flush_scheduled_work();
 		if (i == 0) {
 			iounmap(skt->virt_io + (u32)mips_io_port_base);
 			skt->virt_io = NULL;
@@ -480,7 +478,6 @@
 
 		del_timer_sync(&skt->poll_timer);
 		pcmcia_unregister_socket(&skt->socket);
-		flush_scheduled_work();
 		skt->ops->hw_shutdown(skt);
 		au1x00_pcmcia_config_skt(skt, &dead_socket);
 		iounmap(skt->virt_io + (u32)mips_io_port_base);
diff --git a/drivers/pcmcia/au1000_generic.h b/drivers/pcmcia/au1000_generic.h
index 67530ce..5c36bda 100644
--- a/drivers/pcmcia/au1000_generic.h
+++ b/drivers/pcmcia/au1000_generic.h
@@ -23,7 +23,6 @@
 
 /* include the world */
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 #include <pcmcia/cistpl.h>
 #include "cs_internal.h"
diff --git a/drivers/pcmcia/au1000_pb1x00.c b/drivers/pcmcia/au1000_pb1x00.c
index 807f2d7..b239664 100644
--- a/drivers/pcmcia/au1000_pb1x00.c
+++ b/drivers/pcmcia/au1000_pb1x00.c
@@ -31,7 +31,6 @@
 #include <linux/proc_fs.h>
 #include <linux/types.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 #include <pcmcia/cistpl.h>
 
diff --git a/drivers/pcmcia/cistpl.c b/drivers/pcmcia/cistpl.c
index 91414a0..884a984 100644
--- a/drivers/pcmcia/cistpl.c
+++ b/drivers/pcmcia/cistpl.c
@@ -28,7 +28,6 @@
 #include <asm/unaligned.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/cistpl.h>
 #include "cs_internal.h"
diff --git a/drivers/pcmcia/cs.c b/drivers/pcmcia/cs.c
index 2ec8ac9..d9ea192 100644
--- a/drivers/pcmcia/cs.c
+++ b/drivers/pcmcia/cs.c
@@ -33,7 +33,6 @@
 #include <asm/irq.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -845,7 +844,7 @@
 	return __pcmcia_pm_op(dev, socket_early_resume);
 }
 
-static int pcmcia_socket_dev_resume(struct device *dev)
+static int __used pcmcia_socket_dev_resume(struct device *dev)
 {
 	return __pcmcia_pm_op(dev, socket_late_resume);
 }
diff --git a/drivers/pcmcia/cs_internal.h b/drivers/pcmcia/cs_internal.h
index da055dc..7f1953f 100644
--- a/drivers/pcmcia/cs_internal.h
+++ b/drivers/pcmcia/cs_internal.h
@@ -33,18 +33,9 @@
 typedef struct config_t {
 	struct kref	ref;
 	unsigned int	state;
-	unsigned int	Attributes;
-	unsigned int	IntType;
-	unsigned int	ConfigBase;
-	unsigned char	Status, Pin, Copy, Option, ExtStatus;
-	unsigned int	CardValues;
 
 	struct resource io[MAX_IO_WIN]; /* io ports */
 	struct resource mem[MAX_WIN];   /* mem areas */
-
-	struct {
-		u_int	Attributes;
-	} irq;
 } config_t;
 
 
diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
index 55570d9..100c441 100644
--- a/drivers/pcmcia/ds.c
+++ b/drivers/pcmcia/ds.c
@@ -26,7 +26,6 @@
 #include <linux/dma-mapping.h>
 #include <linux/slab.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/ss.h>
@@ -52,7 +51,7 @@
 
 	if (!p_drv->probe || !p_drv->remove)
 		printk(KERN_DEBUG "pcmcia: %s lacks a requisite callback "
-		       "function\n", p_drv->drv.name);
+		       "function\n", p_drv->name);
 
 	while (did && did->match_flags) {
 		for (i = 0; i < 4; i++) {
@@ -65,7 +64,7 @@
 
 			printk(KERN_DEBUG "pcmcia: %s: invalid hash for "
 			       "product string \"%s\": is 0x%x, should "
-			       "be 0x%x\n", p_drv->drv.name, did->prod_id[i],
+			       "be 0x%x\n", p_drv->name, did->prod_id[i],
 			       did->prod_id_hash[i], hash);
 			printk(KERN_DEBUG "pcmcia: see "
 				"Documentation/pcmcia/devicetable.txt for "
@@ -180,10 +179,11 @@
 	/* initialize common fields */
 	driver->drv.bus = &pcmcia_bus_type;
 	driver->drv.owner = driver->owner;
+	driver->drv.name = driver->name;
 	mutex_init(&driver->dynids.lock);
 	INIT_LIST_HEAD(&driver->dynids.list);
 
-	pr_debug("registering driver %s\n", driver->drv.name);
+	pr_debug("registering driver %s\n", driver->name);
 
 	error = driver_register(&driver->drv);
 	if (error < 0)
@@ -203,7 +203,7 @@
  */
 void pcmcia_unregister_driver(struct pcmcia_driver *driver)
 {
-	pr_debug("unregistering driver %s\n", driver->drv.name);
+	pr_debug("unregistering driver %s\n", driver->name);
 	driver_unregister(&driver->drv);
 	pcmcia_free_dynids(driver);
 }
@@ -264,7 +264,7 @@
 	p_drv = to_pcmcia_drv(dev->driver);
 	s = p_dev->socket;
 
-	dev_dbg(dev, "trying to bind to %s\n", p_drv->drv.name);
+	dev_dbg(dev, "trying to bind to %s\n", p_drv->name);
 
 	if ((!p_drv->probe) || (!p_dev->function_config) ||
 	    (!try_module_get(p_drv->owner))) {
@@ -276,21 +276,28 @@
 	ret = pccard_read_tuple(p_dev->socket, p_dev->func, CISTPL_CONFIG,
 				&cis_config);
 	if (!ret) {
-		p_dev->conf.ConfigBase = cis_config.base;
-		p_dev->conf.Present = cis_config.rmask[0];
+		p_dev->config_base = cis_config.base;
+		p_dev->config_regs = cis_config.rmask[0];
+		dev_dbg(dev, "base %x, regs %x", p_dev->config_base,
+			p_dev->config_regs);
 	} else {
 		dev_printk(KERN_INFO, dev,
 			   "pcmcia: could not parse base and rmask0 of CIS\n");
-		p_dev->conf.ConfigBase = 0;
-		p_dev->conf.Present = 0;
+		p_dev->config_base = 0;
+		p_dev->config_regs = 0;
 	}
 
 	ret = p_drv->probe(p_dev);
 	if (ret) {
 		dev_dbg(dev, "binding to %s failed with %d\n",
-			   p_drv->drv.name, ret);
+			   p_drv->name, ret);
 		goto put_module;
 	}
+	dev_dbg(dev, "%s bound: Vpp %d.%d, idx %x, IRQ %d", p_drv->name,
+		p_dev->vpp/10, p_dev->vpp%10, p_dev->config_index, p_dev->irq);
+	dev_dbg(dev, "resources: ioport %pR %pR iomem %pR %pR %pR",
+		p_dev->resource[0], p_dev->resource[1], p_dev->resource[2],
+		p_dev->resource[3], p_dev->resource[4]);
 
 	mutex_lock(&s->ops_mutex);
 	if ((s->pcmcia_pfc) &&
@@ -374,13 +381,13 @@
 	if (p_dev->_irq || p_dev->_io || p_dev->_locked)
 		dev_printk(KERN_INFO, dev,
 			"pcmcia: driver %s did not release config properly\n",
-			p_drv->drv.name);
+			p_drv->name);
 
 	for (i = 0; i < MAX_WIN; i++)
 		if (p_dev->_win & CLIENT_WIN_REQ(i))
 			dev_printk(KERN_INFO, dev,
 			  "pcmcia: driver %s did not release window properly\n",
-			   p_drv->drv.name);
+			   p_drv->name);
 
 	/* references from pcmcia_probe_device */
 	pcmcia_put_dev(p_dev);
@@ -1136,7 +1143,7 @@
 			dev_printk(KERN_ERR, dev,
 				   "pcmcia: device %s (driver %s) did "
 				   "not want to go to sleep (%d)\n",
-				   p_dev->devname, p_drv->drv.name, ret);
+				   p_dev->devname, p_drv->name, ret);
 			mutex_lock(&p_dev->socket->ops_mutex);
 			p_dev->suspended = 0;
 			mutex_unlock(&p_dev->socket->ops_mutex);
@@ -1178,7 +1185,7 @@
 
 	if (p_dev->device_no == p_dev->func) {
 		dev_dbg(dev, "requesting configuration\n");
-		ret = pcmcia_request_configuration(p_dev, &p_dev->conf);
+		ret = pcmcia_enable_device(p_dev);
 		if (ret)
 			goto out;
 	}
diff --git a/drivers/pcmcia/i82092.c b/drivers/pcmcia/i82092.c
index 05d0879..fc7906e 100644
--- a/drivers/pcmcia/i82092.c
+++ b/drivers/pcmcia/i82092.c
@@ -16,7 +16,6 @@
 #include <linux/device.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 #include <asm/system.h>
 #include <asm/io.h>
diff --git a/drivers/pcmcia/i82365.c b/drivers/pcmcia/i82365.c
index 61746bd..72a033a 100644
--- a/drivers/pcmcia/i82365.c
+++ b/drivers/pcmcia/i82365.c
@@ -51,7 +51,6 @@
 #include <asm/system.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 #include <linux/isapnp.h>
 
diff --git a/drivers/pcmcia/m32r_cfc.c b/drivers/pcmcia/m32r_cfc.c
index 24de499..2adb010 100644
--- a/drivers/pcmcia/m32r_cfc.c
+++ b/drivers/pcmcia/m32r_cfc.c
@@ -27,7 +27,6 @@
 #include <asm/system.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 #undef MAX_IO_WIN	/* FIXME */
 #define MAX_IO_WIN 1
diff --git a/drivers/pcmcia/m32r_pcc.c b/drivers/pcmcia/m32r_pcc.c
index 8e47238..1511ff7 100644
--- a/drivers/pcmcia/m32r_pcc.c
+++ b/drivers/pcmcia/m32r_pcc.c
@@ -28,7 +28,6 @@
 #include <asm/addrspace.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 /* XXX: should be moved into asm/irq.h */
 #define PCC0_IRQ 24
diff --git a/drivers/pcmcia/m8xx_pcmcia.c b/drivers/pcmcia/m8xx_pcmcia.c
index f0ecad9..99d4f23 100644
--- a/drivers/pcmcia/m8xx_pcmcia.c
+++ b/drivers/pcmcia/m8xx_pcmcia.c
@@ -59,7 +59,6 @@
 #include <asm/irq.h>
 #include <asm/fs_pd.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 
 #define pcmcia_info(args...) printk(KERN_INFO "m8xx_pcmcia: "args)
diff --git a/drivers/pcmcia/o2micro.h b/drivers/pcmcia/o2micro.h
index e74beba..5096e92 100644
--- a/drivers/pcmcia/o2micro.h
+++ b/drivers/pcmcia/o2micro.h
@@ -153,14 +153,14 @@
 
 		if (use_speedup) {
 			dev_info(&socket->dev->dev,
-				"O2: enabling read prefetch/write burst\n");
+				"O2: enabling read prefetch/write burst. If you experience problems or performance issues, use the yenta_socket parameter 'o2_speedup=off'\n");
 			config_writeb(socket, O2_RESERVED1,
 				      a | O2_RES_READ_PREFETCH | O2_RES_WRITE_BURST);
 			config_writeb(socket, O2_RESERVED2,
 				      b | O2_RES_READ_PREFETCH | O2_RES_WRITE_BURST);
 		} else {
 			dev_info(&socket->dev->dev,
-				"O2: disabling read prefetch/write burst\n");
+				"O2: disabling read prefetch/write burst. If you experience problems or performance issues, use the yenta_socket parameter 'o2_speedup=on'\n");
 			config_writeb(socket, O2_RESERVED1,
 				      a & ~(O2_RES_READ_PREFETCH | O2_RES_WRITE_BURST));
 			config_writeb(socket, O2_RESERVED2,
diff --git a/drivers/pcmcia/pcmcia_cis.c b/drivers/pcmcia/pcmcia_cis.c
index 0ac54da..e2c9241 100644
--- a/drivers/pcmcia/pcmcia_cis.c
+++ b/drivers/pcmcia/pcmcia_cis.c
@@ -6,7 +6,7 @@
  * are Copyright (C) 1999 David A. Hinds.  All Rights Reserved.
  *
  * Copyright (C) 1999	     David A. Hinds
- * Copyright (C) 2004-2009   Dominik Brodowski
+ * Copyright (C) 2004-2010   Dominik Brodowski
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -22,7 +22,6 @@
 #include <pcmcia/cisreg.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/ds.h>
 #include "cs_internal.h"
 
@@ -126,14 +125,24 @@
 	return ret;
 }
 
+
+/**
+ * pcmcia_io_cfg_data_width() - convert cfgtable to data path width parameter
+ */
+static int pcmcia_io_cfg_data_width(unsigned int flags)
+{
+	if (!(flags & CISTPL_IO_8BIT))
+		return IO_DATA_PATH_WIDTH_16;
+	if (!(flags & CISTPL_IO_16BIT))
+		return IO_DATA_PATH_WIDTH_8;
+	return IO_DATA_PATH_WIDTH_AUTO;
+}
+
+
 struct pcmcia_cfg_mem {
 	struct pcmcia_device *p_dev;
+	int (*conf_check) (struct pcmcia_device *p_dev, void *priv_data);
 	void *priv_data;
-	int (*conf_check) (struct pcmcia_device *p_dev,
-			   cistpl_cftable_entry_t *cfg,
-			   cistpl_cftable_entry_t *dflt,
-			   unsigned int vcc,
-			   void *priv_data);
 	cisparse_t parse;
 	cistpl_cftable_entry_t dflt;
 };
@@ -147,25 +156,102 @@
  */
 static int pcmcia_do_loop_config(tuple_t *tuple, cisparse_t *parse, void *priv)
 {
-	cistpl_cftable_entry_t *cfg = &parse->cftable_entry;
 	struct pcmcia_cfg_mem *cfg_mem = priv;
+	struct pcmcia_device *p_dev = cfg_mem->p_dev;
+	cistpl_cftable_entry_t *cfg = &parse->cftable_entry;
+	cistpl_cftable_entry_t *dflt = &cfg_mem->dflt;
+	unsigned int flags = p_dev->config_flags;
+	unsigned int vcc = p_dev->socket->socket.Vcc;
+
+	dev_dbg(&p_dev->dev, "testing configuration %x, autoconf %x\n",
+		cfg->index, flags);
 
 	/* default values */
-	cfg_mem->p_dev->conf.ConfigIndex = cfg->index;
+	cfg_mem->p_dev->config_index = cfg->index;
 	if (cfg->flags & CISTPL_CFTABLE_DEFAULT)
 		cfg_mem->dflt = *cfg;
 
-	return cfg_mem->conf_check(cfg_mem->p_dev, cfg, &cfg_mem->dflt,
-				   cfg_mem->p_dev->socket->socket.Vcc,
-				   cfg_mem->priv_data);
+	/* check for matching Vcc? */
+	if (flags & CONF_AUTO_CHECK_VCC) {
+		if (cfg->vcc.present & (1 << CISTPL_POWER_VNOM)) {
+			if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM] / 10000)
+				return -ENODEV;
+		} else if (dflt->vcc.present & (1 << CISTPL_POWER_VNOM)) {
+			if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM] / 10000)
+				return -ENODEV;
+		}
+	}
+
+	/* set Vpp? */
+	if (flags & CONF_AUTO_SET_VPP) {
+		if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM))
+			p_dev->vpp = cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
+		else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM))
+			p_dev->vpp =
+				dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
+	}
+
+	/* enable audio? */
+	if ((flags & CONF_AUTO_AUDIO) && (cfg->flags & CISTPL_CFTABLE_AUDIO))
+		p_dev->config_flags |= CONF_ENABLE_SPKR;
+
+
+	/* IO window settings? */
+	if (flags & CONF_AUTO_SET_IO) {
+		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
+		int i = 0;
+
+		p_dev->resource[0]->start = p_dev->resource[0]->end = 0;
+		p_dev->resource[1]->start = p_dev->resource[1]->end = 0;
+		if (io->nwin == 0)
+			return -ENODEV;
+
+		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+		p_dev->resource[0]->flags |=
+					pcmcia_io_cfg_data_width(io->flags);
+		if (io->nwin > 1) {
+			/* For multifunction cards, by convention, we
+			 * configure the network function with window 0,
+			 * and serial with window 1 */
+			i = (io->win[1].len > io->win[0].len);
+			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
+			p_dev->resource[1]->start = io->win[1-i].base;
+			p_dev->resource[1]->end = io->win[1-i].len;
+		}
+		p_dev->resource[0]->start = io->win[i].base;
+		p_dev->resource[0]->end = io->win[i].len;
+		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
+	}
+
+	/* MEM window settings? */
+	if (flags & CONF_AUTO_SET_IOMEM) {
+		/* so far, we only set one memory window */
+		cistpl_mem_t *mem = (cfg->mem.nwin) ? &cfg->mem : &dflt->mem;
+
+		p_dev->resource[2]->start = p_dev->resource[2]->end = 0;
+		if (mem->nwin == 0)
+			return -ENODEV;
+
+		p_dev->resource[2]->start = mem->win[0].host_addr;
+		p_dev->resource[2]->end = mem->win[0].len;
+		if (p_dev->resource[2]->end < 0x1000)
+			p_dev->resource[2]->end = 0x1000;
+		p_dev->card_addr = mem->win[0].card_addr;
+	}
+
+	dev_dbg(&p_dev->dev,
+		"checking configuration %x: %pr %pr %pr (%d lines)\n",
+		p_dev->config_index, p_dev->resource[0], p_dev->resource[1],
+		p_dev->resource[2], p_dev->io_lines);
+
+	return cfg_mem->conf_check(p_dev, cfg_mem->priv_data);
 }
 
 /**
  * pcmcia_loop_config() - loop over configuration options
  * @p_dev:	the struct pcmcia_device which we need to loop for.
  * @conf_check:	function to call for each configuration option.
- *		It gets passed the struct pcmcia_device, the CIS data
- *		describing the configuration option, and private data
+ *		It gets passed the struct pcmcia_device and private data
  *		being passed to pcmcia_loop_config()
  * @priv_data:	private data to be passed to the conf_check function.
  *
@@ -175,9 +261,6 @@
  */
 int pcmcia_loop_config(struct pcmcia_device *p_dev,
 		       int	(*conf_check)	(struct pcmcia_device *p_dev,
-						 cistpl_cftable_entry_t *cfg,
-						 cistpl_cftable_entry_t *dflt,
-						 unsigned int vcc,
 						 void *priv_data),
 		       void *priv_data)
 {
diff --git a/drivers/pcmcia/pcmcia_resource.c b/drivers/pcmcia/pcmcia_resource.c
index 9ba4dad..a9af0d7 100644
--- a/drivers/pcmcia/pcmcia_resource.c
+++ b/drivers/pcmcia/pcmcia_resource.c
@@ -6,7 +6,7 @@
  * are Copyright (C) 1999 David A. Hinds.  All Rights Reserved.
  *
  * Copyright (C) 1999	     David A. Hinds
- * Copyright (C) 2004-2005   Dominik Brodowski
+ * Copyright (C) 2004-2010   Dominik Brodowski
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -26,7 +26,6 @@
 #include <asm/irq.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -56,6 +55,12 @@
 }
 
 
+/**
+ * release_io_space() - release IO ports allocated with alloc_io_space()
+ * @s: pcmcia socket
+ * @res: resource to release
+ *
+ */
 static void release_io_space(struct pcmcia_socket *s, struct resource *res)
 {
 	resource_size_t num = resource_size(res);
@@ -81,9 +86,14 @@
 			}
 		}
 	}
-} /* release_io_space */
+}
 
-/** alloc_io_space
+
+/**
+ * alloc_io_space() - allocate IO ports for use by a PCMCIA device
+ * @s: pcmcia socket
+ * @res: resource to allocate (begin: begin, end: size)
+ * @lines: number of IO lines decoded by the PCMCIA card
  *
  * Special stuff for managing IO windows, because they are scarce
  */
@@ -135,7 +145,7 @@
 	}
 	dev_dbg(&s->dev, "alloc_io_space request result %d: %pR\n", ret, res);
 	return ret;
-} /* alloc_io_space */
+}
 
 
 /**
@@ -168,14 +178,14 @@
 		return -EACCES;
 	}
 
-	addr = (c->ConfigBase + where) >> 1;
+	addr = (p_dev->config_base + where) >> 1;
 
 	ret = accessf(s, 1, addr, 1, val);
 
 	mutex_unlock(&s->ops_mutex);
 
 	return ret;
-} /* pcmcia_access_config */
+}
 
 
 /**
@@ -204,11 +214,20 @@
 EXPORT_SYMBOL(pcmcia_write_config_byte);
 
 
-int pcmcia_map_mem_page(struct pcmcia_device *p_dev, window_handle_t wh,
+/**
+ * pcmcia_map_mem_page() - modify iomem window to point to a different offset
+ * @p_dev: pcmcia device
+ * @res: iomem resource already enabled by pcmcia_request_window()
+ * @offset: card_offset to map
+ *
+ * pcmcia_map_mem_page() modifies what can be read and written by accessing
+ * an iomem range previously enabled by pcmcia_request_window(), by setting
+ * the card_offset value to @offset.
+ */
+int pcmcia_map_mem_page(struct pcmcia_device *p_dev, struct resource *res,
 			unsigned int offset)
 {
 	struct pcmcia_socket *s = p_dev->socket;
-	struct resource *res = wh;
 	unsigned int w;
 	int ret;
 
@@ -223,98 +242,111 @@
 		dev_warn(&p_dev->dev, "failed to set_mem_map\n");
 	mutex_unlock(&s->ops_mutex);
 	return ret;
-} /* pcmcia_map_mem_page */
+}
 EXPORT_SYMBOL(pcmcia_map_mem_page);
 
 
-/** pcmcia_modify_configuration
+/**
+ * pcmcia_fixup_iowidth() - reduce io width to 8bit
+ * @p_dev: pcmcia device
  *
- * Modify a locked socket configuration
+ * pcmcia_fixup_iowidth() allows a PCMCIA device driver to reduce the
+ * IO width to 8bit after having called pcmcia_enable_device()
+ * previously.
  */
-int pcmcia_modify_configuration(struct pcmcia_device *p_dev,
-				modconf_t *mod)
+int pcmcia_fixup_iowidth(struct pcmcia_device *p_dev)
 {
-	struct pcmcia_socket *s;
-	config_t *c;
-	int ret;
-
-	s = p_dev->socket;
+	struct pcmcia_socket *s = p_dev->socket;
+	pccard_io_map io_off = { 0, 0, 0, 0, 1 };
+	pccard_io_map io_on;
+	int i, ret = 0;
 
 	mutex_lock(&s->ops_mutex);
-	c = p_dev->function_config;
 
-	if (!(s->state & SOCKET_PRESENT)) {
-		dev_dbg(&p_dev->dev, "No card present\n");
-		ret = -ENODEV;
-		goto unlock;
-	}
-	if (!(c->state & CONFIG_LOCKED)) {
-		dev_dbg(&p_dev->dev, "Configuration isnt't locked\n");
+	dev_dbg(&p_dev->dev, "fixup iowidth to 8bit\n");
+
+	if (!(s->state & SOCKET_PRESENT) ||
+		!(p_dev->function_config->state & CONFIG_LOCKED)) {
+		dev_dbg(&p_dev->dev, "No card? Config not locked?\n");
 		ret = -EACCES;
 		goto unlock;
 	}
 
-	if (mod->Attributes & (CONF_IRQ_CHANGE_VALID | CONF_VCC_CHANGE_VALID)) {
-		dev_dbg(&p_dev->dev,
-			"changing Vcc or IRQ is not allowed at this time\n");
-		ret = -EINVAL;
-		goto unlock;
+	io_on.speed = io_speed;
+	for (i = 0; i < MAX_IO_WIN; i++) {
+		if (!s->io[i].res)
+			continue;
+		io_off.map = i;
+		io_on.map = i;
+
+		io_on.flags = MAP_ACTIVE | IO_DATA_PATH_WIDTH_8;
+		io_on.start = s->io[i].res->start;
+		io_on.stop = s->io[i].res->end;
+
+		s->ops->set_io_map(s, &io_off);
+		mdelay(40);
+		s->ops->set_io_map(s, &io_on);
 	}
-
-	/* We only allow changing Vpp1 and Vpp2 to the same value */
-	if ((mod->Attributes & CONF_VPP1_CHANGE_VALID) &&
-	    (mod->Attributes & CONF_VPP2_CHANGE_VALID)) {
-		if (mod->Vpp1 != mod->Vpp2) {
-			dev_dbg(&p_dev->dev,
-				"Vpp1 and Vpp2 must be the same\n");
-			ret = -EINVAL;
-			goto unlock;
-		}
-		s->socket.Vpp = mod->Vpp1;
-		if (s->ops->set_socket(s, &s->socket)) {
-			dev_printk(KERN_WARNING, &p_dev->dev,
-				   "Unable to set VPP\n");
-			ret = -EIO;
-			goto unlock;
-		}
-	} else if ((mod->Attributes & CONF_VPP1_CHANGE_VALID) ||
-		   (mod->Attributes & CONF_VPP2_CHANGE_VALID)) {
-		dev_dbg(&p_dev->dev,
-			"changing Vcc is not allowed at this time\n");
-		ret = -EINVAL;
-		goto unlock;
-	}
-
-	if (mod->Attributes & CONF_IO_CHANGE_WIDTH) {
-		pccard_io_map io_off = { 0, 0, 0, 0, 1 };
-		pccard_io_map io_on;
-		int i;
-
-		io_on.speed = io_speed;
-		for (i = 0; i < MAX_IO_WIN; i++) {
-			if (!s->io[i].res)
-				continue;
-			io_off.map = i;
-			io_on.map = i;
-
-			io_on.flags = MAP_ACTIVE | IO_DATA_PATH_WIDTH_8;
-			io_on.start = s->io[i].res->start;
-			io_on.stop = s->io[i].res->end;
-
-			s->ops->set_io_map(s, &io_off);
-			mdelay(40);
-			s->ops->set_io_map(s, &io_on);
-		}
-	}
-	ret = 0;
 unlock:
 	mutex_unlock(&s->ops_mutex);
 
 	return ret;
-} /* modify_configuration */
-EXPORT_SYMBOL(pcmcia_modify_configuration);
+}
+EXPORT_SYMBOL(pcmcia_fixup_iowidth);
 
 
+/**
+ * pcmcia_fixup_vpp() - set Vpp to a new voltage level
+ * @p_dev: pcmcia device
+ * @new_vpp: new Vpp voltage
+ *
+ * pcmcia_fixup_vpp() allows a PCMCIA device driver to set Vpp to
+ * a new voltage level between calls to pcmcia_enable_device()
+ * and pcmcia_disable_device().
+ */
+int pcmcia_fixup_vpp(struct pcmcia_device *p_dev, unsigned char new_vpp)
+{
+	struct pcmcia_socket *s = p_dev->socket;
+	int ret = 0;
+
+	mutex_lock(&s->ops_mutex);
+
+	dev_dbg(&p_dev->dev, "fixup Vpp to %d\n", new_vpp);
+
+	if (!(s->state & SOCKET_PRESENT) ||
+		!(p_dev->function_config->state & CONFIG_LOCKED)) {
+		dev_dbg(&p_dev->dev, "No card? Config not locked?\n");
+		ret = -EACCES;
+		goto unlock;
+	}
+
+	s->socket.Vpp = new_vpp;
+	if (s->ops->set_socket(s, &s->socket)) {
+		dev_warn(&p_dev->dev, "Unable to set VPP\n");
+		ret = -EIO;
+		goto unlock;
+	}
+	p_dev->vpp = new_vpp;
+
+unlock:
+	mutex_unlock(&s->ops_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL(pcmcia_fixup_vpp);
+
+
+/**
+ * pcmcia_release_configuration() - physically disable a PCMCIA device
+ * @p_dev: pcmcia device
+ *
+ * pcmcia_release_configuration() is the 1:1 counterpart to
+ * pcmcia_enable_device(): If a PCMCIA device is no longer used by any
+ * driver, the Vpp voltage is set to 0, IRQs will no longer be generated,
+ * and I/O ranges will be disabled. As pcmcia_release_io() and
+ * pcmcia_release_window() still need to be called, device drivers are
+ * expected to call pcmcia_disable_device() instead.
+ */
 int pcmcia_release_configuration(struct pcmcia_device *p_dev)
 {
 	pccard_io_map io = { 0, 0, 0, 0, 1 };
@@ -327,7 +359,7 @@
 	if (p_dev->_locked) {
 		p_dev->_locked = 0;
 		if (--(s->lock_count) == 0) {
-			s->socket.flags = SS_OUTPUT_ENA;   /* Is this correct? */
+			s->socket.flags = SS_OUTPUT_ENA; /* Is this correct? */
 			s->socket.Vpp = 0;
 			s->socket.io_irq = 0;
 			s->ops->set_socket(s, &s->socket);
@@ -349,16 +381,18 @@
 	mutex_unlock(&s->ops_mutex);
 
 	return 0;
-} /* pcmcia_release_configuration */
+}
 
 
-/** pcmcia_release_io
+/**
+ * pcmcia_release_io() - release I/O allocated by a PCMCIA device
+ * @p_dev: pcmcia device
  *
- * Release_io() releases the I/O ranges allocated by a client.  This
- * may be invoked some time after a card ejection has already dumped
- * the actual socket configuration, so if the client is "stale", we
- * don't bother checking the port ranges against the current socket
- * values.
+ * pcmcia_release_io() releases the I/O ranges allocated by a PCMCIA
+ * device.  This may be invoked some time after a card ejection has
+ * already dumped the actual socket configuration, so if the client is
+ * "stale", we don't bother checking the port ranges against the
+ * current socket values.
  */
 static int pcmcia_release_io(struct pcmcia_device *p_dev)
 {
@@ -387,6 +421,14 @@
 } /* pcmcia_release_io */
 
 
+/**
+ * pcmcia_release_window() - release reserved iomem for PCMCIA devices
+ * @p_dev: pcmcia device
+ * @res: iomem resource to release
+ *
+ * pcmcia_release_window() releases &struct resource *res which was
+ * previously reserved by calling pcmcia_request_window().
+ */
 int pcmcia_release_window(struct pcmcia_device *p_dev, struct resource *res)
 {
 	struct pcmcia_socket *s = p_dev->socket;
@@ -420,6 +462,8 @@
 		kfree(win->res);
 		win->res = NULL;
 	}
+	res->start = res->end = 0;
+	res->flags = IORESOURCE_MEM;
 	p_dev->_win &= ~CLIENT_WIN_REQ(w);
 	mutex_unlock(&s->ops_mutex);
 
@@ -428,23 +472,30 @@
 EXPORT_SYMBOL(pcmcia_release_window);
 
 
-int pcmcia_request_configuration(struct pcmcia_device *p_dev,
-				 config_req_t *req)
+/**
+ * pcmcia_enable_device() - set up and activate a PCMCIA device
+ * @p_dev: the associated PCMCIA device
+ *
+ * pcmcia_enable_device() physically enables a PCMCIA device. It parses
+ * the flags passed to in @flags and stored in @p_dev->flags and sets up
+ * the Vpp voltage, enables the speaker line, I/O ports and store proper
+ * values to configuration registers.
+ */
+int pcmcia_enable_device(struct pcmcia_device *p_dev)
 {
 	int i;
-	u_int base;
+	unsigned int base;
 	struct pcmcia_socket *s = p_dev->socket;
 	config_t *c;
 	pccard_io_map iomap;
+	unsigned char status = 0;
+	unsigned char ext_status = 0;
+	unsigned char option = 0;
+	unsigned int flags = p_dev->config_flags;
 
 	if (!(s->state & SOCKET_PRESENT))
 		return -ENODEV;
 
-	if (req->IntType & INT_CARDBUS) {
-		dev_dbg(&p_dev->dev, "IntType may not be INT_CARDBUS\n");
-		return -EINVAL;
-	}
-
 	mutex_lock(&s->ops_mutex);
 	c = p_dev->function_config;
 	if (c->state & CONFIG_LOCKED) {
@@ -454,7 +505,7 @@
 	}
 
 	/* Do power control.  We don't allow changes in Vcc. */
-	s->socket.Vpp = req->Vpp;
+	s->socket.Vpp = p_dev->vpp;
 	if (s->ops->set_socket(s, &s->socket)) {
 		mutex_unlock(&s->ops_mutex);
 		dev_printk(KERN_WARNING, &p_dev->dev,
@@ -463,64 +514,72 @@
 	}
 
 	/* Pick memory or I/O card, DMA mode, interrupt */
-	c->IntType = req->IntType;
-	c->Attributes = req->Attributes;
-	if (req->IntType & INT_MEMORY_AND_IO)
+	if (p_dev->_io)
 		s->socket.flags |= SS_IOCARD;
-	if (req->IntType & INT_ZOOMED_VIDEO)
-		s->socket.flags |= SS_ZVCARD | SS_IOCARD;
-	if (req->Attributes & CONF_ENABLE_DMA)
-		s->socket.flags |= SS_DMA_MODE;
-	if (req->Attributes & CONF_ENABLE_SPKR)
+	if (flags & CONF_ENABLE_SPKR) {
 		s->socket.flags |= SS_SPKR_ENA;
-	if (req->Attributes & CONF_ENABLE_IRQ)
+		status = CCSR_AUDIO_ENA;
+		if (!(p_dev->config_regs & PRESENT_STATUS))
+			dev_warn(&p_dev->dev, "speaker requested, but "
+					      "PRESENT_STATUS not set!\n");
+	}
+	if (flags & CONF_ENABLE_IRQ)
 		s->socket.io_irq = s->pcmcia_irq;
 	else
 		s->socket.io_irq = 0;
+	if (flags & CONF_ENABLE_ESR) {
+		p_dev->config_regs |= PRESENT_EXT_STATUS;
+		ext_status = ESR_REQ_ATTN_ENA;
+	}
 	s->ops->set_socket(s, &s->socket);
 	s->lock_count++;
 
+	dev_dbg(&p_dev->dev,
+		"enable_device: V %d, flags %x, base %x, regs %x, idx %x\n",
+		p_dev->vpp, flags, p_dev->config_base, p_dev->config_regs,
+		p_dev->config_index);
+
 	/* Set up CIS configuration registers */
-	base = c->ConfigBase = req->ConfigBase;
-	c->CardValues = req->Present;
-	if (req->Present & PRESENT_COPY) {
-		c->Copy = req->Copy;
-		pcmcia_write_cis_mem(s, 1, (base + CISREG_SCR)>>1, 1, &c->Copy);
+	base = p_dev->config_base;
+	if (p_dev->config_regs & PRESENT_COPY) {
+		u16 tmp = 0;
+		dev_dbg(&p_dev->dev, "clearing CISREG_SCR\n");
+		pcmcia_write_cis_mem(s, 1, (base + CISREG_SCR)>>1, 1, &tmp);
 	}
-	if (req->Present & PRESENT_OPTION) {
+	if (p_dev->config_regs & PRESENT_PIN_REPLACE) {
+		u16 tmp = 0;
+		dev_dbg(&p_dev->dev, "clearing CISREG_PRR\n");
+		pcmcia_write_cis_mem(s, 1, (base + CISREG_PRR)>>1, 1, &tmp);
+	}
+	if (p_dev->config_regs & PRESENT_OPTION) {
 		if (s->functions == 1) {
-			c->Option = req->ConfigIndex & COR_CONFIG_MASK;
+			option = p_dev->config_index & COR_CONFIG_MASK;
 		} else {
-			c->Option = req->ConfigIndex & COR_MFC_CONFIG_MASK;
-			c->Option |= COR_FUNC_ENA|COR_IREQ_ENA;
-			if (req->Present & PRESENT_IOBASE_0)
-				c->Option |= COR_ADDR_DECODE;
+			option = p_dev->config_index & COR_MFC_CONFIG_MASK;
+			option |= COR_FUNC_ENA|COR_IREQ_ENA;
+			if (p_dev->config_regs & PRESENT_IOBASE_0)
+				option |= COR_ADDR_DECODE;
 		}
-		if ((req->Attributes & CONF_ENABLE_IRQ) &&
-			!(req->Attributes & CONF_ENABLE_PULSE_IRQ))
-			c->Option |= COR_LEVEL_REQ;
-		pcmcia_write_cis_mem(s, 1, (base + CISREG_COR)>>1, 1, &c->Option);
+		if ((flags & CONF_ENABLE_IRQ) &&
+			!(flags & CONF_ENABLE_PULSE_IRQ))
+			option |= COR_LEVEL_REQ;
+		pcmcia_write_cis_mem(s, 1, (base + CISREG_COR)>>1, 1, &option);
 		mdelay(40);
 	}
-	if (req->Present & PRESENT_STATUS) {
-		c->Status = req->Status;
-		pcmcia_write_cis_mem(s, 1, (base + CISREG_CCSR)>>1, 1, &c->Status);
-	}
-	if (req->Present & PRESENT_PIN_REPLACE) {
-		c->Pin = req->Pin;
-		pcmcia_write_cis_mem(s, 1, (base + CISREG_PRR)>>1, 1, &c->Pin);
-	}
-	if (req->Present & PRESENT_EXT_STATUS) {
-		c->ExtStatus = req->ExtStatus;
-		pcmcia_write_cis_mem(s, 1, (base + CISREG_ESR)>>1, 1, &c->ExtStatus);
-	}
-	if (req->Present & PRESENT_IOBASE_0) {
+	if (p_dev->config_regs & PRESENT_STATUS)
+		pcmcia_write_cis_mem(s, 1, (base + CISREG_CCSR)>>1, 1, &status);
+
+	if (p_dev->config_regs & PRESENT_EXT_STATUS)
+		pcmcia_write_cis_mem(s, 1, (base + CISREG_ESR)>>1, 1,
+					&ext_status);
+
+	if (p_dev->config_regs & PRESENT_IOBASE_0) {
 		u8 b = c->io[0].start & 0xff;
 		pcmcia_write_cis_mem(s, 1, (base + CISREG_IOBASE_0)>>1, 1, &b);
 		b = (c->io[0].start >> 8) & 0xff;
 		pcmcia_write_cis_mem(s, 1, (base + CISREG_IOBASE_1)>>1, 1, &b);
 	}
-	if (req->Present & PRESENT_IOSIZE) {
+	if (p_dev->config_regs & PRESENT_IOSIZE) {
 		u8 b = resource_size(&c->io[0]) + resource_size(&c->io[1]) - 1;
 		pcmcia_write_cis_mem(s, 1, (base + CISREG_IOSIZE)>>1, 1, &b);
 	}
@@ -551,14 +610,15 @@
 	p_dev->_locked = 1;
 	mutex_unlock(&s->ops_mutex);
 	return 0;
-} /* pcmcia_request_configuration */
-EXPORT_SYMBOL(pcmcia_request_configuration);
+} /* pcmcia_enable_device */
+EXPORT_SYMBOL(pcmcia_enable_device);
 
 
 /**
  * pcmcia_request_io() - attempt to reserve port ranges for PCMCIA devices
+ * @p_dev: the associated PCMCIA device
  *
- * pcmcia_request_io() attepts to reserve the IO port ranges specified in
+ * pcmcia_request_io() attempts to reserve the IO port ranges specified in
  * &struct pcmcia_device @p_dev->resource[0] and @p_dev->resource[1]. The
  * "start" value is the requested start of the IO port resource; "end"
  * reflects the number of ports requested. The number of IO lines requested
@@ -622,11 +682,13 @@
 
 /**
  * pcmcia_request_irq() - attempt to request a IRQ for a PCMCIA device
+ * @p_dev: the associated PCMCIA device
+ * @handler: IRQ handler to register
  *
- * pcmcia_request_irq() is a wrapper around request_irq which will allow
+ * pcmcia_request_irq() is a wrapper around request_irq() which allows
  * the PCMCIA core to clean up the registration in pcmcia_disable_device().
  * Drivers are free to use request_irq() directly, but then they need to
- * call free_irq themselfves, too. Also, only IRQF_SHARED capable IRQ
+ * call free_irq() themselfves, too. Also, only %IRQF_SHARED capable IRQ
  * handlers are allowed.
  */
 int __must_check pcmcia_request_irq(struct pcmcia_device *p_dev,
@@ -649,12 +711,14 @@
 
 /**
  * pcmcia_request_exclusive_irq() - attempt to request an exclusive IRQ first
+ * @p_dev: the associated PCMCIA device
+ * @handler: IRQ handler to register
  *
- * pcmcia_request_exclusive_irq() is a wrapper around request_irq which
+ * pcmcia_request_exclusive_irq() is a wrapper around request_irq() which
  * attempts first to request an exclusive IRQ. If it fails, it also accepts
  * a shared IRQ, but prints out a warning. PCMCIA drivers should allow for
  * IRQ sharing and either use request_irq directly (then they need to call
- * free_irq themselves, too), or the pcmcia_request_irq() function.
+ * free_irq() themselves, too), or the pcmcia_request_irq() function.
  */
 int __must_check
 __pcmcia_request_exclusive_irq(struct pcmcia_device *p_dev,
@@ -795,38 +859,47 @@
 }
 
 
-/** pcmcia_request_window
+/**
+ * pcmcia_request_window() - attempt to reserve iomem for PCMCIA devices
+ * @p_dev: the associated PCMCIA device
+ * @res: &struct resource pointing to p_dev->resource[2..5]
+ * @speed: access speed
  *
- * Request_window() establishes a mapping between card memory space
- * and system memory space.
+ * pcmcia_request_window() attepts to reserve an iomem ranges specified in
+ * &struct resource @res pointing to one of the entries in
+ * &struct pcmcia_device @p_dev->resource[2..5]. The "start" value is the
+ * requested start of the IO mem resource; "end" reflects the size
+ * requested.
  */
-int pcmcia_request_window(struct pcmcia_device *p_dev, win_req_t *req, window_handle_t *wh)
+int pcmcia_request_window(struct pcmcia_device *p_dev, struct resource *res,
+			unsigned int speed)
 {
 	struct pcmcia_socket *s = p_dev->socket;
 	pccard_mem_map *win;
 	u_long align;
-	struct resource *res;
 	int w;
 
+	dev_dbg(&p_dev->dev, "request_window %pR %d\n", res, speed);
+
 	if (!(s->state & SOCKET_PRESENT)) {
 		dev_dbg(&p_dev->dev, "No card present\n");
 		return -ENODEV;
 	}
 
 	/* Window size defaults to smallest available */
-	if (req->Size == 0)
-		req->Size = s->map_size;
-	align = (s->features & SS_CAP_MEM_ALIGN) ? req->Size : s->map_size;
-	if (req->Size & (s->map_size-1)) {
+	if (res->end == 0)
+		res->end = s->map_size;
+	align = (s->features & SS_CAP_MEM_ALIGN) ? res->end : s->map_size;
+	if (res->end & (s->map_size-1)) {
 		dev_dbg(&p_dev->dev, "invalid map size\n");
 		return -EINVAL;
 	}
-	if ((req->Base && (s->features & SS_CAP_STATIC_MAP)) ||
-	    (req->Base & (align-1))) {
+	if ((res->start && (s->features & SS_CAP_STATIC_MAP)) ||
+	    (res->start & (align-1))) {
 		dev_dbg(&p_dev->dev, "invalid base address\n");
 		return -EINVAL;
 	}
-	if (req->Base)
+	if (res->start)
 		align = 0;
 
 	/* Allocate system memory window */
@@ -843,7 +916,7 @@
 	win = &s->win[w];
 
 	if (!(s->features & SS_CAP_STATIC_MAP)) {
-		win->res = pcmcia_find_mem_region(req->Base, req->Size, align,
+		win->res = pcmcia_find_mem_region(res->start, res->end, align,
 						0, s);
 		if (!win->res) {
 			dev_dbg(&p_dev->dev, "allocating mem region failed\n");
@@ -855,8 +928,8 @@
 
 	/* Configure the socket controller */
 	win->map = w+1;
-	win->flags = req->Attributes;
-	win->speed = req->AccessSpeed;
+	win->flags = res->flags & WIN_FLAGS_MAP;
+	win->speed = speed;
 	win->card_start = 0;
 
 	if (s->ops->set_mem_map(s, win) != 0) {
@@ -868,17 +941,14 @@
 
 	/* Return window handle */
 	if (s->features & SS_CAP_STATIC_MAP)
-		req->Base = win->static_start;
+		res->start = win->static_start;
 	else
-		req->Base = win->res->start;
+		res->start = win->res->start;
 
 	/* convert to new-style resources */
-	res = p_dev->resource[w + MAX_IO_WIN];
-	res->start = req->Base;
-	res->end = req->Base + req->Size - 1;
-	res->flags &= ~IORESOURCE_BITS;
-	res->flags |= (req->Attributes & WIN_FLAGS_MAP) | (win->map << 2);
-	res->flags |= IORESOURCE_MEM;
+	res->end += res->start - 1;
+	res->flags &= ~WIN_FLAGS_REQ;
+	res->flags |= (win->map << 2) | IORESOURCE_MEM;
 	res->parent = win->res;
 	if (win->res)
 		request_resource(&iomem_resource, res);
@@ -886,15 +956,30 @@
 	dev_dbg(&p_dev->dev, "request_window results in %pR\n", res);
 
 	mutex_unlock(&s->ops_mutex);
-	*wh = res;
 
 	return 0;
 } /* pcmcia_request_window */
 EXPORT_SYMBOL(pcmcia_request_window);
 
+
+/**
+ * pcmcia_disable_device() - disable and clean up a PCMCIA device
+ * @p_dev: the associated PCMCIA device
+ *
+ * pcmcia_disable_device() is the driver-callable counterpart to
+ * pcmcia_enable_device(): If a PCMCIA device is no longer used,
+ * drivers are expected to clean up and disable the device by calling
+ * this function. Any I/O ranges (iomem and ioports) will be released,
+ * the Vpp voltage will be set to 0, and IRQs will no longer be
+ * generated -- at least if there is no other card function (of
+ * multifunction devices) being used.
+ */
 void pcmcia_disable_device(struct pcmcia_device *p_dev)
 {
 	int i;
+
+	dev_dbg(&p_dev->dev, "disabling device\n");
+
 	for (i = 0; i < MAX_WIN; i++) {
 		struct resource *res = p_dev->resource[MAX_IO_WIN + i];
 		if (res->flags & WIN_FLAGS_REQ)
diff --git a/drivers/pcmcia/pd6729.c b/drivers/pcmcia/pd6729.c
index deef665..8cbfa06 100644
--- a/drivers/pcmcia/pd6729.c
+++ b/drivers/pcmcia/pd6729.c
@@ -18,7 +18,6 @@
 #include <linux/io.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 #include <asm/system.h>
 
diff --git a/drivers/pcmcia/rsrc_iodyn.c b/drivers/pcmcia/rsrc_iodyn.c
index 8510c35..523eb69 100644
--- a/drivers/pcmcia/rsrc_iodyn.c
+++ b/drivers/pcmcia/rsrc_iodyn.c
@@ -17,7 +17,6 @@
 #include <linux/kernel.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include "cs_internal.h"
 
diff --git a/drivers/pcmcia/rsrc_mgr.c b/drivers/pcmcia/rsrc_mgr.c
index 4e80421..aa628ed 100644
--- a/drivers/pcmcia/rsrc_mgr.c
+++ b/drivers/pcmcia/rsrc_mgr.c
@@ -17,7 +17,6 @@
 #include <linux/kernel.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include "cs_internal.h"
 
diff --git a/drivers/pcmcia/rsrc_nonstatic.c b/drivers/pcmcia/rsrc_nonstatic.c
index 96f348b..b187555 100644
--- a/drivers/pcmcia/rsrc_nonstatic.c
+++ b/drivers/pcmcia/rsrc_nonstatic.c
@@ -29,7 +29,6 @@
 #include <asm/irq.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include "cs_internal.h"
 
diff --git a/drivers/pcmcia/sa1100_generic.c b/drivers/pcmcia/sa1100_generic.c
index e098514..945857f 100644
--- a/drivers/pcmcia/sa1100_generic.c
+++ b/drivers/pcmcia/sa1100_generic.c
@@ -35,7 +35,6 @@
 #include <linux/slab.h>
 #include <linux/platform_device.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 
 #include <asm/hardware/scoop.h>
diff --git a/drivers/pcmcia/soc_common.c b/drivers/pcmcia/soc_common.c
index 6f1a86b..689e3c0 100644
--- a/drivers/pcmcia/soc_common.c
+++ b/drivers/pcmcia/soc_common.c
@@ -627,8 +627,6 @@
 
 	pcmcia_unregister_socket(&skt->socket);
 
-	flush_scheduled_work();
-
 	skt->ops->hw_shutdown(skt);
 
 	soc_common_pcmcia_config_skt(skt, &dead_socket);
@@ -720,8 +718,6 @@
 	pcmcia_unregister_socket(&skt->socket);
 
  out_err_7:
-	flush_scheduled_work();
-
 	skt->ops->hw_shutdown(skt);
  out_err_6:
 	list_del(&skt->node);
diff --git a/drivers/pcmcia/soc_common.h b/drivers/pcmcia/soc_common.h
index 3fba3a6..bbcd538 100644
--- a/drivers/pcmcia/soc_common.h
+++ b/drivers/pcmcia/soc_common.h
@@ -11,7 +11,6 @@
 
 /* include the world */
 #include <linux/cpufreq.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 #include <pcmcia/cistpl.h>
 
diff --git a/drivers/pcmcia/socket_sysfs.c b/drivers/pcmcia/socket_sysfs.c
index cb0d3ac..71aeed9 100644
--- a/drivers/pcmcia/socket_sysfs.c
+++ b/drivers/pcmcia/socket_sysfs.c
@@ -27,7 +27,6 @@
 #include <asm/irq.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
diff --git a/drivers/pcmcia/tcic.c b/drivers/pcmcia/tcic.c
index be0d841..310160b 100644
--- a/drivers/pcmcia/tcic.c
+++ b/drivers/pcmcia/tcic.c
@@ -49,7 +49,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 #include "tcic.h"
 
diff --git a/drivers/pcmcia/vrc4173_cardu.c b/drivers/pcmcia/vrc4173_cardu.c
index 9b3c158..c6d36b3 100644
--- a/drivers/pcmcia/vrc4173_cardu.c
+++ b/drivers/pcmcia/vrc4173_cardu.c
@@ -461,7 +461,7 @@
 {
 	vrc4173_socket_t *socket;
 	unsigned long start, len, flags;
-	int slot, err;
+	int slot, err, ret;
 
 	slot = vrc4173_cardu_slots++;
 	socket = &cardu_sockets[slot];
@@ -474,43 +474,63 @@
 		return err;
 
 	start = pci_resource_start(dev, 0);
-	if (start == 0)
-		return -ENODEV;
+	if (start == 0) {
+		ret = -ENODEV;
+		goto disable;
+	}
 
 	len = pci_resource_len(dev, 0);
-	if (len == 0)
-		return -ENODEV;
+	if (len == 0) {
+		ret = -ENODEV;
+		goto disable;
+	}
 
-	if (((flags = pci_resource_flags(dev, 0)) & IORESOURCE_MEM) == 0)
-		return -EBUSY;
+	flags = pci_resource_flags(dev, 0);
+	if ((flags & IORESOURCE_MEM) == 0) {
+		ret = -EBUSY;
+		goto disable;
+	}
 
-	if ((err = pci_request_regions(dev, socket->name)) < 0)
-		return err;
+	err = pci_request_regions(dev, socket->name);
+	if (err < 0) {
+		ret = err;
+		goto disable;
+	}
 
 	socket->base = ioremap(start, len);
-	if (socket->base == NULL)
-		return -ENODEV;
+	if (socket->base == NULL) {
+		ret = -ENODEV;
+		goto release;
+	}
 
 	socket->dev = dev;
 
 	socket->pcmcia_socket = pcmcia_register_socket(slot, &cardu_operations, 1);
 	if (socket->pcmcia_socket == NULL) {
-		iounmap(socket->base);
-		socket->base = NULL;
-		return -ENOMEM;
+		ret =  -ENOMEM;
+		goto unmap;
 	}
 
 	if (request_irq(dev->irq, cardu_interrupt, IRQF_SHARED, socket->name, socket) < 0) {
-		pcmcia_unregister_socket(socket->pcmcia_socket);
-		socket->pcmcia_socket = NULL;
-		iounmap(socket->base);
-		socket->base = NULL;
-		return -EBUSY;
+		ret = -EBUSY;
+		goto unregister;
 	}
 
 	printk(KERN_INFO "%s at %#08lx, IRQ %d\n", socket->name, start, dev->irq);
 
 	return 0;
+
+unregister:
+	pcmcia_unregister_socket(socket->pcmcia_socket);
+	socket->pcmcia_socket = NULL;
+unmap:
+	iounmap(socket->base);
+	socket->base = NULL;
+release:
+	pci_release_regions(dev);
+disable:
+	pci_disable_device(dev);
+	return ret;
 }
 
 static int __devinit vrc4173_cardu_setup(char *options)
diff --git a/drivers/pcmcia/xxs1500_ss.c b/drivers/pcmcia/xxs1500_ss.c
index fa88c36..3b67a1b 100644
--- a/drivers/pcmcia/xxs1500_ss.c
+++ b/drivers/pcmcia/xxs1500_ss.c
@@ -17,7 +17,6 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/ss.h>
 #include <pcmcia/cistpl.h>
 
diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c
index 414d9a6..408dbaa 100644
--- a/drivers/pcmcia/yenta_socket.c
+++ b/drivers/pcmcia/yenta_socket.c
@@ -20,7 +20,6 @@
 #include <linux/slab.h>
 
 #include <pcmcia/ss.h>
-#include <pcmcia/cs.h>
 
 #include "yenta_socket.h"
 #include "i82365.h"
diff --git a/drivers/platform/x86/intel_ips.c b/drivers/platform/x86/intel_ips.c
index 9024480..c44a5e8 100644
--- a/drivers/platform/x86/intel_ips.c
+++ b/drivers/platform/x86/intel_ips.c
@@ -51,7 +51,6 @@
  * TODO:
  *   - handle CPU hotplug
  *   - provide turbo enable/disable api
- *   - make sure we can write turbo enable/disable reg based on MISC_EN
  *
  * Related documents:
  *   - CDI 403777, 403778 - Auburndale EDS vol 1 & 2
@@ -230,7 +229,7 @@
 #define THM_TC2		0xac
 #define THM_DTV		0xb0
 #define THM_ITV		0xd8
-#define   ITV_ME_SEQNO_MASK 0x000f0000 /* ME should update every ~200ms */
+#define   ITV_ME_SEQNO_MASK 0x00ff0000 /* ME should update every ~200ms */
 #define   ITV_ME_SEQNO_SHIFT (16)
 #define   ITV_MCH_TEMP_MASK 0x0000ff00
 #define   ITV_MCH_TEMP_SHIFT (8)
@@ -325,6 +324,7 @@
 	bool gpu_preferred;
 	bool poll_turbo_status;
 	bool second_cpu;
+	bool turbo_toggle_allowed;
 	struct ips_mcp_limits *limits;
 
 	/* Optional MCH interfaces for if i915 is in use */
@@ -415,7 +415,7 @@
 	new_limit = cur_limit - 8; /* 1W decrease */
 
 	/* Clamp to SKU TDP limit */
-	if (((new_limit * 10) / 8) < (ips->orig_turbo_limit & TURBO_TDP_MASK))
+	if (new_limit  < (ips->orig_turbo_limit & TURBO_TDP_MASK))
 		new_limit = ips->orig_turbo_limit & TURBO_TDP_MASK;
 
 	thm_writew(THM_MPCPC, (new_limit * 10) / 8);
@@ -461,7 +461,8 @@
 	if (ips->__cpu_turbo_on)
 		return;
 
-	on_each_cpu(do_enable_cpu_turbo, ips, 1);
+	if (ips->turbo_toggle_allowed)
+		on_each_cpu(do_enable_cpu_turbo, ips, 1);
 
 	ips->__cpu_turbo_on = true;
 }
@@ -498,7 +499,8 @@
 	if (!ips->__cpu_turbo_on)
 		return;
 
-	on_each_cpu(do_disable_cpu_turbo, ips, 1);
+	if (ips->turbo_toggle_allowed)
+		on_each_cpu(do_disable_cpu_turbo, ips, 1);
 
 	ips->__cpu_turbo_on = false;
 }
@@ -598,17 +600,29 @@
 {
 	unsigned long flags;
 	bool ret = false;
+	u32 temp_limit;
+	u32 avg_power;
+	const char *msg = "MCP limit exceeded: ";
 
 	spin_lock_irqsave(&ips->turbo_status_lock, flags);
-	if (ips->mcp_avg_temp > (ips->mcp_temp_limit * 100))
-		ret = true;
-	if (ips->cpu_avg_power + ips->mch_avg_power > ips->mcp_power_limit)
-		ret = true;
-	spin_unlock_irqrestore(&ips->turbo_status_lock, flags);
 
-	if (ret)
+	temp_limit = ips->mcp_temp_limit * 100;
+	if (ips->mcp_avg_temp > temp_limit) {
 		dev_info(&ips->dev->dev,
-			 "MCP power or thermal limit exceeded\n");
+			"%sAvg temp %u, limit %u\n", msg, ips->mcp_avg_temp,
+			temp_limit);
+		ret = true;
+	}
+
+	avg_power = ips->cpu_avg_power + ips->mch_avg_power;
+	if (avg_power > ips->mcp_power_limit) {
+		dev_info(&ips->dev->dev,
+			"%sAvg power %u, limit %u\n", msg, avg_power,
+			ips->mcp_power_limit);
+		ret = true;
+	}
+
+	spin_unlock_irqrestore(&ips->turbo_status_lock, flags);
 
 	return ret;
 }
@@ -663,6 +677,27 @@
 }
 
 /**
+ * verify_limits - verify BIOS provided limits
+ * @ips: IPS structure
+ *
+ * BIOS can optionally provide non-default limits for power and temp.  Check
+ * them here and use the defaults if the BIOS values are not provided or
+ * are otherwise unusable.
+ */
+static void verify_limits(struct ips_driver *ips)
+{
+	if (ips->mcp_power_limit < ips->limits->mcp_power_limit ||
+	    ips->mcp_power_limit > 35000)
+		ips->mcp_power_limit = ips->limits->mcp_power_limit;
+
+	if (ips->mcp_temp_limit < ips->limits->core_temp_limit ||
+	    ips->mcp_temp_limit < ips->limits->mch_temp_limit ||
+	    ips->mcp_temp_limit > 150)
+		ips->mcp_temp_limit = min(ips->limits->core_temp_limit,
+					  ips->limits->mch_temp_limit);
+}
+
+/**
  * update_turbo_limits - get various limits & settings from regs
  * @ips: IPS driver struct
  *
@@ -680,12 +715,21 @@
 	u32 hts = thm_readl(THM_HTS);
 
 	ips->cpu_turbo_enabled = !(hts & HTS_PCTD_DIS);
-	ips->gpu_turbo_enabled = !(hts & HTS_GTD_DIS);
+	/* 
+	 * Disable turbo for now, until we can figure out why the power figures
+	 * are wrong
+	 */
+	ips->cpu_turbo_enabled = false;
+
+	if (ips->gpu_busy)
+		ips->gpu_turbo_enabled = !(hts & HTS_GTD_DIS);
+
 	ips->core_power_limit = thm_readw(THM_MPCPC);
 	ips->mch_power_limit = thm_readw(THM_MMGPC);
 	ips->mcp_temp_limit = thm_readw(THM_PTL);
 	ips->mcp_power_limit = thm_readw(THM_MPPC);
 
+	verify_limits(ips);
 	/* Ignore BIOS CPU vs GPU pref */
 }
 
@@ -858,7 +902,7 @@
 	ret = (ret * 1000) / 65535;
 	*last = val;
 
-	return ret;
+	return 0;
 }
 
 static const u16 temp_decay_factor = 2;
@@ -940,7 +984,6 @@
 		kfree(mch_samples);
 		kfree(cpu_samples);
 		kfree(mchp_samples);
-		kthread_stop(ips->adjust);
 		return -ENOMEM;
 	}
 
@@ -948,7 +991,7 @@
 		ITV_ME_SEQNO_SHIFT;
 	seqno_timestamp = get_jiffies_64();
 
-	old_cpu_power = thm_readl(THM_CEC) / 65535;
+	old_cpu_power = thm_readl(THM_CEC);
 	schedule_timeout_interruptible(msecs_to_jiffies(IPS_SAMPLE_PERIOD));
 
 	/* Collect an initial average */
@@ -1150,11 +1193,18 @@
 				STS_GPL_SHIFT;
 			/* ignore EC CPU vs GPU pref */
 			ips->cpu_turbo_enabled = !(sts & STS_PCTD_DIS);
-			ips->gpu_turbo_enabled = !(sts & STS_GTD_DIS);
+			/* 
+			 * Disable turbo for now, until we can figure
+			 * out why the power figures are wrong
+			 */
+			ips->cpu_turbo_enabled = false;
+			if (ips->gpu_busy)
+				ips->gpu_turbo_enabled = !(sts & STS_GTD_DIS);
 			ips->mcp_temp_limit = (sts & STS_PTL_MASK) >>
 				STS_PTL_SHIFT;
 			ips->mcp_power_limit = (tc1 & STS_PPL_MASK) >>
 				STS_PPL_SHIFT;
+			verify_limits(ips);
 			spin_unlock(&ips->turbo_status_lock);
 
 			thm_writeb(THM_SEC, SEC_ACK);
@@ -1333,8 +1383,10 @@
 	 * turbo manually or we'll get an illegal MSR access, even though
 	 * turbo will still be available.
 	 */
-	if (!(misc_en & IA32_MISC_TURBO_EN))
-		; /* add turbo MSR write allowed flag if necessary */
+	if (misc_en & IA32_MISC_TURBO_EN)
+		ips->turbo_toggle_allowed = true;
+	else
+		ips->turbo_toggle_allowed = false;
 
 	if (strstr(boot_cpu_data.x86_model_id, "CPU       M"))
 		limits = &ips_sv_limits;
@@ -1351,9 +1403,10 @@
 	tdp = turbo_power & TURBO_TDP_MASK;
 
 	/* Sanity check TDP against CPU */
-	if (limits->mcp_power_limit != (tdp / 8) * 1000) {
-		dev_warn(&ips->dev->dev, "Warning: CPU TDP doesn't match expected value (found %d, expected %d)\n",
-			 tdp / 8, limits->mcp_power_limit / 1000);
+	if (limits->core_power_limit != (tdp / 8) * 1000) {
+		dev_info(&ips->dev->dev, "CPU TDP doesn't match expected value (found %d, expected %d)\n",
+			 tdp / 8, limits->core_power_limit / 1000);
+		limits->core_power_limit = (tdp / 8) * 1000;
 	}
 
 out:
@@ -1390,7 +1443,7 @@
 	return true;
 
 out_put_busy:
-	symbol_put(i915_gpu_turbo_disable);
+	symbol_put(i915_gpu_busy);
 out_put_lower:
 	symbol_put(i915_gpu_lower);
 out_put_raise:
@@ -1532,23 +1585,28 @@
 	/* Save turbo limits & ratios */
 	rdmsrl(TURBO_POWER_CURRENT_LIMIT, ips->orig_turbo_limit);
 
-	ips_enable_cpu_turbo(ips);
-	ips->cpu_turbo_enabled = true;
+	ips_disable_cpu_turbo(ips);
+	ips->cpu_turbo_enabled = false;
 
-	/* Set up the work queue and monitor/adjust threads */
-	ips->monitor = kthread_run(ips_monitor, ips, "ips-monitor");
-	if (IS_ERR(ips->monitor)) {
-		dev_err(&dev->dev,
-			"failed to create thermal monitor thread, aborting\n");
-		ret = -ENOMEM;
-		goto error_free_irq;
-	}
-
+	/* Create thermal adjust thread */
 	ips->adjust = kthread_create(ips_adjust, ips, "ips-adjust");
 	if (IS_ERR(ips->adjust)) {
 		dev_err(&dev->dev,
 			"failed to create thermal adjust thread, aborting\n");
 		ret = -ENOMEM;
+		goto error_free_irq;
+
+	}
+
+	/*
+	 * Set up the work queue and monitor thread. The monitor thread
+	 * will wake up ips_adjust thread.
+	 */
+	ips->monitor = kthread_run(ips_monitor, ips, "ips-monitor");
+	if (IS_ERR(ips->monitor)) {
+		dev_err(&dev->dev,
+			"failed to create thermal monitor thread, aborting\n");
+		ret = -ENOMEM;
 		goto error_thread_cleanup;
 	}
 
@@ -1566,7 +1624,7 @@
 	return ret;
 
 error_thread_cleanup:
-	kthread_stop(ips->monitor);
+	kthread_stop(ips->adjust);
 error_free_irq:
 	free_irq(ips->dev->irq, ips);
 error_unmap:
diff --git a/drivers/regulator/ad5398.c b/drivers/regulator/ad5398.c
index df1fb53..a4be416 100644
--- a/drivers/regulator/ad5398.c
+++ b/drivers/regulator/ad5398.c
@@ -256,7 +256,6 @@
 
 	regulator_unregister(chip->rdev);
 	kfree(chip);
-	i2c_set_clientdata(client, NULL);
 
 	return 0;
 }
diff --git a/drivers/regulator/isl6271a-regulator.c b/drivers/regulator/isl6271a-regulator.c
index d61ecb8..b8cc638 100644
--- a/drivers/regulator/isl6271a-regulator.c
+++ b/drivers/regulator/isl6271a-regulator.c
@@ -191,8 +191,6 @@
 	struct isl_pmic *pmic = i2c_get_clientdata(i2c);
 	int i;
 
-	i2c_set_clientdata(i2c, NULL);
-
 	for (i = 0; i < 3; i++)
 		regulator_unregister(pmic->rdev[i]);
 
diff --git a/drivers/rtc/rtc-ds3232.c b/drivers/rtc/rtc-ds3232.c
index 9daed8d..9de8516 100644
--- a/drivers/rtc/rtc-ds3232.c
+++ b/drivers/rtc/rtc-ds3232.c
@@ -268,7 +268,6 @@
 		free_irq(client->irq, client);
 
 out_free:
-	i2c_set_clientdata(client, NULL);
 	kfree(ds3232);
 	return ret;
 }
@@ -287,7 +286,6 @@
 	}
 
 	rtc_device_unregister(ds3232->rtc);
-	i2c_set_clientdata(client, NULL);
 	kfree(ds3232);
 	return 0;
 }
diff --git a/drivers/scsi/pcmcia/aha152x_stub.c b/drivers/scsi/pcmcia/aha152x_stub.c
index 61f49bd..e77dd02 100644
--- a/drivers/scsi/pcmcia/aha152x_stub.c
+++ b/drivers/scsi/pcmcia/aha152x_stub.c
@@ -49,7 +49,6 @@
 #include <scsi/scsi_host.h>
 #include "aha152x.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -86,8 +85,6 @@
 static void aha152x_detach(struct pcmcia_device *p_dev);
 static int aha152x_config_cs(struct pcmcia_device *link);
 
-static struct pcmcia_device *dev_list;
-
 static int aha152x_probe(struct pcmcia_device *link)
 {
     scsi_info_t *info;
@@ -100,11 +97,8 @@
     info->p_dev = link;
     link->priv = info;
 
-    link->resource[0]->end = 0x20;
-    link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-    link->conf.Attributes = CONF_ENABLE_IRQ;
-    link->conf.IntType = INT_MEMORY_AND_IO;
-    link->conf.Present = PRESENT_OPTION;
+    link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+    link->config_regs = PRESENT_OPTION;
 
     return aha152x_config_cs(link);
 } /* aha152x_attach */
@@ -123,25 +117,24 @@
 
 /*====================================================================*/
 
-static int aha152x_config_check(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int aha152x_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	p_dev->io_lines = 10;
+
 	/* For New Media T&J, look for a SCSI window */
-	if (cfg->io.win[0].len >= 0x20)
-		p_dev->resource[0]->start = cfg->io.win[0].base;
-	else if ((cfg->io.nwin > 1) &&
-		 (cfg->io.win[1].len >= 0x20))
-		p_dev->resource[0]->start = cfg->io.win[1].base;
-	if ((cfg->io.nwin > 0) &&
-	    (p_dev->resource[0]->start < 0xffff)) {
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -EINVAL;
+	if ((p_dev->resource[0]->end < 0x20) &&
+		(p_dev->resource[1]->end >= 0x20))
+		p_dev->resource[0]->start = p_dev->resource[1]->start;
+
+	if (p_dev->resource[0]->start >= 0xffff)
+		return -EINVAL;
+
+	p_dev->resource[1]->start = p_dev->resource[1]->end = 0;
+	p_dev->resource[0]->end = 0x20;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int aha152x_config_cs(struct pcmcia_device *link)
@@ -160,7 +153,7 @@
     if (!link->irq)
 	    goto failed;
 
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
     
@@ -221,9 +214,7 @@
 
 static struct pcmcia_driver aha152x_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "aha152x_cs",
-	},
+	.name		= "aha152x_cs",
 	.probe		= aha152x_probe,
 	.remove		= aha152x_detach,
 	.id_table       = aha152x_ids,
@@ -238,7 +229,6 @@
 static void __exit exit_aha152x_cs(void)
 {
 	pcmcia_unregister_driver(&aha152x_cs_driver);
-	BUG_ON(dev_list != NULL);
 }
 
 module_init(init_aha152x_cs);
diff --git a/drivers/scsi/pcmcia/fdomain_stub.c b/drivers/scsi/pcmcia/fdomain_stub.c
index 13dbe5c..cd69c26 100644
--- a/drivers/scsi/pcmcia/fdomain_stub.c
+++ b/drivers/scsi/pcmcia/fdomain_stub.c
@@ -46,7 +46,6 @@
 #include <scsi/scsi_host.h>
 #include "fdomain.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -83,11 +82,8 @@
 
 	info->p_dev = link;
 	link->priv = info;
-	link->resource[0]->end = 0x10;
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-	link->conf.Present = PRESENT_OPTION;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+	link->config_regs = PRESENT_OPTION;
 
 	return fdomain_config(link);
 } /* fdomain_attach */
@@ -105,14 +101,12 @@
 
 /*====================================================================*/
 
-static int fdomain_config_check(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int fdomain_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	p_dev->io_lines = 10;
-	p_dev->resource[0]->start = cfg->io.win[0].base;
+	p_dev->resource[0]->end = 0x10;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	return pcmcia_request_io(p_dev);
 }
 
@@ -132,7 +126,7 @@
 
     if (!link->irq)
 	    goto failed;
-    ret = pcmcia_request_configuration(link, &link->conf);
+    ret = pcmcia_enable_device(link);
     if (ret)
 	    goto failed;
 
@@ -194,9 +188,7 @@
 
 static struct pcmcia_driver fdomain_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "fdomain_cs",
-	},
+	.name		= "fdomain_cs",
 	.probe		= fdomain_probe,
 	.remove		= fdomain_detach,
 	.id_table       = fdomain_ids,
diff --git a/drivers/scsi/pcmcia/nsp_cs.c b/drivers/scsi/pcmcia/nsp_cs.c
index dd9b403..9326c2c 100644
--- a/drivers/scsi/pcmcia/nsp_cs.c
+++ b/drivers/scsi/pcmcia/nsp_cs.c
@@ -47,7 +47,6 @@
 #include <scsi/scsi.h>
 #include <scsi/scsi_ioctl.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -1531,15 +1530,6 @@
   PCMCIA functions
 **********************************************************************/
 
-/*======================================================================
-    nsp_cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-======================================================================*/
 static int nsp_cs_probe(struct pcmcia_device *link)
 {
 	scsi_info_t  *info;
@@ -1557,14 +1547,6 @@
 
 	nsp_dbg(NSP_DEBUG_INIT, "info=0x%p", info);
 
-	/* The io structure describes IO port mapping */
-	link->resource[0]->end	 = 0x10;
-	link->resource[0]->flags = IO_DATA_PATH_WIDTH_AUTO;
-
-	/* General socket configuration */
-	link->conf.Attributes	 = CONF_ENABLE_IRQ;
-	link->conf.IntType	 = INT_MEMORY_AND_IO;
-
 	ret = nsp_cs_config(link);
 
 	nsp_dbg(NSP_DEBUG_INIT, "link=0x%p", link);
@@ -1572,12 +1554,6 @@
 } /* nsp_cs_attach */
 
 
-/*======================================================================
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.	 If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-======================================================================*/
 static void nsp_cs_detach(struct pcmcia_device *link)
 {
 	nsp_dbg(NSP_DEBUG_INIT, "in, link=0x%p", link);
@@ -1590,98 +1566,36 @@
 } /* nsp_cs_detach */
 
 
-/*======================================================================
-    nsp_cs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    ethernet device available to the system.
-======================================================================*/
-
-struct nsp_cs_configdata {
-	nsp_hw_data		*data;
-	win_req_t		req;
-};
-
-static int nsp_cs_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cfg,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int nsp_cs_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	struct nsp_cs_configdata *cfg_mem = priv_data;
+	nsp_hw_data		*data = priv_data;
 
-	if (cfg->index == 0)
+	if (p_dev->config_index == 0)
 		return -ENODEV;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
+	/* This reserves IO space but doesn't actually enable it */
+	if (pcmcia_request_io(p_dev) != 0)
+		goto next_entry;
+
+	if (resource_size(p_dev->resource[2])) {
+		p_dev->resource[2]->flags |= (WIN_DATA_WIDTH_16 |
+					WIN_MEMORY_TYPE_CM |
+					WIN_ENABLE);
+		if (p_dev->resource[2]->end < 0x1000)
+			p_dev->resource[2]->end = 0x1000;
+		if (pcmcia_request_window(p_dev, p_dev->resource[2], 0) != 0)
+			goto next_entry;
+		if (pcmcia_map_mem_page(p_dev, p_dev->resource[2],
+						p_dev->card_addr) != 0)
+			goto next_entry;
+
+		data->MmioAddress = (unsigned long)
+			ioremap_nocache(p_dev->resource[2]->start,
+					resource_size(p_dev->resource[2]));
+		data->MmioLength  = resource_size(p_dev->resource[2]);
 	}
-
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-		if (vcc != cfg->vcc.param[CISTPL_POWER_VNOM]/10000)
-			return -ENODEV;
-		else if (dflt->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-			if (vcc != dflt->vcc.param[CISTPL_POWER_VNOM]/10000)
-				return -ENODEV;
-		}
-
-		if (cfg->vpp1.present & (1 << CISTPL_POWER_VNOM)) {
-			p_dev->conf.Vpp =
-				cfg->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-		} else if (dflt->vpp1.present & (1 << CISTPL_POWER_VNOM)) {
-			p_dev->conf.Vpp =
-				dflt->vpp1.param[CISTPL_POWER_VNOM] / 10000;
-		}
-
-		/* Do we need to allocate an interrupt? */
-		p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-		/* IO window settings */
-		p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-		if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-			cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-			p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-			p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-			p_dev->resource[0]->flags |=
-				pcmcia_io_cfg_data_width(io->flags);
-			p_dev->resource[0]->start = io->win[0].base;
-			p_dev->resource[0]->end = io->win[0].len;
-			if (io->nwin > 1) {
-				p_dev->resource[1]->flags =
-					p_dev->resource[0]->flags;
-				p_dev->resource[1]->start = io->win[1].base;
-				p_dev->resource[1]->end = io->win[1].len;
-			}
-			/* This reserves IO space but doesn't actually enable it */
-			if (pcmcia_request_io(p_dev) != 0)
-				goto next_entry;
-		}
-
-		if ((cfg->mem.nwin > 0) || (dflt->mem.nwin > 0)) {
-			cistpl_mem_t	*mem =
-				(cfg->mem.nwin) ? &cfg->mem : &dflt->mem;
-			cfg_mem->req.Attributes = WIN_DATA_WIDTH_16|WIN_MEMORY_TYPE_CM;
-			cfg_mem->req.Attributes |= WIN_ENABLE;
-			cfg_mem->req.Base = mem->win[0].host_addr;
-			cfg_mem->req.Size = mem->win[0].len;
-			if (cfg_mem->req.Size < 0x1000)
-				cfg_mem->req.Size = 0x1000;
-			cfg_mem->req.AccessSpeed = 0;
-			if (pcmcia_request_window(p_dev, &cfg_mem->req, &p_dev->win) != 0)
-				goto next_entry;
-			if (pcmcia_map_mem_page(p_dev, p_dev->win,
-					mem->win[0].card_addr) != 0)
-				goto next_entry;
-
-			cfg_mem->data->MmioAddress = (unsigned long) ioremap_nocache(cfg_mem->req.Base, cfg_mem->req.Size);
-			cfg_mem->data->MmioLength  = cfg_mem->req.Size;
-		}
-		/* If we got this far, we're cool! */
-		return 0;
-	}
+	/* If we got this far, we're cool! */
+	return 0;
 
 next_entry:
 	nsp_dbg(NSP_DEBUG_INIT, "next");
@@ -1693,25 +1607,23 @@
 {
 	int		  ret;
 	scsi_info_t	 *info	 = link->priv;
-	struct nsp_cs_configdata *cfg_mem;
 	struct Scsi_Host *host;
 	nsp_hw_data      *data = &nsp_data_base;
 
 	nsp_dbg(NSP_DEBUG_INIT, "in");
 
-	cfg_mem = kzalloc(sizeof(*cfg_mem), GFP_KERNEL);
-	if (!cfg_mem)
-		return -ENOMEM;
-	cfg_mem->data = data;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_CHECK_VCC |
+		CONF_AUTO_SET_VPP | CONF_AUTO_AUDIO | CONF_AUTO_SET_IOMEM |
+		CONF_AUTO_SET_IO;
 
-	ret = pcmcia_loop_config(link, nsp_cs_config_check, cfg_mem);
+	ret = pcmcia_loop_config(link, nsp_cs_config_check, data);
 	if (ret)
 		goto cs_failed;
 
 	if (pcmcia_request_irq(link, nspintr))
 		goto cs_failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto cs_failed;
 
@@ -1754,41 +1666,16 @@
 
 	info->host = host;
 
-	/* Finally, report what we've done */
-	printk(KERN_INFO "nsp_cs: index 0x%02x: ",
-	       link->conf.ConfigIndex);
-	if (link->conf.Vpp) {
-		printk(", Vpp %d.%d", link->conf.Vpp/10, link->conf.Vpp%10);
-	}
-	if (link->conf.Attributes & CONF_ENABLE_IRQ) {
-		printk(", irq %d", link->irq);
-	}
-	if (link->resource[0])
-		printk(", io %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	if (link->win)
-		printk(", mem 0x%06lx-0x%06lx", cfg_mem->req.Base,
-		       cfg_mem->req.Base+cfg_mem->req.Size-1);
-	printk("\n");
-
-	kfree(cfg_mem);
 	return 0;
 
  cs_failed:
 	nsp_dbg(NSP_DEBUG_INIT, "config fail");
 	nsp_cs_release(link);
-	kfree(cfg_mem);
 
 	return -ENODEV;
 } /* nsp_cs_config */
 
 
-/*======================================================================
-    After a card is removed, nsp_cs_release() will unregister the net
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-======================================================================*/
 static void nsp_cs_release(struct pcmcia_device *link)
 {
 	scsi_info_t *info = link->priv;
@@ -1807,7 +1694,7 @@
 		scsi_remove_host(info->host);
 	}
 
-	if (link->win) {
+	if (resource_size(link->resource[2])) {
 		if (data != NULL) {
 			iounmap((void *)(data->MmioAddress));
 		}
@@ -1877,9 +1764,7 @@
 
 static struct pcmcia_driver nsp_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "nsp_cs",
-	},
+	.name		= "nsp_cs",
 	.probe		= nsp_cs_probe,
 	.remove		= nsp_cs_detach,
 	.id_table	= nsp_cs_ids,
@@ -1889,14 +1774,11 @@
 
 static int __init nsp_cs_init(void)
 {
-	nsp_msg(KERN_INFO, "loading...");
-
 	return pcmcia_register_driver(&nsp_driver);
 }
 
 static void __exit nsp_cs_exit(void)
 {
-	nsp_msg(KERN_INFO, "unloading...");
 	pcmcia_unregister_driver(&nsp_driver);
 }
 
diff --git a/drivers/scsi/pcmcia/qlogic_stub.c b/drivers/scsi/pcmcia/qlogic_stub.c
index eb775f1..9c96ca8 100644
--- a/drivers/scsi/pcmcia/qlogic_stub.c
+++ b/drivers/scsi/pcmcia/qlogic_stub.c
@@ -48,7 +48,6 @@
 #include <scsi/scsi_host.h>
 #include "../qlogicfas408.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/ciscode.h>
@@ -156,11 +155,8 @@
 		return -ENOMEM;
 	info->p_dev = link;
 	link->priv = info;
-	link->resource[0]->end = 16;
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-	link->conf.Present = PRESENT_OPTION;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+	link->config_regs = PRESENT_OPTION;
 
 	return qlogic_config(link);
 }				/* qlogic_attach */
@@ -178,15 +174,11 @@
 
 /*====================================================================*/
 
-static int qlogic_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cfg,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int qlogic_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	p_dev->io_lines = 10;
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 
 	if (p_dev->resource[0]->start == 0)
 		return -ENODEV;
@@ -209,7 +201,7 @@
 	if (!link->irq)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -264,7 +256,7 @@
 {
 	scsi_info_t *info = link->priv;
 
-	pcmcia_request_configuration(link, &link->conf);
+	pcmcia_enable_device(link);
 	if ((info->manf_id == MANFID_MACNICA) ||
 	    (info->manf_id == MANFID_PIONEER) ||
 	    (info->manf_id == 0x0098)) {
@@ -302,9 +294,7 @@
 
 static struct pcmcia_driver qlogic_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
 	.name		= "qlogic_cs",
-	},
 	.probe		= qlogic_probe,
 	.remove		= qlogic_detach,
 	.id_table       = qlogic_ids,
diff --git a/drivers/scsi/pcmcia/sym53c500_cs.c b/drivers/scsi/pcmcia/sym53c500_cs.c
index 321e390..0ae27cb 100644
--- a/drivers/scsi/pcmcia/sym53c500_cs.c
+++ b/drivers/scsi/pcmcia/sym53c500_cs.c
@@ -71,7 +71,6 @@
 #include <scsi/scsi.h>
 #include <scsi/scsi_host.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 #include <pcmcia/ciscode.h>
@@ -684,15 +683,11 @@
      .shost_attrs		= SYM53C500_shost_attrs
 };
 
-static int SYM53C500_config_check(struct pcmcia_device *p_dev,
-				  cistpl_cftable_entry_t *cfg,
-				  cistpl_cftable_entry_t *dflt,
-				  unsigned int vcc,
-				  void *priv_data)
+static int SYM53C500_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	p_dev->io_lines = 10;
-	p_dev->resource[0]->start = cfg->io.win[0].base;
-	p_dev->resource[0]->end = cfg->io.win[0].len;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 
 	if (p_dev->resource[0]->start == 0)
 		return -ENODEV;
@@ -721,7 +716,7 @@
 	if (!link->irq)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -859,10 +854,7 @@
 		return -ENOMEM;
 	info->p_dev = link;
 	link->priv = info;
-	link->resource[0]->end = 16;
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
 	return SYM53C500_config(link);
 } /* SYM53C500_attach */
@@ -881,9 +873,7 @@
 
 static struct pcmcia_driver sym53c500_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "sym53c500_cs",
-	},
+	.name		= "sym53c500_cs",
 	.probe		= SYM53C500_probe,
 	.remove		= SYM53C500_detach,
 	.id_table       = sym53c500_ids,
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index ad0ed21..348fba0 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -1046,13 +1046,13 @@
 
 	/* If the user actually wanted this page, we can skip the rest */
 	if (page == 0)
-		return -EINVAL;
+		return 0;
 
 	for (i = 0; i < min((int)buf[3], buf_len - 4); i++)
 		if (buf[i + 4] == page)
 			goto found;
 
-	if (i < buf[3] && i > buf_len)
+	if (i < buf[3] && i >= buf_len - 4)
 		/* ran off the end of the buffer, give us benefit of doubt */
 		goto found;
 	/* The device claims it doesn't support the requested page */
diff --git a/drivers/serial/ioc3_serial.c b/drivers/serial/ioc3_serial.c
index 93de907..ee43efc 100644
--- a/drivers/serial/ioc3_serial.c
+++ b/drivers/serial/ioc3_serial.c
@@ -2017,6 +2017,7 @@
 	struct ioc3_port *port;
 	struct ioc3_port *ports[PORTS_PER_CARD];
 	int phys_port;
+	int cnt;
 
 	DPRINT_CONFIG(("%s (0x%p, 0x%p)\n", __func__, is, idd));
 
@@ -2044,6 +2045,7 @@
 		if (!port) {
 			printk(KERN_WARNING
 			       "IOC3 serial memory not available for port\n");
+			ret = -ENOMEM;
 			goto out4;
 		}
 		spin_lock_init(&port->ip_lock);
@@ -2146,6 +2148,9 @@
 
 	/* error exits that give back resources */
 out4:
+	for (cnt = 0; cnt < phys_port; cnt++)
+		kfree(ports[cnt]);
+
 	kfree(card_ptr);
 	return ret;
 }
diff --git a/drivers/serial/serial_cs.c b/drivers/serial/serial_cs.c
index 7d475b2..93760b2 100644
--- a/drivers/serial/serial_cs.c
+++ b/drivers/serial/serial_cs.c
@@ -45,7 +45,6 @@
 #include <asm/io.h>
 #include <asm/system.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
@@ -183,10 +182,8 @@
 {
 	struct serial_info *info = link->priv;
 
-	if (info->multi) {
-		link->conf.Present |= PRESENT_EXT_STATUS;
-		link->conf.ExtStatus = ESR_REQ_ATTN_ENA;
-	}
+	if (info->multi)
+		link->config_flags |= CONF_ENABLE_ESR;
 }
 
 static const struct serial_quirk quirks[] = {
@@ -265,13 +262,6 @@
 static int serial_config(struct pcmcia_device * link);
 
 
-/*======================================================================
-
-    After a card is removed, serial_remove() will unregister
-    the serial device(s), and release the PCMCIA configuration.
-    
-======================================================================*/
-
 static void serial_remove(struct pcmcia_device *link)
 {
 	struct serial_info *info = link->priv;
@@ -314,14 +304,6 @@
 	return 0;
 }
 
-/*======================================================================
-
-    serial_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-======================================================================*/
-
 static int serial_probe(struct pcmcia_device *link)
 {
 	struct serial_info *info;
@@ -335,25 +317,13 @@
 	info->p_dev = link;
 	link->priv = info;
 
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	if (do_sound) {
-		link->conf.Attributes |= CONF_ENABLE_SPKR;
-		link->conf.Status = CCSR_AUDIO_ENA;
-	}
-	link->conf.IntType = INT_MEMORY_AND_IO;
+	link->config_flags |= CONF_ENABLE_IRQ;
+	if (do_sound)
+		link->config_flags |= CONF_ENABLE_SPKR;
 
 	return serial_config(link);
 }
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void serial_detach(struct pcmcia_device *link)
 {
 	struct serial_info *info = link->priv;
@@ -361,11 +331,6 @@
 	dev_dbg(&link->dev, "serial_detach\n");
 
 	/*
-	 * Ensure any outstanding scheduled tasks are completed.
-	 */
-	flush_scheduled_work();
-
-	/*
 	 * Ensure that the ports have been released.
 	 */
 	serial_remove(link);
@@ -430,47 +395,45 @@
 	return -ENODEV;
 }
 
-static int simple_config_check(struct pcmcia_device *p_dev,
-			       cistpl_cftable_entry_t *cf,
-			       cistpl_cftable_entry_t *dflt,
-			       unsigned int vcc,
-			       void *priv_data)
+static int simple_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
 	static const int size_table[2] = { 8, 16 };
 	int *try = priv_data;
 
-	if (cf->vpp1.present & (1 << CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			cf->vpp1.param[CISTPL_POWER_VNOM] / 10000;
+	if (p_dev->resource[0]->start == 0)
+		return -ENODEV;
 
-	p_dev->io_lines = ((*try & 0x1) == 0) ?
-			16 : cf->io.flags & CISTPL_IO_LINES_MASK;
+	if ((*try & 0x1) == 0)
+		p_dev->io_lines = 16;
 
-	if ((cf->io.nwin > 0) && (cf->io.win[0].len == size_table[(*try >> 1)])
-	    && (cf->io.win[0].base != 0)) {
-		p_dev->resource[0]->start = cf->io.win[0].base;
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -EINVAL;
+	if (p_dev->resource[0]->end != size_table[(*try >> 1)])
+		return -ENODEV;
+
+	p_dev->resource[0]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int simple_config_check_notpicky(struct pcmcia_device *p_dev,
-					cistpl_cftable_entry_t *cf,
-					cistpl_cftable_entry_t *dflt,
-					unsigned int vcc,
 					void *priv_data)
 {
 	static const unsigned int base[5] = { 0x3f8, 0x2f8, 0x3e8, 0x2e8, 0x0 };
 	int j;
 
-	if ((cf->io.nwin > 0) && ((cf->io.flags & CISTPL_IO_LINES_MASK) <= 3)) {
-		for (j = 0; j < 5; j++) {
-			p_dev->resource[0]->start = base[j];
-			p_dev->io_lines = base[j] ? 16 : 3;
-			if (!pcmcia_request_io(p_dev))
-				return 0;
-		}
+	if (p_dev->io_lines > 3)
+		return -ENODEV;
+
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[0]->end = 8;
+
+	for (j = 0; j < 5; j++) {
+		p_dev->resource[0]->start = base[j];
+		p_dev->io_lines = base[j] ? 16 : 3;
+		if (!pcmcia_request_io(p_dev))
+			return 0;
 	}
 	return -ENODEV;
 }
@@ -480,11 +443,9 @@
 	struct serial_info *info = link->priv;
 	int i = -ENODEV, try;
 
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	link->resource[0]->end = 8;
-
 	/* First pass: look for a config entry that looks normal.
 	 * Two tries: without IO aliases, then with aliases */
+	link->config_flags |= CONF_AUTO_SET_VPP | CONF_AUTO_SET_IO;
 	for (try = 0; try < 4; try++)
 		if (!pcmcia_loop_config(link, simple_config_check, &try))
 			goto found_port;
@@ -500,7 +461,7 @@
 
 found_port:
 	if (info->multi && (info->manfid == MANFID_3COM))
-		link->conf.ConfigIndex &= ~(0x08);
+		link->config_index &= ~(0x08);
 
 	/*
 	 * Apply any configuration quirks.
@@ -508,51 +469,50 @@
 	if (info->quirk && info->quirk->config)
 		info->quirk->config(link);
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		return -1;
 	return setup_serial(link, info, link->resource[0]->start, link->irq);
 }
 
-static int multi_config_check(struct pcmcia_device *p_dev,
-			      cistpl_cftable_entry_t *cf,
-			      cistpl_cftable_entry_t *dflt,
-			      unsigned int vcc,
-			      void *priv_data)
+static int multi_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	int *base2 = priv_data;
+	int *multi = priv_data;
+
+	if (p_dev->resource[1]->end)
+		return -EINVAL;
 
 	/* The quad port cards have bad CIS's, so just look for a
 	   window larger than 8 ports and assume it will be right */
-	if ((cf->io.nwin == 1) && (cf->io.win[0].len > 8)) {
-		p_dev->resource[0]->start = cf->io.win[0].base;
-		p_dev->io_lines = cf->io.flags & CISTPL_IO_LINES_MASK;
-		if (!pcmcia_request_io(p_dev)) {
-			*base2 = p_dev->resource[0]->start + 8;
-			return 0;
-		}
-	}
-	return -ENODEV;
+	if (p_dev->resource[0]->end <= 8)
+		return -EINVAL;
+
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[0]->end = *multi * 8;
+
+	if (pcmcia_request_io(p_dev))
+		return -ENODEV;
+	return 0;
 }
 
 static int multi_config_check_notpicky(struct pcmcia_device *p_dev,
-				       cistpl_cftable_entry_t *cf,
-				       cistpl_cftable_entry_t *dflt,
-				       unsigned int vcc,
 				       void *priv_data)
 {
 	int *base2 = priv_data;
 
-	if (cf->io.nwin == 2) {
-		p_dev->resource[0]->start = cf->io.win[0].base;
-		p_dev->resource[1]->start = cf->io.win[1].base;
-		p_dev->io_lines = cf->io.flags & CISTPL_IO_LINES_MASK;
-		if (!pcmcia_request_io(p_dev)) {
-			*base2 = p_dev->resource[1]->start;
-			return 0;
-		}
-	}
-	return -ENODEV;
+	if (!p_dev->resource[0]->end || !p_dev->resource[1]->end)
+		return -ENODEV;
+
+	p_dev->resource[0]->end = p_dev->resource[1]->end = 8;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+
+	if (pcmcia_request_io(p_dev))
+		return -ENODEV;
+
+	*base2 = p_dev->resource[0]->start + 8;
+	return 0;
 }
 
 static int multi_config(struct pcmcia_device *link)
@@ -560,12 +520,12 @@
 	struct serial_info *info = link->priv;
 	int i, base2 = 0;
 
+	link->config_flags |= CONF_AUTO_SET_IO;
 	/* First, look for a generic full-sized window */
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	link->resource[0]->end = info->multi * 8;
-	if (pcmcia_loop_config(link, multi_config_check, &base2)) {
+	if (!pcmcia_loop_config(link, multi_config_check, &info->multi))
+		base2 = link->resource[0]->start + 8;
+	else {
 		/* If that didn't work, look for two windows */
-		link->resource[0]->end = link->resource[1]->end = 8;
 		info->multi = 2;
 		if (pcmcia_loop_config(link, multi_config_check_notpicky,
 				       &base2)) {
@@ -584,7 +544,7 @@
 	if (info->quirk && info->quirk->config)
 		info->quirk->config(link);
 
-	i = pcmcia_request_configuration(link, &link->conf);
+	i = pcmcia_enable_device(link);
 	if (i != 0)
 		return -ENODEV;
 
@@ -596,11 +556,11 @@
 				info->prodid == PRODID_POSSIO_GCC)) {
 		int err;
 
-		if (link->conf.ConfigIndex == 1 ||
-		    link->conf.ConfigIndex == 3) {
+		if (link->config_index == 1 ||
+		    link->config_index == 3) {
 			err = setup_serial(link, info, base2,
 					link->irq);
-			base2 = link->resource[0]->start;;
+			base2 = link->resource[0]->start;
 		} else {
 			err = setup_serial(link, info, link->resource[0]->start,
 					link->irq);
@@ -624,33 +584,24 @@
 	return 0;
 }
 
-static int serial_check_for_multi(struct pcmcia_device *p_dev,
-				  cistpl_cftable_entry_t *cf,
-				  cistpl_cftable_entry_t *dflt,
-				  unsigned int vcc,
-				  void *priv_data)
+static int serial_check_for_multi(struct pcmcia_device *p_dev,  void *priv_data)
 {
 	struct serial_info *info = p_dev->priv;
 
-	if ((cf->io.nwin == 1) && (cf->io.win[0].len % 8 == 0))
-		info->multi = cf->io.win[0].len >> 3;
+	if (!p_dev->resource[0]->end)
+		return -EINVAL;
 
-	if ((cf->io.nwin == 2) && (cf->io.win[0].len == 8) &&
-		(cf->io.win[1].len == 8))
+	if ((!p_dev->resource[1]->end) && (p_dev->resource[0]->end % 8 == 0))
+		info->multi = p_dev->resource[0]->end >> 3;
+
+	if ((p_dev->resource[1]->end) && (p_dev->resource[0]->end == 8)
+		&& (p_dev->resource[1]->end == 8))
 		info->multi = 2;
 
 	return 0; /* break */
 }
 
 
-/*======================================================================
-
-    serial_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    serial device available to the system.
-
-======================================================================*/
-
 static int serial_config(struct pcmcia_device * link)
 {
 	struct serial_info *info = link->priv;
@@ -894,9 +845,7 @@
 
 static struct pcmcia_driver serial_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "serial_cs",
-	},
+	.name		= "serial_cs",
 	.probe		= serial_probe,
 	.remove		= serial_detach,
 	.id_table	= serial_ids,
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 91c2f4f..9949c25 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -182,12 +182,27 @@
 	  This enables using the Freescale MPC5121 Programmable Serial
 	  Controller in SPI master mode.
 
-config SPI_MPC8xxx
-	tristate "Freescale MPC8xxx SPI controller"
+config SPI_FSL_LIB
+	tristate
 	depends on FSL_SOC
+
+config SPI_FSL_SPI
+	tristate "Freescale SPI controller"
+	depends on FSL_SOC
+	select SPI_FSL_LIB
 	help
-	  This enables using the Freescale MPC8xxx SPI controllers in master
-	  mode.
+	  This enables using the Freescale SPI controllers in master mode.
+	  MPC83xx platform uses the controller in cpu mode or CPM/QE mode.
+	  MPC8569 uses the controller in QE mode, MPC8610 in cpu mode.
+
+config SPI_FSL_ESPI
+	tristate "Freescale eSPI controller"
+	depends on FSL_SOC
+	select SPI_FSL_LIB
+	help
+	  This enables using the Freescale eSPI controllers in master mode.
+	  From MPC8536, 85xx platform uses the controller, and all P10xx,
+	  P20xx, P30xx,P40xx, P50xx uses this controller.
 
 config SPI_OMAP_UWIRE
 	tristate "OMAP1 MicroWire"
@@ -298,6 +313,13 @@
 	help
 	  SPI driver for Freescale STMP37xx/378x SoC SSP interface
 
+config SPI_TOPCLIFF_PCH
+	tristate "Topcliff PCH SPI Controller"
+	depends on PCI
+	help
+	  SPI driver for the Topcliff PCH (Platform Controller Hub) SPI bus
+	  used in some x86 embedded processors.
+
 config SPI_TXX9
 	tristate "Toshiba TXx9 SPI controller"
 	depends on GENERIC_GPIO && CPU_TX49XX
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index e9cbd18..557aaad 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -2,9 +2,7 @@
 # Makefile for kernel SPI drivers.
 #
 
-ifeq ($(CONFIG_SPI_DEBUG),y)
-EXTRA_CFLAGS += -DDEBUG
-endif
+ccflags-$(CONFIG_SPI_DEBUG) := -DDEBUG
 
 # small core, mostly translating board-specific
 # config declarations into driver model code
@@ -34,11 +32,14 @@
 obj-$(CONFIG_SPI_MPC512x_PSC)		+= mpc512x_psc_spi.o
 obj-$(CONFIG_SPI_MPC52xx_PSC)		+= mpc52xx_psc_spi.o
 obj-$(CONFIG_SPI_MPC52xx)		+= mpc52xx_spi.o
-obj-$(CONFIG_SPI_MPC8xxx)		+= spi_mpc8xxx.o
+obj-$(CONFIG_SPI_FSL_LIB)		+= spi_fsl_lib.o
+obj-$(CONFIG_SPI_FSL_ESPI)		+= spi_fsl_espi.o
+obj-$(CONFIG_SPI_FSL_SPI)		+= spi_fsl_spi.o
 obj-$(CONFIG_SPI_PPC4xx)		+= spi_ppc4xx.o
 obj-$(CONFIG_SPI_S3C24XX_GPIO)		+= spi_s3c24xx_gpio.o
 obj-$(CONFIG_SPI_S3C24XX)		+= spi_s3c24xx_hw.o
 obj-$(CONFIG_SPI_S3C64XX)		+= spi_s3c64xx.o
+obj-$(CONFIG_SPI_TOPCLIFF_PCH)		+= spi_topcliff_pch.o
 obj-$(CONFIG_SPI_TXX9)			+= spi_txx9.o
 obj-$(CONFIG_SPI_XILINX)		+= xilinx_spi.o
 obj-$(CONFIG_SPI_XILINX_OF)		+= xilinx_spi_of.o
diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index 4c37c4e2..fb3d1b3 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -27,7 +27,6 @@
 /*
  * TODO:
  * - add timeout on polled transfers
- * - add generic DMA framework support
  */
 
 #include <linux/init.h>
@@ -45,6 +44,9 @@
 #include <linux/amba/pl022.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/scatterlist.h>
 
 /*
  * This macro is used to define some register default values.
@@ -381,6 +383,14 @@
 	enum ssp_reading		read;
 	enum ssp_writing		write;
 	u32				exp_fifo_level;
+	/* DMA settings */
+#ifdef CONFIG_DMA_ENGINE
+	struct dma_chan			*dma_rx_channel;
+	struct dma_chan			*dma_tx_channel;
+	struct sg_table			sgt_rx;
+	struct sg_table			sgt_tx;
+	char				*dummypage;
+#endif
 };
 
 /**
@@ -406,7 +416,7 @@
 	u16 dmacr;
 	u16 cpsr;
 	u8 n_bytes;
-	u8 enable_dma:1;
+	bool enable_dma;
 	enum ssp_reading read;
 	enum ssp_writing write;
 	void (*cs_control) (u32 command);
@@ -763,6 +773,371 @@
 	}
 	return STATE_DONE;
 }
+
+/*
+ * This DMA functionality is only compiled in if we have
+ * access to the generic DMA devices/DMA engine.
+ */
+#ifdef CONFIG_DMA_ENGINE
+static void unmap_free_dma_scatter(struct pl022 *pl022)
+{
+	/* Unmap and free the SG tables */
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	sg_free_table(&pl022->sgt_rx);
+	sg_free_table(&pl022->sgt_tx);
+}
+
+static void dma_callback(void *data)
+{
+	struct pl022 *pl022 = data;
+	struct spi_message *msg = pl022->cur_msg;
+
+	BUG_ON(!pl022->sgt_rx.sgl);
+
+#ifdef VERBOSE_DEBUG
+	/*
+	 * Optionally dump out buffers to inspect contents, this is
+	 * good if you want to convince yourself that the loopback
+	 * read/write contents are the same, when adopting to a new
+	 * DMA engine.
+	 */
+	{
+		struct scatterlist *sg;
+		unsigned int i;
+
+		dma_sync_sg_for_cpu(&pl022->adev->dev,
+				    pl022->sgt_rx.sgl,
+				    pl022->sgt_rx.nents,
+				    DMA_FROM_DEVICE);
+
+		for_each_sg(pl022->sgt_rx.sgl, sg, pl022->sgt_rx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI RX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI RX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+		for_each_sg(pl022->sgt_tx.sgl, sg, pl022->sgt_tx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI TX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI TX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+	}
+#endif
+
+	unmap_free_dma_scatter(pl022);
+
+	/* Update total bytes transfered */
+	msg->actual_length += pl022->cur_transfer->len;
+	if (pl022->cur_transfer->cs_change)
+		pl022->cur_chip->
+			cs_control(SSP_CHIP_DESELECT);
+
+	/* Move to next transfer */
+	msg->state = next_transfer(pl022);
+	tasklet_schedule(&pl022->pump_transfers);
+}
+
+static void setup_dma_scatter(struct pl022 *pl022,
+			      void *buffer,
+			      unsigned int length,
+			      struct sg_table *sgtab)
+{
+	struct scatterlist *sg;
+	int bytesleft = length;
+	void *bufp = buffer;
+	int mapbytes;
+	int i;
+
+	if (buffer) {
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			/*
+			 * If there are less bytes left than what fits
+			 * in the current page (plus page alignment offset)
+			 * we just feed in this, else we stuff in as much
+			 * as we can.
+			 */
+			if (bytesleft < (PAGE_SIZE - offset_in_page(bufp)))
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE - offset_in_page(bufp);
+			sg_set_page(sg, virt_to_page(bufp),
+				    mapbytes, offset_in_page(bufp));
+			bufp += mapbytes;
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX target page @ %p, %d bytes, %d left\n",
+				bufp, mapbytes, bytesleft);
+		}
+	} else {
+		/* Map the dummy buffer on every page */
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			if (bytesleft < PAGE_SIZE)
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE;
+			sg_set_page(sg, virt_to_page(pl022->dummypage),
+				    mapbytes, 0);
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX to dummy page %d bytes, %d left\n",
+				mapbytes, bytesleft);
+
+		}
+	}
+	BUG_ON(bytesleft);
+}
+
+/**
+ * configure_dma - configures the channels for the next transfer
+ * @pl022: SSP driver's private data structure
+ */
+static int configure_dma(struct pl022 *pl022)
+{
+	struct dma_slave_config rx_conf = {
+		.src_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_FROM_DEVICE,
+		.src_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	struct dma_slave_config tx_conf = {
+		.dst_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_TO_DEVICE,
+		.dst_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	unsigned int pages;
+	int ret;
+	int sglen;
+	struct dma_chan *rxchan = pl022->dma_rx_channel;
+	struct dma_chan *txchan = pl022->dma_tx_channel;
+	struct dma_async_tx_descriptor *rxdesc;
+	struct dma_async_tx_descriptor *txdesc;
+	dma_cookie_t cookie;
+
+	/* Check that the channels are available */
+	if (!rxchan || !txchan)
+		return -ENODEV;
+
+	switch (pl022->read) {
+	case READING_NULL:
+		/* Use the same as for writing */
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case READING_U8:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case READING_U16:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case READING_U32:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		break;
+	}
+
+	switch (pl022->write) {
+	case WRITING_NULL:
+		/* Use the same as for reading */
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case WRITING_U8:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case WRITING_U16:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case WRITING_U32:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;;
+		break;
+	}
+
+	/* SPI pecularity: we need to read and write the same width */
+	if (rx_conf.src_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		rx_conf.src_addr_width = tx_conf.dst_addr_width;
+	if (tx_conf.dst_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		tx_conf.dst_addr_width = rx_conf.src_addr_width;
+	BUG_ON(rx_conf.src_addr_width != tx_conf.dst_addr_width);
+
+	rxchan->device->device_control(rxchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &rx_conf);
+	txchan->device->device_control(txchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &tx_conf);
+
+	/* Create sglists for the transfers */
+	pages = (pl022->cur_transfer->len >> PAGE_SHIFT) + 1;
+	dev_dbg(&pl022->adev->dev, "using %d pages for transfer\n", pages);
+
+	ret = sg_alloc_table(&pl022->sgt_rx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_rx_sg;
+
+	ret = sg_alloc_table(&pl022->sgt_tx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_tx_sg;
+
+	/* Fill in the scatterlists for the RX+TX buffers */
+	setup_dma_scatter(pl022, pl022->rx,
+			  pl022->cur_transfer->len, &pl022->sgt_rx);
+	setup_dma_scatter(pl022, pl022->tx,
+			  pl022->cur_transfer->len, &pl022->sgt_tx);
+
+	/* Map DMA buffers */
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+			   pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	if (!sglen)
+		goto err_rx_sgmap;
+
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+			   pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	if (!sglen)
+		goto err_tx_sgmap;
+
+	/* Send both scatterlists */
+	rxdesc = rxchan->device->device_prep_slave_sg(rxchan,
+				      pl022->sgt_rx.sgl,
+				      pl022->sgt_rx.nents,
+				      DMA_FROM_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!rxdesc)
+		goto err_rxdesc;
+
+	txdesc = txchan->device->device_prep_slave_sg(txchan,
+				      pl022->sgt_tx.sgl,
+				      pl022->sgt_tx.nents,
+				      DMA_TO_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!txdesc)
+		goto err_txdesc;
+
+	/* Put the callback on the RX transfer only, that should finish last */
+	rxdesc->callback = dma_callback;
+	rxdesc->callback_param = pl022;
+
+	/* Submit and fire RX and TX with TX last so we're ready to read! */
+	cookie = rxdesc->tx_submit(rxdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_rx;
+	cookie = txdesc->tx_submit(txdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_tx;
+	rxchan->device->device_issue_pending(rxchan);
+	txchan->device->device_issue_pending(txchan);
+
+	return 0;
+
+err_submit_tx:
+err_submit_rx:
+err_txdesc:
+	txchan->device->device_control(txchan, DMA_TERMINATE_ALL, 0);
+err_rxdesc:
+	rxchan->device->device_control(rxchan, DMA_TERMINATE_ALL, 0);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+err_tx_sgmap:
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_tx.nents, DMA_FROM_DEVICE);
+err_rx_sgmap:
+	sg_free_table(&pl022->sgt_tx);
+err_alloc_tx_sg:
+	sg_free_table(&pl022->sgt_rx);
+err_alloc_rx_sg:
+	return -ENOMEM;
+}
+
+static int __init pl022_dma_probe(struct pl022 *pl022)
+{
+	dma_cap_mask_t mask;
+
+	/* Try to acquire a generic DMA engine slave channel */
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+	/*
+	 * We need both RX and TX channels to do DMA, else do none
+	 * of them.
+	 */
+	pl022->dma_rx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_rx_param);
+	if (!pl022->dma_rx_channel) {
+		dev_err(&pl022->adev->dev, "no RX DMA channel!\n");
+		goto err_no_rxchan;
+	}
+
+	pl022->dma_tx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_tx_param);
+	if (!pl022->dma_tx_channel) {
+		dev_err(&pl022->adev->dev, "no TX DMA channel!\n");
+		goto err_no_txchan;
+	}
+
+	pl022->dummypage = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!pl022->dummypage) {
+		dev_err(&pl022->adev->dev, "no DMA dummypage!\n");
+		goto err_no_dummypage;
+	}
+
+	dev_info(&pl022->adev->dev, "setup for DMA on RX %s, TX %s\n",
+		 dma_chan_name(pl022->dma_rx_channel),
+		 dma_chan_name(pl022->dma_tx_channel));
+
+	return 0;
+
+err_no_dummypage:
+	dma_release_channel(pl022->dma_tx_channel);
+err_no_txchan:
+	dma_release_channel(pl022->dma_rx_channel);
+	pl022->dma_rx_channel = NULL;
+err_no_rxchan:
+	return -ENODEV;
+}
+
+static void terminate_dma(struct pl022 *pl022)
+{
+	struct dma_chan *rxchan = pl022->dma_rx_channel;
+	struct dma_chan *txchan = pl022->dma_tx_channel;
+
+	rxchan->device->device_control(rxchan, DMA_TERMINATE_ALL, 0);
+	txchan->device->device_control(txchan, DMA_TERMINATE_ALL, 0);
+	unmap_free_dma_scatter(pl022);
+}
+
+static void pl022_dma_remove(struct pl022 *pl022)
+{
+	if (pl022->busy)
+		terminate_dma(pl022);
+	if (pl022->dma_tx_channel)
+		dma_release_channel(pl022->dma_tx_channel);
+	if (pl022->dma_rx_channel)
+		dma_release_channel(pl022->dma_rx_channel);
+	kfree(pl022->dummypage);
+}
+
+#else
+static inline int configure_dma(struct pl022 *pl022)
+{
+	return -ENODEV;
+}
+
+static inline int pl022_dma_probe(struct pl022 *pl022)
+{
+	return 0;
+}
+
+static inline void pl022_dma_remove(struct pl022 *pl022)
+{
+}
+#endif
+
 /**
  * pl022_interrupt_handler - Interrupt handler for SSP controller
  *
@@ -794,14 +1169,17 @@
 	if (unlikely(!irq_status))
 		return IRQ_NONE;
 
-	/* This handles the error code interrupts */
+	/*
+	 * This handles the FIFO interrupts, the timeout
+	 * interrupts are flatly ignored, they cannot be
+	 * trusted.
+	 */
 	if (unlikely(irq_status & SSP_MIS_MASK_RORMIS)) {
 		/*
 		 * Overrun interrupt - bail out since our Data has been
 		 * corrupted
 		 */
-		dev_err(&pl022->adev->dev,
-			"FIFO overrun\n");
+		dev_err(&pl022->adev->dev, "FIFO overrun\n");
 		if (readw(SSP_SR(pl022->virtbase)) & SSP_SR_MASK_RFF)
 			dev_err(&pl022->adev->dev,
 				"RXFIFO is full\n");
@@ -896,8 +1274,8 @@
 }
 
 /**
- * pump_transfers - Tasklet function which schedules next interrupt transfer
- * when running in interrupt transfer mode.
+ * pump_transfers - Tasklet function which schedules next transfer
+ * when running in interrupt or DMA transfer mode.
  * @data: SSP driver private data structure
  *
  */
@@ -954,65 +1332,23 @@
 	}
 	/* Flush the FIFOs and let's go! */
 	flush(pl022);
+
+	if (pl022->cur_chip->enable_dma) {
+		if (configure_dma(pl022)) {
+			dev_dbg(&pl022->adev->dev,
+				"configuration of DMA failed, fall back to interrupt mode\n");
+			goto err_config_dma;
+		}
+		return;
+	}
+
+err_config_dma:
 	writew(ENABLE_ALL_INTERRUPTS, SSP_IMSC(pl022->virtbase));
 }
 
-/**
- * NOT IMPLEMENTED
- * configure_dma - It configures the DMA pipes for DMA transfers
- * @data: SSP driver's private data structure
- *
- */
-static int configure_dma(void *data)
+static void do_interrupt_dma_transfer(struct pl022 *pl022)
 {
-	struct pl022 *pl022 = data;
-	dev_dbg(&pl022->adev->dev, "configure DMA\n");
-	return -ENOTSUPP;
-}
-
-/**
- * do_dma_transfer - It handles transfers of the current message
- * if it is DMA xfer.
- * NOT FULLY IMPLEMENTED
- * @data: SSP driver's private data structure
- */
-static void do_dma_transfer(void *data)
-{
-	struct pl022 *pl022 = data;
-
-	if (configure_dma(data)) {
-		dev_dbg(&pl022->adev->dev, "configuration of DMA Failed!\n");
-		goto err_config_dma;
-	}
-
-	/* TODO: Implememt DMA setup of pipes here */
-
-	/* Enable target chip, set up transfer */
-	pl022->cur_chip->cs_control(SSP_CHIP_SELECT);
-	if (set_up_next_transfer(pl022, pl022->cur_transfer)) {
-		/* Error path */
-		pl022->cur_msg->state = STATE_ERROR;
-		pl022->cur_msg->status = -EIO;
-		giveback(pl022);
-		return;
-	}
-	/* Enable SSP */
-	writew((readw(SSP_CR1(pl022->virtbase)) | SSP_CR1_MASK_SSE),
-	       SSP_CR1(pl022->virtbase));
-
-	/* TODO: Enable the DMA transfer here */
-	return;
-
- err_config_dma:
-	pl022->cur_msg->state = STATE_ERROR;
-	pl022->cur_msg->status = -EIO;
-	giveback(pl022);
-	return;
-}
-
-static void do_interrupt_transfer(void *data)
-{
-	struct pl022 *pl022 = data;
+	u32 irqflags = ENABLE_ALL_INTERRUPTS;
 
 	/* Enable target chip */
 	pl022->cur_chip->cs_control(SSP_CHIP_SELECT);
@@ -1023,15 +1359,26 @@
 		giveback(pl022);
 		return;
 	}
+	/* If we're using DMA, set up DMA here */
+	if (pl022->cur_chip->enable_dma) {
+		/* Configure DMA transfer */
+		if (configure_dma(pl022)) {
+			dev_dbg(&pl022->adev->dev,
+				"configuration of DMA failed, fall back to interrupt mode\n");
+			goto err_config_dma;
+		}
+		/* Disable interrupts in DMA mode, IRQ from DMA controller */
+		irqflags = DISABLE_ALL_INTERRUPTS;
+	}
+err_config_dma:
 	/* Enable SSP, turn on interrupts */
 	writew((readw(SSP_CR1(pl022->virtbase)) | SSP_CR1_MASK_SSE),
 	       SSP_CR1(pl022->virtbase));
-	writew(ENABLE_ALL_INTERRUPTS, SSP_IMSC(pl022->virtbase));
+	writew(irqflags, SSP_IMSC(pl022->virtbase));
 }
 
-static void do_polling_transfer(void *data)
+static void do_polling_transfer(struct pl022 *pl022)
 {
-	struct pl022 *pl022 = data;
 	struct spi_message *message = NULL;
 	struct spi_transfer *transfer = NULL;
 	struct spi_transfer *previous = NULL;
@@ -1101,7 +1448,7 @@
  *
  * This function checks if there is any spi message in the queue that
  * needs processing and delegate control to appropriate function
- * do_polling_transfer()/do_interrupt_transfer()/do_dma_transfer()
+ * do_polling_transfer()/do_interrupt_dma_transfer()
  * based on the kind of the transfer
  *
  */
@@ -1150,10 +1497,8 @@
 
 	if (pl022->cur_chip->xfer_type == POLLING_TRANSFER)
 		do_polling_transfer(pl022);
-	else if (pl022->cur_chip->xfer_type == INTERRUPT_TRANSFER)
-		do_interrupt_transfer(pl022);
 	else
-		do_dma_transfer(pl022);
+		do_interrupt_dma_transfer(pl022);
 }
 
 
@@ -1248,100 +1593,56 @@
 }
 
 static int verify_controller_parameters(struct pl022 *pl022,
-					struct pl022_config_chip *chip_info)
+				struct pl022_config_chip const *chip_info)
 {
-	if ((chip_info->lbm != LOOPBACK_ENABLED)
-	    && (chip_info->lbm != LOOPBACK_DISABLED)) {
-		dev_err(chip_info->dev,
-			"loopback Mode is configured incorrectly\n");
-		return -EINVAL;
-	}
 	if ((chip_info->iface < SSP_INTERFACE_MOTOROLA_SPI)
 	    || (chip_info->iface > SSP_INTERFACE_UNIDIRECTIONAL)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"interface is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->iface == SSP_INTERFACE_UNIDIRECTIONAL) &&
 	    (!pl022->vendor->unidir)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"unidirectional mode not supported in this "
 			"hardware version\n");
 		return -EINVAL;
 	}
 	if ((chip_info->hierarchy != SSP_MASTER)
 	    && (chip_info->hierarchy != SSP_SLAVE)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"hierarchy is configured incorrectly\n");
 		return -EINVAL;
 	}
-	if (((chip_info->clk_freq).cpsdvsr < CPSDVR_MIN)
-	    || ((chip_info->clk_freq).cpsdvsr > CPSDVR_MAX)) {
-		dev_err(chip_info->dev,
-			"cpsdvsr is configured incorrectly\n");
-		return -EINVAL;
-	}
-	if ((chip_info->endian_rx != SSP_RX_MSB)
-	    && (chip_info->endian_rx != SSP_RX_LSB)) {
-		dev_err(chip_info->dev,
-			"RX FIFO endianess is configured incorrectly\n");
-		return -EINVAL;
-	}
-	if ((chip_info->endian_tx != SSP_TX_MSB)
-	    && (chip_info->endian_tx != SSP_TX_LSB)) {
-		dev_err(chip_info->dev,
-			"TX FIFO endianess is configured incorrectly\n");
-		return -EINVAL;
-	}
-	if ((chip_info->data_size < SSP_DATA_BITS_4)
-	    || (chip_info->data_size > SSP_DATA_BITS_32)) {
-		dev_err(chip_info->dev,
-			"DATA Size is configured incorrectly\n");
-		return -EINVAL;
-	}
 	if ((chip_info->com_mode != INTERRUPT_TRANSFER)
 	    && (chip_info->com_mode != DMA_TRANSFER)
 	    && (chip_info->com_mode != POLLING_TRANSFER)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"Communication mode is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->rx_lev_trig < SSP_RX_1_OR_MORE_ELEM)
 	    || (chip_info->rx_lev_trig > SSP_RX_32_OR_MORE_ELEM)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"RX FIFO Trigger Level is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->tx_lev_trig < SSP_TX_1_OR_MORE_EMPTY_LOC)
 	    || (chip_info->tx_lev_trig > SSP_TX_32_OR_MORE_EMPTY_LOC)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"TX FIFO Trigger Level is configured incorrectly\n");
 		return -EINVAL;
 	}
-	if (chip_info->iface == SSP_INTERFACE_MOTOROLA_SPI) {
-		if ((chip_info->clk_phase != SSP_CLK_FIRST_EDGE)
-		    && (chip_info->clk_phase != SSP_CLK_SECOND_EDGE)) {
-			dev_err(chip_info->dev,
-				"Clock Phase is configured incorrectly\n");
-			return -EINVAL;
-		}
-		if ((chip_info->clk_pol != SSP_CLK_POL_IDLE_LOW)
-		    && (chip_info->clk_pol != SSP_CLK_POL_IDLE_HIGH)) {
-			dev_err(chip_info->dev,
-				"Clock Polarity is configured incorrectly\n");
-			return -EINVAL;
-		}
-	}
 	if (chip_info->iface == SSP_INTERFACE_NATIONAL_MICROWIRE) {
 		if ((chip_info->ctrl_len < SSP_BITS_4)
 		    || (chip_info->ctrl_len > SSP_BITS_32)) {
-			dev_err(chip_info->dev,
+			dev_err(&pl022->adev->dev,
 				"CTRL LEN is configured incorrectly\n");
 			return -EINVAL;
 		}
 		if ((chip_info->wait_state != SSP_MWIRE_WAIT_ZERO)
 		    && (chip_info->wait_state != SSP_MWIRE_WAIT_ONE)) {
-			dev_err(chip_info->dev,
+			dev_err(&pl022->adev->dev,
 				"Wait State is configured incorrectly\n");
 			return -EINVAL;
 		}
@@ -1350,24 +1651,20 @@
 			if ((chip_info->duplex !=
 			     SSP_MICROWIRE_CHANNEL_FULL_DUPLEX)
 			    && (chip_info->duplex !=
-				SSP_MICROWIRE_CHANNEL_HALF_DUPLEX))
-				dev_err(chip_info->dev,
+				SSP_MICROWIRE_CHANNEL_HALF_DUPLEX)) {
+				dev_err(&pl022->adev->dev,
 					"Microwire duplex mode is configured incorrectly\n");
 				return -EINVAL;
+			}
 		} else {
 			if (chip_info->duplex != SSP_MICROWIRE_CHANNEL_FULL_DUPLEX)
-				dev_err(chip_info->dev,
+				dev_err(&pl022->adev->dev,
 					"Microwire half duplex mode requested,"
 					" but this is only available in the"
 					" ST version of PL022\n");
 			return -EINVAL;
 		}
 	}
-	if (chip_info->cs_control == NULL) {
-		dev_warn(chip_info->dev,
-			"Chip Select Function is NULL for this chip\n");
-		chip_info->cs_control = null_cs_control;
-	}
 	return 0;
 }
 
@@ -1467,22 +1764,24 @@
 	return 0;
 }
 
-/**
- * NOT IMPLEMENTED
- * process_dma_info - Processes the DMA info provided by client drivers
- * @chip_info: chip info provided by client device
- * @chip: Runtime state maintained by the SSP controller for each spi device
- *
- * This function processes and stores DMA config provided by client driver
- * into the runtime state maintained by the SSP controller driver
+
+/*
+ * A piece of default chip info unless the platform
+ * supplies it.
  */
-static int process_dma_info(struct pl022_config_chip *chip_info,
-			    struct chip_data *chip)
-{
-	dev_err(chip_info->dev,
-		"cannot process DMA info, DMA not implemented!\n");
-	return -ENOTSUPP;
-}
+static const struct pl022_config_chip pl022_default_chip_info = {
+	.com_mode = POLLING_TRANSFER,
+	.iface = SSP_INTERFACE_MOTOROLA_SPI,
+	.hierarchy = SSP_SLAVE,
+	.slave_tx_disable = DO_NOT_DRIVE_TX,
+	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
+	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
+	.ctrl_len = SSP_BITS_8,
+	.wait_state = SSP_MWIRE_WAIT_ZERO,
+	.duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
+	.cs_control = null_cs_control,
+};
+
 
 /**
  * pl022_setup - setup function registered to SPI master framework
@@ -1496,23 +1795,15 @@
  * controller hardware here, that is not done until the actual transfer
  * commence.
  */
-
-/* FIXME: JUST GUESSING the spi->mode bits understood by this driver */
-#define MODEBITS	(SPI_CPOL | SPI_CPHA | SPI_CS_HIGH \
-			| SPI_LSB_FIRST | SPI_LOOP)
-
 static int pl022_setup(struct spi_device *spi)
 {
-	struct pl022_config_chip *chip_info;
+	struct pl022_config_chip const *chip_info;
 	struct chip_data *chip;
+	struct ssp_clock_params clk_freq;
 	int status = 0;
 	struct pl022 *pl022 = spi_master_get_devdata(spi->master);
-
-	if (spi->mode & ~MODEBITS) {
-		dev_dbg(&spi->dev, "unsupported mode bits %x\n",
-			spi->mode & ~MODEBITS);
-		return -EINVAL;
-	}
+	unsigned int bits = spi->bits_per_word;
+	u32 tmp;
 
 	if (!spi->max_speed_hz)
 		return -EINVAL;
@@ -1535,48 +1826,13 @@
 	chip_info = spi->controller_data;
 
 	if (chip_info == NULL) {
+		chip_info = &pl022_default_chip_info;
 		/* spi_board_info.controller_data not is supplied */
 		dev_dbg(&spi->dev,
 			"using default controller_data settings\n");
-
-		chip_info =
-			kzalloc(sizeof(struct pl022_config_chip), GFP_KERNEL);
-
-		if (!chip_info) {
-			dev_err(&spi->dev,
-				"cannot allocate controller data\n");
-			status = -ENOMEM;
-			goto err_first_setup;
-		}
-
-		dev_dbg(&spi->dev, "allocated memory for controller data\n");
-
-		/* Pointer back to the SPI device */
-		chip_info->dev = &spi->dev;
-		/*
-		 * Set controller data default values:
-		 * Polling is supported by default
-		 */
-		chip_info->lbm = LOOPBACK_DISABLED;
-		chip_info->com_mode = POLLING_TRANSFER;
-		chip_info->iface = SSP_INTERFACE_MOTOROLA_SPI;
-		chip_info->hierarchy = SSP_SLAVE;
-		chip_info->slave_tx_disable = DO_NOT_DRIVE_TX;
-		chip_info->endian_tx = SSP_TX_LSB;
-		chip_info->endian_rx = SSP_RX_LSB;
-		chip_info->data_size = SSP_DATA_BITS_12;
-		chip_info->rx_lev_trig = SSP_RX_1_OR_MORE_ELEM;
-		chip_info->tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC;
-		chip_info->clk_phase = SSP_CLK_SECOND_EDGE;
-		chip_info->clk_pol = SSP_CLK_POL_IDLE_LOW;
-		chip_info->ctrl_len = SSP_BITS_8;
-		chip_info->wait_state = SSP_MWIRE_WAIT_ZERO;
-		chip_info->duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX;
-		chip_info->cs_control = null_cs_control;
-	} else {
+	} else
 		dev_dbg(&spi->dev,
 			"using user supplied controller_data settings\n");
-	}
 
 	/*
 	 * We can override with custom divisors, else we use the board
@@ -1586,29 +1842,48 @@
 	    && (0 == chip_info->clk_freq.scr)) {
 		status = calculate_effective_freq(pl022,
 						  spi->max_speed_hz,
-						  &chip_info->clk_freq);
+						  &clk_freq);
 		if (status < 0)
 			goto err_config_params;
 	} else {
-		if ((chip_info->clk_freq.cpsdvsr % 2) != 0)
-			chip_info->clk_freq.cpsdvsr =
-				chip_info->clk_freq.cpsdvsr - 1;
+		memcpy(&clk_freq, &chip_info->clk_freq, sizeof(clk_freq));
+		if ((clk_freq.cpsdvsr % 2) != 0)
+			clk_freq.cpsdvsr =
+				clk_freq.cpsdvsr - 1;
 	}
+	if ((clk_freq.cpsdvsr < CPSDVR_MIN)
+	    || (clk_freq.cpsdvsr > CPSDVR_MAX)) {
+		dev_err(&spi->dev,
+			"cpsdvsr is configured incorrectly\n");
+		goto err_config_params;
+	}
+
+
 	status = verify_controller_parameters(pl022, chip_info);
 	if (status) {
 		dev_err(&spi->dev, "controller data is incorrect");
 		goto err_config_params;
 	}
+
 	/* Now set controller state based on controller data */
 	chip->xfer_type = chip_info->com_mode;
-	chip->cs_control = chip_info->cs_control;
+	if (!chip_info->cs_control) {
+		chip->cs_control = null_cs_control;
+		dev_warn(&spi->dev,
+			 "chip select function is NULL for this chip\n");
+	} else
+		chip->cs_control = chip_info->cs_control;
 
-	if (chip_info->data_size <= 8) {
-		dev_dbg(&spi->dev, "1 <= n <=8 bits per word\n");
+	if (bits <= 3) {
+		/* PL022 doesn't support less than 4-bits */
+		status = -ENOTSUPP;
+		goto err_config_params;
+	} else if (bits <= 8) {
+		dev_dbg(&spi->dev, "4 <= n <=8 bits per word\n");
 		chip->n_bytes = 1;
 		chip->read = READING_U8;
 		chip->write = WRITING_U8;
-	} else if (chip_info->data_size <= 16) {
+	} else if (bits <= 16) {
 		dev_dbg(&spi->dev, "9 <= n <= 16 bits per word\n");
 		chip->n_bytes = 2;
 		chip->read = READING_U16;
@@ -1625,6 +1900,7 @@
 			dev_err(&spi->dev,
 				"a standard pl022 can only handle "
 				"1 <= n <= 16 bit words\n");
+			status = -ENOTSUPP;
 			goto err_config_params;
 		}
 	}
@@ -1636,9 +1912,8 @@
 	chip->cpsr = 0;
 	if ((chip_info->com_mode == DMA_TRANSFER)
 	    && ((pl022->master_info)->enable_dma)) {
-		chip->enable_dma = 1;
+		chip->enable_dma = true;
 		dev_dbg(&spi->dev, "DMA mode set in controller state\n");
-		status = process_dma_info(chip_info, chip);
 		if (status < 0)
 			goto err_config_params;
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_ENABLED,
@@ -1646,7 +1921,7 @@
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_ENABLED,
 			       SSP_DMACR_MASK_TXDMAE, 1);
 	} else {
-		chip->enable_dma = 0;
+		chip->enable_dma = false;
 		dev_dbg(&spi->dev, "DMA mode NOT set in controller state\n");
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_DISABLED,
 			       SSP_DMACR_MASK_RXDMAE, 0);
@@ -1654,10 +1929,12 @@
 			       SSP_DMACR_MASK_TXDMAE, 1);
 	}
 
-	chip->cpsr = chip_info->clk_freq.cpsdvsr;
+	chip->cpsr = clk_freq.cpsdvsr;
 
 	/* Special setup for the ST micro extended control registers */
 	if (pl022->vendor->extended_cr) {
+		u32 etx;
+
 		if (pl022->vendor->pl023) {
 			/* These bits are only in the PL023 */
 			SSP_WRITE_BITS(chip->cr1, chip_info->clkdelay,
@@ -1673,29 +1950,51 @@
 			SSP_WRITE_BITS(chip->cr1, chip_info->wait_state,
 				       SSP_CR1_MASK_MWAIT_ST, 6);
 		}
-		SSP_WRITE_BITS(chip->cr0, chip_info->data_size,
+		SSP_WRITE_BITS(chip->cr0, bits - 1,
 			       SSP_CR0_MASK_DSS_ST, 0);
-		SSP_WRITE_BITS(chip->cr1, chip_info->endian_rx,
-			       SSP_CR1_MASK_RENDN_ST, 4);
-		SSP_WRITE_BITS(chip->cr1, chip_info->endian_tx,
-			       SSP_CR1_MASK_TENDN_ST, 5);
+
+		if (spi->mode & SPI_LSB_FIRST) {
+			tmp = SSP_RX_LSB;
+			etx = SSP_TX_LSB;
+		} else {
+			tmp = SSP_RX_MSB;
+			etx = SSP_TX_MSB;
+		}
+		SSP_WRITE_BITS(chip->cr1, tmp, SSP_CR1_MASK_RENDN_ST, 4);
+		SSP_WRITE_BITS(chip->cr1, etx, SSP_CR1_MASK_TENDN_ST, 5);
 		SSP_WRITE_BITS(chip->cr1, chip_info->rx_lev_trig,
 			       SSP_CR1_MASK_RXIFLSEL_ST, 7);
 		SSP_WRITE_BITS(chip->cr1, chip_info->tx_lev_trig,
 			       SSP_CR1_MASK_TXIFLSEL_ST, 10);
 	} else {
-		SSP_WRITE_BITS(chip->cr0, chip_info->data_size,
+		SSP_WRITE_BITS(chip->cr0, bits - 1,
 			       SSP_CR0_MASK_DSS, 0);
 		SSP_WRITE_BITS(chip->cr0, chip_info->iface,
 			       SSP_CR0_MASK_FRF, 4);
 	}
+
 	/* Stuff that is common for all versions */
-	SSP_WRITE_BITS(chip->cr0, chip_info->clk_pol, SSP_CR0_MASK_SPO, 6);
-	SSP_WRITE_BITS(chip->cr0, chip_info->clk_phase, SSP_CR0_MASK_SPH, 7);
-	SSP_WRITE_BITS(chip->cr0, chip_info->clk_freq.scr, SSP_CR0_MASK_SCR, 8);
+	if (spi->mode & SPI_CPOL)
+		tmp = SSP_CLK_POL_IDLE_HIGH;
+	else
+		tmp = SSP_CLK_POL_IDLE_LOW;
+	SSP_WRITE_BITS(chip->cr0, tmp, SSP_CR0_MASK_SPO, 6);
+
+	if (spi->mode & SPI_CPHA)
+		tmp = SSP_CLK_SECOND_EDGE;
+	else
+		tmp = SSP_CLK_FIRST_EDGE;
+	SSP_WRITE_BITS(chip->cr0, tmp, SSP_CR0_MASK_SPH, 7);
+
+	SSP_WRITE_BITS(chip->cr0, clk_freq.scr, SSP_CR0_MASK_SCR, 8);
 	/* Loopback is available on all versions except PL023 */
-	if (!pl022->vendor->pl023)
-		SSP_WRITE_BITS(chip->cr1, chip_info->lbm, SSP_CR1_MASK_LBM, 0);
+	if (!pl022->vendor->pl023) {
+		if (spi->mode & SPI_LOOP)
+			tmp = LOOPBACK_ENABLED;
+		else
+			tmp = LOOPBACK_DISABLED;
+		SSP_WRITE_BITS(chip->cr1, tmp, SSP_CR1_MASK_LBM, 0);
+	}
 	SSP_WRITE_BITS(chip->cr1, SSP_DISABLED, SSP_CR1_MASK_SSE, 1);
 	SSP_WRITE_BITS(chip->cr1, chip_info->hierarchy, SSP_CR1_MASK_MS, 2);
 	SSP_WRITE_BITS(chip->cr1, chip_info->slave_tx_disable, SSP_CR1_MASK_SOD, 3);
@@ -1704,7 +2003,7 @@
 	spi_set_ctldata(spi, chip);
 	return status;
  err_config_params:
- err_first_setup:
+	spi_set_ctldata(spi, NULL);
 	kfree(chip);
 	return status;
 }
@@ -1766,12 +2065,21 @@
 	master->setup = pl022_setup;
 	master->transfer = pl022_transfer;
 
+	/*
+	 * Supports mode 0-3, loopback, and active low CS. Transfers are
+	 * always MS bit first on the original pl022.
+	 */
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH | SPI_LOOP;
+	if (pl022->vendor->extended_cr)
+		master->mode_bits |= SPI_LSB_FIRST;
+
 	dev_dbg(&adev->dev, "BUSNO: %d\n", master->bus_num);
 
 	status = amba_request_regions(adev, NULL);
 	if (status)
 		goto err_no_ioregion;
 
+	pl022->phybase = adev->res.start;
 	pl022->virtbase = ioremap(adev->res.start, resource_size(&adev->res));
 	if (pl022->virtbase == NULL) {
 		status = -ENOMEM;
@@ -1798,6 +2106,14 @@
 		dev_err(&adev->dev, "probe - cannot get IRQ (%d)\n", status);
 		goto err_no_irq;
 	}
+
+	/* Get DMA channels */
+	if (platform_info->enable_dma) {
+		status = pl022_dma_probe(pl022);
+		if (status != 0)
+			goto err_no_dma;
+	}
+
 	/* Initialize and start queue */
 	status = init_queue(pl022);
 	if (status != 0) {
@@ -1826,6 +2142,8 @@
  err_start_queue:
  err_init_queue:
 	destroy_queue(pl022);
+	pl022_dma_remove(pl022);
+ err_no_dma:
 	free_irq(adev->irq[0], pl022);
  err_no_irq:
 	clk_put(pl022->clk);
@@ -1856,6 +2174,7 @@
 		return status;
 	}
 	load_ssp_default_config(pl022);
+	pl022_dma_remove(pl022);
 	free_irq(adev->irq[0], pl022);
 	clk_disable(pl022->clk);
 	clk_put(pl022->clk);
diff --git a/drivers/spi/atmel_spi.c b/drivers/spi/atmel_spi.c
index c4e0442..154529a 100644
--- a/drivers/spi/atmel_spi.c
+++ b/drivers/spi/atmel_spi.c
@@ -654,6 +654,8 @@
 	struct spi_transfer	*xfer;
 	unsigned long		flags;
 	struct device		*controller = spi->master->dev.parent;
+	u8			bits;
+	struct atmel_spi_device	*asd;
 
 	as = spi_master_get_devdata(spi->master);
 
@@ -672,8 +674,18 @@
 			return -EINVAL;
 		}
 
+		if (xfer->bits_per_word) {
+			asd = spi->controller_state;
+			bits = (asd->csr >> 4) & 0xf;
+			if (bits != xfer->bits_per_word - 8) {
+				dev_dbg(&spi->dev, "you can't yet change "
+					 "bits_per_word in transfers\n");
+				return -ENOPROTOOPT;
+			}
+		}
+
 		/* FIXME implement these protocol options!! */
-		if (xfer->bits_per_word || xfer->speed_hz) {
+		if (xfer->speed_hz) {
 			dev_dbg(&spi->dev, "no protocol options yet\n");
 			return -ENOPROTOOPT;
 		}
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index b3a94ca..2a651e6 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -296,6 +296,19 @@
 	return 0;
 }
 
+static int mcspi_wait_for_reg_bit(void __iomem *reg, unsigned long bit)
+{
+	unsigned long timeout;
+
+	timeout = jiffies + msecs_to_jiffies(1000);
+	while (!(__raw_readl(reg) & bit)) {
+		if (time_after(jiffies, timeout))
+			return -1;
+		cpu_relax();
+	}
+	return 0;
+}
+
 static unsigned
 omap2_mcspi_txrx_dma(struct spi_device *spi, struct spi_transfer *xfer)
 {
@@ -309,11 +322,14 @@
 	u32			l;
 	u8			* rx;
 	const u8		* tx;
+	void __iomem		*chstat_reg;
 
 	mcspi = spi_master_get_devdata(spi->master);
 	mcspi_dma = &mcspi->dma_channels[spi->chip_select];
 	l = mcspi_cached_chconf0(spi);
 
+	chstat_reg = cs->base + OMAP2_MCSPI_CHSTAT0;
+
 	count = xfer->len;
 	c = count;
 	word_len = cs->word_len;
@@ -382,6 +398,16 @@
 	if (tx != NULL) {
 		wait_for_completion(&mcspi_dma->dma_tx_completion);
 		dma_unmap_single(NULL, xfer->tx_dma, count, DMA_TO_DEVICE);
+
+		/* for TX_ONLY mode, be sure all words have shifted out */
+		if (rx == NULL) {
+			if (mcspi_wait_for_reg_bit(chstat_reg,
+						OMAP2_MCSPI_CHSTAT_TXS) < 0)
+				dev_err(&spi->dev, "TXS timed out\n");
+			else if (mcspi_wait_for_reg_bit(chstat_reg,
+						OMAP2_MCSPI_CHSTAT_EOT) < 0)
+				dev_err(&spi->dev, "EOT timed out\n");
+		}
 	}
 
 	if (rx != NULL) {
@@ -435,19 +461,6 @@
 	return count;
 }
 
-static int mcspi_wait_for_reg_bit(void __iomem *reg, unsigned long bit)
-{
-	unsigned long timeout;
-
-	timeout = jiffies + msecs_to_jiffies(1000);
-	while (!(__raw_readl(reg) & bit)) {
-		if (time_after(jiffies, timeout))
-			return -1;
-		cpu_relax();
-	}
-	return 0;
-}
-
 static unsigned
 omap2_mcspi_txrx_pio(struct spi_device *spi, struct spi_transfer *xfer)
 {
@@ -489,10 +502,8 @@
 					dev_err(&spi->dev, "TXS timed out\n");
 					goto out;
 				}
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "write-%d %02x\n",
+				dev_vdbg(&spi->dev, "write-%d %02x\n",
 						word_len, *tx);
-#endif
 				__raw_writel(*tx++, tx_reg);
 			}
 			if (rx != NULL) {
@@ -506,10 +517,8 @@
 				    (l & OMAP2_MCSPI_CHCONF_TURBO)) {
 					omap2_mcspi_set_enable(spi, 0);
 					*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-					dev_dbg(&spi->dev, "read-%d %02x\n",
+					dev_vdbg(&spi->dev, "read-%d %02x\n",
 						    word_len, *(rx - 1));
-#endif
 					if (mcspi_wait_for_reg_bit(chstat_reg,
 						OMAP2_MCSPI_CHSTAT_RXS) < 0) {
 						dev_err(&spi->dev,
@@ -522,10 +531,8 @@
 				}
 
 				*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "read-%d %02x\n",
+				dev_vdbg(&spi->dev, "read-%d %02x\n",
 						word_len, *(rx - 1));
-#endif
 			}
 		} while (c);
 	} else if (word_len <= 16) {
@@ -542,10 +549,8 @@
 					dev_err(&spi->dev, "TXS timed out\n");
 					goto out;
 				}
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "write-%d %04x\n",
+				dev_vdbg(&spi->dev, "write-%d %04x\n",
 						word_len, *tx);
-#endif
 				__raw_writel(*tx++, tx_reg);
 			}
 			if (rx != NULL) {
@@ -559,10 +564,8 @@
 				    (l & OMAP2_MCSPI_CHCONF_TURBO)) {
 					omap2_mcspi_set_enable(spi, 0);
 					*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-					dev_dbg(&spi->dev, "read-%d %04x\n",
+					dev_vdbg(&spi->dev, "read-%d %04x\n",
 						    word_len, *(rx - 1));
-#endif
 					if (mcspi_wait_for_reg_bit(chstat_reg,
 						OMAP2_MCSPI_CHSTAT_RXS) < 0) {
 						dev_err(&spi->dev,
@@ -575,10 +578,8 @@
 				}
 
 				*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "read-%d %04x\n",
+				dev_vdbg(&spi->dev, "read-%d %04x\n",
 						word_len, *(rx - 1));
-#endif
 			}
 		} while (c);
 	} else if (word_len <= 32) {
@@ -595,10 +596,8 @@
 					dev_err(&spi->dev, "TXS timed out\n");
 					goto out;
 				}
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "write-%d %08x\n",
+				dev_vdbg(&spi->dev, "write-%d %08x\n",
 						word_len, *tx);
-#endif
 				__raw_writel(*tx++, tx_reg);
 			}
 			if (rx != NULL) {
@@ -612,10 +611,8 @@
 				    (l & OMAP2_MCSPI_CHCONF_TURBO)) {
 					omap2_mcspi_set_enable(spi, 0);
 					*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-					dev_dbg(&spi->dev, "read-%d %08x\n",
+					dev_vdbg(&spi->dev, "read-%d %08x\n",
 						    word_len, *(rx - 1));
-#endif
 					if (mcspi_wait_for_reg_bit(chstat_reg,
 						OMAP2_MCSPI_CHSTAT_RXS) < 0) {
 						dev_err(&spi->dev,
@@ -628,10 +625,8 @@
 				}
 
 				*rx++ = __raw_readl(rx_reg);
-#ifdef VERBOSE
-				dev_dbg(&spi->dev, "read-%d %08x\n",
+				dev_vdbg(&spi->dev, "read-%d %08x\n",
 						word_len, *(rx - 1));
-#endif
 			}
 		} while (c);
 	}
@@ -644,6 +639,12 @@
 		} else if (mcspi_wait_for_reg_bit(chstat_reg,
 				OMAP2_MCSPI_CHSTAT_EOT) < 0)
 			dev_err(&spi->dev, "EOT timed out\n");
+
+		/* disable chan to purge rx datas received in TX_ONLY transfer,
+		 * otherwise these rx datas will affect the direct following
+		 * RX_ONLY transfer.
+		 */
+		omap2_mcspi_set_enable(spi, 0);
 	}
 out:
 	omap2_mcspi_set_enable(spi, 1);
diff --git a/drivers/spi/orion_spi.c b/drivers/spi/orion_spi.c
index 3aea50d..0b677dc 100644
--- a/drivers/spi/orion_spi.c
+++ b/drivers/spi/orion_spi.c
@@ -404,7 +404,7 @@
 			goto msg_rejected;
 		}
 
-		if ((t != NULL) && t->bits_per_word)
+		if (t->bits_per_word)
 			bits_per_word = t->bits_per_word;
 
 		if ((bits_per_word != 8) && (bits_per_word != 16)) {
@@ -415,7 +415,7 @@
 			goto msg_rejected;
 		}
 		/*make sure buffer length is even when working in 16 bit mode*/
-		if ((t != NULL) && (t->bits_per_word == 16) && (t->len & 1)) {
+		if ((t->bits_per_word == 16) && (t->len & 1)) {
 			dev_err(&spi->dev,
 				"message rejected : "
 				"odd data length (%d) while in 16 bit mode\n",
diff --git a/drivers/spi/spi_bfin5xx.c b/drivers/spi/spi_bfin5xx.c
index 10a6dc3..ab483a0 100644
--- a/drivers/spi/spi_bfin5xx.c
+++ b/drivers/spi/spi_bfin5xx.c
@@ -1,7 +1,7 @@
 /*
  * Blackfin On-Chip SPI Driver
  *
- * Copyright 2004-2007 Analog Devices Inc.
+ * Copyright 2004-2010 Analog Devices Inc.
  *
  * Enter bugs at http://blackfin.uclinux.org/
  *
@@ -41,13 +41,16 @@
 #define RUNNING_STATE	((void *)1)
 #define DONE_STATE	((void *)2)
 #define ERROR_STATE	((void *)-1)
-#define QUEUE_RUNNING	0
-#define QUEUE_STOPPED	1
 
-/* Value to send if no TX value is supplied */
-#define SPI_IDLE_TXVAL 0x0000
+struct bfin_spi_master_data;
 
-struct driver_data {
+struct bfin_spi_transfer_ops {
+	void (*write) (struct bfin_spi_master_data *);
+	void (*read) (struct bfin_spi_master_data *);
+	void (*duplex) (struct bfin_spi_master_data *);
+};
+
+struct bfin_spi_master_data {
 	/* Driver model hookup */
 	struct platform_device *pdev;
 
@@ -69,7 +72,7 @@
 	spinlock_t lock;
 	struct list_head queue;
 	int busy;
-	int run;
+	bool running;
 
 	/* Message Transfer pump */
 	struct tasklet_struct pump_transfers;
@@ -77,7 +80,7 @@
 	/* Current message transfer state info */
 	struct spi_message *cur_msg;
 	struct spi_transfer *cur_transfer;
-	struct chip_data *cur_chip;
+	struct bfin_spi_slave_data *cur_chip;
 	size_t len_in_bytes;
 	size_t len;
 	void *tx;
@@ -92,38 +95,37 @@
 	dma_addr_t rx_dma;
 	dma_addr_t tx_dma;
 
+	int irq_requested;
+	int spi_irq;
+
 	size_t rx_map_len;
 	size_t tx_map_len;
 	u8 n_bytes;
+	u16 ctrl_reg;
+	u16 flag_reg;
+
 	int cs_change;
-	void (*write) (struct driver_data *);
-	void (*read) (struct driver_data *);
-	void (*duplex) (struct driver_data *);
+	const struct bfin_spi_transfer_ops *ops;
 };
 
-struct chip_data {
+struct bfin_spi_slave_data {
 	u16 ctl_reg;
 	u16 baud;
 	u16 flag;
 
 	u8 chip_select_num;
-	u8 n_bytes;
-	u8 width;		/* 0 or 1 */
 	u8 enable_dma;
-	u8 bits_per_word;	/* 8 or 16 */
-	u8 cs_change_per_word;
 	u16 cs_chg_udelay;	/* Some devices require > 255usec delay */
 	u32 cs_gpio;
 	u16 idle_tx_val;
-	void (*write) (struct driver_data *);
-	void (*read) (struct driver_data *);
-	void (*duplex) (struct driver_data *);
+	u8 pio_interrupt;	/* use spi data irq */
+	const struct bfin_spi_transfer_ops *ops;
 };
 
 #define DEFINE_SPI_REG(reg, off) \
-static inline u16 read_##reg(struct driver_data *drv_data) \
+static inline u16 read_##reg(struct bfin_spi_master_data *drv_data) \
 	{ return bfin_read16(drv_data->regs_base + off); } \
-static inline void write_##reg(struct driver_data *drv_data, u16 v) \
+static inline void write_##reg(struct bfin_spi_master_data *drv_data, u16 v) \
 	{ bfin_write16(drv_data->regs_base + off, v); }
 
 DEFINE_SPI_REG(CTRL, 0x00)
@@ -134,7 +136,7 @@
 DEFINE_SPI_REG(BAUD, 0x14)
 DEFINE_SPI_REG(SHAW, 0x18)
 
-static void bfin_spi_enable(struct driver_data *drv_data)
+static void bfin_spi_enable(struct bfin_spi_master_data *drv_data)
 {
 	u16 cr;
 
@@ -142,7 +144,7 @@
 	write_CTRL(drv_data, (cr | BIT_CTL_ENABLE));
 }
 
-static void bfin_spi_disable(struct driver_data *drv_data)
+static void bfin_spi_disable(struct bfin_spi_master_data *drv_data)
 {
 	u16 cr;
 
@@ -165,7 +167,7 @@
 	return spi_baud;
 }
 
-static int bfin_spi_flush(struct driver_data *drv_data)
+static int bfin_spi_flush(struct bfin_spi_master_data *drv_data)
 {
 	unsigned long limit = loops_per_jiffy << 1;
 
@@ -179,13 +181,12 @@
 }
 
 /* Chip select operation functions for cs_change flag */
-static void bfin_spi_cs_active(struct driver_data *drv_data, struct chip_data *chip)
+static void bfin_spi_cs_active(struct bfin_spi_master_data *drv_data, struct bfin_spi_slave_data *chip)
 {
-	if (likely(chip->chip_select_num)) {
+	if (likely(chip->chip_select_num < MAX_CTRL_CS)) {
 		u16 flag = read_FLAG(drv_data);
 
-		flag |= chip->flag;
-		flag &= ~(chip->flag << 8);
+		flag &= ~chip->flag;
 
 		write_FLAG(drv_data, flag);
 	} else {
@@ -193,13 +194,13 @@
 	}
 }
 
-static void bfin_spi_cs_deactive(struct driver_data *drv_data, struct chip_data *chip)
+static void bfin_spi_cs_deactive(struct bfin_spi_master_data *drv_data,
+                                 struct bfin_spi_slave_data *chip)
 {
-	if (likely(chip->chip_select_num)) {
+	if (likely(chip->chip_select_num < MAX_CTRL_CS)) {
 		u16 flag = read_FLAG(drv_data);
 
-		flag &= ~chip->flag;
-		flag |= (chip->flag << 8);
+		flag |= chip->flag;
 
 		write_FLAG(drv_data, flag);
 	} else {
@@ -211,16 +212,43 @@
 		udelay(chip->cs_chg_udelay);
 }
 
-/* stop controller and re-config current chip*/
-static void bfin_spi_restore_state(struct driver_data *drv_data)
+/* enable or disable the pin muxed by GPIO and SPI CS to work as SPI CS */
+static inline void bfin_spi_cs_enable(struct bfin_spi_master_data *drv_data,
+                                      struct bfin_spi_slave_data *chip)
 {
-	struct chip_data *chip = drv_data->cur_chip;
+	if (chip->chip_select_num < MAX_CTRL_CS) {
+		u16 flag = read_FLAG(drv_data);
+
+		flag |= (chip->flag >> 8);
+
+		write_FLAG(drv_data, flag);
+	}
+}
+
+static inline void bfin_spi_cs_disable(struct bfin_spi_master_data *drv_data,
+                                       struct bfin_spi_slave_data *chip)
+{
+	if (chip->chip_select_num < MAX_CTRL_CS) {
+		u16 flag = read_FLAG(drv_data);
+
+		flag &= ~(chip->flag >> 8);
+
+		write_FLAG(drv_data, flag);
+	}
+}
+
+/* stop controller and re-config current chip*/
+static void bfin_spi_restore_state(struct bfin_spi_master_data *drv_data)
+{
+	struct bfin_spi_slave_data *chip = drv_data->cur_chip;
 
 	/* Clear status and disable clock */
 	write_STAT(drv_data, BIT_STAT_CLR);
 	bfin_spi_disable(drv_data);
 	dev_dbg(&drv_data->pdev->dev, "restoring spi ctl state\n");
 
+	SSYNC();
+
 	/* Load the registers */
 	write_CTRL(drv_data, chip->ctl_reg);
 	write_BAUD(drv_data, chip->baud);
@@ -230,49 +258,12 @@
 }
 
 /* used to kick off transfer in rx mode and read unwanted RX data */
-static inline void bfin_spi_dummy_read(struct driver_data *drv_data)
+static inline void bfin_spi_dummy_read(struct bfin_spi_master_data *drv_data)
 {
 	(void) read_RDBR(drv_data);
 }
 
-static void bfin_spi_null_writer(struct driver_data *drv_data)
-{
-	u8 n_bytes = drv_data->n_bytes;
-	u16 tx_val = drv_data->cur_chip->idle_tx_val;
-
-	/* clear RXS (we check for RXS inside the loop) */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->tx < drv_data->tx_end) {
-		write_TDBR(drv_data, tx_val);
-		drv_data->tx += n_bytes;
-		/* wait until transfer finished.
-		   checking SPIF or TXS may not guarantee transfer completion */
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		/* discard RX data and clear RXS */
-		bfin_spi_dummy_read(drv_data);
-	}
-}
-
-static void bfin_spi_null_reader(struct driver_data *drv_data)
-{
-	u8 n_bytes = drv_data->n_bytes;
-	u16 tx_val = drv_data->cur_chip->idle_tx_val;
-
-	/* discard old RX data and clear RXS */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->rx < drv_data->rx_end) {
-		write_TDBR(drv_data, tx_val);
-		drv_data->rx += n_bytes;
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		bfin_spi_dummy_read(drv_data);
-	}
-}
-
-static void bfin_spi_u8_writer(struct driver_data *drv_data)
+static void bfin_spi_u8_writer(struct bfin_spi_master_data *drv_data)
 {
 	/* clear RXS (we check for RXS inside the loop) */
 	bfin_spi_dummy_read(drv_data);
@@ -288,25 +279,7 @@
 	}
 }
 
-static void bfin_spi_u8_cs_chg_writer(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
-
-	/* clear RXS (we check for RXS inside the loop) */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->tx < drv_data->tx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, (*(u8 *) (drv_data->tx++)));
-		/* make sure transfer finished before deactiving CS */
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		bfin_spi_dummy_read(drv_data);
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-static void bfin_spi_u8_reader(struct driver_data *drv_data)
+static void bfin_spi_u8_reader(struct bfin_spi_master_data *drv_data)
 {
 	u16 tx_val = drv_data->cur_chip->idle_tx_val;
 
@@ -321,25 +294,7 @@
 	}
 }
 
-static void bfin_spi_u8_cs_chg_reader(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
-	u16 tx_val = chip->idle_tx_val;
-
-	/* discard old RX data and clear RXS */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->rx < drv_data->rx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, tx_val);
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		*(u8 *) (drv_data->rx++) = read_RDBR(drv_data);
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-static void bfin_spi_u8_duplex(struct driver_data *drv_data)
+static void bfin_spi_u8_duplex(struct bfin_spi_master_data *drv_data)
 {
 	/* discard old RX data and clear RXS */
 	bfin_spi_dummy_read(drv_data);
@@ -352,24 +307,13 @@
 	}
 }
 
-static void bfin_spi_u8_cs_chg_duplex(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
+static const struct bfin_spi_transfer_ops bfin_bfin_spi_transfer_ops_u8 = {
+	.write  = bfin_spi_u8_writer,
+	.read   = bfin_spi_u8_reader,
+	.duplex = bfin_spi_u8_duplex,
+};
 
-	/* discard old RX data and clear RXS */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->rx < drv_data->rx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, (*(u8 *) (drv_data->tx++)));
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		*(u8 *) (drv_data->rx++) = read_RDBR(drv_data);
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-static void bfin_spi_u16_writer(struct driver_data *drv_data)
+static void bfin_spi_u16_writer(struct bfin_spi_master_data *drv_data)
 {
 	/* clear RXS (we check for RXS inside the loop) */
 	bfin_spi_dummy_read(drv_data);
@@ -386,26 +330,7 @@
 	}
 }
 
-static void bfin_spi_u16_cs_chg_writer(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
-
-	/* clear RXS (we check for RXS inside the loop) */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->tx < drv_data->tx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, (*(u16 *) (drv_data->tx)));
-		drv_data->tx += 2;
-		/* make sure transfer finished before deactiving CS */
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		bfin_spi_dummy_read(drv_data);
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-static void bfin_spi_u16_reader(struct driver_data *drv_data)
+static void bfin_spi_u16_reader(struct bfin_spi_master_data *drv_data)
 {
 	u16 tx_val = drv_data->cur_chip->idle_tx_val;
 
@@ -421,26 +346,7 @@
 	}
 }
 
-static void bfin_spi_u16_cs_chg_reader(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
-	u16 tx_val = chip->idle_tx_val;
-
-	/* discard old RX data and clear RXS */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->rx < drv_data->rx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, tx_val);
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		*(u16 *) (drv_data->rx) = read_RDBR(drv_data);
-		drv_data->rx += 2;
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-static void bfin_spi_u16_duplex(struct driver_data *drv_data)
+static void bfin_spi_u16_duplex(struct bfin_spi_master_data *drv_data)
 {
 	/* discard old RX data and clear RXS */
 	bfin_spi_dummy_read(drv_data);
@@ -455,27 +361,14 @@
 	}
 }
 
-static void bfin_spi_u16_cs_chg_duplex(struct driver_data *drv_data)
-{
-	struct chip_data *chip = drv_data->cur_chip;
+static const struct bfin_spi_transfer_ops bfin_bfin_spi_transfer_ops_u16 = {
+	.write  = bfin_spi_u16_writer,
+	.read   = bfin_spi_u16_reader,
+	.duplex = bfin_spi_u16_duplex,
+};
 
-	/* discard old RX data and clear RXS */
-	bfin_spi_dummy_read(drv_data);
-
-	while (drv_data->rx < drv_data->rx_end) {
-		bfin_spi_cs_active(drv_data, chip);
-		write_TDBR(drv_data, (*(u16 *) (drv_data->tx)));
-		drv_data->tx += 2;
-		while (!(read_STAT(drv_data) & BIT_STAT_RXS))
-			cpu_relax();
-		*(u16 *) (drv_data->rx) = read_RDBR(drv_data);
-		drv_data->rx += 2;
-		bfin_spi_cs_deactive(drv_data, chip);
-	}
-}
-
-/* test if ther is more transfer to be done */
-static void *bfin_spi_next_transfer(struct driver_data *drv_data)
+/* test if there is more transfer to be done */
+static void *bfin_spi_next_transfer(struct bfin_spi_master_data *drv_data)
 {
 	struct spi_message *msg = drv_data->cur_msg;
 	struct spi_transfer *trans = drv_data->cur_transfer;
@@ -494,9 +387,9 @@
  * caller already set message->status;
  * dma and pio irqs are blocked give finished message back
  */
-static void bfin_spi_giveback(struct driver_data *drv_data)
+static void bfin_spi_giveback(struct bfin_spi_master_data *drv_data)
 {
-	struct chip_data *chip = drv_data->cur_chip;
+	struct bfin_spi_slave_data *chip = drv_data->cur_chip;
 	struct spi_transfer *last_transfer;
 	unsigned long flags;
 	struct spi_message *msg;
@@ -525,10 +418,83 @@
 		msg->complete(msg->context);
 }
 
+/* spi data irq handler */
+static irqreturn_t bfin_spi_pio_irq_handler(int irq, void *dev_id)
+{
+	struct bfin_spi_master_data *drv_data = dev_id;
+	struct bfin_spi_slave_data *chip = drv_data->cur_chip;
+	struct spi_message *msg = drv_data->cur_msg;
+	int n_bytes = drv_data->n_bytes;
+
+	/* wait until transfer finished. */
+	while (!(read_STAT(drv_data) & BIT_STAT_RXS))
+		cpu_relax();
+
+	if ((drv_data->tx && drv_data->tx >= drv_data->tx_end) ||
+		(drv_data->rx && drv_data->rx >= (drv_data->rx_end - n_bytes))) {
+		/* last read */
+		if (drv_data->rx) {
+			dev_dbg(&drv_data->pdev->dev, "last read\n");
+			if (n_bytes == 2)
+				*(u16 *) (drv_data->rx) = read_RDBR(drv_data);
+			else if (n_bytes == 1)
+				*(u8 *) (drv_data->rx) = read_RDBR(drv_data);
+			drv_data->rx += n_bytes;
+		}
+
+		msg->actual_length += drv_data->len_in_bytes;
+		if (drv_data->cs_change)
+			bfin_spi_cs_deactive(drv_data, chip);
+		/* Move to next transfer */
+		msg->state = bfin_spi_next_transfer(drv_data);
+
+		disable_irq_nosync(drv_data->spi_irq);
+
+		/* Schedule transfer tasklet */
+		tasklet_schedule(&drv_data->pump_transfers);
+		return IRQ_HANDLED;
+	}
+
+	if (drv_data->rx && drv_data->tx) {
+		/* duplex */
+		dev_dbg(&drv_data->pdev->dev, "duplex: write_TDBR\n");
+		if (drv_data->n_bytes == 2) {
+			*(u16 *) (drv_data->rx) = read_RDBR(drv_data);
+			write_TDBR(drv_data, (*(u16 *) (drv_data->tx)));
+		} else if (drv_data->n_bytes == 1) {
+			*(u8 *) (drv_data->rx) = read_RDBR(drv_data);
+			write_TDBR(drv_data, (*(u8 *) (drv_data->tx)));
+		}
+	} else if (drv_data->rx) {
+		/* read */
+		dev_dbg(&drv_data->pdev->dev, "read: write_TDBR\n");
+		if (drv_data->n_bytes == 2)
+			*(u16 *) (drv_data->rx) = read_RDBR(drv_data);
+		else if (drv_data->n_bytes == 1)
+			*(u8 *) (drv_data->rx) = read_RDBR(drv_data);
+		write_TDBR(drv_data, chip->idle_tx_val);
+	} else if (drv_data->tx) {
+		/* write */
+		dev_dbg(&drv_data->pdev->dev, "write: write_TDBR\n");
+		bfin_spi_dummy_read(drv_data);
+		if (drv_data->n_bytes == 2)
+			write_TDBR(drv_data, (*(u16 *) (drv_data->tx)));
+		else if (drv_data->n_bytes == 1)
+			write_TDBR(drv_data, (*(u8 *) (drv_data->tx)));
+	}
+
+	if (drv_data->tx)
+		drv_data->tx += n_bytes;
+	if (drv_data->rx)
+		drv_data->rx += n_bytes;
+
+	return IRQ_HANDLED;
+}
+
 static irqreturn_t bfin_spi_dma_irq_handler(int irq, void *dev_id)
 {
-	struct driver_data *drv_data = dev_id;
-	struct chip_data *chip = drv_data->cur_chip;
+	struct bfin_spi_master_data *drv_data = dev_id;
+	struct bfin_spi_slave_data *chip = drv_data->cur_chip;
 	struct spi_message *msg = drv_data->cur_msg;
 	unsigned long timeout;
 	unsigned short dmastat = get_dma_curr_irqstat(drv_data->dma_channel);
@@ -540,10 +506,6 @@
 
 	clear_dma_irqstat(drv_data->dma_channel);
 
-	/* Wait for DMA to complete */
-	while (get_dma_curr_irqstat(drv_data->dma_channel) & DMA_RUN)
-		cpu_relax();
-
 	/*
 	 * wait for the last transaction shifted out.  HRM states:
 	 * at this point there may still be data in the SPI DMA FIFO waiting
@@ -551,8 +513,8 @@
 	 * register until it goes low for 2 successive reads
 	 */
 	if (drv_data->tx != NULL) {
-		while ((read_STAT(drv_data) & TXS) ||
-		       (read_STAT(drv_data) & TXS))
+		while ((read_STAT(drv_data) & BIT_STAT_TXS) ||
+		       (read_STAT(drv_data) & BIT_STAT_TXS))
 			cpu_relax();
 	}
 
@@ -561,14 +523,14 @@
 		dmastat, read_STAT(drv_data));
 
 	timeout = jiffies + HZ;
-	while (!(read_STAT(drv_data) & SPIF))
+	while (!(read_STAT(drv_data) & BIT_STAT_SPIF))
 		if (!time_before(jiffies, timeout)) {
 			dev_warn(&drv_data->pdev->dev, "timeout waiting for SPIF");
 			break;
 		} else
 			cpu_relax();
 
-	if ((dmastat & DMA_ERR) && (spistat & RBSY)) {
+	if ((dmastat & DMA_ERR) && (spistat & BIT_STAT_RBSY)) {
 		msg->state = ERROR_STATE;
 		dev_err(&drv_data->pdev->dev, "dma receive: fifo/buffer overflow\n");
 	} else {
@@ -588,20 +550,20 @@
 	dev_dbg(&drv_data->pdev->dev,
 		"disable dma channel irq%d\n",
 		drv_data->dma_channel);
-	dma_disable_irq(drv_data->dma_channel);
+	dma_disable_irq_nosync(drv_data->dma_channel);
 
 	return IRQ_HANDLED;
 }
 
 static void bfin_spi_pump_transfers(unsigned long data)
 {
-	struct driver_data *drv_data = (struct driver_data *)data;
+	struct bfin_spi_master_data *drv_data = (struct bfin_spi_master_data *)data;
 	struct spi_message *message = NULL;
 	struct spi_transfer *transfer = NULL;
 	struct spi_transfer *previous = NULL;
-	struct chip_data *chip = NULL;
-	u8 width;
-	u16 cr, dma_width, dma_config;
+	struct bfin_spi_slave_data *chip = NULL;
+	unsigned int bits_per_word;
+	u16 cr, cr_width, dma_width, dma_config;
 	u32 tranf_success = 1;
 	u8 full_duplex = 0;
 
@@ -639,7 +601,7 @@
 			udelay(previous->delay_usecs);
 	}
 
-	/* Setup the transfer state based on the type of transfer */
+	/* Flush any existing transfers that may be sitting in the hardware */
 	if (bfin_spi_flush(drv_data) == 0) {
 		dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
 		message->status = -EIO;
@@ -679,52 +641,31 @@
 	drv_data->cs_change = transfer->cs_change;
 
 	/* Bits per word setup */
-	switch (transfer->bits_per_word) {
-	case 8:
+	bits_per_word = transfer->bits_per_word ? : message->spi->bits_per_word;
+	if (bits_per_word == 8) {
 		drv_data->n_bytes = 1;
-		width = CFG_SPI_WORDSIZE8;
-		drv_data->read = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_reader : bfin_spi_u8_reader;
-		drv_data->write = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_writer : bfin_spi_u8_writer;
-		drv_data->duplex = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_duplex : bfin_spi_u8_duplex;
-		break;
-
-	case 16:
+		drv_data->len = transfer->len;
+		cr_width = 0;
+		drv_data->ops = &bfin_bfin_spi_transfer_ops_u8;
+	} else if (bits_per_word == 16) {
 		drv_data->n_bytes = 2;
-		width = CFG_SPI_WORDSIZE16;
-		drv_data->read = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_reader : bfin_spi_u16_reader;
-		drv_data->write = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_writer : bfin_spi_u16_writer;
-		drv_data->duplex = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_duplex : bfin_spi_u16_duplex;
-		break;
-
-	default:
-		/* No change, the same as default setting */
-		drv_data->n_bytes = chip->n_bytes;
-		width = chip->width;
-		drv_data->write = drv_data->tx ? chip->write : bfin_spi_null_writer;
-		drv_data->read = drv_data->rx ? chip->read : bfin_spi_null_reader;
-		drv_data->duplex = chip->duplex ? chip->duplex : bfin_spi_null_writer;
-		break;
+		drv_data->len = (transfer->len) >> 1;
+		cr_width = BIT_CTL_WORDSIZE;
+		drv_data->ops = &bfin_bfin_spi_transfer_ops_u16;
+	} else {
+		dev_err(&drv_data->pdev->dev, "transfer: unsupported bits_per_word\n");
+		message->status = -EINVAL;
+		bfin_spi_giveback(drv_data);
+		return;
 	}
-	cr = (read_CTRL(drv_data) & (~BIT_CTL_TIMOD));
-	cr |= (width << 8);
+	cr = read_CTRL(drv_data) & ~(BIT_CTL_TIMOD | BIT_CTL_WORDSIZE);
+	cr |= cr_width;
 	write_CTRL(drv_data, cr);
 
-	if (width == CFG_SPI_WORDSIZE16) {
-		drv_data->len = (transfer->len) >> 1;
-	} else {
-		drv_data->len = transfer->len;
-	}
 	dev_dbg(&drv_data->pdev->dev,
-		"transfer: drv_data->write is %p, chip->write is %p, null_wr is %p\n",
-		drv_data->write, chip->write, bfin_spi_null_writer);
+		"transfer: drv_data->ops is %p, chip->ops is %p, u8_ops is %p\n",
+		drv_data->ops, chip->ops, &bfin_bfin_spi_transfer_ops_u8);
 
-	/* speed and width has been set on per message */
 	message->state = RUNNING_STATE;
 	dma_config = 0;
 
@@ -735,13 +676,11 @@
 		write_BAUD(drv_data, chip->baud);
 
 	write_STAT(drv_data, BIT_STAT_CLR);
-	cr = (read_CTRL(drv_data) & (~BIT_CTL_TIMOD));
-	if (drv_data->cs_change)
-		bfin_spi_cs_active(drv_data, chip);
+	bfin_spi_cs_active(drv_data, chip);
 
 	dev_dbg(&drv_data->pdev->dev,
 		"now pumping a transfer: width is %d, len is %d\n",
-		width, transfer->len);
+		cr_width, transfer->len);
 
 	/*
 	 * Try to map dma buffer and do a dma transfer.  If successful use,
@@ -760,7 +699,7 @@
 		/* config dma channel */
 		dev_dbg(&drv_data->pdev->dev, "doing dma transfer\n");
 		set_dma_x_count(drv_data->dma_channel, drv_data->len);
-		if (width == CFG_SPI_WORDSIZE16) {
+		if (cr_width == BIT_CTL_WORDSIZE) {
 			set_dma_x_modify(drv_data->dma_channel, 2);
 			dma_width = WDSIZE_16;
 		} else {
@@ -846,73 +785,100 @@
 		dma_enable_irq(drv_data->dma_channel);
 		local_irq_restore(flags);
 
-	} else {
-		/* IO mode write then read */
-		dev_dbg(&drv_data->pdev->dev, "doing IO transfer\n");
-
-		/* we always use SPI_WRITE mode. SPI_READ mode
-		   seems to have problems with setting up the
-		   output value in TDBR prior to the transfer. */
-		write_CTRL(drv_data, (cr | CFG_SPI_WRITE));
-
-		if (full_duplex) {
-			/* full duplex mode */
-			BUG_ON((drv_data->tx_end - drv_data->tx) !=
-			       (drv_data->rx_end - drv_data->rx));
-			dev_dbg(&drv_data->pdev->dev,
-				"IO duplex: cr is 0x%x\n", cr);
-
-			drv_data->duplex(drv_data);
-
-			if (drv_data->tx != drv_data->tx_end)
-				tranf_success = 0;
-		} else if (drv_data->tx != NULL) {
-			/* write only half duplex */
-			dev_dbg(&drv_data->pdev->dev,
-				"IO write: cr is 0x%x\n", cr);
-
-			drv_data->write(drv_data);
-
-			if (drv_data->tx != drv_data->tx_end)
-				tranf_success = 0;
-		} else if (drv_data->rx != NULL) {
-			/* read only half duplex */
-			dev_dbg(&drv_data->pdev->dev,
-				"IO read: cr is 0x%x\n", cr);
-
-			drv_data->read(drv_data);
-			if (drv_data->rx != drv_data->rx_end)
-				tranf_success = 0;
-		}
-
-		if (!tranf_success) {
-			dev_dbg(&drv_data->pdev->dev,
-				"IO write error!\n");
-			message->state = ERROR_STATE;
-		} else {
-			/* Update total byte transfered */
-			message->actual_length += drv_data->len_in_bytes;
-			/* Move to next transfer of this msg */
-			message->state = bfin_spi_next_transfer(drv_data);
-			if (drv_data->cs_change)
-				bfin_spi_cs_deactive(drv_data, chip);
-		}
-		/* Schedule next transfer tasklet */
-		tasklet_schedule(&drv_data->pump_transfers);
+		return;
 	}
+
+	/*
+	 * We always use SPI_WRITE mode (transfer starts with TDBR write).
+	 * SPI_READ mode (transfer starts with RDBR read) seems to have
+	 * problems with setting up the output value in TDBR prior to the
+	 * start of the transfer.
+	 */
+	write_CTRL(drv_data, cr | BIT_CTL_TXMOD);
+
+	if (chip->pio_interrupt) {
+		/* SPI irq should have been disabled by now */
+
+		/* discard old RX data and clear RXS */
+		bfin_spi_dummy_read(drv_data);
+
+		/* start transfer */
+		if (drv_data->tx == NULL)
+			write_TDBR(drv_data, chip->idle_tx_val);
+		else {
+			if (bits_per_word == 8)
+				write_TDBR(drv_data, (*(u8 *) (drv_data->tx)));
+			else
+				write_TDBR(drv_data, (*(u16 *) (drv_data->tx)));
+			drv_data->tx += drv_data->n_bytes;
+		}
+
+		/* once TDBR is empty, interrupt is triggered */
+		enable_irq(drv_data->spi_irq);
+		return;
+	}
+
+	/* IO mode */
+	dev_dbg(&drv_data->pdev->dev, "doing IO transfer\n");
+
+	if (full_duplex) {
+		/* full duplex mode */
+		BUG_ON((drv_data->tx_end - drv_data->tx) !=
+		       (drv_data->rx_end - drv_data->rx));
+		dev_dbg(&drv_data->pdev->dev,
+			"IO duplex: cr is 0x%x\n", cr);
+
+		drv_data->ops->duplex(drv_data);
+
+		if (drv_data->tx != drv_data->tx_end)
+			tranf_success = 0;
+	} else if (drv_data->tx != NULL) {
+		/* write only half duplex */
+		dev_dbg(&drv_data->pdev->dev,
+			"IO write: cr is 0x%x\n", cr);
+
+		drv_data->ops->write(drv_data);
+
+		if (drv_data->tx != drv_data->tx_end)
+			tranf_success = 0;
+	} else if (drv_data->rx != NULL) {
+		/* read only half duplex */
+		dev_dbg(&drv_data->pdev->dev,
+			"IO read: cr is 0x%x\n", cr);
+
+		drv_data->ops->read(drv_data);
+		if (drv_data->rx != drv_data->rx_end)
+			tranf_success = 0;
+	}
+
+	if (!tranf_success) {
+		dev_dbg(&drv_data->pdev->dev,
+			"IO write error!\n");
+		message->state = ERROR_STATE;
+	} else {
+		/* Update total byte transfered */
+		message->actual_length += drv_data->len_in_bytes;
+		/* Move to next transfer of this msg */
+		message->state = bfin_spi_next_transfer(drv_data);
+		if (drv_data->cs_change)
+			bfin_spi_cs_deactive(drv_data, chip);
+	}
+
+	/* Schedule next transfer tasklet */
+	tasklet_schedule(&drv_data->pump_transfers);
 }
 
 /* pop a msg from queue and kick off real transfer */
 static void bfin_spi_pump_messages(struct work_struct *work)
 {
-	struct driver_data *drv_data;
+	struct bfin_spi_master_data *drv_data;
 	unsigned long flags;
 
-	drv_data = container_of(work, struct driver_data, pump_messages);
+	drv_data = container_of(work, struct bfin_spi_master_data, pump_messages);
 
 	/* Lock queue and check for queue work */
 	spin_lock_irqsave(&drv_data->lock, flags);
-	if (list_empty(&drv_data->queue) || drv_data->run == QUEUE_STOPPED) {
+	if (list_empty(&drv_data->queue) || !drv_data->running) {
 		/* pumper kicked off but no work to do */
 		drv_data->busy = 0;
 		spin_unlock_irqrestore(&drv_data->lock, flags);
@@ -962,12 +928,12 @@
  */
 static int bfin_spi_transfer(struct spi_device *spi, struct spi_message *msg)
 {
-	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
+	struct bfin_spi_master_data *drv_data = spi_master_get_devdata(spi->master);
 	unsigned long flags;
 
 	spin_lock_irqsave(&drv_data->lock, flags);
 
-	if (drv_data->run == QUEUE_STOPPED) {
+	if (!drv_data->running) {
 		spin_unlock_irqrestore(&drv_data->lock, flags);
 		return -ESHUTDOWN;
 	}
@@ -979,7 +945,7 @@
 	dev_dbg(&spi->dev, "adding an msg in transfer() \n");
 	list_add_tail(&msg->queue, &drv_data->queue);
 
-	if (drv_data->run == QUEUE_RUNNING && !drv_data->busy)
+	if (drv_data->running && !drv_data->busy)
 		queue_work(drv_data->workqueue, &drv_data->pump_messages);
 
 	spin_unlock_irqrestore(&drv_data->lock, flags);
@@ -1003,147 +969,184 @@
 	P_SPI2_SSEL6, P_SPI2_SSEL7},
 };
 
-/* first setup for new devices */
+/* setup for devices (may be called multiple times -- not just first setup) */
 static int bfin_spi_setup(struct spi_device *spi)
 {
-	struct bfin5xx_spi_chip *chip_info = NULL;
-	struct chip_data *chip;
-	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
-	int ret;
-
-	if (spi->bits_per_word != 8 && spi->bits_per_word != 16)
-		return -EINVAL;
+	struct bfin5xx_spi_chip *chip_info;
+	struct bfin_spi_slave_data *chip = NULL;
+	struct bfin_spi_master_data *drv_data = spi_master_get_devdata(spi->master);
+	u16 bfin_ctl_reg;
+	int ret = -EINVAL;
 
 	/* Only alloc (or use chip_info) on first setup */
+	chip_info = NULL;
 	chip = spi_get_ctldata(spi);
 	if (chip == NULL) {
-		chip = kzalloc(sizeof(struct chip_data), GFP_KERNEL);
-		if (!chip)
-			return -ENOMEM;
+		chip = kzalloc(sizeof(*chip), GFP_KERNEL);
+		if (!chip) {
+			dev_err(&spi->dev, "cannot allocate chip data\n");
+			ret = -ENOMEM;
+			goto error;
+		}
 
 		chip->enable_dma = 0;
 		chip_info = spi->controller_data;
 	}
 
+	/* Let people set non-standard bits directly */
+	bfin_ctl_reg = BIT_CTL_OPENDRAIN | BIT_CTL_EMISO |
+		BIT_CTL_PSSE | BIT_CTL_GM | BIT_CTL_SZ;
+
 	/* chip_info isn't always needed */
 	if (chip_info) {
 		/* Make sure people stop trying to set fields via ctl_reg
 		 * when they should actually be using common SPI framework.
-		 * Currently we let through: WOM EMISO PSSE GM SZ TIMOD.
+		 * Currently we let through: WOM EMISO PSSE GM SZ.
 		 * Not sure if a user actually needs/uses any of these,
 		 * but let's assume (for now) they do.
 		 */
-		if (chip_info->ctl_reg & (SPE|MSTR|CPOL|CPHA|LSBF|SIZE)) {
+		if (chip_info->ctl_reg & ~bfin_ctl_reg) {
 			dev_err(&spi->dev, "do not set bits in ctl_reg "
 				"that the SPI framework manages\n");
-			return -EINVAL;
+			goto error;
 		}
-
 		chip->enable_dma = chip_info->enable_dma != 0
 		    && drv_data->master_info->enable_dma;
 		chip->ctl_reg = chip_info->ctl_reg;
-		chip->bits_per_word = chip_info->bits_per_word;
-		chip->cs_change_per_word = chip_info->cs_change_per_word;
 		chip->cs_chg_udelay = chip_info->cs_chg_udelay;
-		chip->cs_gpio = chip_info->cs_gpio;
 		chip->idle_tx_val = chip_info->idle_tx_val;
+		chip->pio_interrupt = chip_info->pio_interrupt;
+		spi->bits_per_word = chip_info->bits_per_word;
+	} else {
+		/* force a default base state */
+		chip->ctl_reg &= bfin_ctl_reg;
+	}
+
+	if (spi->bits_per_word != 8 && spi->bits_per_word != 16) {
+		dev_err(&spi->dev, "%d bits_per_word is not supported\n",
+				spi->bits_per_word);
+		goto error;
 	}
 
 	/* translate common spi framework into our register */
-	if (spi->mode & SPI_CPOL)
-		chip->ctl_reg |= CPOL;
-	if (spi->mode & SPI_CPHA)
-		chip->ctl_reg |= CPHA;
-	if (spi->mode & SPI_LSB_FIRST)
-		chip->ctl_reg |= LSBF;
-	/* we dont support running in slave mode (yet?) */
-	chip->ctl_reg |= MSTR;
-
-	/*
-	 * if any one SPI chip is registered and wants DMA, request the
-	 * DMA channel for it
-	 */
-	if (chip->enable_dma && !drv_data->dma_requested) {
-		/* register dma irq handler */
-		if (request_dma(drv_data->dma_channel, "BFIN_SPI_DMA") < 0) {
-			dev_dbg(&spi->dev,
-				"Unable to request BlackFin SPI DMA channel\n");
-			return -ENODEV;
-		}
-		if (set_dma_callback(drv_data->dma_channel,
-		    bfin_spi_dma_irq_handler, drv_data) < 0) {
-			dev_dbg(&spi->dev, "Unable to set dma callback\n");
-			return -EPERM;
-		}
-		dma_disable_irq(drv_data->dma_channel);
-		drv_data->dma_requested = 1;
+	if (spi->mode & ~(SPI_CPOL | SPI_CPHA | SPI_LSB_FIRST)) {
+		dev_err(&spi->dev, "unsupported spi modes detected\n");
+		goto error;
 	}
+	if (spi->mode & SPI_CPOL)
+		chip->ctl_reg |= BIT_CTL_CPOL;
+	if (spi->mode & SPI_CPHA)
+		chip->ctl_reg |= BIT_CTL_CPHA;
+	if (spi->mode & SPI_LSB_FIRST)
+		chip->ctl_reg |= BIT_CTL_LSBF;
+	/* we dont support running in slave mode (yet?) */
+	chip->ctl_reg |= BIT_CTL_MASTER;
 
 	/*
 	 * Notice: for blackfin, the speed_hz is the value of register
 	 * SPI_BAUD, not the real baudrate
 	 */
 	chip->baud = hz_to_spi_baud(spi->max_speed_hz);
-	chip->flag = 1 << (spi->chip_select);
 	chip->chip_select_num = spi->chip_select;
+	if (chip->chip_select_num < MAX_CTRL_CS) {
+		if (!(spi->mode & SPI_CPHA))
+			dev_warn(&spi->dev, "Warning: SPI CPHA not set:"
+				" Slave Select not under software control!\n"
+				" See Documentation/blackfin/bfin-spi-notes.txt");
 
-	if (chip->chip_select_num == 0) {
+		chip->flag = (1 << spi->chip_select) << 8;
+	} else
+		chip->cs_gpio = chip->chip_select_num - MAX_CTRL_CS;
+
+	if (chip->enable_dma && chip->pio_interrupt) {
+		dev_err(&spi->dev, "enable_dma is set, "
+				"do not set pio_interrupt\n");
+		goto error;
+	}
+	/*
+	 * if any one SPI chip is registered and wants DMA, request the
+	 * DMA channel for it
+	 */
+	if (chip->enable_dma && !drv_data->dma_requested) {
+		/* register dma irq handler */
+		ret = request_dma(drv_data->dma_channel, "BFIN_SPI_DMA");
+		if (ret) {
+			dev_err(&spi->dev,
+				"Unable to request BlackFin SPI DMA channel\n");
+			goto error;
+		}
+		drv_data->dma_requested = 1;
+
+		ret = set_dma_callback(drv_data->dma_channel,
+			bfin_spi_dma_irq_handler, drv_data);
+		if (ret) {
+			dev_err(&spi->dev, "Unable to set dma callback\n");
+			goto error;
+		}
+		dma_disable_irq(drv_data->dma_channel);
+	}
+
+	if (chip->pio_interrupt && !drv_data->irq_requested) {
+		ret = request_irq(drv_data->spi_irq, bfin_spi_pio_irq_handler,
+			IRQF_DISABLED, "BFIN_SPI", drv_data);
+		if (ret) {
+			dev_err(&spi->dev, "Unable to register spi IRQ\n");
+			goto error;
+		}
+		drv_data->irq_requested = 1;
+		/* we use write mode, spi irq has to be disabled here */
+		disable_irq(drv_data->spi_irq);
+	}
+
+	if (chip->chip_select_num >= MAX_CTRL_CS) {
 		ret = gpio_request(chip->cs_gpio, spi->modalias);
 		if (ret) {
-			if (drv_data->dma_requested)
-				free_dma(drv_data->dma_channel);
-			return ret;
+			dev_err(&spi->dev, "gpio_request() error\n");
+			goto pin_error;
 		}
 		gpio_direction_output(chip->cs_gpio, 1);
 	}
 
-	switch (chip->bits_per_word) {
-	case 8:
-		chip->n_bytes = 1;
-		chip->width = CFG_SPI_WORDSIZE8;
-		chip->read = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_reader : bfin_spi_u8_reader;
-		chip->write = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_writer : bfin_spi_u8_writer;
-		chip->duplex = chip->cs_change_per_word ?
-			bfin_spi_u8_cs_chg_duplex : bfin_spi_u8_duplex;
-		break;
-
-	case 16:
-		chip->n_bytes = 2;
-		chip->width = CFG_SPI_WORDSIZE16;
-		chip->read = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_reader : bfin_spi_u16_reader;
-		chip->write = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_writer : bfin_spi_u16_writer;
-		chip->duplex = chip->cs_change_per_word ?
-			bfin_spi_u16_cs_chg_duplex : bfin_spi_u16_duplex;
-		break;
-
-	default:
-		dev_err(&spi->dev, "%d bits_per_word is not supported\n",
-				chip->bits_per_word);
-		if (chip_info)
-			kfree(chip);
-		return -ENODEV;
-	}
-
 	dev_dbg(&spi->dev, "setup spi chip %s, width is %d, dma is %d\n",
-			spi->modalias, chip->width, chip->enable_dma);
+			spi->modalias, spi->bits_per_word, chip->enable_dma);
 	dev_dbg(&spi->dev, "ctl_reg is 0x%x, flag_reg is 0x%x\n",
 			chip->ctl_reg, chip->flag);
 
 	spi_set_ctldata(spi, chip);
 
 	dev_dbg(&spi->dev, "chip select number is %d\n", chip->chip_select_num);
-	if ((chip->chip_select_num > 0)
-		&& (chip->chip_select_num <= spi->master->num_chipselect))
-		peripheral_request(ssel[spi->master->bus_num]
-			[chip->chip_select_num-1], spi->modalias);
+	if (chip->chip_select_num < MAX_CTRL_CS) {
+		ret = peripheral_request(ssel[spi->master->bus_num]
+		                         [chip->chip_select_num-1], spi->modalias);
+		if (ret) {
+			dev_err(&spi->dev, "peripheral_request() error\n");
+			goto pin_error;
+		}
+	}
 
+	bfin_spi_cs_enable(drv_data, chip);
 	bfin_spi_cs_deactive(drv_data, chip);
 
 	return 0;
+
+ pin_error:
+	if (chip->chip_select_num >= MAX_CTRL_CS)
+		gpio_free(chip->cs_gpio);
+	else
+		peripheral_free(ssel[spi->master->bus_num]
+			[chip->chip_select_num - 1]);
+ error:
+	if (chip) {
+		if (drv_data->dma_requested)
+			free_dma(drv_data->dma_channel);
+		drv_data->dma_requested = 0;
+
+		kfree(chip);
+		/* prevent free 'chip' twice */
+		spi_set_ctldata(spi, NULL);
+	}
+
+	return ret;
 }
 
 /*
@@ -1152,28 +1155,30 @@
  */
 static void bfin_spi_cleanup(struct spi_device *spi)
 {
-	struct chip_data *chip = spi_get_ctldata(spi);
+	struct bfin_spi_slave_data *chip = spi_get_ctldata(spi);
+	struct bfin_spi_master_data *drv_data = spi_master_get_devdata(spi->master);
 
 	if (!chip)
 		return;
 
-	if ((chip->chip_select_num > 0)
-		&& (chip->chip_select_num <= spi->master->num_chipselect))
+	if (chip->chip_select_num < MAX_CTRL_CS) {
 		peripheral_free(ssel[spi->master->bus_num]
 					[chip->chip_select_num-1]);
-
-	if (chip->chip_select_num == 0)
+		bfin_spi_cs_disable(drv_data, chip);
+	} else
 		gpio_free(chip->cs_gpio);
 
 	kfree(chip);
+	/* prevent free 'chip' twice */
+	spi_set_ctldata(spi, NULL);
 }
 
-static inline int bfin_spi_init_queue(struct driver_data *drv_data)
+static inline int bfin_spi_init_queue(struct bfin_spi_master_data *drv_data)
 {
 	INIT_LIST_HEAD(&drv_data->queue);
 	spin_lock_init(&drv_data->lock);
 
-	drv_data->run = QUEUE_STOPPED;
+	drv_data->running = false;
 	drv_data->busy = 0;
 
 	/* init transfer tasklet */
@@ -1190,18 +1195,18 @@
 	return 0;
 }
 
-static inline int bfin_spi_start_queue(struct driver_data *drv_data)
+static inline int bfin_spi_start_queue(struct bfin_spi_master_data *drv_data)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&drv_data->lock, flags);
 
-	if (drv_data->run == QUEUE_RUNNING || drv_data->busy) {
+	if (drv_data->running || drv_data->busy) {
 		spin_unlock_irqrestore(&drv_data->lock, flags);
 		return -EBUSY;
 	}
 
-	drv_data->run = QUEUE_RUNNING;
+	drv_data->running = true;
 	drv_data->cur_msg = NULL;
 	drv_data->cur_transfer = NULL;
 	drv_data->cur_chip = NULL;
@@ -1212,7 +1217,7 @@
 	return 0;
 }
 
-static inline int bfin_spi_stop_queue(struct driver_data *drv_data)
+static inline int bfin_spi_stop_queue(struct bfin_spi_master_data *drv_data)
 {
 	unsigned long flags;
 	unsigned limit = 500;
@@ -1226,7 +1231,7 @@
 	 * execution path (pump_messages) would be required to call wake_up or
 	 * friends on every SPI message. Do this instead
 	 */
-	drv_data->run = QUEUE_STOPPED;
+	drv_data->running = false;
 	while (!list_empty(&drv_data->queue) && drv_data->busy && limit--) {
 		spin_unlock_irqrestore(&drv_data->lock, flags);
 		msleep(10);
@@ -1241,7 +1246,7 @@
 	return status;
 }
 
-static inline int bfin_spi_destroy_queue(struct driver_data *drv_data)
+static inline int bfin_spi_destroy_queue(struct bfin_spi_master_data *drv_data)
 {
 	int status;
 
@@ -1259,14 +1264,14 @@
 	struct device *dev = &pdev->dev;
 	struct bfin5xx_spi_master *platform_info;
 	struct spi_master *master;
-	struct driver_data *drv_data = 0;
+	struct bfin_spi_master_data *drv_data;
 	struct resource *res;
 	int status = 0;
 
 	platform_info = dev->platform_data;
 
 	/* Allocate master with space for drv_data */
-	master = spi_alloc_master(dev, sizeof(struct driver_data) + 16);
+	master = spi_alloc_master(dev, sizeof(*drv_data));
 	if (!master) {
 		dev_err(&pdev->dev, "can not alloc spi_master\n");
 		return -ENOMEM;
@@ -1302,11 +1307,19 @@
 		goto out_error_ioremap;
 	}
 
-	drv_data->dma_channel = platform_get_irq(pdev, 0);
-	if (drv_data->dma_channel < 0) {
+	res = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+	if (res == NULL) {
 		dev_err(dev, "No DMA channel specified\n");
 		status = -ENOENT;
-		goto out_error_no_dma_ch;
+		goto out_error_free_io;
+	}
+	drv_data->dma_channel = res->start;
+
+	drv_data->spi_irq = platform_get_irq(pdev, 0);
+	if (drv_data->spi_irq < 0) {
+		dev_err(dev, "No spi pio irq specified\n");
+		status = -ENOENT;
+		goto out_error_free_io;
 	}
 
 	/* Initial and start queue */
@@ -1328,6 +1341,12 @@
 		goto out_error_queue_alloc;
 	}
 
+	/* Reset SPI registers. If these registers were used by the boot loader,
+	 * the sky may fall on your head if you enable the dma controller.
+	 */
+	write_CTRL(drv_data, BIT_CTL_CPHA | BIT_CTL_MASTER);
+	write_FLAG(drv_data, 0xFF00);
+
 	/* Register with the SPI framework */
 	platform_set_drvdata(pdev, drv_data);
 	status = spi_register_master(master);
@@ -1343,7 +1362,7 @@
 
 out_error_queue_alloc:
 	bfin_spi_destroy_queue(drv_data);
-out_error_no_dma_ch:
+out_error_free_io:
 	iounmap((void *) drv_data->regs_base);
 out_error_ioremap:
 out_error_get_res:
@@ -1355,7 +1374,7 @@
 /* stop hardware and remove the driver */
 static int __devexit bfin_spi_remove(struct platform_device *pdev)
 {
-	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	struct bfin_spi_master_data *drv_data = platform_get_drvdata(pdev);
 	int status = 0;
 
 	if (!drv_data)
@@ -1375,6 +1394,11 @@
 			free_dma(drv_data->dma_channel);
 	}
 
+	if (drv_data->irq_requested) {
+		free_irq(drv_data->spi_irq, drv_data);
+		drv_data->irq_requested = 0;
+	}
+
 	/* Disconnect from the SPI framework */
 	spi_unregister_master(drv_data->master);
 
@@ -1389,26 +1413,32 @@
 #ifdef CONFIG_PM
 static int bfin_spi_suspend(struct platform_device *pdev, pm_message_t state)
 {
-	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	struct bfin_spi_master_data *drv_data = platform_get_drvdata(pdev);
 	int status = 0;
 
 	status = bfin_spi_stop_queue(drv_data);
 	if (status != 0)
 		return status;
 
-	/* stop hardware */
-	bfin_spi_disable(drv_data);
+	drv_data->ctrl_reg = read_CTRL(drv_data);
+	drv_data->flag_reg = read_FLAG(drv_data);
+
+	/*
+	 * reset SPI_CTL and SPI_FLG registers
+	 */
+	write_CTRL(drv_data, BIT_CTL_CPHA | BIT_CTL_MASTER);
+	write_FLAG(drv_data, 0xFF00);
 
 	return 0;
 }
 
 static int bfin_spi_resume(struct platform_device *pdev)
 {
-	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	struct bfin_spi_master_data *drv_data = platform_get_drvdata(pdev);
 	int status = 0;
 
-	/* Enable the SPI interface */
-	bfin_spi_enable(drv_data);
+	write_CTRL(drv_data, drv_data->ctrl_reg);
+	write_FLAG(drv_data, drv_data->flag_reg);
 
 	/* Start the queue running */
 	status = bfin_spi_start_queue(drv_data);
@@ -1439,7 +1469,7 @@
 {
 	return platform_driver_probe(&bfin_spi_driver, bfin_spi_probe);
 }
-module_init(bfin_spi_init);
+subsys_initcall(bfin_spi_init);
 
 static void __exit bfin_spi_exit(void)
 {
diff --git a/drivers/spi/spi_fsl_espi.c b/drivers/spi/spi_fsl_espi.c
new file mode 100644
index 0000000..e3b4f64
--- /dev/null
+++ b/drivers/spi/spi_fsl_espi.c
@@ -0,0 +1,748 @@
+/*
+ * Freescale eSPI controller driver.
+ *
+ * Copyright 2010 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/spi/spi.h>
+#include <linux/platform_device.h>
+#include <linux/fsl_devices.h>
+#include <linux/mm.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/of_spi.h>
+#include <linux/interrupt.h>
+#include <linux/err.h>
+#include <sysdev/fsl_soc.h>
+
+#include "spi_fsl_lib.h"
+
+/* eSPI Controller registers */
+struct fsl_espi_reg {
+	__be32 mode;		/* 0x000 - eSPI mode register */
+	__be32 event;		/* 0x004 - eSPI event register */
+	__be32 mask;		/* 0x008 - eSPI mask register */
+	__be32 command;		/* 0x00c - eSPI command register */
+	__be32 transmit;	/* 0x010 - eSPI transmit FIFO access register*/
+	__be32 receive;		/* 0x014 - eSPI receive FIFO access register*/
+	u8 res[8];		/* 0x018 - 0x01c reserved */
+	__be32 csmode[4];	/* 0x020 - 0x02c eSPI cs mode register */
+};
+
+struct fsl_espi_transfer {
+	const void *tx_buf;
+	void *rx_buf;
+	unsigned len;
+	unsigned n_tx;
+	unsigned n_rx;
+	unsigned actual_length;
+	int status;
+};
+
+/* eSPI Controller mode register definitions */
+#define SPMODE_ENABLE		(1 << 31)
+#define SPMODE_LOOP		(1 << 30)
+#define SPMODE_TXTHR(x)		((x) << 8)
+#define SPMODE_RXTHR(x)		((x) << 0)
+
+/* eSPI Controller CS mode register definitions */
+#define CSMODE_CI_INACTIVEHIGH	(1 << 31)
+#define CSMODE_CP_BEGIN_EDGECLK	(1 << 30)
+#define CSMODE_REV		(1 << 29)
+#define CSMODE_DIV16		(1 << 28)
+#define CSMODE_PM(x)		((x) << 24)
+#define CSMODE_POL_1		(1 << 20)
+#define CSMODE_LEN(x)		((x) << 16)
+#define CSMODE_BEF(x)		((x) << 12)
+#define CSMODE_AFT(x)		((x) << 8)
+#define CSMODE_CG(x)		((x) << 3)
+
+/* Default mode/csmode for eSPI controller */
+#define SPMODE_INIT_VAL (SPMODE_TXTHR(4) | SPMODE_RXTHR(3))
+#define CSMODE_INIT_VAL (CSMODE_POL_1 | CSMODE_BEF(0) \
+		| CSMODE_AFT(0) | CSMODE_CG(1))
+
+/* SPIE register values */
+#define	SPIE_NE		0x00000200	/* Not empty */
+#define	SPIE_NF		0x00000100	/* Not full */
+
+/* SPIM register values */
+#define	SPIM_NE		0x00000200	/* Not empty */
+#define	SPIM_NF		0x00000100	/* Not full */
+#define SPIE_RXCNT(reg)     ((reg >> 24) & 0x3F)
+#define SPIE_TXCNT(reg)     ((reg >> 16) & 0x3F)
+
+/* SPCOM register values */
+#define SPCOM_CS(x)		((x) << 30)
+#define SPCOM_TRANLEN(x)	((x) << 0)
+#define	SPCOM_TRANLEN_MAX	0xFFFF	/* Max transaction length */
+
+static void fsl_espi_change_mode(struct spi_device *spi)
+{
+	struct mpc8xxx_spi *mspi = spi_master_get_devdata(spi->master);
+	struct spi_mpc8xxx_cs *cs = spi->controller_state;
+	struct fsl_espi_reg *reg_base = mspi->reg_base;
+	__be32 __iomem *mode = &reg_base->csmode[spi->chip_select];
+	__be32 __iomem *espi_mode = &reg_base->mode;
+	u32 tmp;
+	unsigned long flags;
+
+	/* Turn off IRQs locally to minimize time that SPI is disabled. */
+	local_irq_save(flags);
+
+	/* Turn off SPI unit prior changing mode */
+	tmp = mpc8xxx_spi_read_reg(espi_mode);
+	mpc8xxx_spi_write_reg(espi_mode, tmp & ~SPMODE_ENABLE);
+	mpc8xxx_spi_write_reg(mode, cs->hw_mode);
+	mpc8xxx_spi_write_reg(espi_mode, tmp);
+
+	local_irq_restore(flags);
+}
+
+static u32 fsl_espi_tx_buf_lsb(struct mpc8xxx_spi *mpc8xxx_spi)
+{
+	u32 data;
+	u16 data_h;
+	u16 data_l;
+	const u32 *tx = mpc8xxx_spi->tx;
+
+	if (!tx)
+		return 0;
+
+	data = *tx++ << mpc8xxx_spi->tx_shift;
+	data_l = data & 0xffff;
+	data_h = (data >> 16) & 0xffff;
+	swab16s(&data_l);
+	swab16s(&data_h);
+	data = data_h | data_l;
+
+	mpc8xxx_spi->tx = tx;
+	return data;
+}
+
+static int fsl_espi_setup_transfer(struct spi_device *spi,
+					struct spi_transfer *t)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
+	int bits_per_word = 0;
+	u8 pm;
+	u32 hz = 0;
+	struct spi_mpc8xxx_cs *cs = spi->controller_state;
+
+	if (t) {
+		bits_per_word = t->bits_per_word;
+		hz = t->speed_hz;
+	}
+
+	/* spi_transfer level calls that work per-word */
+	if (!bits_per_word)
+		bits_per_word = spi->bits_per_word;
+
+	/* Make sure its a bit width we support [4..16] */
+	if ((bits_per_word < 4) || (bits_per_word > 16))
+		return -EINVAL;
+
+	if (!hz)
+		hz = spi->max_speed_hz;
+
+	cs->rx_shift = 0;
+	cs->tx_shift = 0;
+	cs->get_rx = mpc8xxx_spi_rx_buf_u32;
+	cs->get_tx = mpc8xxx_spi_tx_buf_u32;
+	if (bits_per_word <= 8) {
+		cs->rx_shift = 8 - bits_per_word;
+	} else if (bits_per_word <= 16) {
+		cs->rx_shift = 16 - bits_per_word;
+		if (spi->mode & SPI_LSB_FIRST)
+			cs->get_tx = fsl_espi_tx_buf_lsb;
+	} else {
+		return -EINVAL;
+	}
+
+	mpc8xxx_spi->rx_shift = cs->rx_shift;
+	mpc8xxx_spi->tx_shift = cs->tx_shift;
+	mpc8xxx_spi->get_rx = cs->get_rx;
+	mpc8xxx_spi->get_tx = cs->get_tx;
+
+	bits_per_word = bits_per_word - 1;
+
+	/* mask out bits we are going to set */
+	cs->hw_mode &= ~(CSMODE_LEN(0xF) | CSMODE_DIV16 | CSMODE_PM(0xF));
+
+	cs->hw_mode |= CSMODE_LEN(bits_per_word);
+
+	if ((mpc8xxx_spi->spibrg / hz) > 64) {
+		cs->hw_mode |= CSMODE_DIV16;
+		pm = (mpc8xxx_spi->spibrg - 1) / (hz * 64) + 1;
+
+		WARN_ONCE(pm > 16, "%s: Requested speed is too low: %d Hz. "
+			  "Will use %d Hz instead.\n", dev_name(&spi->dev),
+			  hz, mpc8xxx_spi->spibrg / 1024);
+		if (pm > 16)
+			pm = 16;
+	} else {
+		pm = (mpc8xxx_spi->spibrg - 1) / (hz * 4) + 1;
+	}
+	if (pm)
+		pm--;
+
+	cs->hw_mode |= CSMODE_PM(pm);
+
+	fsl_espi_change_mode(spi);
+	return 0;
+}
+
+static int fsl_espi_cpu_bufs(struct mpc8xxx_spi *mspi, struct spi_transfer *t,
+		unsigned int len)
+{
+	u32 word;
+	struct fsl_espi_reg *reg_base = mspi->reg_base;
+
+	mspi->count = len;
+
+	/* enable rx ints */
+	mpc8xxx_spi_write_reg(&reg_base->mask, SPIM_NE);
+
+	/* transmit word */
+	word = mspi->get_tx(mspi);
+	mpc8xxx_spi_write_reg(&reg_base->transmit, word);
+
+	return 0;
+}
+
+static int fsl_espi_bufs(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
+	struct fsl_espi_reg *reg_base = mpc8xxx_spi->reg_base;
+	unsigned int len = t->len;
+	u8 bits_per_word;
+	int ret;
+
+	bits_per_word = spi->bits_per_word;
+	if (t->bits_per_word)
+		bits_per_word = t->bits_per_word;
+
+	mpc8xxx_spi->len = t->len;
+	len = roundup(len, 4) / 4;
+
+	mpc8xxx_spi->tx = t->tx_buf;
+	mpc8xxx_spi->rx = t->rx_buf;
+
+	INIT_COMPLETION(mpc8xxx_spi->done);
+
+	/* Set SPCOM[CS] and SPCOM[TRANLEN] field */
+	if ((t->len - 1) > SPCOM_TRANLEN_MAX) {
+		dev_err(mpc8xxx_spi->dev, "Transaction length (%d)"
+				" beyond the SPCOM[TRANLEN] field\n", t->len);
+		return -EINVAL;
+	}
+	mpc8xxx_spi_write_reg(&reg_base->command,
+		(SPCOM_CS(spi->chip_select) | SPCOM_TRANLEN(t->len - 1)));
+
+	ret = fsl_espi_cpu_bufs(mpc8xxx_spi, t, len);
+	if (ret)
+		return ret;
+
+	wait_for_completion(&mpc8xxx_spi->done);
+
+	/* disable rx ints */
+	mpc8xxx_spi_write_reg(&reg_base->mask, 0);
+
+	return mpc8xxx_spi->count;
+}
+
+static void fsl_espi_addr2cmd(unsigned int addr, u8 *cmd)
+{
+	if (cmd[1] && cmd[2] && cmd[3]) {
+		cmd[1] = (u8)(addr >> 16);
+		cmd[2] = (u8)(addr >> 8);
+		cmd[3] = (u8)(addr >> 0);
+	}
+}
+
+static unsigned int fsl_espi_cmd2addr(u8 *cmd)
+{
+	if (cmd[1] && cmd[2] && cmd[3])
+		return cmd[1] << 16 | cmd[2] << 8 | cmd[3] << 0;
+
+	return 0;
+}
+
+static void fsl_espi_do_trans(struct spi_message *m,
+				struct fsl_espi_transfer *tr)
+{
+	struct spi_device *spi = m->spi;
+	struct mpc8xxx_spi *mspi = spi_master_get_devdata(spi->master);
+	struct fsl_espi_transfer *espi_trans = tr;
+	struct spi_message message;
+	struct spi_transfer *t, *first, trans;
+	int status = 0;
+
+	spi_message_init(&message);
+	memset(&trans, 0, sizeof(trans));
+
+	first = list_first_entry(&m->transfers, struct spi_transfer,
+			transfer_list);
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		if ((first->bits_per_word != t->bits_per_word) ||
+			(first->speed_hz != t->speed_hz)) {
+			espi_trans->status = -EINVAL;
+			dev_err(mspi->dev, "bits_per_word/speed_hz should be"
+					" same for the same SPI transfer\n");
+			return;
+		}
+
+		trans.speed_hz = t->speed_hz;
+		trans.bits_per_word = t->bits_per_word;
+		trans.delay_usecs = max(first->delay_usecs, t->delay_usecs);
+	}
+
+	trans.len = espi_trans->len;
+	trans.tx_buf = espi_trans->tx_buf;
+	trans.rx_buf = espi_trans->rx_buf;
+	spi_message_add_tail(&trans, &message);
+
+	list_for_each_entry(t, &message.transfers, transfer_list) {
+		if (t->bits_per_word || t->speed_hz) {
+			status = -EINVAL;
+
+			status = fsl_espi_setup_transfer(spi, t);
+			if (status < 0)
+				break;
+		}
+
+		if (t->len)
+			status = fsl_espi_bufs(spi, t);
+
+		if (status) {
+			status = -EMSGSIZE;
+			break;
+		}
+
+		if (t->delay_usecs)
+			udelay(t->delay_usecs);
+	}
+
+	espi_trans->status = status;
+	fsl_espi_setup_transfer(spi, NULL);
+}
+
+static void fsl_espi_cmd_trans(struct spi_message *m,
+				struct fsl_espi_transfer *trans, u8 *rx_buff)
+{
+	struct spi_transfer *t;
+	u8 *local_buf;
+	int i = 0;
+	struct fsl_espi_transfer *espi_trans = trans;
+
+	local_buf = kzalloc(SPCOM_TRANLEN_MAX, GFP_KERNEL);
+	if (!local_buf) {
+		espi_trans->status = -ENOMEM;
+		return;
+	}
+
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		if (t->tx_buf) {
+			memcpy(local_buf + i, t->tx_buf, t->len);
+			i += t->len;
+		}
+	}
+
+	espi_trans->tx_buf = local_buf;
+	espi_trans->rx_buf = local_buf + espi_trans->n_tx;
+	fsl_espi_do_trans(m, espi_trans);
+
+	espi_trans->actual_length = espi_trans->len;
+	kfree(local_buf);
+}
+
+static void fsl_espi_rw_trans(struct spi_message *m,
+				struct fsl_espi_transfer *trans, u8 *rx_buff)
+{
+	struct fsl_espi_transfer *espi_trans = trans;
+	unsigned int n_tx = espi_trans->n_tx;
+	unsigned int n_rx = espi_trans->n_rx;
+	struct spi_transfer *t;
+	u8 *local_buf;
+	u8 *rx_buf = rx_buff;
+	unsigned int trans_len;
+	unsigned int addr;
+	int i, pos, loop;
+
+	local_buf = kzalloc(SPCOM_TRANLEN_MAX, GFP_KERNEL);
+	if (!local_buf) {
+		espi_trans->status = -ENOMEM;
+		return;
+	}
+
+	for (pos = 0, loop = 0; pos < n_rx; pos += trans_len, loop++) {
+		trans_len = n_rx - pos;
+		if (trans_len > SPCOM_TRANLEN_MAX - n_tx)
+			trans_len = SPCOM_TRANLEN_MAX - n_tx;
+
+		i = 0;
+		list_for_each_entry(t, &m->transfers, transfer_list) {
+			if (t->tx_buf) {
+				memcpy(local_buf + i, t->tx_buf, t->len);
+				i += t->len;
+			}
+		}
+
+		addr = fsl_espi_cmd2addr(local_buf);
+		addr += pos;
+		fsl_espi_addr2cmd(addr, local_buf);
+
+		espi_trans->n_tx = n_tx;
+		espi_trans->n_rx = trans_len;
+		espi_trans->len = trans_len + n_tx;
+		espi_trans->tx_buf = local_buf;
+		espi_trans->rx_buf = local_buf + n_tx;
+		fsl_espi_do_trans(m, espi_trans);
+
+		memcpy(rx_buf + pos, espi_trans->rx_buf + n_tx, trans_len);
+
+		if (loop > 0)
+			espi_trans->actual_length += espi_trans->len - n_tx;
+		else
+			espi_trans->actual_length += espi_trans->len;
+	}
+
+	kfree(local_buf);
+}
+
+static void fsl_espi_do_one_msg(struct spi_message *m)
+{
+	struct spi_transfer *t;
+	u8 *rx_buf = NULL;
+	unsigned int n_tx = 0;
+	unsigned int n_rx = 0;
+	struct fsl_espi_transfer espi_trans;
+
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		if (t->tx_buf)
+			n_tx += t->len;
+		if (t->rx_buf) {
+			n_rx += t->len;
+			rx_buf = t->rx_buf;
+		}
+	}
+
+	espi_trans.n_tx = n_tx;
+	espi_trans.n_rx = n_rx;
+	espi_trans.len = n_tx + n_rx;
+	espi_trans.actual_length = 0;
+	espi_trans.status = 0;
+
+	if (!rx_buf)
+		fsl_espi_cmd_trans(m, &espi_trans, NULL);
+	else
+		fsl_espi_rw_trans(m, &espi_trans, rx_buf);
+
+	m->actual_length = espi_trans.actual_length;
+	m->status = espi_trans.status;
+	m->complete(m->context);
+}
+
+static int fsl_espi_setup(struct spi_device *spi)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi;
+	struct fsl_espi_reg *reg_base;
+	int retval;
+	u32 hw_mode;
+	u32 loop_mode;
+	struct spi_mpc8xxx_cs *cs = spi->controller_state;
+
+	if (!spi->max_speed_hz)
+		return -EINVAL;
+
+	if (!cs) {
+		cs = kzalloc(sizeof *cs, GFP_KERNEL);
+		if (!cs)
+			return -ENOMEM;
+		spi->controller_state = cs;
+	}
+
+	mpc8xxx_spi = spi_master_get_devdata(spi->master);
+	reg_base = mpc8xxx_spi->reg_base;
+
+	hw_mode = cs->hw_mode; /* Save orginal settings */
+	cs->hw_mode = mpc8xxx_spi_read_reg(
+			&reg_base->csmode[spi->chip_select]);
+	/* mask out bits we are going to set */
+	cs->hw_mode &= ~(CSMODE_CP_BEGIN_EDGECLK | CSMODE_CI_INACTIVEHIGH
+			 | CSMODE_REV);
+
+	if (spi->mode & SPI_CPHA)
+		cs->hw_mode |= CSMODE_CP_BEGIN_EDGECLK;
+	if (spi->mode & SPI_CPOL)
+		cs->hw_mode |= CSMODE_CI_INACTIVEHIGH;
+	if (!(spi->mode & SPI_LSB_FIRST))
+		cs->hw_mode |= CSMODE_REV;
+
+	/* Handle the loop mode */
+	loop_mode = mpc8xxx_spi_read_reg(&reg_base->mode);
+	loop_mode &= ~SPMODE_LOOP;
+	if (spi->mode & SPI_LOOP)
+		loop_mode |= SPMODE_LOOP;
+	mpc8xxx_spi_write_reg(&reg_base->mode, loop_mode);
+
+	retval = fsl_espi_setup_transfer(spi, NULL);
+	if (retval < 0) {
+		cs->hw_mode = hw_mode; /* Restore settings */
+		return retval;
+	}
+	return 0;
+}
+
+void fsl_espi_cpu_irq(struct mpc8xxx_spi *mspi, u32 events)
+{
+	struct fsl_espi_reg *reg_base = mspi->reg_base;
+
+	/* We need handle RX first */
+	if (events & SPIE_NE) {
+		u32 rx_data;
+
+		/* Spin until RX is done */
+		while (SPIE_RXCNT(events) < min(4, mspi->len)) {
+			cpu_relax();
+			events = mpc8xxx_spi_read_reg(&reg_base->event);
+		}
+		mspi->len -= 4;
+
+		rx_data = mpc8xxx_spi_read_reg(&reg_base->receive);
+
+		if (mspi->rx)
+			mspi->get_rx(rx_data, mspi);
+	}
+
+	if (!(events & SPIE_NF)) {
+		int ret;
+
+		/* spin until TX is done */
+		ret = spin_event_timeout(((events = mpc8xxx_spi_read_reg(
+				&reg_base->event)) & SPIE_NF) == 0, 1000, 0);
+		if (!ret) {
+			dev_err(mspi->dev, "tired waiting for SPIE_NF\n");
+			return;
+		}
+	}
+
+	/* Clear the events */
+	mpc8xxx_spi_write_reg(&reg_base->event, events);
+
+	mspi->count -= 1;
+	if (mspi->count) {
+		u32 word = mspi->get_tx(mspi);
+
+		mpc8xxx_spi_write_reg(&reg_base->transmit, word);
+	} else {
+		complete(&mspi->done);
+	}
+}
+
+static irqreturn_t fsl_espi_irq(s32 irq, void *context_data)
+{
+	struct mpc8xxx_spi *mspi = context_data;
+	struct fsl_espi_reg *reg_base = mspi->reg_base;
+	irqreturn_t ret = IRQ_NONE;
+	u32 events;
+
+	/* Get interrupt events(tx/rx) */
+	events = mpc8xxx_spi_read_reg(&reg_base->event);
+	if (events)
+		ret = IRQ_HANDLED;
+
+	dev_vdbg(mspi->dev, "%s: events %x\n", __func__, events);
+
+	fsl_espi_cpu_irq(mspi, events);
+
+	return ret;
+}
+
+static void fsl_espi_remove(struct mpc8xxx_spi *mspi)
+{
+	iounmap(mspi->reg_base);
+}
+
+static struct spi_master * __devinit fsl_espi_probe(struct device *dev,
+		struct resource *mem, unsigned int irq)
+{
+	struct fsl_spi_platform_data *pdata = dev->platform_data;
+	struct spi_master *master;
+	struct mpc8xxx_spi *mpc8xxx_spi;
+	struct fsl_espi_reg *reg_base;
+	u32 regval;
+	int i, ret = 0;
+
+	master = spi_alloc_master(dev, sizeof(struct mpc8xxx_spi));
+	if (!master) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	dev_set_drvdata(dev, master);
+
+	ret = mpc8xxx_spi_probe(dev, mem, irq);
+	if (ret)
+		goto err_probe;
+
+	master->setup = fsl_espi_setup;
+
+	mpc8xxx_spi = spi_master_get_devdata(master);
+	mpc8xxx_spi->spi_do_one_msg = fsl_espi_do_one_msg;
+	mpc8xxx_spi->spi_remove = fsl_espi_remove;
+
+	mpc8xxx_spi->reg_base = ioremap(mem->start, resource_size(mem));
+	if (!mpc8xxx_spi->reg_base) {
+		ret = -ENOMEM;
+		goto err_probe;
+	}
+
+	reg_base = mpc8xxx_spi->reg_base;
+
+	/* Register for SPI Interrupt */
+	ret = request_irq(mpc8xxx_spi->irq, fsl_espi_irq,
+			  0, "fsl_espi", mpc8xxx_spi);
+	if (ret)
+		goto free_irq;
+
+	if (mpc8xxx_spi->flags & SPI_QE_CPU_MODE) {
+		mpc8xxx_spi->rx_shift = 16;
+		mpc8xxx_spi->tx_shift = 24;
+	}
+
+	/* SPI controller initializations */
+	mpc8xxx_spi_write_reg(&reg_base->mode, 0);
+	mpc8xxx_spi_write_reg(&reg_base->mask, 0);
+	mpc8xxx_spi_write_reg(&reg_base->command, 0);
+	mpc8xxx_spi_write_reg(&reg_base->event, 0xffffffff);
+
+	/* Init eSPI CS mode register */
+	for (i = 0; i < pdata->max_chipselect; i++)
+		mpc8xxx_spi_write_reg(&reg_base->csmode[i], CSMODE_INIT_VAL);
+
+	/* Enable SPI interface */
+	regval = pdata->initial_spmode | SPMODE_INIT_VAL | SPMODE_ENABLE;
+
+	mpc8xxx_spi_write_reg(&reg_base->mode, regval);
+
+	ret = spi_register_master(master);
+	if (ret < 0)
+		goto unreg_master;
+
+	dev_info(dev, "at 0x%p (irq = %d)\n", reg_base, mpc8xxx_spi->irq);
+
+	return master;
+
+unreg_master:
+	free_irq(mpc8xxx_spi->irq, mpc8xxx_spi);
+free_irq:
+	iounmap(mpc8xxx_spi->reg_base);
+err_probe:
+	spi_master_put(master);
+err:
+	return ERR_PTR(ret);
+}
+
+static int of_fsl_espi_get_chipselects(struct device *dev)
+{
+	struct device_node *np = dev->of_node;
+	struct fsl_spi_platform_data *pdata = dev->platform_data;
+	const u32 *prop;
+	int len;
+
+	prop = of_get_property(np, "fsl,espi-num-chipselects", &len);
+	if (!prop || len < sizeof(*prop)) {
+		dev_err(dev, "No 'fsl,espi-num-chipselects' property\n");
+		return -EINVAL;
+	}
+
+	pdata->max_chipselect = *prop;
+	pdata->cs_control = NULL;
+
+	return 0;
+}
+
+static int __devinit of_fsl_espi_probe(struct platform_device *ofdev,
+					const struct of_device_id *ofid)
+{
+	struct device *dev = &ofdev->dev;
+	struct device_node *np = ofdev->dev.of_node;
+	struct spi_master *master;
+	struct resource mem;
+	struct resource irq;
+	int ret = -ENOMEM;
+
+	ret = of_mpc8xxx_spi_probe(ofdev, ofid);
+	if (ret)
+		return ret;
+
+	ret = of_fsl_espi_get_chipselects(dev);
+	if (ret)
+		goto err;
+
+	ret = of_address_to_resource(np, 0, &mem);
+	if (ret)
+		goto err;
+
+	ret = of_irq_to_resource(np, 0, &irq);
+	if (!ret) {
+		ret = -EINVAL;
+		goto err;
+	}
+
+	master = fsl_espi_probe(dev, &mem, irq.start);
+	if (IS_ERR(master)) {
+		ret = PTR_ERR(master);
+		goto err;
+	}
+
+	return 0;
+
+err:
+	return ret;
+}
+
+static int __devexit of_fsl_espi_remove(struct platform_device *dev)
+{
+	return mpc8xxx_spi_remove(&dev->dev);
+}
+
+static const struct of_device_id of_fsl_espi_match[] = {
+	{ .compatible = "fsl,mpc8536-espi" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, of_fsl_espi_match);
+
+static struct of_platform_driver fsl_espi_driver = {
+	.driver = {
+		.name = "fsl_espi",
+		.owner = THIS_MODULE,
+		.of_match_table = of_fsl_espi_match,
+	},
+	.probe		= of_fsl_espi_probe,
+	.remove		= __devexit_p(of_fsl_espi_remove),
+};
+
+static int __init fsl_espi_init(void)
+{
+	return of_register_platform_driver(&fsl_espi_driver);
+}
+module_init(fsl_espi_init);
+
+static void __exit fsl_espi_exit(void)
+{
+	of_unregister_platform_driver(&fsl_espi_driver);
+}
+module_exit(fsl_espi_exit);
+
+MODULE_AUTHOR("Mingkai Hu");
+MODULE_DESCRIPTION("Enhanced Freescale SPI Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi_fsl_lib.c b/drivers/spi/spi_fsl_lib.c
new file mode 100644
index 0000000..5cd741f
--- /dev/null
+++ b/drivers/spi/spi_fsl_lib.c
@@ -0,0 +1,237 @@
+/*
+ * Freescale SPI/eSPI controller driver library.
+ *
+ * Maintainer: Kumar Gala
+ *
+ * Copyright (C) 2006 Polycom, Inc.
+ *
+ * CPM SPI and QE buffer descriptors mode support:
+ * Copyright (c) 2009  MontaVista Software, Inc.
+ * Author: Anton Vorontsov <avorontsov@ru.mvista.com>
+ *
+ * Copyright 2010 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/fsl_devices.h>
+#include <linux/dma-mapping.h>
+#include <linux/mm.h>
+#include <linux/of_platform.h>
+#include <linux/of_spi.h>
+#include <sysdev/fsl_soc.h>
+
+#include "spi_fsl_lib.h"
+
+#define MPC8XXX_SPI_RX_BUF(type) 					  \
+void mpc8xxx_spi_rx_buf_##type(u32 data, struct mpc8xxx_spi *mpc8xxx_spi) \
+{									  \
+	type *rx = mpc8xxx_spi->rx;					  \
+	*rx++ = (type)(data >> mpc8xxx_spi->rx_shift);			  \
+	mpc8xxx_spi->rx = rx;						  \
+}
+
+#define MPC8XXX_SPI_TX_BUF(type)				\
+u32 mpc8xxx_spi_tx_buf_##type(struct mpc8xxx_spi *mpc8xxx_spi)	\
+{								\
+	u32 data;						\
+	const type *tx = mpc8xxx_spi->tx;			\
+	if (!tx)						\
+		return 0;					\
+	data = *tx++ << mpc8xxx_spi->tx_shift;			\
+	mpc8xxx_spi->tx = tx;					\
+	return data;						\
+}
+
+MPC8XXX_SPI_RX_BUF(u8)
+MPC8XXX_SPI_RX_BUF(u16)
+MPC8XXX_SPI_RX_BUF(u32)
+MPC8XXX_SPI_TX_BUF(u8)
+MPC8XXX_SPI_TX_BUF(u16)
+MPC8XXX_SPI_TX_BUF(u32)
+
+struct mpc8xxx_spi_probe_info *to_of_pinfo(struct fsl_spi_platform_data *pdata)
+{
+	return container_of(pdata, struct mpc8xxx_spi_probe_info, pdata);
+}
+
+void mpc8xxx_spi_work(struct work_struct *work)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi = container_of(work, struct mpc8xxx_spi,
+						       work);
+
+	spin_lock_irq(&mpc8xxx_spi->lock);
+	while (!list_empty(&mpc8xxx_spi->queue)) {
+		struct spi_message *m = container_of(mpc8xxx_spi->queue.next,
+						   struct spi_message, queue);
+
+		list_del_init(&m->queue);
+		spin_unlock_irq(&mpc8xxx_spi->lock);
+
+		if (mpc8xxx_spi->spi_do_one_msg)
+			mpc8xxx_spi->spi_do_one_msg(m);
+
+		spin_lock_irq(&mpc8xxx_spi->lock);
+	}
+	spin_unlock_irq(&mpc8xxx_spi->lock);
+}
+
+int mpc8xxx_spi_transfer(struct spi_device *spi,
+				struct spi_message *m)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
+	unsigned long flags;
+
+	m->actual_length = 0;
+	m->status = -EINPROGRESS;
+
+	spin_lock_irqsave(&mpc8xxx_spi->lock, flags);
+	list_add_tail(&m->queue, &mpc8xxx_spi->queue);
+	queue_work(mpc8xxx_spi->workqueue, &mpc8xxx_spi->work);
+	spin_unlock_irqrestore(&mpc8xxx_spi->lock, flags);
+
+	return 0;
+}
+
+void mpc8xxx_spi_cleanup(struct spi_device *spi)
+{
+	kfree(spi->controller_state);
+}
+
+const char *mpc8xxx_spi_strmode(unsigned int flags)
+{
+	if (flags & SPI_QE_CPU_MODE) {
+		return "QE CPU";
+	} else if (flags & SPI_CPM_MODE) {
+		if (flags & SPI_QE)
+			return "QE";
+		else if (flags & SPI_CPM2)
+			return "CPM2";
+		else
+			return "CPM1";
+	}
+	return "CPU";
+}
+
+int mpc8xxx_spi_probe(struct device *dev, struct resource *mem,
+			unsigned int irq)
+{
+	struct fsl_spi_platform_data *pdata = dev->platform_data;
+	struct spi_master *master;
+	struct mpc8xxx_spi *mpc8xxx_spi;
+	int ret = 0;
+
+	master = dev_get_drvdata(dev);
+
+	/* the spi->mode bits understood by this driver: */
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH
+			| SPI_LSB_FIRST | SPI_LOOP;
+
+	master->transfer = mpc8xxx_spi_transfer;
+	master->cleanup = mpc8xxx_spi_cleanup;
+	master->dev.of_node = dev->of_node;
+
+	mpc8xxx_spi = spi_master_get_devdata(master);
+	mpc8xxx_spi->dev = dev;
+	mpc8xxx_spi->get_rx = mpc8xxx_spi_rx_buf_u8;
+	mpc8xxx_spi->get_tx = mpc8xxx_spi_tx_buf_u8;
+	mpc8xxx_spi->flags = pdata->flags;
+	mpc8xxx_spi->spibrg = pdata->sysclk;
+	mpc8xxx_spi->irq = irq;
+
+	mpc8xxx_spi->rx_shift = 0;
+	mpc8xxx_spi->tx_shift = 0;
+
+	init_completion(&mpc8xxx_spi->done);
+
+	master->bus_num = pdata->bus_num;
+	master->num_chipselect = pdata->max_chipselect;
+
+	spin_lock_init(&mpc8xxx_spi->lock);
+	init_completion(&mpc8xxx_spi->done);
+	INIT_WORK(&mpc8xxx_spi->work, mpc8xxx_spi_work);
+	INIT_LIST_HEAD(&mpc8xxx_spi->queue);
+
+	mpc8xxx_spi->workqueue = create_singlethread_workqueue(
+		dev_name(master->dev.parent));
+	if (mpc8xxx_spi->workqueue == NULL) {
+		ret = -EBUSY;
+		goto err;
+	}
+
+	return 0;
+
+err:
+	return ret;
+}
+
+int __devexit mpc8xxx_spi_remove(struct device *dev)
+{
+	struct mpc8xxx_spi *mpc8xxx_spi;
+	struct spi_master *master;
+
+	master = dev_get_drvdata(dev);
+	mpc8xxx_spi = spi_master_get_devdata(master);
+
+	flush_workqueue(mpc8xxx_spi->workqueue);
+	destroy_workqueue(mpc8xxx_spi->workqueue);
+	spi_unregister_master(master);
+
+	free_irq(mpc8xxx_spi->irq, mpc8xxx_spi);
+
+	if (mpc8xxx_spi->spi_remove)
+		mpc8xxx_spi->spi_remove(mpc8xxx_spi);
+
+	return 0;
+}
+
+int __devinit of_mpc8xxx_spi_probe(struct platform_device *ofdev,
+					const struct of_device_id *ofid)
+{
+	struct device *dev = &ofdev->dev;
+	struct device_node *np = ofdev->dev.of_node;
+	struct mpc8xxx_spi_probe_info *pinfo;
+	struct fsl_spi_platform_data *pdata;
+	const void *prop;
+	int ret = -ENOMEM;
+
+	pinfo = kzalloc(sizeof(*pinfo), GFP_KERNEL);
+	if (!pinfo)
+		return -ENOMEM;
+
+	pdata = &pinfo->pdata;
+	dev->platform_data = pdata;
+
+	/* Allocate bus num dynamically. */
+	pdata->bus_num = -1;
+
+	/* SPI controller is either clocked from QE or SoC clock. */
+	pdata->sysclk = get_brgfreq();
+	if (pdata->sysclk == -1) {
+		pdata->sysclk = fsl_get_sys_freq();
+		if (pdata->sysclk == -1) {
+			ret = -ENODEV;
+			goto err;
+		}
+	}
+
+	prop = of_get_property(np, "mode", NULL);
+	if (prop && !strcmp(prop, "cpu-qe"))
+		pdata->flags = SPI_QE_CPU_MODE;
+	else if (prop && !strcmp(prop, "qe"))
+		pdata->flags = SPI_CPM_MODE | SPI_QE;
+	else if (of_device_is_compatible(np, "fsl,cpm2-spi"))
+		pdata->flags = SPI_CPM_MODE | SPI_CPM2;
+	else if (of_device_is_compatible(np, "fsl,cpm1-spi"))
+		pdata->flags = SPI_CPM_MODE | SPI_CPM1;
+
+	return 0;
+
+err:
+	kfree(pinfo);
+	return ret;
+}
diff --git a/drivers/spi/spi_fsl_lib.h b/drivers/spi/spi_fsl_lib.h
new file mode 100644
index 0000000..281e060
--- /dev/null
+++ b/drivers/spi/spi_fsl_lib.h
@@ -0,0 +1,124 @@
+/*
+ * Freescale SPI/eSPI controller driver library.
+ *
+ * Maintainer: Kumar Gala
+ *
+ * Copyright 2010 Freescale Semiconductor, Inc.
+ * Copyright (C) 2006 Polycom, Inc.
+ *
+ * CPM SPI and QE buffer descriptors mode support:
+ * Copyright (c) 2009  MontaVista Software, Inc.
+ * Author: Anton Vorontsov <avorontsov@ru.mvista.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+#ifndef __SPI_FSL_LIB_H__
+#define __SPI_FSL_LIB_H__
+
+#include <asm/io.h>
+
+/* SPI/eSPI Controller driver's private data. */
+struct mpc8xxx_spi {
+	struct device *dev;
+	void *reg_base;
+
+	/* rx & tx bufs from the spi_transfer */
+	const void *tx;
+	void *rx;
+#ifdef CONFIG_SPI_FSL_ESPI
+	int len;
+#endif
+
+	int subblock;
+	struct spi_pram __iomem *pram;
+	struct cpm_buf_desc __iomem *tx_bd;
+	struct cpm_buf_desc __iomem *rx_bd;
+
+	struct spi_transfer *xfer_in_progress;
+
+	/* dma addresses for CPM transfers */
+	dma_addr_t tx_dma;
+	dma_addr_t rx_dma;
+	bool map_tx_dma;
+	bool map_rx_dma;
+
+	dma_addr_t dma_dummy_tx;
+	dma_addr_t dma_dummy_rx;
+
+	/* functions to deal with different sized buffers */
+	void (*get_rx) (u32 rx_data, struct mpc8xxx_spi *);
+	u32(*get_tx) (struct mpc8xxx_spi *);
+
+	/* hooks for different controller driver */
+	void (*spi_do_one_msg) (struct spi_message *m);
+	void (*spi_remove) (struct mpc8xxx_spi *mspi);
+
+	unsigned int count;
+	unsigned int irq;
+
+	unsigned nsecs;		/* (clock cycle time)/2 */
+
+	u32 spibrg;		/* SPIBRG input clock */
+	u32 rx_shift;		/* RX data reg shift when in qe mode */
+	u32 tx_shift;		/* TX data reg shift when in qe mode */
+
+	unsigned int flags;
+
+	struct workqueue_struct *workqueue;
+	struct work_struct work;
+
+	struct list_head queue;
+	spinlock_t lock;
+
+	struct completion done;
+};
+
+struct spi_mpc8xxx_cs {
+	/* functions to deal with different sized buffers */
+	void (*get_rx) (u32 rx_data, struct mpc8xxx_spi *);
+	u32 (*get_tx) (struct mpc8xxx_spi *);
+	u32 rx_shift;		/* RX data reg shift when in qe mode */
+	u32 tx_shift;		/* TX data reg shift when in qe mode */
+	u32 hw_mode;		/* Holds HW mode register settings */
+};
+
+static inline void mpc8xxx_spi_write_reg(__be32 __iomem *reg, u32 val)
+{
+	out_be32(reg, val);
+}
+
+static inline u32 mpc8xxx_spi_read_reg(__be32 __iomem *reg)
+{
+	return in_be32(reg);
+}
+
+struct mpc8xxx_spi_probe_info {
+	struct fsl_spi_platform_data pdata;
+	int *gpios;
+	bool *alow_flags;
+};
+
+extern u32 mpc8xxx_spi_tx_buf_u8(struct mpc8xxx_spi *mpc8xxx_spi);
+extern u32 mpc8xxx_spi_tx_buf_u16(struct mpc8xxx_spi *mpc8xxx_spi);
+extern u32 mpc8xxx_spi_tx_buf_u32(struct mpc8xxx_spi *mpc8xxx_spi);
+extern void mpc8xxx_spi_rx_buf_u8(u32 data, struct mpc8xxx_spi *mpc8xxx_spi);
+extern void mpc8xxx_spi_rx_buf_u16(u32 data, struct mpc8xxx_spi *mpc8xxx_spi);
+extern void mpc8xxx_spi_rx_buf_u32(u32 data, struct mpc8xxx_spi *mpc8xxx_spi);
+
+extern struct mpc8xxx_spi_probe_info *to_of_pinfo(
+		struct fsl_spi_platform_data *pdata);
+extern int mpc8xxx_spi_bufs(struct mpc8xxx_spi *mspi,
+		struct spi_transfer *t, unsigned int len);
+extern int mpc8xxx_spi_transfer(struct spi_device *spi, struct spi_message *m);
+extern void mpc8xxx_spi_cleanup(struct spi_device *spi);
+extern const char *mpc8xxx_spi_strmode(unsigned int flags);
+extern int mpc8xxx_spi_probe(struct device *dev, struct resource *mem,
+		unsigned int irq);
+extern int mpc8xxx_spi_remove(struct device *dev);
+extern int of_mpc8xxx_spi_probe(struct platform_device *ofdev,
+				const struct of_device_id *ofid);
+
+#endif /* __SPI_FSL_LIB_H__ */
diff --git a/drivers/spi/spi_mpc8xxx.c b/drivers/spi/spi_fsl_spi.c
similarity index 64%
rename from drivers/spi/spi_mpc8xxx.c
rename to drivers/spi/spi_fsl_spi.c
index 1dd86b8..7ca52d3 100644
--- a/drivers/spi/spi_mpc8xxx.c
+++ b/drivers/spi/spi_fsl_spi.c
@@ -1,9 +1,10 @@
 /*
- * MPC8xxx SPI controller driver.
+ * Freescale SPI controller driver.
  *
  * Maintainer: Kumar Gala
  *
  * Copyright (C) 2006 Polycom, Inc.
+ * Copyright 2010 Freescale Semiconductor, Inc.
  *
  * CPM SPI and QE buffer descriptors mode support:
  * Copyright (c) 2009  MontaVista Software, Inc.
@@ -15,18 +16,11 @@
  * option) any later version.
  */
 #include <linux/module.h>
-#include <linux/init.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/bug.h>
-#include <linux/errno.h>
-#include <linux/err.h>
-#include <linux/io.h>
-#include <linux/completion.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/irq.h>
-#include <linux/device.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_bitbang.h>
 #include <linux/platform_device.h>
@@ -38,12 +32,12 @@
 #include <linux/of_platform.h>
 #include <linux/gpio.h>
 #include <linux/of_gpio.h>
-#include <linux/slab.h>
 
 #include <sysdev/fsl_soc.h>
 #include <asm/cpm.h>
 #include <asm/qe.h>
-#include <asm/irq.h>
+
+#include "spi_fsl_lib.h"
 
 /* CPM1 and CPM2 are mutually exclusive. */
 #ifdef CONFIG_CPM1
@@ -55,7 +49,7 @@
 #endif
 
 /* SPI Controller registers */
-struct mpc8xxx_spi_reg {
+struct fsl_spi_reg {
 	u8 res1[0x20];
 	__be32 mode;
 	__be32 event;
@@ -80,7 +74,7 @@
 
 /*
  * Default for SPI Mode:
- * 	SPI MODE 0 (inactive low, phase middle, MSB, 8-bit length, slow clk
+ *	SPI MODE 0 (inactive low, phase middle, MSB, 8-bit length, slow clk
  */
 #define	SPMODE_INIT_VAL (SPMODE_CI_INACTIVEHIGH | SPMODE_DIV16 | SPMODE_REV | \
 			 SPMODE_MS | SPMODE_LEN(7) | SPMODE_PM(0xf))
@@ -102,112 +96,16 @@
 #define	SPI_PRAM_SIZE	0x100
 #define	SPI_MRBLR	((unsigned int)PAGE_SIZE)
 
-/* SPI Controller driver's private data. */
-struct mpc8xxx_spi {
-	struct device *dev;
-	struct mpc8xxx_spi_reg __iomem *base;
+static void *fsl_dummy_rx;
+static DEFINE_MUTEX(fsl_dummy_rx_lock);
+static int fsl_dummy_rx_refcnt;
 
-	/* rx & tx bufs from the spi_transfer */
-	const void *tx;
-	void *rx;
-
-	int subblock;
-	struct spi_pram __iomem *pram;
-	struct cpm_buf_desc __iomem *tx_bd;
-	struct cpm_buf_desc __iomem *rx_bd;
-
-	struct spi_transfer *xfer_in_progress;
-
-	/* dma addresses for CPM transfers */
-	dma_addr_t tx_dma;
-	dma_addr_t rx_dma;
-	bool map_tx_dma;
-	bool map_rx_dma;
-
-	dma_addr_t dma_dummy_tx;
-	dma_addr_t dma_dummy_rx;
-
-	/* functions to deal with different sized buffers */
-	void (*get_rx) (u32 rx_data, struct mpc8xxx_spi *);
-	u32(*get_tx) (struct mpc8xxx_spi *);
-
-	unsigned int count;
-	unsigned int irq;
-
-	unsigned nsecs;		/* (clock cycle time)/2 */
-
-	u32 spibrg;		/* SPIBRG input clock */
-	u32 rx_shift;		/* RX data reg shift when in qe mode */
-	u32 tx_shift;		/* TX data reg shift when in qe mode */
-
-	unsigned int flags;
-
-	struct workqueue_struct *workqueue;
-	struct work_struct work;
-
-	struct list_head queue;
-	spinlock_t lock;
-
-	struct completion done;
-};
-
-static void *mpc8xxx_dummy_rx;
-static DEFINE_MUTEX(mpc8xxx_dummy_rx_lock);
-static int mpc8xxx_dummy_rx_refcnt;
-
-struct spi_mpc8xxx_cs {
-	/* functions to deal with different sized buffers */
-	void (*get_rx) (u32 rx_data, struct mpc8xxx_spi *);
-	u32 (*get_tx) (struct mpc8xxx_spi *);
-	u32 rx_shift;		/* RX data reg shift when in qe mode */
-	u32 tx_shift;		/* TX data reg shift when in qe mode */
-	u32 hw_mode;		/* Holds HW mode register settings */
-};
-
-static inline void mpc8xxx_spi_write_reg(__be32 __iomem *reg, u32 val)
-{
-	out_be32(reg, val);
-}
-
-static inline u32 mpc8xxx_spi_read_reg(__be32 __iomem *reg)
-{
-	return in_be32(reg);
-}
-
-#define MPC83XX_SPI_RX_BUF(type) 					  \
-static									  \
-void mpc8xxx_spi_rx_buf_##type(u32 data, struct mpc8xxx_spi *mpc8xxx_spi) \
-{									  \
-	type *rx = mpc8xxx_spi->rx;					  \
-	*rx++ = (type)(data >> mpc8xxx_spi->rx_shift);			  \
-	mpc8xxx_spi->rx = rx;						  \
-}
-
-#define MPC83XX_SPI_TX_BUF(type)				\
-static								\
-u32 mpc8xxx_spi_tx_buf_##type(struct mpc8xxx_spi *mpc8xxx_spi)	\
-{								\
-	u32 data;						\
-	const type *tx = mpc8xxx_spi->tx;			\
-	if (!tx)						\
-		return 0;					\
-	data = *tx++ << mpc8xxx_spi->tx_shift;			\
-	mpc8xxx_spi->tx = tx;					\
-	return data;						\
-}
-
-MPC83XX_SPI_RX_BUF(u8)
-MPC83XX_SPI_RX_BUF(u16)
-MPC83XX_SPI_RX_BUF(u32)
-MPC83XX_SPI_TX_BUF(u8)
-MPC83XX_SPI_TX_BUF(u16)
-MPC83XX_SPI_TX_BUF(u32)
-
-static void mpc8xxx_spi_change_mode(struct spi_device *spi)
+static void fsl_spi_change_mode(struct spi_device *spi)
 {
 	struct mpc8xxx_spi *mspi = spi_master_get_devdata(spi->master);
 	struct spi_mpc8xxx_cs *cs = spi->controller_state;
-	__be32 __iomem *mode = &mspi->base->mode;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
+	__be32 __iomem *mode = &reg_base->mode;
 	unsigned long flags;
 
 	if (cs->hw_mode == mpc8xxx_spi_read_reg(mode))
@@ -238,7 +136,7 @@
 	local_irq_restore(flags);
 }
 
-static void mpc8xxx_spi_chipselect(struct spi_device *spi, int value)
+static void fsl_spi_chipselect(struct spi_device *spi, int value)
 {
 	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
 	struct fsl_spi_platform_data *pdata = spi->dev.parent->platform_data;
@@ -256,18 +154,17 @@
 		mpc8xxx_spi->get_rx = cs->get_rx;
 		mpc8xxx_spi->get_tx = cs->get_tx;
 
-		mpc8xxx_spi_change_mode(spi);
+		fsl_spi_change_mode(spi);
 
 		if (pdata->cs_control)
 			pdata->cs_control(spi, pol);
 	}
 }
 
-static int
-mspi_apply_cpu_mode_quirks(struct spi_mpc8xxx_cs *cs,
-			   struct spi_device *spi,
-			   struct mpc8xxx_spi *mpc8xxx_spi,
-			   int bits_per_word)
+static int mspi_apply_cpu_mode_quirks(struct spi_mpc8xxx_cs *cs,
+				struct spi_device *spi,
+				struct mpc8xxx_spi *mpc8xxx_spi,
+				int bits_per_word)
 {
 	cs->rx_shift = 0;
 	cs->tx_shift = 0;
@@ -307,10 +204,9 @@
 	return bits_per_word;
 }
 
-static int
-mspi_apply_qe_mode_quirks(struct spi_mpc8xxx_cs *cs,
-			  struct spi_device *spi,
-			  int bits_per_word)
+static int mspi_apply_qe_mode_quirks(struct spi_mpc8xxx_cs *cs,
+				struct spi_device *spi,
+				int bits_per_word)
 {
 	/* QE uses Little Endian for words > 8
 	 * so transform all words > 8 into 8 bits
@@ -326,13 +222,13 @@
 	return bits_per_word;
 }
 
-static
-int mpc8xxx_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
+static int fsl_spi_setup_transfer(struct spi_device *spi,
+					struct spi_transfer *t)
 {
 	struct mpc8xxx_spi *mpc8xxx_spi;
-	int bits_per_word;
+	int bits_per_word = 0;
 	u8 pm;
-	u32 hz;
+	u32 hz = 0;
 	struct spi_mpc8xxx_cs	*cs = spi->controller_state;
 
 	mpc8xxx_spi = spi_master_get_devdata(spi->master);
@@ -340,9 +236,6 @@
 	if (t) {
 		bits_per_word = t->bits_per_word;
 		hz = t->speed_hz;
-	} else {
-		bits_per_word = 0;
-		hz = 0;
 	}
 
 	/* spi_transfer level calls that work per-word */
@@ -388,23 +281,25 @@
 			  hz, mpc8xxx_spi->spibrg / 1024);
 		if (pm > 16)
 			pm = 16;
-	} else
+	} else {
 		pm = (mpc8xxx_spi->spibrg - 1) / (hz * 4) + 1;
+	}
 	if (pm)
 		pm--;
 
 	cs->hw_mode |= SPMODE_PM(pm);
 
-	mpc8xxx_spi_change_mode(spi);
+	fsl_spi_change_mode(spi);
 	return 0;
 }
 
-static void mpc8xxx_spi_cpm_bufs_start(struct mpc8xxx_spi *mspi)
+static void fsl_spi_cpm_bufs_start(struct mpc8xxx_spi *mspi)
 {
 	struct cpm_buf_desc __iomem *tx_bd = mspi->tx_bd;
 	struct cpm_buf_desc __iomem *rx_bd = mspi->rx_bd;
 	unsigned int xfer_len = min(mspi->count, SPI_MRBLR);
 	unsigned int xfer_ofs;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
 
 	xfer_ofs = mspi->xfer_in_progress->len - mspi->count;
 
@@ -424,13 +319,14 @@
 				 BD_SC_LAST);
 
 	/* start transfer */
-	mpc8xxx_spi_write_reg(&mspi->base->command, SPCOM_STR);
+	mpc8xxx_spi_write_reg(&reg_base->command, SPCOM_STR);
 }
 
-static int mpc8xxx_spi_cpm_bufs(struct mpc8xxx_spi *mspi,
+static int fsl_spi_cpm_bufs(struct mpc8xxx_spi *mspi,
 				struct spi_transfer *t, bool is_dma_mapped)
 {
 	struct device *dev = mspi->dev;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
 
 	if (is_dma_mapped) {
 		mspi->map_tx_dma = 0;
@@ -475,13 +371,13 @@
 	}
 
 	/* enable rx ints */
-	mpc8xxx_spi_write_reg(&mspi->base->mask, SPIE_RXB);
+	mpc8xxx_spi_write_reg(&reg_base->mask, SPIE_RXB);
 
 	mspi->xfer_in_progress = t;
 	mspi->count = t->len;
 
 	/* start CPM transfers */
-	mpc8xxx_spi_cpm_bufs_start(mspi);
+	fsl_spi_cpm_bufs_start(mspi);
 
 	return 0;
 
@@ -491,7 +387,7 @@
 	return -ENOMEM;
 }
 
-static void mpc8xxx_spi_cpm_bufs_complete(struct mpc8xxx_spi *mspi)
+static void fsl_spi_cpm_bufs_complete(struct mpc8xxx_spi *mspi)
 {
 	struct device *dev = mspi->dev;
 	struct spi_transfer *t = mspi->xfer_in_progress;
@@ -503,31 +399,34 @@
 	mspi->xfer_in_progress = NULL;
 }
 
-static int mpc8xxx_spi_cpu_bufs(struct mpc8xxx_spi *mspi,
+static int fsl_spi_cpu_bufs(struct mpc8xxx_spi *mspi,
 				struct spi_transfer *t, unsigned int len)
 {
 	u32 word;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
 
 	mspi->count = len;
 
 	/* enable rx ints */
-	mpc8xxx_spi_write_reg(&mspi->base->mask, SPIM_NE);
+	mpc8xxx_spi_write_reg(&reg_base->mask, SPIM_NE);
 
 	/* transmit word */
 	word = mspi->get_tx(mspi);
-	mpc8xxx_spi_write_reg(&mspi->base->transmit, word);
+	mpc8xxx_spi_write_reg(&reg_base->transmit, word);
 
 	return 0;
 }
 
-static int mpc8xxx_spi_bufs(struct spi_device *spi, struct spi_transfer *t,
+static int fsl_spi_bufs(struct spi_device *spi, struct spi_transfer *t,
 			    bool is_dma_mapped)
 {
 	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
+	struct fsl_spi_reg *reg_base;
 	unsigned int len = t->len;
 	u8 bits_per_word;
 	int ret;
 
+	reg_base = mpc8xxx_spi->reg_base;
 	bits_per_word = spi->bits_per_word;
 	if (t->bits_per_word)
 		bits_per_word = t->bits_per_word;
@@ -551,24 +450,24 @@
 	INIT_COMPLETION(mpc8xxx_spi->done);
 
 	if (mpc8xxx_spi->flags & SPI_CPM_MODE)
-		ret = mpc8xxx_spi_cpm_bufs(mpc8xxx_spi, t, is_dma_mapped);
+		ret = fsl_spi_cpm_bufs(mpc8xxx_spi, t, is_dma_mapped);
 	else
-		ret = mpc8xxx_spi_cpu_bufs(mpc8xxx_spi, t, len);
+		ret = fsl_spi_cpu_bufs(mpc8xxx_spi, t, len);
 	if (ret)
 		return ret;
 
 	wait_for_completion(&mpc8xxx_spi->done);
 
 	/* disable rx ints */
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->mask, 0);
+	mpc8xxx_spi_write_reg(&reg_base->mask, 0);
 
 	if (mpc8xxx_spi->flags & SPI_CPM_MODE)
-		mpc8xxx_spi_cpm_bufs_complete(mpc8xxx_spi);
+		fsl_spi_cpm_bufs_complete(mpc8xxx_spi);
 
 	return mpc8xxx_spi->count;
 }
 
-static void mpc8xxx_spi_do_one_msg(struct spi_message *m)
+static void fsl_spi_do_one_msg(struct spi_message *m)
 {
 	struct spi_device *spi = m->spi;
 	struct spi_transfer *t;
@@ -584,18 +483,18 @@
 			status = -EINVAL;
 
 			if (cs_change)
-				status = mpc8xxx_spi_setup_transfer(spi, t);
+				status = fsl_spi_setup_transfer(spi, t);
 			if (status < 0)
 				break;
 		}
 
 		if (cs_change) {
-			mpc8xxx_spi_chipselect(spi, BITBANG_CS_ACTIVE);
+			fsl_spi_chipselect(spi, BITBANG_CS_ACTIVE);
 			ndelay(nsecs);
 		}
 		cs_change = t->cs_change;
 		if (t->len)
-			status = mpc8xxx_spi_bufs(spi, t, m->is_dma_mapped);
+			status = fsl_spi_bufs(spi, t, m->is_dma_mapped);
 		if (status) {
 			status = -EMSGSIZE;
 			break;
@@ -607,7 +506,7 @@
 
 		if (cs_change) {
 			ndelay(nsecs);
-			mpc8xxx_spi_chipselect(spi, BITBANG_CS_INACTIVE);
+			fsl_spi_chipselect(spi, BITBANG_CS_INACTIVE);
 			ndelay(nsecs);
 		}
 	}
@@ -617,35 +516,16 @@
 
 	if (status || !cs_change) {
 		ndelay(nsecs);
-		mpc8xxx_spi_chipselect(spi, BITBANG_CS_INACTIVE);
+		fsl_spi_chipselect(spi, BITBANG_CS_INACTIVE);
 	}
 
-	mpc8xxx_spi_setup_transfer(spi, NULL);
+	fsl_spi_setup_transfer(spi, NULL);
 }
 
-static void mpc8xxx_spi_work(struct work_struct *work)
-{
-	struct mpc8xxx_spi *mpc8xxx_spi = container_of(work, struct mpc8xxx_spi,
-						       work);
-
-	spin_lock_irq(&mpc8xxx_spi->lock);
-	while (!list_empty(&mpc8xxx_spi->queue)) {
-		struct spi_message *m = container_of(mpc8xxx_spi->queue.next,
-						   struct spi_message, queue);
-
-		list_del_init(&m->queue);
-		spin_unlock_irq(&mpc8xxx_spi->lock);
-
-		mpc8xxx_spi_do_one_msg(m);
-
-		spin_lock_irq(&mpc8xxx_spi->lock);
-	}
-	spin_unlock_irq(&mpc8xxx_spi->lock);
-}
-
-static int mpc8xxx_spi_setup(struct spi_device *spi)
+static int fsl_spi_setup(struct spi_device *spi)
 {
 	struct mpc8xxx_spi *mpc8xxx_spi;
+	struct fsl_spi_reg *reg_base;
 	int retval;
 	u32 hw_mode;
 	struct spi_mpc8xxx_cs	*cs = spi->controller_state;
@@ -661,8 +541,10 @@
 	}
 	mpc8xxx_spi = spi_master_get_devdata(spi->master);
 
+	reg_base = mpc8xxx_spi->reg_base;
+
 	hw_mode = cs->hw_mode; /* Save original settings */
-	cs->hw_mode = mpc8xxx_spi_read_reg(&mpc8xxx_spi->base->mode);
+	cs->hw_mode = mpc8xxx_spi_read_reg(&reg_base->mode);
 	/* mask out bits we are going to set */
 	cs->hw_mode &= ~(SPMODE_CP_BEGIN_EDGECLK | SPMODE_CI_INACTIVEHIGH
 			 | SPMODE_REV | SPMODE_LOOP);
@@ -676,7 +558,7 @@
 	if (spi->mode & SPI_LOOP)
 		cs->hw_mode |= SPMODE_LOOP;
 
-	retval = mpc8xxx_spi_setup_transfer(spi, NULL);
+	retval = fsl_spi_setup_transfer(spi, NULL);
 	if (retval < 0) {
 		cs->hw_mode = hw_mode; /* Restore settings */
 		return retval;
@@ -684,9 +566,10 @@
 	return 0;
 }
 
-static void mpc8xxx_spi_cpm_irq(struct mpc8xxx_spi *mspi, u32 events)
+static void fsl_spi_cpm_irq(struct mpc8xxx_spi *mspi, u32 events)
 {
 	u16 len;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
 
 	dev_dbg(mspi->dev, "%s: bd datlen %d, count %d\n", __func__,
 		in_be16(&mspi->rx_bd->cbd_datlen), mspi->count);
@@ -698,20 +581,22 @@
 	}
 
 	/* Clear the events */
-	mpc8xxx_spi_write_reg(&mspi->base->event, events);
+	mpc8xxx_spi_write_reg(&reg_base->event, events);
 
 	mspi->count -= len;
 	if (mspi->count)
-		mpc8xxx_spi_cpm_bufs_start(mspi);
+		fsl_spi_cpm_bufs_start(mspi);
 	else
 		complete(&mspi->done);
 }
 
-static void mpc8xxx_spi_cpu_irq(struct mpc8xxx_spi *mspi, u32 events)
+static void fsl_spi_cpu_irq(struct mpc8xxx_spi *mspi, u32 events)
 {
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
+
 	/* We need handle RX first */
 	if (events & SPIE_NE) {
-		u32 rx_data = mpc8xxx_spi_read_reg(&mspi->base->receive);
+		u32 rx_data = mpc8xxx_spi_read_reg(&reg_base->receive);
 
 		if (mspi->rx)
 			mspi->get_rx(rx_data, mspi);
@@ -720,102 +605,80 @@
 	if ((events & SPIE_NF) == 0)
 		/* spin until TX is done */
 		while (((events =
-			mpc8xxx_spi_read_reg(&mspi->base->event)) &
+			mpc8xxx_spi_read_reg(&reg_base->event)) &
 						SPIE_NF) == 0)
 			cpu_relax();
 
 	/* Clear the events */
-	mpc8xxx_spi_write_reg(&mspi->base->event, events);
+	mpc8xxx_spi_write_reg(&reg_base->event, events);
 
 	mspi->count -= 1;
 	if (mspi->count) {
 		u32 word = mspi->get_tx(mspi);
 
-		mpc8xxx_spi_write_reg(&mspi->base->transmit, word);
+		mpc8xxx_spi_write_reg(&reg_base->transmit, word);
 	} else {
 		complete(&mspi->done);
 	}
 }
 
-static irqreturn_t mpc8xxx_spi_irq(s32 irq, void *context_data)
+static irqreturn_t fsl_spi_irq(s32 irq, void *context_data)
 {
 	struct mpc8xxx_spi *mspi = context_data;
 	irqreturn_t ret = IRQ_NONE;
 	u32 events;
+	struct fsl_spi_reg *reg_base = mspi->reg_base;
 
 	/* Get interrupt events(tx/rx) */
-	events = mpc8xxx_spi_read_reg(&mspi->base->event);
+	events = mpc8xxx_spi_read_reg(&reg_base->event);
 	if (events)
 		ret = IRQ_HANDLED;
 
 	dev_dbg(mspi->dev, "%s: events %x\n", __func__, events);
 
 	if (mspi->flags & SPI_CPM_MODE)
-		mpc8xxx_spi_cpm_irq(mspi, events);
+		fsl_spi_cpm_irq(mspi, events);
 	else
-		mpc8xxx_spi_cpu_irq(mspi, events);
+		fsl_spi_cpu_irq(mspi, events);
 
 	return ret;
 }
 
-static int mpc8xxx_spi_transfer(struct spi_device *spi,
-				struct spi_message *m)
+static void *fsl_spi_alloc_dummy_rx(void)
 {
-	struct mpc8xxx_spi *mpc8xxx_spi = spi_master_get_devdata(spi->master);
-	unsigned long flags;
+	mutex_lock(&fsl_dummy_rx_lock);
 
-	m->actual_length = 0;
-	m->status = -EINPROGRESS;
+	if (!fsl_dummy_rx)
+		fsl_dummy_rx = kmalloc(SPI_MRBLR, GFP_KERNEL);
+	if (fsl_dummy_rx)
+		fsl_dummy_rx_refcnt++;
 
-	spin_lock_irqsave(&mpc8xxx_spi->lock, flags);
-	list_add_tail(&m->queue, &mpc8xxx_spi->queue);
-	queue_work(mpc8xxx_spi->workqueue, &mpc8xxx_spi->work);
-	spin_unlock_irqrestore(&mpc8xxx_spi->lock, flags);
+	mutex_unlock(&fsl_dummy_rx_lock);
 
-	return 0;
+	return fsl_dummy_rx;
 }
 
-
-static void mpc8xxx_spi_cleanup(struct spi_device *spi)
+static void fsl_spi_free_dummy_rx(void)
 {
-	kfree(spi->controller_state);
-}
+	mutex_lock(&fsl_dummy_rx_lock);
 
-static void *mpc8xxx_spi_alloc_dummy_rx(void)
-{
-	mutex_lock(&mpc8xxx_dummy_rx_lock);
-
-	if (!mpc8xxx_dummy_rx)
-		mpc8xxx_dummy_rx = kmalloc(SPI_MRBLR, GFP_KERNEL);
-	if (mpc8xxx_dummy_rx)
-		mpc8xxx_dummy_rx_refcnt++;
-
-	mutex_unlock(&mpc8xxx_dummy_rx_lock);
-
-	return mpc8xxx_dummy_rx;
-}
-
-static void mpc8xxx_spi_free_dummy_rx(void)
-{
-	mutex_lock(&mpc8xxx_dummy_rx_lock);
-
-	switch (mpc8xxx_dummy_rx_refcnt) {
+	switch (fsl_dummy_rx_refcnt) {
 	case 0:
 		WARN_ON(1);
 		break;
 	case 1:
-		kfree(mpc8xxx_dummy_rx);
-		mpc8xxx_dummy_rx = NULL;
+		kfree(fsl_dummy_rx);
+		fsl_dummy_rx = NULL;
 		/* fall through */
 	default:
-		mpc8xxx_dummy_rx_refcnt--;
+		fsl_dummy_rx_refcnt--;
 		break;
 	}
 
-	mutex_unlock(&mpc8xxx_dummy_rx_lock);
+	mutex_unlock(&fsl_dummy_rx_lock);
 }
 
-static unsigned long mpc8xxx_spi_cpm_get_pram(struct mpc8xxx_spi *mspi)
+static unsigned long fsl_spi_cpm_get_pram(struct mpc8xxx_spi *mspi)
 {
 	struct device *dev = mspi->dev;
 	struct device_node *np = dev->of_node;
@@ -869,7 +732,7 @@
 	return pram_ofs;
 }
 
-static int mpc8xxx_spi_cpm_init(struct mpc8xxx_spi *mspi)
+static int fsl_spi_cpm_init(struct mpc8xxx_spi *mspi)
 {
 	struct device *dev = mspi->dev;
 	struct device_node *np = dev->of_node;
@@ -881,7 +744,7 @@
 	if (!(mspi->flags & SPI_CPM_MODE))
 		return 0;
 
-	if (!mpc8xxx_spi_alloc_dummy_rx())
+	if (!fsl_spi_alloc_dummy_rx())
 		return -ENOMEM;
 
 	if (mspi->flags & SPI_QE) {
@@ -902,7 +765,7 @@
 		}
 	}
 
-	pram_ofs = mpc8xxx_spi_cpm_get_pram(mspi);
+	pram_ofs = fsl_spi_cpm_get_pram(mspi);
 	if (IS_ERR_VALUE(pram_ofs)) {
 		dev_err(dev, "can't allocate spi parameter ram\n");
 		goto err_pram;
@@ -922,7 +785,7 @@
 		goto err_dummy_tx;
 	}
 
-	mspi->dma_dummy_rx = dma_map_single(dev, mpc8xxx_dummy_rx, SPI_MRBLR,
+	mspi->dma_dummy_rx = dma_map_single(dev, fsl_dummy_rx, SPI_MRBLR,
 					    DMA_FROM_DEVICE);
 	if (dma_mapping_error(dev, mspi->dma_dummy_rx)) {
 		dev_err(dev, "unable to map dummy rx buffer\n");
@@ -960,11 +823,11 @@
 err_bds:
 	cpm_muram_free(pram_ofs);
 err_pram:
-	mpc8xxx_spi_free_dummy_rx();
+	fsl_spi_free_dummy_rx();
 	return -ENOMEM;
 }
 
-static void mpc8xxx_spi_cpm_free(struct mpc8xxx_spi *mspi)
+static void fsl_spi_cpm_free(struct mpc8xxx_spi *mspi)
 {
 	struct device *dev = mspi->dev;
 
@@ -972,30 +835,22 @@
 	dma_unmap_single(dev, mspi->dma_dummy_tx, PAGE_SIZE, DMA_TO_DEVICE);
 	cpm_muram_free(cpm_muram_offset(mspi->tx_bd));
 	cpm_muram_free(cpm_muram_offset(mspi->pram));
-	mpc8xxx_spi_free_dummy_rx();
+	fsl_spi_free_dummy_rx();
 }
 
-static const char *mpc8xxx_spi_strmode(unsigned int flags)
+static void fsl_spi_remove(struct mpc8xxx_spi *mspi)
 {
-	if (flags & SPI_QE_CPU_MODE) {
-		return "QE CPU";
-	} else if (flags & SPI_CPM_MODE) {
-		if (flags & SPI_QE)
-			return "QE";
-		else if (flags & SPI_CPM2)
-			return "CPM2";
-		else
-			return "CPM1";
-	}
-	return "CPU";
+	iounmap(mspi->reg_base);
+	fsl_spi_cpm_free(mspi);
 }
 
-static struct spi_master * __devinit
-mpc8xxx_spi_probe(struct device *dev, struct resource *mem, unsigned int irq)
+static struct spi_master * __devinit fsl_spi_probe(struct device *dev,
+		struct resource *mem, unsigned int irq)
 {
 	struct fsl_spi_platform_data *pdata = dev->platform_data;
 	struct spi_master *master;
 	struct mpc8xxx_spi *mpc8xxx_spi;
+	struct fsl_spi_reg *reg_base;
 	u32 regval;
 	int ret = 0;
 
@@ -1007,132 +862,77 @@
 
 	dev_set_drvdata(dev, master);
 
-	/* the spi->mode bits understood by this driver: */
-	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH
-			| SPI_LSB_FIRST | SPI_LOOP;
+	ret = mpc8xxx_spi_probe(dev, mem, irq);
+	if (ret)
+		goto err_probe;
 
-	master->setup = mpc8xxx_spi_setup;
-	master->transfer = mpc8xxx_spi_transfer;
-	master->cleanup = mpc8xxx_spi_cleanup;
-	master->dev.of_node = dev->of_node;
+	master->setup = fsl_spi_setup;
 
 	mpc8xxx_spi = spi_master_get_devdata(master);
-	mpc8xxx_spi->dev = dev;
-	mpc8xxx_spi->get_rx = mpc8xxx_spi_rx_buf_u8;
-	mpc8xxx_spi->get_tx = mpc8xxx_spi_tx_buf_u8;
-	mpc8xxx_spi->flags = pdata->flags;
-	mpc8xxx_spi->spibrg = pdata->sysclk;
+	mpc8xxx_spi->spi_do_one_msg = fsl_spi_do_one_msg;
+	mpc8xxx_spi->spi_remove = fsl_spi_remove;
 
-	ret = mpc8xxx_spi_cpm_init(mpc8xxx_spi);
+
+	ret = fsl_spi_cpm_init(mpc8xxx_spi);
 	if (ret)
 		goto err_cpm_init;
 
-	mpc8xxx_spi->rx_shift = 0;
-	mpc8xxx_spi->tx_shift = 0;
 	if (mpc8xxx_spi->flags & SPI_QE_CPU_MODE) {
 		mpc8xxx_spi->rx_shift = 16;
 		mpc8xxx_spi->tx_shift = 24;
 	}
 
-	init_completion(&mpc8xxx_spi->done);
-
-	mpc8xxx_spi->base = ioremap(mem->start, resource_size(mem));
-	if (mpc8xxx_spi->base == NULL) {
+	mpc8xxx_spi->reg_base = ioremap(mem->start, resource_size(mem));
+	if (mpc8xxx_spi->reg_base == NULL) {
 		ret = -ENOMEM;
 		goto err_ioremap;
 	}
 
-	mpc8xxx_spi->irq = irq;
-
 	/* Register for SPI Interrupt */
-	ret = request_irq(mpc8xxx_spi->irq, mpc8xxx_spi_irq,
-			  0, "mpc8xxx_spi", mpc8xxx_spi);
+	ret = request_irq(mpc8xxx_spi->irq, fsl_spi_irq,
+			  0, "fsl_spi", mpc8xxx_spi);
 
 	if (ret != 0)
-		goto unmap_io;
+		goto free_irq;
 
-	master->bus_num = pdata->bus_num;
-	master->num_chipselect = pdata->max_chipselect;
+	reg_base = mpc8xxx_spi->reg_base;
 
 	/* SPI controller initializations */
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->mode, 0);
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->mask, 0);
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->command, 0);
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->event, 0xffffffff);
+	mpc8xxx_spi_write_reg(&reg_base->mode, 0);
+	mpc8xxx_spi_write_reg(&reg_base->mask, 0);
+	mpc8xxx_spi_write_reg(&reg_base->command, 0);
+	mpc8xxx_spi_write_reg(&reg_base->event, 0xffffffff);
 
 	/* Enable SPI interface */
 	regval = pdata->initial_spmode | SPMODE_INIT_VAL | SPMODE_ENABLE;
 	if (mpc8xxx_spi->flags & SPI_QE_CPU_MODE)
 		regval |= SPMODE_OP;
 
-	mpc8xxx_spi_write_reg(&mpc8xxx_spi->base->mode, regval);
-	spin_lock_init(&mpc8xxx_spi->lock);
-	init_completion(&mpc8xxx_spi->done);
-	INIT_WORK(&mpc8xxx_spi->work, mpc8xxx_spi_work);
-	INIT_LIST_HEAD(&mpc8xxx_spi->queue);
-
-	mpc8xxx_spi->workqueue = create_singlethread_workqueue(
-		dev_name(master->dev.parent));
-	if (mpc8xxx_spi->workqueue == NULL) {
-		ret = -EBUSY;
-		goto free_irq;
-	}
+	mpc8xxx_spi_write_reg(&reg_base->mode, regval);
 
 	ret = spi_register_master(master);
 	if (ret < 0)
 		goto unreg_master;
 
-	dev_info(dev, "at 0x%p (irq = %d), %s mode\n", mpc8xxx_spi->base,
+	dev_info(dev, "at 0x%p (irq = %d), %s mode\n", reg_base,
 		 mpc8xxx_spi->irq, mpc8xxx_spi_strmode(mpc8xxx_spi->flags));
 
 	return master;
 
 unreg_master:
-	destroy_workqueue(mpc8xxx_spi->workqueue);
-free_irq:
 	free_irq(mpc8xxx_spi->irq, mpc8xxx_spi);
-unmap_io:
-	iounmap(mpc8xxx_spi->base);
+free_irq:
+	iounmap(mpc8xxx_spi->reg_base);
 err_ioremap:
-	mpc8xxx_spi_cpm_free(mpc8xxx_spi);
+	fsl_spi_cpm_free(mpc8xxx_spi);
 err_cpm_init:
+err_probe:
 	spi_master_put(master);
 err:
 	return ERR_PTR(ret);
 }
 
-static int __devexit mpc8xxx_spi_remove(struct device *dev)
-{
-	struct mpc8xxx_spi *mpc8xxx_spi;
-	struct spi_master *master;
-
-	master = dev_get_drvdata(dev);
-	mpc8xxx_spi = spi_master_get_devdata(master);
-
-	flush_workqueue(mpc8xxx_spi->workqueue);
-	destroy_workqueue(mpc8xxx_spi->workqueue);
-	spi_unregister_master(master);
-
-	free_irq(mpc8xxx_spi->irq, mpc8xxx_spi);
-	iounmap(mpc8xxx_spi->base);
-	mpc8xxx_spi_cpm_free(mpc8xxx_spi);
-
-	return 0;
-}
-
-struct mpc8xxx_spi_probe_info {
-	struct fsl_spi_platform_data pdata;
-	int *gpios;
-	bool *alow_flags;
-};
-
-static struct mpc8xxx_spi_probe_info *
-to_of_pinfo(struct fsl_spi_platform_data *pdata)
-{
-	return container_of(pdata, struct mpc8xxx_spi_probe_info, pdata);
-}
-
-static void mpc8xxx_spi_cs_control(struct spi_device *spi, bool on)
+static void fsl_spi_cs_control(struct spi_device *spi, bool on)
 {
 	struct device *dev = spi->dev.parent;
 	struct mpc8xxx_spi_probe_info *pinfo = to_of_pinfo(dev->platform_data);
@@ -1143,7 +943,7 @@
 	gpio_set_value(gpio, on ^ alow);
 }
 
-static int of_mpc8xxx_spi_get_chipselects(struct device *dev)
+static int of_fsl_spi_get_chipselects(struct device *dev)
 {
 	struct device_node *np = dev->of_node;
 	struct fsl_spi_platform_data *pdata = dev->platform_data;
@@ -1204,7 +1004,7 @@
 	}
 
 	pdata->max_chipselect = ngpios;
-	pdata->cs_control = mpc8xxx_spi_cs_control;
+	pdata->cs_control = fsl_spi_cs_control;
 
 	return 0;
 
@@ -1223,7 +1023,7 @@
 	return ret;
 }
 
-static int of_mpc8xxx_spi_free_chipselects(struct device *dev)
+static int of_fsl_spi_free_chipselects(struct device *dev)
 {
 	struct fsl_spi_platform_data *pdata = dev->platform_data;
 	struct mpc8xxx_spi_probe_info *pinfo = to_of_pinfo(pdata);
@@ -1242,50 +1042,21 @@
 	return 0;
 }
 
-static int __devinit of_mpc8xxx_spi_probe(struct platform_device *ofdev,
-					  const struct of_device_id *ofid)
+static int __devinit of_fsl_spi_probe(struct platform_device *ofdev,
+					const struct of_device_id *ofid)
 {
 	struct device *dev = &ofdev->dev;
 	struct device_node *np = ofdev->dev.of_node;
-	struct mpc8xxx_spi_probe_info *pinfo;
-	struct fsl_spi_platform_data *pdata;
 	struct spi_master *master;
 	struct resource mem;
 	struct resource irq;
-	const void *prop;
 	int ret = -ENOMEM;
 
-	pinfo = kzalloc(sizeof(*pinfo), GFP_KERNEL);
-	if (!pinfo)
-		return -ENOMEM;
+	ret = of_mpc8xxx_spi_probe(ofdev, ofid);
+	if (ret)
+		return ret;
 
-	pdata = &pinfo->pdata;
-	dev->platform_data = pdata;
-
-	/* Allocate bus num dynamically. */
-	pdata->bus_num = -1;
-
-	/* SPI controller is either clocked from QE or SoC clock. */
-	pdata->sysclk = get_brgfreq();
-	if (pdata->sysclk == -1) {
-		pdata->sysclk = fsl_get_sys_freq();
-		if (pdata->sysclk == -1) {
-			ret = -ENODEV;
-			goto err_clk;
-		}
-	}
-
-	prop = of_get_property(np, "mode", NULL);
-	if (prop && !strcmp(prop, "cpu-qe"))
-		pdata->flags = SPI_QE_CPU_MODE;
-	else if (prop && !strcmp(prop, "qe"))
-		pdata->flags = SPI_CPM_MODE | SPI_QE;
-	else if (of_device_is_compatible(np, "fsl,cpm2-spi"))
-		pdata->flags = SPI_CPM_MODE | SPI_CPM2;
-	else if (of_device_is_compatible(np, "fsl,cpm1-spi"))
-		pdata->flags = SPI_CPM_MODE | SPI_CPM1;
-
-	ret = of_mpc8xxx_spi_get_chipselects(dev);
+	ret = of_fsl_spi_get_chipselects(dev);
 	if (ret)
 		goto err;
 
@@ -1299,7 +1070,7 @@
 		goto err;
 	}
 
-	master = mpc8xxx_spi_probe(dev, &mem, irq.start);
+	master = fsl_spi_probe(dev, &mem, irq.start);
 	if (IS_ERR(master)) {
 		ret = PTR_ERR(master);
 		goto err;
@@ -1308,42 +1079,40 @@
 	return 0;
 
 err:
-	of_mpc8xxx_spi_free_chipselects(dev);
-err_clk:
-	kfree(pinfo);
+	of_fsl_spi_free_chipselects(dev);
 	return ret;
 }
 
-static int __devexit of_mpc8xxx_spi_remove(struct platform_device *ofdev)
+static int __devexit of_fsl_spi_remove(struct platform_device *ofdev)
 {
 	int ret;
 
 	ret = mpc8xxx_spi_remove(&ofdev->dev);
 	if (ret)
 		return ret;
-	of_mpc8xxx_spi_free_chipselects(&ofdev->dev);
+	of_fsl_spi_free_chipselects(&ofdev->dev);
 	return 0;
 }
 
-static const struct of_device_id of_mpc8xxx_spi_match[] = {
+static const struct of_device_id of_fsl_spi_match[] = {
 	{ .compatible = "fsl,spi" },
-	{},
+	{}
 };
-MODULE_DEVICE_TABLE(of, of_mpc8xxx_spi_match);
+MODULE_DEVICE_TABLE(of, of_fsl_spi_match);
 
-static struct of_platform_driver of_mpc8xxx_spi_driver = {
+static struct of_platform_driver of_fsl_spi_driver = {
 	.driver = {
-		.name = "mpc8xxx_spi",
+		.name = "fsl_spi",
 		.owner = THIS_MODULE,
-		.of_match_table = of_mpc8xxx_spi_match,
+		.of_match_table = of_fsl_spi_match,
 	},
-	.probe		= of_mpc8xxx_spi_probe,
-	.remove		= __devexit_p(of_mpc8xxx_spi_remove),
+	.probe		= of_fsl_spi_probe,
+	.remove		= __devexit_p(of_fsl_spi_remove),
 };
 
 #ifdef CONFIG_MPC832x_RDB
 /*
- * 				XXX XXX XXX
+ * XXX XXX XXX
  * This is "legacy" platform driver, was used by the MPC8323E-RDB boards
  * only. The driver should go away soon, since newer MPC8323E-RDB's device
  * tree can work with OpenFirmware driver. But for now we support old trees
@@ -1366,7 +1135,7 @@
 	if (irq <= 0)
 		return -EINVAL;
 
-	master = mpc8xxx_spi_probe(&pdev->dev, mem, irq);
+	master = fsl_spi_probe(&pdev->dev, mem, irq);
 	if (IS_ERR(master))
 		return PTR_ERR(master);
 	return 0;
@@ -1405,21 +1174,20 @@
 static void __exit legacy_driver_unregister(void) {}
 #endif /* CONFIG_MPC832x_RDB */
 
-static int __init mpc8xxx_spi_init(void)
+static int __init fsl_spi_init(void)
 {
 	legacy_driver_register();
-	return of_register_platform_driver(&of_mpc8xxx_spi_driver);
+	return of_register_platform_driver(&of_fsl_spi_driver);
 }
+module_init(fsl_spi_init);
 
-static void __exit mpc8xxx_spi_exit(void)
+static void __exit fsl_spi_exit(void)
 {
-	of_unregister_platform_driver(&of_mpc8xxx_spi_driver);
+	of_unregister_platform_driver(&of_fsl_spi_driver);
 	legacy_driver_unregister();
 }
-
-module_init(mpc8xxx_spi_init);
-module_exit(mpc8xxx_spi_exit);
+module_exit(fsl_spi_exit);
 
 MODULE_AUTHOR("Kumar Gala");
-MODULE_DESCRIPTION("Simple MPC8xxx SPI Driver");
+MODULE_DESCRIPTION("Simple Freescale SPI Driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi_s3c64xx.c b/drivers/spi/spi_s3c64xx.c
index c3038da..795828b 100644
--- a/drivers/spi/spi_s3c64xx.c
+++ b/drivers/spi/spi_s3c64xx.c
@@ -261,15 +261,25 @@
 		chcfg |= S3C64XX_SPI_CH_TXCH_ON;
 		if (dma_mode) {
 			modecfg |= S3C64XX_SPI_MODE_TXDMA_ON;
-			s3c2410_dma_config(sdd->tx_dmach, 1);
+			s3c2410_dma_config(sdd->tx_dmach, sdd->cur_bpw / 8);
 			s3c2410_dma_enqueue(sdd->tx_dmach, (void *)sdd,
 						xfer->tx_dma, xfer->len);
 			s3c2410_dma_ctrl(sdd->tx_dmach, S3C2410_DMAOP_START);
 		} else {
-			unsigned char *buf = (unsigned char *) xfer->tx_buf;
-			int i = 0;
-			while (i < xfer->len)
-				writeb(buf[i++], regs + S3C64XX_SPI_TX_DATA);
+			switch (sdd->cur_bpw) {
+			case 32:
+				iowrite32_rep(regs + S3C64XX_SPI_TX_DATA,
+					xfer->tx_buf, xfer->len / 4);
+				break;
+			case 16:
+				iowrite16_rep(regs + S3C64XX_SPI_TX_DATA,
+					xfer->tx_buf, xfer->len / 2);
+				break;
+			default:
+				iowrite8_rep(regs + S3C64XX_SPI_TX_DATA,
+					xfer->tx_buf, xfer->len);
+				break;
+			}
 		}
 	}
 
@@ -286,7 +296,7 @@
 			writel(((xfer->len * 8 / sdd->cur_bpw) & 0xffff)
 					| S3C64XX_SPI_PACKET_CNT_EN,
 					regs + S3C64XX_SPI_PACKET_CNT);
-			s3c2410_dma_config(sdd->rx_dmach, 1);
+			s3c2410_dma_config(sdd->rx_dmach, sdd->cur_bpw / 8);
 			s3c2410_dma_enqueue(sdd->rx_dmach, (void *)sdd,
 						xfer->rx_dma, xfer->len);
 			s3c2410_dma_ctrl(sdd->rx_dmach, S3C2410_DMAOP_START);
@@ -366,20 +376,26 @@
 				return -EIO;
 		}
 	} else {
-		unsigned char *buf;
-		int i;
-
 		/* If it was only Tx */
 		if (xfer->rx_buf == NULL) {
 			sdd->state &= ~TXBUSY;
 			return 0;
 		}
 
-		i = 0;
-		buf = xfer->rx_buf;
-		while (i < xfer->len)
-			buf[i++] = readb(regs + S3C64XX_SPI_RX_DATA);
-
+		switch (sdd->cur_bpw) {
+		case 32:
+			ioread32_rep(regs + S3C64XX_SPI_RX_DATA,
+				xfer->rx_buf, xfer->len / 4);
+			break;
+		case 16:
+			ioread16_rep(regs + S3C64XX_SPI_RX_DATA,
+				xfer->rx_buf, xfer->len / 2);
+			break;
+		default:
+			ioread8_rep(regs + S3C64XX_SPI_RX_DATA,
+				xfer->rx_buf, xfer->len);
+			break;
+		}
 		sdd->state &= ~RXBUSY;
 	}
 
@@ -399,13 +415,18 @@
 
 static void s3c64xx_spi_config(struct s3c64xx_spi_driver_data *sdd)
 {
+	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
 	void __iomem *regs = sdd->regs;
 	u32 val;
 
 	/* Disable Clock */
-	val = readl(regs + S3C64XX_SPI_CLK_CFG);
-	val &= ~S3C64XX_SPI_ENCLK_ENABLE;
-	writel(val, regs + S3C64XX_SPI_CLK_CFG);
+	if (sci->clk_from_cmu) {
+		clk_disable(sdd->src_clk);
+	} else {
+		val = readl(regs + S3C64XX_SPI_CLK_CFG);
+		val &= ~S3C64XX_SPI_ENCLK_ENABLE;
+		writel(val, regs + S3C64XX_SPI_CLK_CFG);
+	}
 
 	/* Set Polarity and Phase */
 	val = readl(regs + S3C64XX_SPI_CH_CFG);
@@ -429,29 +450,39 @@
 	switch (sdd->cur_bpw) {
 	case 32:
 		val |= S3C64XX_SPI_MODE_BUS_TSZ_WORD;
+		val |= S3C64XX_SPI_MODE_CH_TSZ_WORD;
 		break;
 	case 16:
 		val |= S3C64XX_SPI_MODE_BUS_TSZ_HALFWORD;
+		val |= S3C64XX_SPI_MODE_CH_TSZ_HALFWORD;
 		break;
 	default:
 		val |= S3C64XX_SPI_MODE_BUS_TSZ_BYTE;
+		val |= S3C64XX_SPI_MODE_CH_TSZ_BYTE;
 		break;
 	}
-	val |= S3C64XX_SPI_MODE_CH_TSZ_BYTE; /* Always 8bits wide */
 
 	writel(val, regs + S3C64XX_SPI_MODE_CFG);
 
-	/* Configure Clock */
-	val = readl(regs + S3C64XX_SPI_CLK_CFG);
-	val &= ~S3C64XX_SPI_PSR_MASK;
-	val |= ((clk_get_rate(sdd->src_clk) / sdd->cur_speed / 2 - 1)
-			& S3C64XX_SPI_PSR_MASK);
-	writel(val, regs + S3C64XX_SPI_CLK_CFG);
+	if (sci->clk_from_cmu) {
+		/* Configure Clock */
+		/* There is half-multiplier before the SPI */
+		clk_set_rate(sdd->src_clk, sdd->cur_speed * 2);
+		/* Enable Clock */
+		clk_enable(sdd->src_clk);
+	} else {
+		/* Configure Clock */
+		val = readl(regs + S3C64XX_SPI_CLK_CFG);
+		val &= ~S3C64XX_SPI_PSR_MASK;
+		val |= ((clk_get_rate(sdd->src_clk) / sdd->cur_speed / 2 - 1)
+				& S3C64XX_SPI_PSR_MASK);
+		writel(val, regs + S3C64XX_SPI_CLK_CFG);
 
-	/* Enable Clock */
-	val = readl(regs + S3C64XX_SPI_CLK_CFG);
-	val |= S3C64XX_SPI_ENCLK_ENABLE;
-	writel(val, regs + S3C64XX_SPI_CLK_CFG);
+		/* Enable Clock */
+		val = readl(regs + S3C64XX_SPI_CLK_CFG);
+		val |= S3C64XX_SPI_ENCLK_ENABLE;
+		writel(val, regs + S3C64XX_SPI_CLK_CFG);
+	}
 }
 
 static void s3c64xx_spi_dma_rxcb(struct s3c2410_dma_chan *chan, void *buf_id,
@@ -499,6 +530,7 @@
 static int s3c64xx_spi_map_mssg(struct s3c64xx_spi_driver_data *sdd,
 						struct spi_message *msg)
 {
+	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
 	struct device *dev = &sdd->pdev->dev;
 	struct spi_transfer *xfer;
 
@@ -514,6 +546,9 @@
 	/* Map until end or first fail */
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
 
+		if (xfer->len <= ((sci->fifo_lvl_mask >> 1) + 1))
+			continue;
+
 		if (xfer->tx_buf != NULL) {
 			xfer->tx_dma = dma_map_single(dev,
 					(void *)xfer->tx_buf, xfer->len,
@@ -545,6 +580,7 @@
 static void s3c64xx_spi_unmap_mssg(struct s3c64xx_spi_driver_data *sdd,
 						struct spi_message *msg)
 {
+	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
 	struct device *dev = &sdd->pdev->dev;
 	struct spi_transfer *xfer;
 
@@ -553,6 +589,9 @@
 
 	list_for_each_entry(xfer, &msg->transfers, transfer_list) {
 
+		if (xfer->len <= ((sci->fifo_lvl_mask >> 1) + 1))
+			continue;
+
 		if (xfer->rx_buf != NULL
 				&& xfer->rx_dma != XFER_DMAADDR_INVALID)
 			dma_unmap_single(dev, xfer->rx_dma,
@@ -608,6 +647,14 @@
 		bpw = xfer->bits_per_word ? : spi->bits_per_word;
 		speed = xfer->speed_hz ? : spi->max_speed_hz;
 
+		if (xfer->len % (bpw / 8)) {
+			dev_err(&spi->dev,
+				"Xfer length(%u) not a multiple of word size(%u)\n",
+				xfer->len, bpw / 8);
+			status = -EIO;
+			goto out;
+		}
+
 		if (bpw != sdd->cur_bpw || speed != sdd->cur_speed) {
 			sdd->cur_bpw = bpw;
 			sdd->cur_speed = speed;
@@ -798,7 +845,6 @@
 	struct s3c64xx_spi_driver_data *sdd;
 	struct s3c64xx_spi_info *sci;
 	struct spi_message *msg;
-	u32 psr, speed;
 	unsigned long flags;
 	int err = 0;
 
@@ -841,31 +887,36 @@
 	}
 
 	/* Check if we can provide the requested rate */
-	speed = clk_get_rate(sdd->src_clk) / 2 / (0 + 1); /* Max possible */
+	if (!sci->clk_from_cmu) {
+		u32 psr, speed;
 
-	if (spi->max_speed_hz > speed)
-		spi->max_speed_hz = speed;
+		/* Max possible */
+		speed = clk_get_rate(sdd->src_clk) / 2 / (0 + 1);
 
-	psr = clk_get_rate(sdd->src_clk) / 2 / spi->max_speed_hz - 1;
-	psr &= S3C64XX_SPI_PSR_MASK;
-	if (psr == S3C64XX_SPI_PSR_MASK)
-		psr--;
+		if (spi->max_speed_hz > speed)
+			spi->max_speed_hz = speed;
 
-	speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
-	if (spi->max_speed_hz < speed) {
-		if (psr+1 < S3C64XX_SPI_PSR_MASK) {
-			psr++;
-		} else {
-			err = -EINVAL;
-			goto setup_exit;
+		psr = clk_get_rate(sdd->src_clk) / 2 / spi->max_speed_hz - 1;
+		psr &= S3C64XX_SPI_PSR_MASK;
+		if (psr == S3C64XX_SPI_PSR_MASK)
+			psr--;
+
+		speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
+		if (spi->max_speed_hz < speed) {
+			if (psr+1 < S3C64XX_SPI_PSR_MASK) {
+				psr++;
+			} else {
+				err = -EINVAL;
+				goto setup_exit;
+			}
 		}
-	}
 
-	speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
-	if (spi->max_speed_hz >= speed)
-		spi->max_speed_hz = speed;
-	else
-		err = -EINVAL;
+		speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
+		if (spi->max_speed_hz >= speed)
+			spi->max_speed_hz = speed;
+		else
+			err = -EINVAL;
+	}
 
 setup_exit:
 
@@ -888,7 +939,8 @@
 	/* Disable Interrupts - we use Polling if not DMA mode */
 	writel(0, regs + S3C64XX_SPI_INT_EN);
 
-	writel(sci->src_clk_nr << S3C64XX_SPI_CLKSEL_SRCSHFT,
+	if (!sci->clk_from_cmu)
+		writel(sci->src_clk_nr << S3C64XX_SPI_CLKSEL_SRCSHFT,
 				regs + S3C64XX_SPI_CLK_CFG);
 	writel(0, regs + S3C64XX_SPI_MODE_CFG);
 	writel(0, regs + S3C64XX_SPI_PACKET_CNT);
diff --git a/drivers/spi/spi_topcliff_pch.c b/drivers/spi/spi_topcliff_pch.c
new file mode 100644
index 0000000..58e187f
--- /dev/null
+++ b/drivers/spi/spi_topcliff_pch.c
@@ -0,0 +1,1303 @@
+/*
+ * SPI bus driver for the Topcliff PCH used by Intel SoCs
+ *
+ * Copyright (C) 2010 OKI SEMICONDUCTOR Co., LTD.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307, USA.
+ */
+
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/wait.h>
+#include <linux/spi/spi.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <linux/spi/spidev.h>
+#include <linux/module.h>
+#include <linux/device.h>
+
+/* Register offsets */
+#define PCH_SPCR		0x00	/* SPI control register */
+#define PCH_SPBRR		0x04	/* SPI baud rate register */
+#define PCH_SPSR		0x08	/* SPI status register */
+#define PCH_SPDWR		0x0C	/* SPI write data register */
+#define PCH_SPDRR		0x10	/* SPI read data register */
+#define PCH_SSNXCR		0x18	/* SSN Expand Control Register */
+#define PCH_SRST		0x1C	/* SPI reset register */
+
+#define PCH_SPSR_TFD		0x000007C0
+#define PCH_SPSR_RFD		0x0000F800
+
+#define PCH_READABLE(x)		(((x) & PCH_SPSR_RFD)>>11)
+#define PCH_WRITABLE(x)		(((x) & PCH_SPSR_TFD)>>6)
+
+#define PCH_RX_THOLD		7
+#define PCH_RX_THOLD_MAX	15
+
+#define PCH_MAX_BAUDRATE	5000000
+#define PCH_MAX_FIFO_DEPTH	16
+
+#define STATUS_RUNNING		1
+#define STATUS_EXITING		2
+#define PCH_SLEEP_TIME		10
+
+#define PCH_ADDRESS_SIZE	0x20
+
+#define SSN_LOW			0x02U
+#define SSN_NO_CONTROL		0x00U
+#define PCH_MAX_CS		0xFF
+#define PCI_DEVICE_ID_GE_SPI	0x8816
+
+#define SPCR_SPE_BIT		(1 << 0)
+#define SPCR_MSTR_BIT		(1 << 1)
+#define SPCR_LSBF_BIT		(1 << 4)
+#define SPCR_CPHA_BIT		(1 << 5)
+#define SPCR_CPOL_BIT		(1 << 6)
+#define SPCR_TFIE_BIT		(1 << 8)
+#define SPCR_RFIE_BIT		(1 << 9)
+#define SPCR_FIE_BIT		(1 << 10)
+#define SPCR_ORIE_BIT		(1 << 11)
+#define SPCR_MDFIE_BIT		(1 << 12)
+#define SPCR_FICLR_BIT		(1 << 24)
+#define SPSR_TFI_BIT		(1 << 0)
+#define SPSR_RFI_BIT		(1 << 1)
+#define SPSR_FI_BIT		(1 << 2)
+#define SPBRR_SIZE_BIT		(1 << 10)
+
+#define PCH_ALL			(SPCR_TFIE_BIT|SPCR_RFIE_BIT|SPCR_FIE_BIT|SPCR_ORIE_BIT|SPCR_MDFIE_BIT)
+
+#define SPCR_RFIC_FIELD		20
+#define SPCR_TFIC_FIELD		16
+
+#define SPSR_INT_BITS		0x1F
+#define MASK_SPBRR_SPBR_BITS	(~((1 << 10) - 1))
+#define MASK_RFIC_SPCR_BITS	(~(0xf << 20))
+#define MASK_TFIC_SPCR_BITS	(~(0xf000f << 12))
+
+#define PCH_CLOCK_HZ		50000000
+#define PCH_MAX_SPBR		1023
+
+
+/**
+ * struct pch_spi_data - Holds the SPI channel specific details
+ * @io_remap_addr:		The remapped PCI base address
+ * @master:			Pointer to the SPI master structure
+ * @work:			Reference to work queue handler
+ * @wk:				Workqueue for carrying out execution of the
+ *				requests
+ * @wait:			Wait queue for waking up upon receiving an
+ *				interrupt.
+ * @transfer_complete:		Status of SPI Transfer
+ * @bcurrent_msg_processing:	Status flag for message processing
+ * @lock:			Lock for protecting this structure
+ * @queue:			SPI Message queue
+ * @status:			Status of the SPI driver
+ * @bpw_len:			Length of data to be transferred in bits per
+ *				word
+ * @transfer_active:		Flag showing active transfer
+ * @tx_index:			Transmit data count; for bookkeeping during
+ *				transfer
+ * @rx_index:			Receive data count; for bookkeeping during
+ *				transfer
+ * @tx_buff:			Buffer for data to be transmitted
+ * @rx_index:			Buffer for Received data
+ * @n_curnt_chip:		The chip number that this SPI driver currently
+ *				operates on
+ * @current_chip:		Reference to the current chip that this SPI
+ *				driver currently operates on
+ * @current_msg:		The current message that this SPI driver is
+ *				handling
+ * @cur_trans:			The current transfer that this SPI driver is
+ *				handling
+ * @board_dat:			Reference to the SPI device data structure
+ */
+struct pch_spi_data {
+	void __iomem *io_remap_addr;
+	struct spi_master *master;
+	struct work_struct work;
+	struct workqueue_struct *wk;
+	wait_queue_head_t wait;
+	u8 transfer_complete;
+	u8 bcurrent_msg_processing;
+	spinlock_t lock;
+	struct list_head queue;
+	u8 status;
+	u32 bpw_len;
+	u8 transfer_active;
+	u32 tx_index;
+	u32 rx_index;
+	u16 *pkt_tx_buff;
+	u16 *pkt_rx_buff;
+	u8 n_curnt_chip;
+	struct spi_device *current_chip;
+	struct spi_message *current_msg;
+	struct spi_transfer *cur_trans;
+	struct pch_spi_board_data *board_dat;
+};
+
+/**
+ * struct pch_spi_board_data - Holds the SPI device specific details
+ * @pdev:		Pointer to the PCI device
+ * @irq_reg_sts:	Status of IRQ registration
+ * @pci_req_sts:	Status of pci_request_regions
+ * @suspend_sts:	Status of suspend
+ * @data:		Pointer to SPI channel data structure
+ */
+struct pch_spi_board_data {
+	struct pci_dev *pdev;
+	u8 irq_reg_sts;
+	u8 pci_req_sts;
+	u8 suspend_sts;
+	struct pch_spi_data *data;
+};
+
+static struct pci_device_id pch_spi_pcidev_id[] = {
+	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_GE_SPI)},
+	{0,}
+};
+
+/**
+ * pch_spi_writereg() - Performs  register writes
+ * @master:	Pointer to struct spi_master.
+ * @idx:	Register offset.
+ * @val:	Value to be written to register.
+ */
+static inline void pch_spi_writereg(struct spi_master *master, int idx, u32 val)
+{
+	struct pch_spi_data *data = spi_master_get_devdata(master);
+	iowrite32(val, (data->io_remap_addr + idx));
+}
+
+/**
+ * pch_spi_readreg() - Performs register reads
+ * @master:	Pointer to struct spi_master.
+ * @idx:	Register offset.
+ */
+static inline u32 pch_spi_readreg(struct spi_master *master, int idx)
+{
+	struct pch_spi_data *data = spi_master_get_devdata(master);
+	return ioread32(data->io_remap_addr + idx);
+}
+
+static inline void pch_spi_setclr_reg(struct spi_master *master, int idx,
+				      u32 set, u32 clr)
+{
+	u32 tmp = pch_spi_readreg(master, idx);
+	tmp = (tmp & ~clr) | set;
+	pch_spi_writereg(master, idx, tmp);
+}
+
+static void pch_spi_set_master_mode(struct spi_master *master)
+{
+	pch_spi_setclr_reg(master, PCH_SPCR, SPCR_MSTR_BIT, 0);
+}
+
+/**
+ * pch_spi_clear_fifo() - Clears the Transmit and Receive FIFOs
+ * @master:	Pointer to struct spi_master.
+ */
+static void pch_spi_clear_fifo(struct spi_master *master)
+{
+	pch_spi_setclr_reg(master, PCH_SPCR, SPCR_FICLR_BIT, 0);
+	pch_spi_setclr_reg(master, PCH_SPCR, 0, SPCR_FICLR_BIT);
+}
+
+static void pch_spi_handler_sub(struct pch_spi_data *data, u32 reg_spsr_val,
+				void __iomem *io_remap_addr)
+{
+	u32 n_read, tx_index, rx_index, bpw_len;
+	u16 *pkt_rx_buffer, *pkt_tx_buff;
+	int read_cnt;
+	u32 reg_spcr_val;
+	void __iomem *spsr;
+	void __iomem *spdrr;
+	void __iomem *spdwr;
+
+	spsr = io_remap_addr + PCH_SPSR;
+	iowrite32(reg_spsr_val, spsr);
+
+	if (data->transfer_active) {
+		rx_index = data->rx_index;
+		tx_index = data->tx_index;
+		bpw_len = data->bpw_len;
+		pkt_rx_buffer = data->pkt_rx_buff;
+		pkt_tx_buff = data->pkt_tx_buff;
+
+		spdrr = io_remap_addr + PCH_SPDRR;
+		spdwr = io_remap_addr + PCH_SPDWR;
+
+		n_read = PCH_READABLE(reg_spsr_val);
+
+		for (read_cnt = 0; (read_cnt < n_read); read_cnt++) {
+			pkt_rx_buffer[rx_index++] = ioread32(spdrr);
+			if (tx_index < bpw_len)
+				iowrite32(pkt_tx_buff[tx_index++], spdwr);
+		}
+
+		/* disable RFI if not needed */
+		if ((bpw_len - rx_index) <= PCH_MAX_FIFO_DEPTH) {
+			reg_spcr_val = ioread32(io_remap_addr + PCH_SPCR);
+			reg_spcr_val &= ~SPCR_RFIE_BIT; /* disable RFI */
+
+			/* reset rx threshold */
+			reg_spcr_val &= MASK_RFIC_SPCR_BITS;
+			reg_spcr_val |= (PCH_RX_THOLD_MAX << SPCR_RFIC_FIELD);
+			iowrite32(((reg_spcr_val) &= (~(SPCR_RFIE_BIT))),
+				 (io_remap_addr + PCH_SPCR));
+		}
+
+		/* update counts */
+		data->tx_index = tx_index;
+		data->rx_index = rx_index;
+
+	}
+
+	/* if transfer complete interrupt */
+	if (reg_spsr_val & SPSR_FI_BIT) {
+		/* disable FI & RFI interrupts */
+		pch_spi_setclr_reg(data->master, PCH_SPCR, 0,
+				   SPCR_FIE_BIT | SPCR_TFIE_BIT);
+
+		/* transfer is completed;inform pch_spi_process_messages */
+		data->transfer_complete = true;
+		wake_up(&data->wait);
+	}
+}
+
+/**
+ * pch_spi_handler() - Interrupt handler
+ * @irq:	The interrupt number.
+ * @dev_id:	Pointer to struct pch_spi_board_data.
+ */
+static irqreturn_t pch_spi_handler(int irq, void *dev_id)
+{
+	u32 reg_spsr_val;
+	struct pch_spi_data *data;
+	void __iomem *spsr;
+	void __iomem *io_remap_addr;
+	irqreturn_t ret = IRQ_NONE;
+	struct pch_spi_board_data *board_dat = dev_id;
+
+	if (board_dat->suspend_sts) {
+		dev_dbg(&board_dat->pdev->dev,
+			"%s returning due to suspend\n", __func__);
+		return IRQ_NONE;
+	}
+
+	data = board_dat->data;
+	io_remap_addr = data->io_remap_addr;
+	spsr = io_remap_addr + PCH_SPSR;
+
+	reg_spsr_val = ioread32(spsr);
+
+	/* Check if the interrupt is for SPI device */
+	if (reg_spsr_val & (SPSR_FI_BIT | SPSR_RFI_BIT)) {
+		pch_spi_handler_sub(data, reg_spsr_val, io_remap_addr);
+		ret = IRQ_HANDLED;
+	}
+
+	dev_dbg(&board_dat->pdev->dev, "%s EXIT return value=%d\n",
+		__func__, ret);
+
+	return ret;
+}
+
+/**
+ * pch_spi_set_baud_rate() - Sets SPBR field in SPBRR
+ * @master:	Pointer to struct spi_master.
+ * @speed_hz:	Baud rate.
+ */
+static void pch_spi_set_baud_rate(struct spi_master *master, u32 speed_hz)
+{
+	u32 n_spbr = PCH_CLOCK_HZ / (speed_hz * 2);
+
+	/* if baud rate is less than we can support limit it */
+	if (n_spbr > PCH_MAX_SPBR)
+		n_spbr = PCH_MAX_SPBR;
+
+	pch_spi_setclr_reg(master, PCH_SPBRR, n_spbr, ~MASK_SPBRR_SPBR_BITS);
+}
+
+/**
+ * pch_spi_set_bits_per_word() - Sets SIZE field in SPBRR
+ * @master:		Pointer to struct spi_master.
+ * @bits_per_word:	Bits per word for SPI transfer.
+ */
+static void pch_spi_set_bits_per_word(struct spi_master *master,
+				      u8 bits_per_word)
+{
+	if (bits_per_word == 8)
+		pch_spi_setclr_reg(master, PCH_SPBRR, 0, SPBRR_SIZE_BIT);
+	else
+		pch_spi_setclr_reg(master, PCH_SPBRR, SPBRR_SIZE_BIT, 0);
+}
+
+/**
+ * pch_spi_setup_transfer() - Configures the PCH SPI hardware for transfer
+ * @spi:	Pointer to struct spi_device.
+ */
+static void pch_spi_setup_transfer(struct spi_device *spi)
+{
+	u32 flags = 0;
+
+	dev_dbg(&spi->dev, "%s SPBRR content =%x setting baud rate=%d\n",
+		__func__, pch_spi_readreg(spi->master, PCH_SPBRR),
+		spi->max_speed_hz);
+	pch_spi_set_baud_rate(spi->master, spi->max_speed_hz);
+
+	/* set bits per word */
+	pch_spi_set_bits_per_word(spi->master, spi->bits_per_word);
+
+	if (!(spi->mode & SPI_LSB_FIRST))
+		flags |= SPCR_LSBF_BIT;
+	if (spi->mode & SPI_CPOL)
+		flags |= SPCR_CPOL_BIT;
+	if (spi->mode & SPI_CPHA)
+		flags |= SPCR_CPHA_BIT;
+	pch_spi_setclr_reg(spi->master, PCH_SPCR, flags,
+			   (SPCR_LSBF_BIT | SPCR_CPOL_BIT | SPCR_CPHA_BIT));
+
+	/* Clear the FIFO by toggling  FICLR to 1 and back to 0 */
+	pch_spi_clear_fifo(spi->master);
+}
+
+/**
+ * pch_spi_reset() - Clears SPI registers
+ * @master:	Pointer to struct spi_master.
+ */
+static void pch_spi_reset(struct spi_master *master)
+{
+	/* write 1 to reset SPI */
+	pch_spi_writereg(master, PCH_SRST, 0x1);
+
+	/* clear reset */
+	pch_spi_writereg(master, PCH_SRST, 0x0);
+}
+
+static int pch_spi_setup(struct spi_device *pspi)
+{
+	/* check bits per word */
+	if (pspi->bits_per_word == 0) {
+		pspi->bits_per_word = 8;
+		dev_dbg(&pspi->dev, "%s 8 bits per word\n", __func__);
+	}
+
+	if ((pspi->bits_per_word != 8) && (pspi->bits_per_word != 16)) {
+		dev_err(&pspi->dev, "%s Invalid bits per word\n", __func__);
+		return -EINVAL;
+	}
+
+	/* Check baud rate setting */
+	/* if baud rate of chip is greater than
+	   max we can support,return error */
+	if ((pspi->max_speed_hz) > PCH_MAX_BAUDRATE)
+		pspi->max_speed_hz = PCH_MAX_BAUDRATE;
+
+	dev_dbg(&pspi->dev, "%s MODE = %x\n", __func__,
+		(pspi->mode) & (SPI_CPOL | SPI_CPHA));
+
+	return 0;
+}
+
+static int pch_spi_transfer(struct spi_device *pspi, struct spi_message *pmsg)
+{
+
+	struct spi_transfer *transfer;
+	struct pch_spi_data *data = spi_master_get_devdata(pspi->master);
+	int retval;
+	unsigned long flags;
+
+	/* validate spi message and baud rate */
+	if (unlikely(list_empty(&pmsg->transfers) == 1)) {
+		dev_err(&pspi->dev, "%s list empty\n", __func__);
+		retval = -EINVAL;
+		goto err_out;
+	}
+
+	if (unlikely(pspi->max_speed_hz == 0)) {
+		dev_err(&pspi->dev, "%s pch_spi_tranfer maxspeed=%d\n",
+			__func__, pspi->max_speed_hz);
+		retval = -EINVAL;
+		goto err_out;
+	}
+
+	dev_dbg(&pspi->dev, "%s Transfer List not empty. "
+		"Transfer Speed is set.\n", __func__);
+
+	/* validate Tx/Rx buffers and Transfer length */
+	list_for_each_entry(transfer, &pmsg->transfers, transfer_list) {
+		if (!transfer->tx_buf && !transfer->rx_buf) {
+			dev_err(&pspi->dev,
+				"%s Tx and Rx buffer NULL\n", __func__);
+			retval = -EINVAL;
+			goto err_out;
+		}
+
+		if (!transfer->len) {
+			dev_err(&pspi->dev, "%s Transfer length invalid\n",
+				__func__);
+			retval = -EINVAL;
+			goto err_out;
+		}
+
+		dev_dbg(&pspi->dev, "%s Tx/Rx buffer valid. Transfer length"
+			" valid\n", __func__);
+
+		/* if baud rate hs been specified validate the same */
+		if (transfer->speed_hz > PCH_MAX_BAUDRATE)
+			transfer->speed_hz = PCH_MAX_BAUDRATE;
+
+		/* if bits per word has been specified validate the same */
+		if (transfer->bits_per_word) {
+			if ((transfer->bits_per_word != 8)
+			    && (transfer->bits_per_word != 16)) {
+				retval = -EINVAL;
+				dev_err(&pspi->dev,
+					"%s Invalid bits per word\n", __func__);
+				goto err_out;
+			}
+		}
+	}
+
+	spin_lock_irqsave(&data->lock, flags);
+
+	/* We won't process any messages if we have been asked to terminate */
+	if (data->status == STATUS_EXITING) {
+		dev_err(&pspi->dev, "%s status = STATUS_EXITING.\n", __func__);
+		retval = -ESHUTDOWN;
+		goto err_return_spinlock;
+	}
+
+	/* If suspended ,return -EINVAL */
+	if (data->board_dat->suspend_sts) {
+		dev_err(&pspi->dev, "%s suspend; returning EINVAL\n", __func__);
+		retval = -EINVAL;
+		goto err_return_spinlock;
+	}
+
+	/* set status of message */
+	pmsg->actual_length = 0;
+	dev_dbg(&pspi->dev, "%s - pmsg->status =%d\n", __func__, pmsg->status);
+
+	pmsg->status = -EINPROGRESS;
+
+	/* add message to queue */
+	list_add_tail(&pmsg->queue, &data->queue);
+	dev_dbg(&pspi->dev, "%s - Invoked list_add_tail\n", __func__);
+
+	/* schedule work queue to run */
+	queue_work(data->wk, &data->work);
+	dev_dbg(&pspi->dev, "%s - Invoked queue work\n", __func__);
+
+	retval = 0;
+
+err_return_spinlock:
+	spin_unlock_irqrestore(&data->lock, flags);
+err_out:
+	dev_dbg(&pspi->dev, "%s RETURN=%d\n", __func__, retval);
+	return retval;
+}
+
+static inline void pch_spi_select_chip(struct pch_spi_data *data,
+				       struct spi_device *pspi)
+{
+	if (data->current_chip != NULL) {
+		if (pspi->chip_select != data->n_curnt_chip) {
+			dev_dbg(&pspi->dev, "%s : different slave\n", __func__);
+			data->current_chip = NULL;
+		}
+	}
+
+	data->current_chip = pspi;
+
+	data->n_curnt_chip = data->current_chip->chip_select;
+
+	dev_dbg(&pspi->dev, "%s :Invoking pch_spi_setup_transfer\n", __func__);
+	pch_spi_setup_transfer(pspi);
+}
+
+static void pch_spi_set_tx(struct pch_spi_data *data, int *bpw,
+			   struct spi_message **ppmsg)
+{
+	int size;
+	u32 n_writes;
+	int j;
+	struct spi_message *pmsg;
+	const u8 *tx_buf;
+	const u16 *tx_sbuf;
+
+	pmsg = *ppmsg;
+
+	/* set baud rate if needed */
+	if (data->cur_trans->speed_hz) {
+		dev_dbg(&data->master->dev, "%s:setting baud rate\n", __func__);
+		pch_spi_set_baud_rate(data->master, data->cur_trans->speed_hz);
+	}
+
+	/* set bits per word if needed */
+	if (data->cur_trans->bits_per_word &&
+	    (data->current_msg->spi->bits_per_word != data->cur_trans->bits_per_word)) {
+		dev_dbg(&data->master->dev, "%s:set bits per word\n", __func__);
+		pch_spi_set_bits_per_word(data->master,
+					  data->cur_trans->bits_per_word);
+		*bpw = data->cur_trans->bits_per_word;
+	} else {
+		*bpw = data->current_msg->spi->bits_per_word;
+	}
+
+	/* reset Tx/Rx index */
+	data->tx_index = 0;
+	data->rx_index = 0;
+
+	data->bpw_len = data->cur_trans->len / (*bpw / 8);
+
+	/* find alloc size */
+	size = data->cur_trans->len * sizeof(*data->pkt_tx_buff);
+
+	/* allocate memory for pkt_tx_buff & pkt_rx_buffer */
+	data->pkt_tx_buff = kzalloc(size, GFP_KERNEL);
+	if (data->pkt_tx_buff != NULL) {
+		data->pkt_rx_buff = kzalloc(size, GFP_KERNEL);
+		if (!data->pkt_rx_buff)
+			kfree(data->pkt_tx_buff);
+	}
+
+	if (!data->pkt_rx_buff) {
+		/* flush queue and set status of all transfers to -ENOMEM */
+		dev_err(&data->master->dev, "%s :kzalloc failed\n", __func__);
+		list_for_each_entry(pmsg, data->queue.next, queue) {
+			pmsg->status = -ENOMEM;
+
+			if (pmsg->complete != 0)
+				pmsg->complete(pmsg->context);
+
+			/* delete from queue */
+			list_del_init(&pmsg->queue);
+		}
+		return;
+	}
+
+	/* copy Tx Data */
+	if (data->cur_trans->tx_buf != NULL) {
+		if (*bpw == 8) {
+			tx_buf = data->cur_trans->tx_buf;
+			for (j = 0; j < data->bpw_len; j++)
+				data->pkt_tx_buff[j] = *tx_buf++;
+		} else {
+			tx_sbuf = data->cur_trans->tx_buf;
+			for (j = 0; j < data->bpw_len; j++)
+				data->pkt_tx_buff[j] = *tx_sbuf++;
+		}
+	}
+
+	/* if len greater than PCH_MAX_FIFO_DEPTH, write 16,else len bytes */
+	n_writes = data->bpw_len;
+	if (n_writes > PCH_MAX_FIFO_DEPTH)
+		n_writes = PCH_MAX_FIFO_DEPTH;
+
+	dev_dbg(&data->master->dev, "\n%s:Pulling down SSN low - writing "
+		"0x2 to SSNXCR\n", __func__);
+	pch_spi_writereg(data->master, PCH_SSNXCR, SSN_LOW);
+
+	for (j = 0; j < n_writes; j++)
+		pch_spi_writereg(data->master, PCH_SPDWR, data->pkt_tx_buff[j]);
+
+	/* update tx_index */
+	data->tx_index = j;
+
+	/* reset transfer complete flag */
+	data->transfer_complete = false;
+	data->transfer_active = true;
+}
+
+
+static void pch_spi_nomore_transfer(struct pch_spi_data *data,
+						struct spi_message *pmsg)
+{
+	dev_dbg(&data->master->dev, "%s called\n", __func__);
+	/* Invoke complete callback
+	 * [To the spi core..indicating end of transfer] */
+	data->current_msg->status = 0;
+
+	if (data->current_msg->complete != 0) {
+		dev_dbg(&data->master->dev,
+			"%s:Invoking callback of SPI core\n", __func__);
+		data->current_msg->complete(data->current_msg->context);
+	}
+
+	/* update status in global variable */
+	data->bcurrent_msg_processing = false;
+
+	dev_dbg(&data->master->dev,
+		"%s:data->bcurrent_msg_processing = false\n", __func__);
+
+	data->current_msg = NULL;
+	data->cur_trans = NULL;
+
+	/* check if we have items in list and not suspending
+	 * return 1 if list empty */
+	if ((list_empty(&data->queue) == 0) &&
+	    (!data->board_dat->suspend_sts) &&
+	    (data->status != STATUS_EXITING)) {
+		/* We have some more work to do (either there is more tranint
+		 * bpw;sfer requests in the current message or there are
+		 *more messages)
+		 */
+		dev_dbg(&data->master->dev, "%s:Invoke queue_work\n", __func__);
+		queue_work(data->wk, &data->work);
+	} else if (data->board_dat->suspend_sts ||
+		   data->status == STATUS_EXITING) {
+		dev_dbg(&data->master->dev,
+			"%s suspend/remove initiated, flushing queue\n",
+			__func__);
+		list_for_each_entry(pmsg, data->queue.next, queue) {
+			pmsg->status = -EIO;
+
+			if (pmsg->complete)
+				pmsg->complete(pmsg->context);
+
+			/* delete from queue */
+			list_del_init(&pmsg->queue);
+		}
+	}
+}
+
+static void pch_spi_set_ir(struct pch_spi_data *data)
+{
+	/* enable interrupts */
+	if ((data->bpw_len) > PCH_MAX_FIFO_DEPTH) {
+		/* set receive threhold to PCH_RX_THOLD */
+		pch_spi_setclr_reg(data->master, PCH_SPCR,
+				   PCH_RX_THOLD << SPCR_TFIC_FIELD,
+				   ~MASK_TFIC_SPCR_BITS);
+		/* enable FI and RFI interrupts */
+		pch_spi_setclr_reg(data->master, PCH_SPCR,
+				   SPCR_RFIE_BIT | SPCR_TFIE_BIT, 0);
+	} else {
+		/* set receive threhold to maximum */
+		pch_spi_setclr_reg(data->master, PCH_SPCR,
+				   PCH_RX_THOLD_MAX << SPCR_TFIC_FIELD,
+				   ~MASK_TFIC_SPCR_BITS);
+		/* enable FI interrupt */
+		pch_spi_setclr_reg(data->master, PCH_SPCR, SPCR_FIE_BIT, 0);
+	}
+
+	dev_dbg(&data->master->dev,
+		"%s:invoking pch_spi_set_enable to enable SPI\n", __func__);
+
+	/* SPI set enable */
+	pch_spi_setclr_reg(data->current_chip->master, PCH_SPCR, SPCR_SPE_BIT, 0);
+
+	/* Wait until the transfer completes; go to sleep after
+				 initiating the transfer. */
+	dev_dbg(&data->master->dev,
+		"%s:waiting for transfer to get over\n", __func__);
+
+	wait_event_interruptible(data->wait, data->transfer_complete);
+
+	pch_spi_writereg(data->master, PCH_SSNXCR, SSN_NO_CONTROL);
+	dev_dbg(&data->master->dev,
+		"%s:no more control over SSN-writing 0 to SSNXCR.", __func__);
+
+	data->transfer_active = false;
+	dev_dbg(&data->master->dev,
+		"%s set data->transfer_active = false\n", __func__);
+
+	/* clear all interrupts */
+	pch_spi_writereg(data->master, PCH_SPSR,
+			 pch_spi_readreg(data->master, PCH_SPSR));
+	/* disable interrupts */
+	pch_spi_setclr_reg(data->master, PCH_SPCR, 0, PCH_ALL);
+}
+
+static void pch_spi_copy_rx_data(struct pch_spi_data *data, int bpw)
+{
+	int j;
+	u8 *rx_buf;
+	u16 *rx_sbuf;
+
+	/* copy Rx Data */
+	if (!data->cur_trans->rx_buf)
+		return;
+
+	if (bpw == 8) {
+		rx_buf = data->cur_trans->rx_buf;
+		for (j = 0; j < data->bpw_len; j++)
+			*rx_buf++ = data->pkt_rx_buff[j] & 0xFF;
+	} else {
+		rx_sbuf = data->cur_trans->rx_buf;
+		for (j = 0; j < data->bpw_len; j++)
+			*rx_sbuf++ = data->pkt_rx_buff[j];
+	}
+}
+
+
+static void pch_spi_process_messages(struct work_struct *pwork)
+{
+	struct spi_message *pmsg;
+	struct pch_spi_data *data;
+	int bpw;
+
+	data = container_of(pwork, struct pch_spi_data, work);
+	dev_dbg(&data->master->dev, "%s data initialized\n", __func__);
+
+	spin_lock(&data->lock);
+
+	/* check if suspend has been initiated;if yes flush queue */
+	if (data->board_dat->suspend_sts || (data->status == STATUS_EXITING)) {
+		dev_dbg(&data->master->dev,
+			"%s suspend/remove initiated,flushing queue\n",
+			__func__);
+
+		list_for_each_entry(pmsg, data->queue.next, queue) {
+			pmsg->status = -EIO;
+
+			if (pmsg->complete != 0) {
+				spin_unlock(&data->lock);
+				pmsg->complete(pmsg->context);
+				spin_lock(&data->lock);
+			}
+
+			/* delete from queue */
+			list_del_init(&pmsg->queue);
+		}
+
+		spin_unlock(&data->lock);
+		return;
+	}
+
+	data->bcurrent_msg_processing = true;
+	dev_dbg(&data->master->dev,
+		"%s Set data->bcurrent_msg_processing= true\n", __func__);
+
+	/* Get the message from the queue and delete it from there. */
+	data->current_msg = list_entry(data->queue.next, struct spi_message,
+					queue);
+
+	list_del_init(&data->current_msg->queue);
+
+	data->current_msg->status = 0;
+
+	pch_spi_select_chip(data, data->current_msg->spi);
+
+	spin_unlock(&data->lock);
+
+	do {
+		/* If we are already processing a message get the next
+		transfer structure from the message otherwise retrieve
+		the 1st transfer request from the message. */
+		spin_lock(&data->lock);
+
+		if (data->cur_trans == NULL) {
+			data->cur_trans =
+			    list_entry(data->current_msg->transfers.
+				       next, struct spi_transfer,
+				       transfer_list);
+			dev_dbg(&data->master->dev,
+				"%s :Getting 1st transfer message\n", __func__);
+		} else {
+			data->cur_trans =
+			    list_entry(data->cur_trans->transfer_list.next,
+				       struct spi_transfer,
+				       transfer_list);
+			dev_dbg(&data->master->dev,
+				"%s :Getting next transfer message\n",
+				__func__);
+		}
+
+		spin_unlock(&data->lock);
+
+		pch_spi_set_tx(data, &bpw, &pmsg);
+
+		/* Control interrupt*/
+		pch_spi_set_ir(data);
+
+		/* Disable SPI transfer */
+		pch_spi_setclr_reg(data->current_chip->master, PCH_SPCR, 0,
+				   SPCR_SPE_BIT);
+
+		/* clear FIFO */
+		pch_spi_clear_fifo(data->master);
+
+		/* copy Rx Data */
+		pch_spi_copy_rx_data(data, bpw);
+
+		/* free memory */
+		kfree(data->pkt_rx_buff);
+		data->pkt_rx_buff = NULL;
+
+		kfree(data->pkt_tx_buff);
+		data->pkt_tx_buff = NULL;
+
+		/* increment message count */
+		data->current_msg->actual_length += data->cur_trans->len;
+
+		dev_dbg(&data->master->dev,
+			"%s:data->current_msg->actual_length=%d\n",
+			__func__, data->current_msg->actual_length);
+
+		/* check for delay */
+		if (data->cur_trans->delay_usecs) {
+			dev_dbg(&data->master->dev, "%s:"
+				"delay in usec=%d\n", __func__,
+				data->cur_trans->delay_usecs);
+			udelay(data->cur_trans->delay_usecs);
+		}
+
+		spin_lock(&data->lock);
+
+		/* No more transfer in this message. */
+		if ((data->cur_trans->transfer_list.next) ==
+		    &(data->current_msg->transfers)) {
+			pch_spi_nomore_transfer(data, pmsg);
+		}
+
+		spin_unlock(&data->lock);
+
+	} while (data->cur_trans != NULL);
+}
+
+static void pch_spi_free_resources(struct pch_spi_board_data *board_dat)
+{
+	dev_dbg(&board_dat->pdev->dev, "%s ENTRY\n", __func__);
+
+	/* free workqueue */
+	if (board_dat->data->wk != NULL) {
+		destroy_workqueue(board_dat->data->wk);
+		board_dat->data->wk = NULL;
+		dev_dbg(&board_dat->pdev->dev,
+			"%s destroy_workqueue invoked successfully\n",
+			__func__);
+	}
+
+	/* disable interrupts & free IRQ */
+	if (board_dat->irq_reg_sts) {
+		/* disable interrupts */
+		pch_spi_setclr_reg(board_dat->data->master, PCH_SPCR, 0,
+				   PCH_ALL);
+
+		/* free IRQ */
+		free_irq(board_dat->pdev->irq, board_dat);
+
+		dev_dbg(&board_dat->pdev->dev,
+			"%s free_irq invoked successfully\n", __func__);
+
+		board_dat->irq_reg_sts = false;
+	}
+
+	/* unmap PCI base address */
+	if (board_dat->data->io_remap_addr != 0) {
+		pci_iounmap(board_dat->pdev, board_dat->data->io_remap_addr);
+
+		board_dat->data->io_remap_addr = 0;
+
+		dev_dbg(&board_dat->pdev->dev,
+			"%s pci_iounmap invoked successfully\n", __func__);
+	}
+
+	/* release PCI region */
+	if (board_dat->pci_req_sts) {
+		pci_release_regions(board_dat->pdev);
+		dev_dbg(&board_dat->pdev->dev,
+			"%s pci_release_regions invoked successfully\n",
+			__func__);
+		board_dat->pci_req_sts = false;
+	}
+}
+
+static int pch_spi_get_resources(struct pch_spi_board_data *board_dat)
+{
+	void __iomem *io_remap_addr;
+	int retval;
+	dev_dbg(&board_dat->pdev->dev, "%s ENTRY\n", __func__);
+
+	/* create workqueue */
+	board_dat->data->wk = create_singlethread_workqueue(KBUILD_MODNAME);
+	if (!board_dat->data->wk) {
+		dev_err(&board_dat->pdev->dev,
+			"%s create_singlet hread_workqueue failed\n", __func__);
+		retval = -EBUSY;
+		goto err_return;
+	}
+
+	dev_dbg(&board_dat->pdev->dev,
+		"%s create_singlethread_workqueue success\n", __func__);
+
+	retval = pci_request_regions(board_dat->pdev, KBUILD_MODNAME);
+	if (retval != 0) {
+		dev_err(&board_dat->pdev->dev,
+			"%s request_region failed\n", __func__);
+		goto err_return;
+	}
+
+	board_dat->pci_req_sts = true;
+
+	io_remap_addr = pci_iomap(board_dat->pdev, 1, 0);
+	if (io_remap_addr == 0) {
+		dev_err(&board_dat->pdev->dev,
+			"%s pci_iomap failed\n", __func__);
+		retval = -ENOMEM;
+		goto err_return;
+	}
+
+	/* calculate base address for all channels */
+	board_dat->data->io_remap_addr = io_remap_addr;
+
+	/* reset PCH SPI h/w */
+	pch_spi_reset(board_dat->data->master);
+	dev_dbg(&board_dat->pdev->dev,
+		"%s pch_spi_reset invoked successfully\n", __func__);
+
+	/* register IRQ */
+	retval = request_irq(board_dat->pdev->irq, pch_spi_handler,
+			     IRQF_SHARED, KBUILD_MODNAME, board_dat);
+	if (retval != 0) {
+		dev_err(&board_dat->pdev->dev,
+			"%s request_irq failed\n", __func__);
+		goto err_return;
+	}
+
+	dev_dbg(&board_dat->pdev->dev, "%s request_irq returned=%d\n",
+		__func__, retval);
+
+	board_dat->irq_reg_sts = true;
+	dev_dbg(&board_dat->pdev->dev, "%s data->irq_reg_sts=true\n", __func__);
+
+err_return:
+	if (retval != 0) {
+		dev_err(&board_dat->pdev->dev,
+			"%s FAIL:invoking pch_spi_free_resources\n", __func__);
+		pch_spi_free_resources(board_dat);
+	}
+
+	dev_dbg(&board_dat->pdev->dev, "%s Return=%d\n", __func__, retval);
+
+	return retval;
+}
+
+static int pch_spi_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+
+	struct spi_master *master;
+
+	struct pch_spi_board_data *board_dat;
+	int retval;
+
+	dev_dbg(&pdev->dev, "%s ENTRY\n", __func__);
+
+	/* allocate memory for private data */
+	board_dat = kzalloc(sizeof(struct pch_spi_board_data), GFP_KERNEL);
+	if (board_dat == NULL) {
+		dev_err(&pdev->dev,
+			" %s memory allocation for private data failed\n",
+			__func__);
+		retval = -ENOMEM;
+		goto err_kmalloc;
+	}
+
+	dev_dbg(&pdev->dev,
+		"%s memory allocation for private data success\n", __func__);
+
+	/* enable PCI device */
+	retval = pci_enable_device(pdev);
+	if (retval != 0) {
+		dev_err(&pdev->dev, "%s pci_enable_device FAILED\n", __func__);
+
+		goto err_pci_en_device;
+	}
+
+	dev_dbg(&pdev->dev, "%s pci_enable_device returned=%d\n",
+		__func__, retval);
+
+	board_dat->pdev = pdev;
+
+	/* alllocate memory for SPI master */
+	master = spi_alloc_master(&pdev->dev, sizeof(struct pch_spi_data));
+	if (master == NULL) {
+		retval = -ENOMEM;
+		dev_err(&pdev->dev, "%s Fail.\n", __func__);
+		goto err_spi_alloc_master;
+	}
+
+	dev_dbg(&pdev->dev,
+		"%s spi_alloc_master returned non NULL\n", __func__);
+
+	/* initialize members of SPI master */
+	master->bus_num = -1;
+	master->num_chipselect = PCH_MAX_CS;
+	master->setup = pch_spi_setup;
+	master->transfer = pch_spi_transfer;
+	dev_dbg(&pdev->dev,
+		"%s transfer member of SPI master initialized\n", __func__);
+
+	board_dat->data = spi_master_get_devdata(master);
+
+	board_dat->data->master = master;
+	board_dat->data->n_curnt_chip = 255;
+	board_dat->data->board_dat = board_dat;
+	board_dat->data->status = STATUS_RUNNING;
+
+	INIT_LIST_HEAD(&board_dat->data->queue);
+	spin_lock_init(&board_dat->data->lock);
+	INIT_WORK(&board_dat->data->work, pch_spi_process_messages);
+	init_waitqueue_head(&board_dat->data->wait);
+
+	/* allocate resources for PCH SPI */
+	retval = pch_spi_get_resources(board_dat);
+	if (retval) {
+		dev_err(&pdev->dev, "%s fail(retval=%d)\n", __func__, retval);
+		goto err_spi_get_resources;
+	}
+
+	dev_dbg(&pdev->dev, "%s pch_spi_get_resources returned=%d\n",
+		__func__, retval);
+
+	/* save private data in dev */
+	pci_set_drvdata(pdev, board_dat);
+	dev_dbg(&pdev->dev, "%s invoked pci_set_drvdata\n", __func__);
+
+	/* set master mode */
+	pch_spi_set_master_mode(master);
+	dev_dbg(&pdev->dev,
+		"%s invoked pch_spi_set_master_mode\n", __func__);
+
+	/* Register the controller with the SPI core. */
+	retval = spi_register_master(master);
+	if (retval != 0) {
+		dev_err(&pdev->dev,
+			"%s spi_register_master FAILED\n", __func__);
+		goto err_spi_reg_master;
+	}
+
+	dev_dbg(&pdev->dev, "%s spi_register_master returned=%d\n",
+		__func__, retval);
+
+
+	return 0;
+
+err_spi_reg_master:
+	spi_unregister_master(master);
+err_spi_get_resources:
+err_spi_alloc_master:
+	spi_master_put(master);
+	pci_disable_device(pdev);
+err_pci_en_device:
+	kfree(board_dat);
+err_kmalloc:
+	return retval;
+}
+
+static void pch_spi_remove(struct pci_dev *pdev)
+{
+	struct pch_spi_board_data *board_dat = pci_get_drvdata(pdev);
+	int count;
+
+	dev_dbg(&pdev->dev, "%s ENTRY\n", __func__);
+
+	if (!board_dat) {
+		dev_err(&pdev->dev,
+			"%s pci_get_drvdata returned NULL\n", __func__);
+		return;
+	}
+
+	/* check for any pending messages; no action is taken if the queue
+	 * is still full; but at least we tried.  Unload anyway */
+	count = 500;
+	spin_lock(&board_dat->data->lock);
+	board_dat->data->status = STATUS_EXITING;
+	while ((list_empty(&board_dat->data->queue) == 0) && --count) {
+		dev_dbg(&board_dat->pdev->dev, "%s :queue not empty\n",
+			__func__);
+		spin_unlock(&board_dat->data->lock);
+		msleep(PCH_SLEEP_TIME);
+		spin_lock(&board_dat->data->lock);
+	}
+	spin_unlock(&board_dat->data->lock);
+
+	/* Free resources allocated for PCH SPI */
+	pch_spi_free_resources(board_dat);
+
+	spi_unregister_master(board_dat->data->master);
+
+	/* free memory for private data */
+	kfree(board_dat);
+
+	pci_set_drvdata(pdev, NULL);
+
+	/* disable PCI device */
+	pci_disable_device(pdev);
+
+	dev_dbg(&pdev->dev, "%s invoked pci_disable_device\n", __func__);
+}
+
+#ifdef CONFIG_PM
+static int pch_spi_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	u8 count;
+	int retval;
+
+	struct pch_spi_board_data *board_dat = pci_get_drvdata(pdev);
+
+	dev_dbg(&pdev->dev, "%s ENTRY\n", __func__);
+
+	if (!board_dat) {
+		dev_err(&pdev->dev,
+			"%s pci_get_drvdata returned NULL\n", __func__);
+		return -EFAULT;
+	}
+
+	retval = 0;
+	board_dat->suspend_sts = true;
+
+	/* check if the current message is processed:
+	   Only after thats done the transfer will be suspended */
+	count = 255;
+	while ((--count) > 0) {
+		if (!(board_dat->data->bcurrent_msg_processing)) {
+			dev_dbg(&pdev->dev, "%s board_dat->data->bCurrent_"
+				"msg_processing = false\n", __func__);
+			break;
+		} else {
+			dev_dbg(&pdev->dev, "%s board_dat->data->bCurrent_msg_"
+				"processing = true\n", __func__);
+		}
+		msleep(PCH_SLEEP_TIME);
+	}
+
+	/* Free IRQ */
+	if (board_dat->irq_reg_sts) {
+		/* disable all interrupts */
+		pch_spi_setclr_reg(board_dat->data->master, PCH_SPCR, 0,
+				   PCH_ALL);
+		pch_spi_reset(board_dat->data->master);
+
+		free_irq(board_dat->pdev->irq, board_dat);
+
+		board_dat->irq_reg_sts = false;
+		dev_dbg(&pdev->dev,
+			"%s free_irq invoked successfully.\n", __func__);
+	}
+
+	/* save config space */
+	retval = pci_save_state(pdev);
+
+	if (retval == 0) {
+		dev_dbg(&pdev->dev, "%s pci_save_state returned=%d\n",
+			__func__, retval);
+		/* disable PM notifications */
+		pci_enable_wake(pdev, PCI_D3hot, 0);
+		dev_dbg(&pdev->dev,
+			"%s pci_enable_wake invoked successfully\n", __func__);
+		/* disable PCI device */
+		pci_disable_device(pdev);
+		dev_dbg(&pdev->dev,
+			"%s pci_disable_device invoked successfully\n",
+			__func__);
+		/* move device to D3hot  state */
+		pci_set_power_state(pdev, PCI_D3hot);
+		dev_dbg(&pdev->dev,
+			"%s pci_set_power_state invoked successfully\n",
+			__func__);
+	} else {
+		dev_err(&pdev->dev, "%s pci_save_state failed\n", __func__);
+	}
+
+	dev_dbg(&pdev->dev, "%s return=%d\n", __func__, retval);
+
+	return retval;
+}
+
+static int pch_spi_resume(struct pci_dev *pdev)
+{
+	int retval;
+
+	struct pch_spi_board_data *board = pci_get_drvdata(pdev);
+	dev_dbg(&pdev->dev, "%s ENTRY\n", __func__);
+
+	if (!board) {
+		dev_err(&pdev->dev,
+			"%s pci_get_drvdata returned NULL\n", __func__);
+		return -EFAULT;
+	}
+
+	/* move device to DO power state */
+	pci_set_power_state(pdev, PCI_D0);
+
+	/* restore state */
+	pci_restore_state(pdev);
+
+	retval = pci_enable_device(pdev);
+	if (retval < 0) {
+		dev_err(&pdev->dev,
+			"%s pci_enable_device failed\n", __func__);
+	} else {
+		/* disable PM notifications */
+		pci_enable_wake(pdev, PCI_D3hot, 0);
+
+		/* register IRQ handler */
+		if (!board->irq_reg_sts) {
+			/* register IRQ */
+			retval = request_irq(board->pdev->irq, pch_spi_handler,
+					     IRQF_SHARED, KBUILD_MODNAME,
+					     board);
+			if (retval < 0) {
+				dev_err(&pdev->dev,
+					"%s request_irq failed\n", __func__);
+				return retval;
+			}
+			board->irq_reg_sts = true;
+
+			/* reset PCH SPI h/w */
+			pch_spi_reset(board->data->master);
+			pch_spi_set_master_mode(board->data->master);
+
+			/* set suspend status to false */
+			board->suspend_sts = false;
+
+		}
+	}
+
+	dev_dbg(&pdev->dev, "%s returning=%d\n", __func__, retval);
+
+	return retval;
+}
+#else
+#define pch_spi_suspend NULL
+#define pch_spi_resume NULL
+
+#endif
+
+static struct pci_driver pch_spi_pcidev = {
+	.name = "pch_spi",
+	.id_table = pch_spi_pcidev_id,
+	.probe = pch_spi_probe,
+	.remove = pch_spi_remove,
+	.suspend = pch_spi_suspend,
+	.resume = pch_spi_resume,
+};
+
+static int __init pch_spi_init(void)
+{
+	return pci_register_driver(&pch_spi_pcidev);
+}
+module_init(pch_spi_init);
+
+static void __exit pch_spi_exit(void)
+{
+	pci_unregister_driver(&pch_spi_pcidev);
+}
+module_exit(pch_spi_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Topcliff PCH SPI PCI Driver");
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index 7892ac1..c68b3dc 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -20,7 +20,6 @@
 #include <linux/mmc/sdio_func.h>
 #include <linux/slab.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c
index 526682d..c7345db 100644
--- a/drivers/ssb/pcmcia.c
+++ b/drivers/ssb/pcmcia.c
@@ -13,7 +13,6 @@
 #include <linux/io.h>
 #include <linux/etherdevice.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ciscode.h>
 #include <pcmcia/ds.h>
diff --git a/drivers/ssb/scan.c b/drivers/ssb/scan.c
index 9738cad..ee079ab 100644
--- a/drivers/ssb/scan.c
+++ b/drivers/ssb/scan.c
@@ -17,7 +17,6 @@
 #include <linux/pci.h>
 #include <linux/io.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
diff --git a/drivers/staging/comedi/drivers/cb_das16_cs.c b/drivers/staging/comedi/drivers/cb_das16_cs.c
index f8ede11..0345b4c 100644
--- a/drivers/staging/comedi/drivers/cb_das16_cs.c
+++ b/drivers/staging/comedi/drivers/cb_das16_cs.c
@@ -37,7 +37,6 @@
 #include <linux/delay.h>
 #include <linux/pci.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -692,10 +691,6 @@
 	local->link = link;
 	link->priv = local;
 
-	/* Initialize the pcmcia_device structure */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	cur_dev = link;
 
 	das16cs_pcmcia_config(link);
@@ -715,37 +710,12 @@
 
 
 static int das16cs_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
 				void *priv_data)
 {
-	if (cfg->index == 0)
+	if (p_dev->config_index == 0)
 		return -EINVAL;
 
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		return pcmcia_request_io(p_dev);
-	}
-
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
 static void das16cs_pcmcia_config(struct pcmcia_device *link)
@@ -754,6 +724,9 @@
 
 	dev_dbg(&link->dev, "das16cs_pcmcia_config\n");
 
+	/* Do we need to allocate an interrupt? */
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, das16cs_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -763,25 +736,10 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %u", link->irq);
-	if (link->resource[0])
-		printk(", io %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(", io %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -832,9 +790,7 @@
 	.resume = das16cs_pcmcia_resume,
 	.id_table = das16cs_id_table,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "cb_das16_cs",
-		},
+	.name = "cb_das16_cs",
 };
 
 static int __init init_das16cs_pcmcia_cs(void)
diff --git a/drivers/staging/comedi/drivers/das08_cs.c b/drivers/staging/comedi/drivers/das08_cs.c
index 48d9fb1..0b32a2d 100644
--- a/drivers/staging/comedi/drivers/das08_cs.c
+++ b/drivers/staging/comedi/drivers/das08_cs.c
@@ -48,7 +48,6 @@
 #include "das08.h"
 
 /* pcmcia includes */
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -115,40 +114,15 @@
 static int das08_pcmcia_suspend(struct pcmcia_device *p_dev);
 static int das08_pcmcia_resume(struct pcmcia_device *p_dev);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static int das08_pcmcia_attach(struct pcmcia_device *);
 static void das08_pcmcia_detach(struct pcmcia_device *);
 
-/*
-   You'll also need to prototype all the functions that will actually
-   be used to talk to your device.  See 'memory_cs' for a good example
-   of a fully self-sufficient driver; the other drivers rely more or
-   less on other parts of the kernel.
-*/
-
 struct local_info_t {
 	struct pcmcia_device *link;
 	int stop;
 	struct bus_operations *bus;
 };
 
-/*======================================================================
-
-    das08_pcmcia_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int das08_pcmcia_attach(struct pcmcia_device *link)
 {
 	struct local_info_t *local;
@@ -162,16 +136,6 @@
 	local->link = link;
 	link->priv = local;
 
-	/*
-	   General socket configuration defaults can go here.  In this
-	   client, we assume very little, and rely on the CIS for almost
-	   everything.  In most clients, many details (i.e., number, sizes,
-	   and attributes of IO windows) are fixed by the nature of the
-	   device, and can be hard-wired here.
-	 */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	cur_dev = link;
 
 	das08_pcmcia_config(link);
@@ -179,15 +143,6 @@
 	return 0;
 }				/* das08_pcmcia_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void das08_pcmcia_detach(struct pcmcia_device *link)
 {
 
@@ -203,53 +158,22 @@
 
 
 static int das08_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
 				void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		return pcmcia_request_io(p_dev);
-	}
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
-
-/*======================================================================
-
-    das08_pcmcia_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
 static void das08_pcmcia_config(struct pcmcia_device *link)
 {
 	int ret;
 
 	dev_dbg(&link->dev, "das08_pcmcia_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, das08_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -259,25 +183,10 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %u", link->irq);
-	if (link->resource[0])
-		printk(", io %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -285,32 +194,12 @@
 
 }				/* das08_pcmcia_config */
 
-/*======================================================================
-
-    After a card is removed, das08_pcmcia_release() will unregister the
-    device, and release the PCMCIA configuration.  If the device is
-    still open, this will be postponed until it is closed.
-
-======================================================================*/
-
 static void das08_pcmcia_release(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "das08_pcmcia_release\n");
 	pcmcia_disable_device(link);
 }				/* das08_pcmcia_release */
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.
-
-    When a CARD_REMOVAL event is received, we immediately set a
-    private flag to block future accesses to this device.  All the
-    functions that actually access the device should check this flag
-    to make sure the card is still present.
-
-======================================================================*/
-
 static int das08_pcmcia_suspend(struct pcmcia_device *link)
 {
 	struct local_info_t *local = link->priv;
@@ -348,9 +237,7 @@
 	.resume = das08_pcmcia_resume,
 	.id_table = das08_cs_id_table,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "pcm-das08",
-		},
+	.name = "pcm-das08",
 };
 
 static int __init init_das08_pcmcia_cs(void)
diff --git a/drivers/staging/comedi/drivers/ni_daq_700.c b/drivers/staging/comedi/drivers/ni_daq_700.c
index cc15666..fc772a8 100644
--- a/drivers/staging/comedi/drivers/ni_daq_700.c
+++ b/drivers/staging/comedi/drivers/ni_daq_700.c
@@ -47,7 +47,6 @@
 
 #include <linux/ioport.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -435,47 +434,20 @@
 	return 0;
 };
 
-/* PCMCIA crap -- watch your words, please! */
-
 static void dio700_config(struct pcmcia_device *link);
 static void dio700_release(struct pcmcia_device *link);
 static int dio700_cs_suspend(struct pcmcia_device *p_dev);
 static int dio700_cs_resume(struct pcmcia_device *p_dev);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static int dio700_cs_attach(struct pcmcia_device *);
 static void dio700_cs_detach(struct pcmcia_device *);
 
-/*
-   You'll also need to prototype all the functions that will actually
-   be used to talk to your device.  See 'memory_cs' for a good example
-   of a fully self-sufficient driver; the other drivers rely more or
-   less on other parts of the kernel.
-*/
-
 struct local_info_t {
 	struct pcmcia_device *link;
 	int stop;
 	struct bus_operations *bus;
 };
 
-/*======================================================================
-
-    dio700_cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int dio700_cs_attach(struct pcmcia_device *link)
 {
 	struct local_info_t *local;
@@ -491,16 +463,6 @@
 	local->link = link;
 	link->priv = local;
 
-	/*
-	   General socket configuration defaults can go here.  In this
-	   client, we assume very little, and rely on the CIS for almost
-	   everything.  In most clients, many details (i.e., number, sizes,
-	   and attributes of IO windows) are fixed by the nature of the
-	   device, and can be hard-wired here.
-	 */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	pcmcia_cur_dev = link;
 
 	dio700_config(link);
@@ -508,15 +470,6 @@
 	return 0;
 }				/* dio700_cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void dio700_cs_detach(struct pcmcia_device *link)
 {
 
@@ -532,54 +485,13 @@
 
 }				/* dio700_cs_detach */
 
-/*======================================================================
-
-    dio700_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
 static int dio700_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
 				void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		if (pcmcia_request_io(p_dev) != 0)
-			return -ENODEV;
-	}
-
-	/* If we got this far, we're cool! */
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
 static void dio700_config(struct pcmcia_device *link)
@@ -591,6 +503,9 @@
 
 	dev_dbg(&link->dev, "dio700_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_AUDIO |
+		CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, dio700_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -600,25 +515,10 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret != 0)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %d", link->irq);
-	if (link->resource[0])
-		printk(", io %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -634,18 +534,6 @@
 	pcmcia_disable_device(link);
 }				/* dio700_release */
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.
-
-    When a CARD_REMOVAL event is received, we immediately set a
-    private flag to block future accesses to this device.  All the
-    functions that actually access the device should check this flag
-    to make sure the card is still present.
-
-======================================================================*/
-
 static int dio700_cs_suspend(struct pcmcia_device *link)
 {
 	struct local_info_t *local = link->priv;
@@ -685,9 +573,7 @@
 	.resume = dio700_cs_resume,
 	.id_table = dio700_cs_ids,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "ni_daq_700",
-		},
+	.name = "ni_daq_700",
 };
 
 static int __init init_dio700_cs(void)
diff --git a/drivers/staging/comedi/drivers/ni_daq_dio24.c b/drivers/staging/comedi/drivers/ni_daq_dio24.c
index 773ae20..c9c2858 100644
--- a/drivers/staging/comedi/drivers/ni_daq_dio24.c
+++ b/drivers/staging/comedi/drivers/ni_daq_dio24.c
@@ -48,7 +48,6 @@
 
 #include "8255.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -187,47 +186,20 @@
 	return 0;
 };
 
-/* PCMCIA crap -- watch your words! */
-
 static void dio24_config(struct pcmcia_device *link);
 static void dio24_release(struct pcmcia_device *link);
 static int dio24_cs_suspend(struct pcmcia_device *p_dev);
 static int dio24_cs_resume(struct pcmcia_device *p_dev);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static int dio24_cs_attach(struct pcmcia_device *);
 static void dio24_cs_detach(struct pcmcia_device *);
 
-/*
-   You'll also need to prototype all the functions that will actually
-   be used to talk to your device.  See 'memory_cs' for a good example
-   of a fully self-sufficient driver; the other drivers rely more or
-   less on other parts of the kernel.
-*/
-
 struct local_info_t {
 	struct pcmcia_device *link;
 	int stop;
 	struct bus_operations *bus;
 };
 
-/*======================================================================
-
-    dio24_cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int dio24_cs_attach(struct pcmcia_device *link)
 {
 	struct local_info_t *local;
@@ -243,16 +215,6 @@
 	local->link = link;
 	link->priv = local;
 
-	/*
-	   General socket configuration defaults can go here.  In this
-	   client, we assume very little, and rely on the CIS for almost
-	   everything.  In most clients, many details (i.e., number, sizes,
-	   and attributes of IO windows) are fixed by the nature of the
-	   device, and can be hard-wired here.
-	 */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	pcmcia_cur_dev = link;
 
 	dio24_config(link);
@@ -260,15 +222,6 @@
 	return 0;
 }				/* dio24_cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void dio24_cs_detach(struct pcmcia_device *link)
 {
 
@@ -284,54 +237,13 @@
 
 }				/* dio24_cs_detach */
 
-/*======================================================================
-
-    dio24_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
 static int dio24_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
 				void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		if (pcmcia_request_io(p_dev) != 0)
-			return -ENODEV;
-	}
-
-	/* If we got this far, we're cool! */
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
 static void dio24_config(struct pcmcia_device *link)
@@ -342,6 +254,9 @@
 
 	dev_dbg(&link->dev, "dio24_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_AUDIO |
+		CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, dio24_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -351,25 +266,10 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %d", link->irq);
-	if (link->resource[0])
-		printk(" & %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -385,18 +285,6 @@
 	pcmcia_disable_device(link);
 }				/* dio24_release */
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.
-
-    When a CARD_REMOVAL event is received, we immediately set a
-    private flag to block future accesses to this device.  All the
-    functions that actually access the device should check this flag
-    to make sure the card is still present.
-
-======================================================================*/
-
 static int dio24_cs_suspend(struct pcmcia_device *link)
 {
 	struct local_info_t *local = link->priv;
@@ -435,9 +323,7 @@
 	.resume = dio24_cs_resume,
 	.id_table = dio24_cs_ids,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "ni_daq_dio24",
-		},
+	.name = "ni_daq_dio24",
 };
 
 static int __init init_dio24_cs(void)
diff --git a/drivers/staging/comedi/drivers/ni_labpc_cs.c b/drivers/staging/comedi/drivers/ni_labpc_cs.c
index 68c4ecb..6facbc8 100644
--- a/drivers/staging/comedi/drivers/ni_labpc_cs.c
+++ b/drivers/staging/comedi/drivers/ni_labpc_cs.c
@@ -71,7 +71,6 @@
 #include "comedi_fc.h"
 #include "ni_labpc.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -153,59 +152,20 @@
 	return labpc_common_attach(dev, iobase, irq, 0);
 }
 
-/*====================================================================*/
-
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card
-   insertion and ejection events.  They are invoked from the dummy
-   event handler.
-
-   Kernel version 2.6.16 upwards uses suspend() and resume() functions
-   instead of an event() function.
-*/
-
 static void labpc_config(struct pcmcia_device *link);
 static void labpc_release(struct pcmcia_device *link);
 static int labpc_cs_suspend(struct pcmcia_device *p_dev);
 static int labpc_cs_resume(struct pcmcia_device *p_dev);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static int labpc_cs_attach(struct pcmcia_device *);
 static void labpc_cs_detach(struct pcmcia_device *);
 
-/*
-   You'll also need to prototype all the functions that will actually
-   be used to talk to your device.  See 'memory_cs' for a good example
-   of a fully self-sufficient driver; the other drivers rely more or
-   less on other parts of the kernel.
-*/
-
 struct local_info_t {
 	struct pcmcia_device *link;
 	int stop;
 	struct bus_operations *bus;
 };
 
-/*======================================================================
-
-    labpc_cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int labpc_cs_attach(struct pcmcia_device *link)
 {
 	struct local_info_t *local;
@@ -219,16 +179,6 @@
 	local->link = link;
 	link->priv = local;
 
-	/*
-	   General socket configuration defaults can go here.  In this
-	   client, we assume very little, and rely on the CIS for almost
-	   everything.  In most clients, many details (i.e., number, sizes,
-	   and attributes of IO windows) are fixed by the nature of the
-	   device, and can be hard-wired here.
-	 */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	pcmcia_cur_dev = link;
 
 	labpc_config(link);
@@ -236,15 +186,6 @@
 	return 0;
 }				/* labpc_cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void labpc_cs_detach(struct pcmcia_device *link)
 {
 	dev_dbg(&link->dev, "labpc_cs_detach\n");
@@ -263,54 +204,13 @@
 
 }				/* labpc_cs_detach */
 
-/*======================================================================
-
-    labpc_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
 static int labpc_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
 				void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Does this card need audio output? */
-	if (cfg->flags & CISTPL_CFTABLE_AUDIO) {
-		p_dev->conf.Attributes |= CONF_ENABLE_SPKR;
-		p_dev->conf.Status = CCSR_AUDIO_ENA;
-	}
-
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ | CONF_ENABLE_PULSE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		/* This reserves IO space but doesn't actually enable it */
-		if (pcmcia_request_io(p_dev) != 0)
-			return -ENODEV;
-	}
-
-	/* If we got this far, we're cool! */
-	return 0;
+	return pcmcia_request_io(p_dev);
 }
 
 
@@ -320,6 +220,9 @@
 
 	dev_dbg(&link->dev, "labpc_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_ENABLE_PULSE_IRQ |
+		CONF_AUTO_AUDIO | CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, labpc_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -329,25 +232,10 @@
 	if (!link->irq)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %d", link->irq);
-	if (link->resource[0])
-		printk(" & %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -362,18 +250,6 @@
 	pcmcia_disable_device(link);
 }				/* labpc_release */
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.
-
-    When a CARD_REMOVAL event is received, we immediately set a
-    private flag to block future accesses to this device.  All the
-    functions that actually access the device should check this flag
-    to make sure the card is still present.
-
-======================================================================*/
-
 static int labpc_cs_suspend(struct pcmcia_device *link)
 {
 	struct local_info_t *local = link->priv;
@@ -391,8 +267,6 @@
 	return 0;
 }				/* labpc_cs_resume */
 
-/*====================================================================*/
-
 static struct pcmcia_device_id labpc_cs_ids[] = {
 	/* N.B. These IDs should match those in labpc_cs_boards (ni_labpc.c) */
 	PCMCIA_DEVICE_MANF_CARD(0x010b, 0x0103),	/* daqcard-1200 */
@@ -411,9 +285,7 @@
 	.resume = labpc_cs_resume,
 	.id_table = labpc_cs_ids,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "daqcard-1200",
-		},
+	.name = "daqcard-1200",
 };
 
 static int __init init_labpc_cs(void)
diff --git a/drivers/staging/comedi/drivers/ni_mio_cs.c b/drivers/staging/comedi/drivers/ni_mio_cs.c
index 1f24263..4956327 100644
--- a/drivers/staging/comedi/drivers/ni_mio_cs.c
+++ b/drivers/staging/comedi/drivers/ni_mio_cs.c
@@ -48,7 +48,6 @@
 #include "ni_stc.h"
 #include "8255.h"
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -263,11 +262,6 @@
 
 static int cs_attach(struct pcmcia_device *link)
 {
-	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
-	link->resource[0]->end = 16;
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	cur_dev = link;
 
 	mio_cs_config(link);
@@ -301,16 +295,12 @@
 }
 
 
-static int mio_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int mio_pcmcia_config_loop(struct pcmcia_device *p_dev, void *priv_data)
 {
 	int base, ret;
 
-	p_dev->resource[0]->end = cfg->io.win[0].len;
-	p_dev->io_lines = cfg->io.flags & CISTPL_IO_LINES_MASK;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_16;
 
 	for (base = 0x000; base < 0x400; base += 0x20) {
 		p_dev->resource[0]->start = base;
@@ -327,6 +317,7 @@
 	int ret;
 
 	DPRINTK("mio_cs_config(link=%p)\n", link);
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
 
 	ret = pcmcia_loop_config(link, mio_pcmcia_config_loop, NULL);
 	if (ret) {
@@ -337,7 +328,7 @@
 	if (!link->irq)
 		dev_info(&link->dev, "no IRQ available\n");
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 }
 
 static int mio_cs_attach(struct comedi_device *dev, struct comedi_devconfig *it)
@@ -446,9 +437,7 @@
 	.resume = &mio_cs_resume,
 	.id_table = ni_mio_cs_ids,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "ni_mio_cs",
-		},
+	.name = "ni_mio_cs",
 };
 
 int init_module(void)
diff --git a/drivers/staging/comedi/drivers/quatech_daqp_cs.c b/drivers/staging/comedi/drivers/quatech_daqp_cs.c
index bf489d7..ebba9bb 100644
--- a/drivers/staging/comedi/drivers/quatech_daqp_cs.c
+++ b/drivers/staging/comedi/drivers/quatech_daqp_cs.c
@@ -50,7 +50,6 @@
 #include "../comedidev.h"
 #include <linux/semaphore.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -969,43 +968,14 @@
 
 ======================================================================*/
 
-/*
-   The event() function is this driver's Card Services event handler.
-   It will be called by Card Services when an appropriate card status
-   event is received.  The config() and release() entry points are
-   used to configure or release a socket, in response to card
-   insertion and ejection events.
-
-   Kernel version 2.6.16 upwards uses suspend() and resume() functions
-   instead of an event() function.
-*/
-
 static void daqp_cs_config(struct pcmcia_device *link);
 static void daqp_cs_release(struct pcmcia_device *link);
 static int daqp_cs_suspend(struct pcmcia_device *p_dev);
 static int daqp_cs_resume(struct pcmcia_device *p_dev);
 
-/*
-   The attach() and detach() entry points are used to create and destroy
-   "instances" of the driver, where each instance represents everything
-   needed to manage one actual PCMCIA card.
-*/
-
 static int daqp_cs_attach(struct pcmcia_device *);
 static void daqp_cs_detach(struct pcmcia_device *);
 
-/*======================================================================
-
-    daqp_cs_attach() creates an "instance" of the driver, allocating
-    local data structures for one device.  The device is registered
-    with Card Services.
-
-    The dev_link structure is initialized, but we don't actually
-    configure the card at this point -- we wait until we receive a
-    card insertion event.
-
-======================================================================*/
-
 static int daqp_cs_attach(struct pcmcia_device *link)
 {
 	struct local_info_t *local;
@@ -1031,30 +1001,11 @@
 	local->link = link;
 	link->priv = local;
 
-	/*
-	   General socket configuration defaults can go here.  In this
-	   client, we assume very little, and rely on the CIS for almost
-	   everything.  In most clients, many details (i.e., number, sizes,
-	   and attributes of IO windows) are fixed by the nature of the
-	   device, and can be hard-wired here.
-	 */
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	daqp_cs_config(link);
 
 	return 0;
 }				/* daqp_cs_attach */
 
-/*======================================================================
-
-    This deletes a driver "instance".  The device is de-registered
-    with Card Services.  If it has been released, all local data
-    structures are freed.  Otherwise, the structures will be freed
-    when the device is released.
-
-======================================================================*/
-
 static void daqp_cs_detach(struct pcmcia_device *link)
 {
 	struct local_info_t *dev = link->priv;
@@ -1070,45 +1021,11 @@
 
 }				/* daqp_cs_detach */
 
-/*======================================================================
-
-    daqp_cs_config() is scheduled to run after a CARD_INSERTION event
-    is received, to configure the PCMCIA socket, and to make the
-    device available to the system.
-
-======================================================================*/
-
-
-static int daqp_pcmcia_config_loop(struct pcmcia_device *p_dev,
-				cistpl_cftable_entry_t *cfg,
-				cistpl_cftable_entry_t *dflt,
-				unsigned int vcc,
-				void *priv_data)
+static int daqp_pcmcia_config_loop(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Do we need to allocate an interrupt? */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-		p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
-		p_dev->resource[0]->flags |=
-			pcmcia_io_cfg_data_width(io->flags);
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		if (io->nwin > 1) {
-			p_dev->resource[1]->flags = p_dev->resource[0]->flags;
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-	}
-
-	/* This reserves IO space but doesn't actually enable it */
 	return pcmcia_request_io(p_dev);
 }
 
@@ -1118,6 +1035,8 @@
 
 	dev_dbg(&link->dev, "daqp_cs_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO;
+
 	ret = pcmcia_loop_config(link, daqp_pcmcia_config_loop, NULL);
 	if (ret) {
 		dev_warn(&link->dev, "no configuration found\n");
@@ -1128,25 +1047,10 @@
 	if (ret)
 		goto failed;
 
-	/*
-	   This actually configures the PCMCIA socket -- setting up
-	   the I/O windows and the interrupt mapping, and putting the
-	   card and host interface into "Memory and IO" mode.
-	 */
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	/* Finally, report what we've done */
-	dev_info(&link->dev, "index 0x%02x", link->conf.ConfigIndex);
-	if (link->conf.Attributes & CONF_ENABLE_IRQ)
-		printk(", irq %u", link->irq);
-	if (link->resource[0])
-		printk(" & %pR", link->resource[0]);
-	if (link->resource[1])
-		printk(" & %pR", link->resource[1]);
-	printk("\n");
-
 	return;
 
 failed:
@@ -1161,18 +1065,6 @@
 	pcmcia_disable_device(link);
 }				/* daqp_cs_release */
 
-/*======================================================================
-
-    The card status event handler.  Mostly, this schedules other
-    stuff to run after an event is received.
-
-    When a CARD_REMOVAL event is received, we immediately set a
-    private flag to block future accesses to this device.  All the
-    functions that actually access the device should check this flag
-    to make sure the card is still present.
-
-======================================================================*/
-
 static int daqp_cs_suspend(struct pcmcia_device *link)
 {
 	struct local_info_t *local = link->priv;
@@ -1212,9 +1104,7 @@
 	.resume = daqp_cs_resume,
 	.id_table = daqp_cs_id_table,
 	.owner = THIS_MODULE,
-	.drv = {
-		.name = "quatech_daqp_cs",
-		},
+	.name = "quatech_daqp_cs",
 };
 
 int __init init_module(void)
diff --git a/drivers/staging/tm6000/Kconfig b/drivers/staging/tm6000/Kconfig
index c725356..de7ebb9 100644
--- a/drivers/staging/tm6000/Kconfig
+++ b/drivers/staging/tm6000/Kconfig
@@ -1,6 +1,6 @@
 config VIDEO_TM6000
 	tristate "TV Master TM5600/6000/6010 driver"
-	depends on VIDEO_DEV && I2C && INPUT && USB && EXPERIMENTAL
+	depends on VIDEO_DEV && I2C && INPUT && IR_CORE && USB && EXPERIMENTAL
 	select VIDEO_TUNER
 	select MEDIA_TUNER_XC2028
 	select MEDIA_TUNER_XC5000
diff --git a/drivers/staging/tm6000/tm6000-input.c b/drivers/staging/tm6000/tm6000-input.c
index 32f7a0a..54f7667 100644
--- a/drivers/staging/tm6000/tm6000-input.c
+++ b/drivers/staging/tm6000/tm6000-input.c
@@ -46,7 +46,7 @@
 	}
 
 struct tm6000_ir_poll_result {
-	u8 rc_data[4];
+	u16 rc_data;
 };
 
 struct tm6000_IR {
@@ -60,9 +60,9 @@
 	int			polling;
 	struct delayed_work	work;
 	u8			wait:1;
+	u8			key:1;
 	struct urb		*int_urb;
 	u8			*urb_data;
-	u8			key:1;
 
 	int (*get_key) (struct tm6000_IR *, struct tm6000_ir_poll_result *);
 
@@ -122,13 +122,14 @@
 
 	if (urb->status != 0)
 		printk(KERN_INFO "not ready\n");
-	else if (urb->actual_length > 0)
+	else if (urb->actual_length > 0) {
 		memcpy(ir->urb_data, urb->transfer_buffer, urb->actual_length);
 
-	dprintk("data %02x %02x %02x %02x\n", ir->urb_data[0],
-	ir->urb_data[1], ir->urb_data[2], ir->urb_data[3]);
+		dprintk("data %02x %02x %02x %02x\n", ir->urb_data[0],
+			ir->urb_data[1], ir->urb_data[2], ir->urb_data[3]);
 
-	ir->key = 1;
+		ir->key = 1;
+	}
 
 	rc = usb_submit_urb(urb, GFP_ATOMIC);
 }
@@ -140,30 +141,47 @@
 	int rc;
 	u8 buf[2];
 
-	if (ir->wait && !&dev->int_in) {
-		poll_result->rc_data[0] = 0xff;
+	if (ir->wait && !&dev->int_in)
 		return 0;
-	}
 
 	if (&dev->int_in) {
-		poll_result->rc_data[0] = ir->urb_data[0];
-		poll_result->rc_data[1] = ir->urb_data[1];
+		if (ir->ir.ir_type == IR_TYPE_RC5)
+			poll_result->rc_data = ir->urb_data[0];
+		else
+			poll_result->rc_data = ir->urb_data[0] | ir->urb_data[1] << 8;
 	} else {
 		tm6000_set_reg(dev, REQ_04_EN_DISABLE_MCU_INT, 2, 0);
 		msleep(10);
 		tm6000_set_reg(dev, REQ_04_EN_DISABLE_MCU_INT, 2, 1);
 		msleep(10);
 
-		rc = tm6000_read_write_usb(dev, USB_DIR_IN | USB_TYPE_VENDOR |
-		 USB_RECIP_DEVICE, REQ_02_GET_IR_CODE, 0, 0, buf, 1);
+		if (ir->ir.ir_type == IR_TYPE_RC5) {
+			rc = tm6000_read_write_usb(dev, USB_DIR_IN |
+				USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+				REQ_02_GET_IR_CODE, 0, 0, buf, 1);
 
-		msleep(10);
+			msleep(10);
 
-		dprintk("read data=%02x\n", buf[0]);
-		if (rc < 0)
-			return rc;
+			dprintk("read data=%02x\n", buf[0]);
+			if (rc < 0)
+				return rc;
 
-		poll_result->rc_data[0] = buf[0];
+			poll_result->rc_data = buf[0];
+		} else {
+			rc = tm6000_read_write_usb(dev, USB_DIR_IN |
+				USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+				REQ_02_GET_IR_CODE, 0, 0, buf, 2);
+
+			msleep(10);
+
+			dprintk("read data=%04x\n", buf[0] | buf[1] << 8);
+			if (rc < 0)
+				return rc;
+
+			poll_result->rc_data = buf[0] | buf[1] << 8;
+		}
+		if ((poll_result->rc_data & 0x00ff) != 0xff)
+			ir->key = 1;
 	}
 	return 0;
 }
@@ -180,12 +198,11 @@
 		return;
 	}
 
-	dprintk("ir->get_key result data=%02x %02x\n",
-		poll_result.rc_data[0], poll_result.rc_data[1]);
+	dprintk("ir->get_key result data=%04x\n", poll_result.rc_data);
 
-	if (poll_result.rc_data[0] != 0xff && ir->key == 1) {
+	if (ir->key) {
 		ir_input_keydown(ir->input->input_dev, &ir->ir,
-			poll_result.rc_data[0] | poll_result.rc_data[1] << 8);
+				(u32)poll_result.rc_data);
 
 		ir_input_nokey(ir->input->input_dev, &ir->ir);
 		ir->key = 0;
diff --git a/drivers/staging/wlags49_h2/wl_cs.c b/drivers/staging/wlags49_h2/wl_cs.c
index 19c3354..6555891 100644
--- a/drivers/staging/wlags49_h2/wl_cs.c
+++ b/drivers/staging/wlags49_h2/wl_cs.c
@@ -83,7 +83,6 @@
 #include <linux/if_arp.h>
 #include <linux/ioport.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
@@ -147,10 +146,9 @@
 
 	link->resource[0]->end  = HCF_NUM_IO_PORTS;
 	link->resource[0]->flags= IO_DATA_PATH_WIDTH_16;
-	link->conf.Attributes   = CONF_ENABLE_IRQ;
-	link->conf.IntType      = INT_MEMORY_AND_IO;
-	link->conf.ConfigIndex  = 5;
-	link->conf.Present      = PRESENT_OPTION;
+	link->config_flags     |= CONF_ENABLE_IRQ;
+	link->config_index      = 5;
+	link->config_regs       = PRESENT_OPTION;
 
 	link->priv = dev;
 	lp = wl_priv(dev);
@@ -165,27 +163,6 @@
 
 
 
-/*******************************************************************************
- *	wl_adapter_detach()
- *******************************************************************************
- *
- *  DESCRIPTION:
- *
- *      This deletes a driver "instance". The device is de-registered with Card
- *  Services. If it has been released, then the net device is unregistered, and
- *  all local data structures are freed. Otherwise, the structures will be
- *  freed when the device is released.
- *
- *  PARAMETERS:
- *
- *      link    - pointer to the dev_link_t structure representing the device to
- *                detach
- *
- *  RETURNS:
- *
- *      N/A
- *
- ******************************************************************************/
 static void wl_adapter_detach(struct pcmcia_device *link)
 {
 	struct net_device   *dev = link->priv;
@@ -209,26 +186,6 @@
 /*============================================================================*/
 
 
-/*******************************************************************************
- *	wl_adapter_release()
- *******************************************************************************
- *
- *  DESCRIPTION:
- *
- *      After a card is removed, this routine will release the PCMCIA
- *  configuration. If the device is still open, this will be postponed until it
- *  is closed.
- *
- *  PARAMETERS:
- *
- *      arg - a u_long representing a pointer to a dev_link_t structure for the
- *            device to be released.
- *
- *  RETURNS:
- *
- *      N/A
- *
- ******************************************************************************/
 void wl_adapter_release(struct pcmcia_device *link)
 {
 	DBG_FUNC("wl_adapter_release");
@@ -268,26 +225,6 @@
 	return 0;
 } /* wl_adapter_resume */
 
-/*******************************************************************************
- *	wl_adapter_insert()
- *******************************************************************************
- *
- *  DESCRIPTION:
- *
- *      wl_adapter_insert() is scheduled to run after a CARD_INSERTION event is
- *  received, to configure the PCMCIA socket, and to make the ethernet device
- *  available to the system.
- *
- *  PARAMETERS:
- *
- *      link    - pointer to the dev_link_t structure representing the device to
- *                insert
- *
- *  RETURNS:
- *
- *      N/A
- *
- ******************************************************************************/
 void wl_adapter_insert(struct pcmcia_device *link)
 {
 	struct net_device *dev;
@@ -302,7 +239,7 @@
 	dev     = link->priv;
 
 	/* Do we need to allocate an interrupt? */
-	link->conf.Attributes |= CONF_ENABLE_IRQ;
+	link->config_flags |= CONF_ENABLE_IRQ;
 	link->io_lines = 6;
 
 	ret = pcmcia_request_io(link);
@@ -313,7 +250,7 @@
 	if (ret != 0)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret != 0)
 		goto failed;
 
@@ -457,9 +394,7 @@
 
 static struct pcmcia_driver wlags49_driver = {
 	.owner	    = THIS_MODULE,
-	.drv	    = {
-		.name = DRIVER_NAME,
-	},
+	.name	    = DRIVER_NAME,
 	.probe	    = wl_adapter_attach,
 	.remove	    = wl_adapter_detach,
 	.id_table   = wl_adapter_ids,
diff --git a/drivers/staging/wlags49_h2/wl_internal.h b/drivers/staging/wlags49_h2/wl_internal.h
index 02f0a20..cd129b3 100644
--- a/drivers/staging/wlags49_h2/wl_internal.h
+++ b/drivers/staging/wlags49_h2/wl_internal.h
@@ -69,7 +69,6 @@
  ******************************************************************************/
 #include <linux/version.h>
 #ifdef BUS_PCMCIA
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ciscode.h>
diff --git a/drivers/staging/wlags49_h2/wl_main.c b/drivers/staging/wlags49_h2/wl_main.c
index 88d0d47..8e3536a 100644
--- a/drivers/staging/wlags49_h2/wl_main.c
+++ b/drivers/staging/wlags49_h2/wl_main.c
@@ -414,25 +414,6 @@
 #endif /* HCF_STA */
 
 
-/*******************************************************************************
- *	wl_insert()
- *******************************************************************************
- *
- *  DESCRIPTION:
- *
- *      wl_insert() is scheduled to run after a CARD_INSERTION event is
- *  received, to configure the PCMCIA socket, and to make the ethernet device
- *  available to the system.
- *
- *  PARAMETERS:
- *
- *      dev - a pointer to the net_device struct of the wireless device
- *
- *  RETURNS:
- *
- *      TRUE or FALSE
- *
- ******************************************************************************/
 int wl_insert( struct net_device *dev )
 {
 	int                     result = 0;
diff --git a/drivers/telephony/ixj_pcmcia.c b/drivers/telephony/ixj_pcmcia.c
index a1900e5..d005b9e 100644
--- a/drivers/telephony/ixj_pcmcia.c
+++ b/drivers/telephony/ixj_pcmcia.c
@@ -8,7 +8,6 @@
 #include <linux/errno.h>	/* error codes */
 #include <linux/slab.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
@@ -32,9 +31,6 @@
 {
 	dev_dbg(&p_dev->dev, "ixj_attach()\n");
 	/* Create new ixj device */
-	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
-	p_dev->conf.IntType = INT_MEMORY_AND_IO;
 	p_dev->priv = kzalloc(sizeof(struct ixj_info_t), GFP_KERNEL);
 	if (!p_dev->priv) {
 		return -ENOMEM;
@@ -111,40 +107,31 @@
 	return;
 }
 
-static int ixj_config_check(struct pcmcia_device *p_dev,
-			    cistpl_cftable_entry_t *cfg,
-			    cistpl_cftable_entry_t *dflt,
-			    unsigned int vcc,
-			    void *priv_data)
+static int ixj_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-		p_dev->io_lines = 3;
-		if (io->nwin == 2) {
-			p_dev->resource[1]->start = io->win[1].base;
-			p_dev->resource[1]->end = io->win[1].len;
-		}
-		if (!pcmcia_request_io(p_dev))
-			return 0;
-	}
-	return -ENODEV;
+	p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->resource[1]->flags &= ~IO_DATA_PATH_WIDTH;
+	p_dev->resource[1]->flags |= IO_DATA_PATH_WIDTH_8;
+	p_dev->io_lines = 3;
+
+	return pcmcia_request_io(p_dev);
 }
 
 static int ixj_config(struct pcmcia_device * link)
 {
 	IXJ *j;
 	ixj_info_t *info;
-	cistpl_cftable_entry_t dflt = { 0 };
 
 	info = link->priv;
 	dev_dbg(&link->dev, "ixj_config\n");
 
-	if (pcmcia_loop_config(link, ixj_config_check, &dflt))
+	link->config_flags = CONF_AUTO_SET_IO;
+
+	if (pcmcia_loop_config(link, ixj_config_check, NULL))
 		goto failed;
 
-	if (pcmcia_request_configuration(link, &link->conf))
+	if (pcmcia_enable_device(link))
 		goto failed;
 
 	/*
@@ -178,9 +165,7 @@
 
 static struct pcmcia_driver ixj_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "ixj_cs",
-	},
+	.name		= "ixj_cs",
 	.probe		= ixj_probe,
 	.remove		= ixj_detach,
 	.id_table	= ixj_ids,
diff --git a/drivers/usb/host/sl811_cs.c b/drivers/usb/host/sl811_cs.c
index 0e13a00..3775c03 100644
--- a/drivers/usb/host/sl811_cs.c
+++ b/drivers/usb/host/sl811_cs.c
@@ -20,7 +20,6 @@
 #include <linux/ioport.h>
 #include <linux/platform_device.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/cisreg.h>
 #include <pcmcia/ds.h>
@@ -132,49 +131,12 @@
 	platform_device_unregister(&platform_dev);
 }
 
-static int sl811_cs_config_check(struct pcmcia_device *p_dev,
-				 cistpl_cftable_entry_t *cfg,
-				 cistpl_cftable_entry_t *dflt,
-				 unsigned int vcc,
-				 void *priv_data)
+static int sl811_cs_config_check(struct pcmcia_device *p_dev, void *priv_data)
 {
-	if (cfg->index == 0)
-		return -ENODEV;
+	if (p_dev->config_index == 0)
+		return -EINVAL;
 
-	/* Use power settings for Vcc and Vpp if present */
-	/*  Note that the CIS values need to be rescaled */
-	if (cfg->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-		if (cfg->vcc.param[CISTPL_POWER_VNOM]/10000 != vcc)
-			return -ENODEV;
-	} else if (dflt->vcc.present & (1<<CISTPL_POWER_VNOM)) {
-		if (dflt->vcc.param[CISTPL_POWER_VNOM]/10000 != vcc)
-			return -ENODEV;
-		}
-
-	if (cfg->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			cfg->vpp1.param[CISTPL_POWER_VNOM]/10000;
-	else if (dflt->vpp1.present & (1<<CISTPL_POWER_VNOM))
-		p_dev->conf.Vpp =
-			dflt->vpp1.param[CISTPL_POWER_VNOM]/10000;
-
-	/* we need an interrupt */
-	p_dev->conf.Attributes |= CONF_ENABLE_IRQ;
-
-	/* IO window settings */
-	p_dev->resource[0]->end = p_dev->resource[1]->end = 0;
-	if ((cfg->io.nwin > 0) || (dflt->io.nwin > 0)) {
-		cistpl_io_t *io = (cfg->io.nwin) ? &cfg->io : &dflt->io;
-		p_dev->io_lines = io->flags & CISTPL_IO_LINES_MASK;
-
-		p_dev->resource[0]->flags |= IO_DATA_PATH_WIDTH_8;
-		p_dev->resource[0]->start = io->win[0].base;
-		p_dev->resource[0]->end = io->win[0].len;
-
-		return pcmcia_request_io(p_dev);
-	}
-	pcmcia_disable_device(p_dev);
-	return -ENODEV;
+	return pcmcia_request_io(p_dev);
 }
 
 
@@ -185,6 +147,9 @@
 
 	dev_dbg(&link->dev, "sl811_cs_config\n");
 
+	link->config_flags |= CONF_ENABLE_IRQ |	CONF_AUTO_SET_VPP |
+		CONF_AUTO_CHECK_VCC | CONF_AUTO_SET_IO;
+
 	if (pcmcia_loop_config(link, sl811_cs_config_check, NULL))
 		goto failed;
 
@@ -195,18 +160,10 @@
 	if (!link->irq)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
-	dev_info(&link->dev, "index 0x%02x: ",
-		link->conf.ConfigIndex);
-	if (link->conf.Vpp)
-		printk(", Vpp %d.%d", link->conf.Vpp/10, link->conf.Vpp%10);
-	printk(", irq %d", link->irq);
-	printk(", io %pR", link->resource[0]);
-	printk("\n");
-
 	if (sl811_hc_init(parent, link->resource[0]->start, link->irq)
 			< 0) {
 failed:
@@ -227,9 +184,6 @@
 	local->p_dev = link;
 	link->priv = local;
 
-	link->conf.Attributes = 0;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-
 	return sl811_cs_config(link);
 }
 
@@ -241,9 +195,7 @@
 
 static struct pcmcia_driver sl811_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "sl811_cs",
-	},
+	.name		= "sl811_cs",
 	.probe		= sl811_cs_probe,
 	.remove		= sl811_cs_detach,
 	.id_table	= sl811_ids,
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 7c80082..17927b1 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -127,7 +127,10 @@
 	size_t len, total_len = 0;
 	int err, wmem;
 	size_t hdr_size;
-	struct socket *sock = rcu_dereference(vq->private_data);
+	struct socket *sock;
+
+	sock = rcu_dereference_check(vq->private_data,
+				     lockdep_is_held(&vq->mutex));
 	if (!sock)
 		return;
 
@@ -582,7 +585,10 @@
 static void vhost_net_enable_vq(struct vhost_net *n,
 				struct vhost_virtqueue *vq)
 {
-	struct socket *sock = vq->private_data;
+	struct socket *sock;
+
+	sock = rcu_dereference_protected(vq->private_data,
+					 lockdep_is_held(&vq->mutex));
 	if (!sock)
 		return;
 	if (vq == n->vqs + VHOST_NET_VQ_TX) {
@@ -598,7 +604,8 @@
 	struct socket *sock;
 
 	mutex_lock(&vq->mutex);
-	sock = vq->private_data;
+	sock = rcu_dereference_protected(vq->private_data,
+					 lockdep_is_held(&vq->mutex));
 	vhost_net_disable_vq(n, vq);
 	rcu_assign_pointer(vq->private_data, NULL);
 	mutex_unlock(&vq->mutex);
@@ -736,7 +743,8 @@
 	}
 
 	/* start polling new socket */
-	oldsock = vq->private_data;
+	oldsock = rcu_dereference_protected(vq->private_data,
+					    lockdep_is_held(&vq->mutex));
 	if (sock != oldsock) {
                 vhost_net_disable_vq(n, vq);
                 rcu_assign_pointer(vq->private_data, sock);
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index dd3d6f7..8b5a1b3 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -320,7 +320,7 @@
 	vhost_dev_cleanup(dev);
 
 	memory->nregions = 0;
-	dev->memory = memory;
+	RCU_INIT_POINTER(dev->memory, memory);
 	return 0;
 }
 
@@ -352,8 +352,9 @@
 		fput(dev->log_file);
 	dev->log_file = NULL;
 	/* No one will access memory at this point */
-	kfree(dev->memory);
-	dev->memory = NULL;
+	kfree(rcu_dereference_protected(dev->memory,
+					lockdep_is_held(&dev->mutex)));
+	RCU_INIT_POINTER(dev->memory, NULL);
 	if (dev->mm)
 		mmput(dev->mm);
 	dev->mm = NULL;
@@ -440,14 +441,22 @@
 /* Caller should have device mutex but not vq mutex */
 int vhost_log_access_ok(struct vhost_dev *dev)
 {
-	return memory_access_ok(dev, dev->memory, 1);
+	struct vhost_memory *mp;
+
+	mp = rcu_dereference_protected(dev->memory,
+				       lockdep_is_held(&dev->mutex));
+	return memory_access_ok(dev, mp, 1);
 }
 
 /* Verify access for write logging. */
 /* Caller should have vq mutex and device mutex */
 static int vq_log_access_ok(struct vhost_virtqueue *vq, void __user *log_base)
 {
-	return vq_memory_access_ok(log_base, vq->dev->memory,
+	struct vhost_memory *mp;
+
+	mp = rcu_dereference_protected(vq->dev->memory,
+				       lockdep_is_held(&vq->mutex));
+	return vq_memory_access_ok(log_base, mp,
 			    vhost_has_feature(vq->dev, VHOST_F_LOG_ALL)) &&
 		(!vq->log_used || log_access_ok(log_base, vq->log_addr,
 					sizeof *vq->used +
@@ -487,7 +496,8 @@
 		kfree(newmem);
 		return -EFAULT;
 	}
-	oldmem = d->memory;
+	oldmem = rcu_dereference_protected(d->memory,
+					   lockdep_is_held(&d->mutex));
 	rcu_assign_pointer(d->memory, newmem);
 	synchronize_rcu();
 	kfree(oldmem);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index afd7729..af3c11d 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -106,7 +106,7 @@
 	 * vhost_work execution acts instead of rcu_read_lock() and the end of
 	 * vhost_work execution acts instead of rcu_read_lock().
 	 * Writers use virtqueue mutex. */
-	void *private_data;
+	void __rcu *private_data;
 	/* Log write descriptors */
 	void __user *log_base;
 	struct vhost_log log[VHOST_NET_MAX_SG];
@@ -116,7 +116,7 @@
 	/* Readers use RCU to access memory table pointer
 	 * log base pointer and features.
 	 * Writers use mutex below.*/
-	struct vhost_memory *memory;
+	struct vhost_memory __rcu *memory;
 	struct mm_struct *mm;
 	struct mutex mutex;
 	unsigned acked_features;
@@ -173,7 +173,11 @@
 
 static inline int vhost_has_feature(struct vhost_dev *dev, int bit)
 {
-	unsigned acked_features = rcu_dereference(dev->acked_features);
+	unsigned acked_features;
+
+	acked_features =
+		rcu_dereference_index_check(dev->acked_features,
+					    lockdep_is_held(&dev->mutex));
 	return acked_features & (1 << bit);
 }
 
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 13365ba..7d24b0d 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -338,30 +338,29 @@
 
 static int find_unbound_irq(void)
 {
-	int irq;
-	struct irq_desc *desc;
+	struct irq_data *data;
+	int irq, res;
 
 	for (irq = 0; irq < nr_irqs; irq++) {
-		desc = irq_to_desc(irq);
+		data = irq_get_irq_data(irq);
 		/* only 0->15 have init'd desc; handle irq > 16 */
-		if (desc == NULL)
+		if (!data)
 			break;
-		if (desc->chip == &no_irq_chip)
+		if (data->chip == &no_irq_chip)
 			break;
-		if (desc->chip != &xen_dynamic_chip)
+		if (data->chip != &xen_dynamic_chip)
 			continue;
 		if (irq_info[irq].type == IRQT_UNBOUND)
-			break;
+			return irq;
 	}
 
 	if (irq == nr_irqs)
 		panic("No available IRQ to bind to: increase nr_irqs!\n");
 
-	desc = irq_to_desc_alloc_node(irq, 0);
-	if (WARN_ON(desc == NULL))
-		return -1;
+	res = irq_alloc_desc_at(irq, 0);
 
-	dynamic_irq_init_keep_chip_data(irq);
+	if (WARN_ON(res != irq))
+		return -1;
 
 	return irq;
 }
@@ -495,7 +494,7 @@
 	if (irq_info[irq].type != IRQT_UNBOUND) {
 		irq_info[irq] = mk_unbound_info();
 
-		dynamic_irq_cleanup(irq);
+		irq_free_desc(irq);
 	}
 
 	spin_unlock(&irq_mapping_update_lock);
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 33c4e7e..9581ea9 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -109,8 +109,8 @@
 {
 	struct affs_inode_info *ei = (struct affs_inode_info *) foo;
 
-	init_MUTEX(&ei->i_link_lock);
-	init_MUTEX(&ei->i_ext_lock);
+	sema_init(&ei->i_link_lock, 1);
+	sema_init(&ei->i_ext_lock, 1);
 	inode_init_once(&ei->vfs_inode);
 }
 
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index f96eff0..a6395bd 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -134,10 +134,6 @@
 		if (!dump_write(file, dump_start, dump_size))
 			goto end_coredump;
 	}
-/* Finally dump the task struct.  Not be used by gdb, but could be useful */
-	set_fs(KERNEL_DS);
-	if (!dump_write(file, current, sizeof(*current)))
-		goto end_coredump;
 end_coredump:
 	set_fs(fs);
 	return has_dumped;
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 0fcd264..9eb134e 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -1,9 +1,11 @@
 config CEPH_FS
         tristate "Ceph distributed file system (EXPERIMENTAL)"
 	depends on INET && EXPERIMENTAL
+	select CEPH_LIB
 	select LIBCRC32C
 	select CRYPTO_AES
 	select CRYPTO
+	default n
 	help
 	  Choose Y or M here to include support for mounting the
 	  experimental Ceph distributed file system.  Ceph is an extremely
@@ -14,15 +16,3 @@
 
 	  If unsure, say N.
 
-config CEPH_FS_PRETTYDEBUG
-	bool "Include file:line in ceph debug output"
-	depends on CEPH_FS
-	default n
-	help
-	  If you say Y here, debug output will include a filename and
-	  line to aid debugging.  This icnreases kernel size and slows
-	  execution slightly when debug call sites are enabled (e.g.,
-	  via CONFIG_DYNAMIC_DEBUG).
-
-	  If unsure, say N.
-
diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile
index 278e117..9e6c4f2 100644
--- a/fs/ceph/Makefile
+++ b/fs/ceph/Makefile
@@ -8,15 +8,8 @@
 
 ceph-objs := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \
 	export.o caps.o snap.o xattr.o \
-	messenger.o msgpool.o buffer.o pagelist.o \
-	mds_client.o mdsmap.o \
-	mon_client.o \
-	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
-	debugfs.o \
-	auth.o auth_none.o \
-	crypto.o armor.o \
-	auth_x.o \
-	ceph_fs.o ceph_strings.o ceph_hash.o ceph_frag.o
+	mds_client.o mdsmap.o strings.o ceph_frag.o \
+	debugfs.o
 
 else
 #Otherwise we were called directly from the command
diff --git a/fs/ceph/README b/fs/ceph/README
deleted file mode 100644
index 18352fa..0000000
--- a/fs/ceph/README
+++ /dev/null
@@ -1,20 +0,0 @@
-#
-# The following files are shared by (and manually synchronized
-# between) the Ceph userland and kernel client.
-#
-# userland                  kernel
-src/include/ceph_fs.h	    fs/ceph/ceph_fs.h
-src/include/ceph_fs.cc	    fs/ceph/ceph_fs.c
-src/include/msgr.h	    fs/ceph/msgr.h
-src/include/rados.h	    fs/ceph/rados.h
-src/include/ceph_strings.cc fs/ceph/ceph_strings.c
-src/include/ceph_frag.h	    fs/ceph/ceph_frag.h
-src/include/ceph_frag.cc    fs/ceph/ceph_frag.c
-src/include/ceph_hash.h	    fs/ceph/ceph_hash.h
-src/include/ceph_hash.cc    fs/ceph/ceph_hash.c
-src/crush/crush.c	    fs/ceph/crush/crush.c
-src/crush/crush.h	    fs/ceph/crush/crush.h
-src/crush/mapper.c	    fs/ceph/crush/mapper.c
-src/crush/mapper.h	    fs/ceph/crush/mapper.h
-src/crush/hash.h	    fs/ceph/crush/hash.h
-src/crush/hash.c	    fs/ceph/crush/hash.c
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index efbc604..51bcc5c 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/backing-dev.h>
 #include <linux/fs.h>
@@ -10,7 +10,8 @@
 #include <linux/task_io_accounting_ops.h>
 
 #include "super.h"
-#include "osd_client.h"
+#include "mds_client.h"
+#include <linux/ceph/osd_client.h>
 
 /*
  * Ceph address space ops.
@@ -193,7 +194,8 @@
 {
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->osdc;
+	struct ceph_osd_client *osdc = 
+		&ceph_inode_to_client(inode)->client->osdc;
 	int err = 0;
 	u64 len = PAGE_CACHE_SIZE;
 
@@ -265,7 +267,8 @@
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_inode_to_client(inode)->client->osdc;
 	int rc = 0;
 	struct page **pages;
 	loff_t offset;
@@ -365,7 +368,7 @@
 {
 	struct inode *inode;
 	struct ceph_inode_info *ci;
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	struct ceph_osd_client *osdc;
 	loff_t page_off = page->index << PAGE_CACHE_SHIFT;
 	int len = PAGE_CACHE_SIZE;
@@ -383,8 +386,8 @@
 	}
 	inode = page->mapping->host;
 	ci = ceph_inode(inode);
-	client = ceph_inode_to_client(inode);
-	osdc = &client->osdc;
+	fsc = ceph_inode_to_client(inode);
+	osdc = &fsc->client->osdc;
 
 	/* verify this is a writeable snap context */
 	snapc = (void *)page->private;
@@ -414,10 +417,10 @@
 	dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
 	     inode, page, page->index, page_off, len, snapc);
 
-	writeback_stat = atomic_long_inc_return(&client->writeback_count);
+	writeback_stat = atomic_long_inc_return(&fsc->writeback_count);
 	if (writeback_stat >
-	    CONGESTION_ON_THRESH(client->mount_args->congestion_kb))
-		set_bdi_congested(&client->backing_dev_info, BLK_RW_ASYNC);
+	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
+		set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
 
 	set_page_writeback(page);
 	err = ceph_osdc_writepages(osdc, ceph_vino(inode),
@@ -496,7 +499,7 @@
 	struct address_space *mapping = inode->i_mapping;
 	__s32 rc = -EIO;
 	u64 bytes = 0;
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	long writeback_stat;
 	unsigned issued = ceph_caps_issued(ci);
 
@@ -529,10 +532,10 @@
 		WARN_ON(!PageUptodate(page));
 
 		writeback_stat =
-			atomic_long_dec_return(&client->writeback_count);
+			atomic_long_dec_return(&fsc->writeback_count);
 		if (writeback_stat <
-		    CONGESTION_OFF_THRESH(client->mount_args->congestion_kb))
-			clear_bdi_congested(&client->backing_dev_info,
+		    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
+			clear_bdi_congested(&fsc->backing_dev_info,
 					    BLK_RW_ASYNC);
 
 		ceph_put_snap_context((void *)page->private);
@@ -569,13 +572,13 @@
  * mempool.  we avoid the mempool if we can because req->r_num_pages
  * may be less than the maximum write size.
  */
-static void alloc_page_vec(struct ceph_client *client,
+static void alloc_page_vec(struct ceph_fs_client *fsc,
 			   struct ceph_osd_request *req)
 {
 	req->r_pages = kmalloc(sizeof(struct page *) * req->r_num_pages,
 			       GFP_NOFS);
 	if (!req->r_pages) {
-		req->r_pages = mempool_alloc(client->wb_pagevec_pool, GFP_NOFS);
+		req->r_pages = mempool_alloc(fsc->wb_pagevec_pool, GFP_NOFS);
 		req->r_pages_from_pool = 1;
 		WARN_ON(!req->r_pages);
 	}
@@ -590,7 +593,7 @@
 	struct inode *inode = mapping->host;
 	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	pgoff_t index, start, end;
 	int range_whole = 0;
 	int should_loop = 1;
@@ -617,13 +620,13 @@
 	     wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
 	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
 
-	client = ceph_inode_to_client(inode);
-	if (client->mount_state == CEPH_MOUNT_SHUTDOWN) {
+	fsc = ceph_inode_to_client(inode);
+	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
 		pr_warning("writepage_start %p on forced umount\n", inode);
 		return -EIO; /* we're in a forced umount, don't write! */
 	}
-	if (client->mount_args->wsize && client->mount_args->wsize < wsize)
-		wsize = client->mount_args->wsize;
+	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
+		wsize = fsc->mount_options->wsize;
 	if (wsize < PAGE_CACHE_SIZE)
 		wsize = PAGE_CACHE_SIZE;
 	max_pages_ever = wsize >> PAGE_CACHE_SHIFT;
@@ -769,7 +772,7 @@
 				offset = (unsigned long long)page->index
 					<< PAGE_CACHE_SHIFT;
 				len = wsize;
-				req = ceph_osdc_new_request(&client->osdc,
+				req = ceph_osdc_new_request(&fsc->client->osdc,
 					    &ci->i_layout,
 					    ceph_vino(inode),
 					    offset, &len,
@@ -782,7 +785,7 @@
 					    &inode->i_mtime, true, 1);
 				max_pages = req->r_num_pages;
 
-				alloc_page_vec(client, req);
+				alloc_page_vec(fsc, req);
 				req->r_callback = writepages_finish;
 				req->r_inode = inode;
 			}
@@ -794,10 +797,10 @@
 			     inode, page, page->index);
 
 			writeback_stat =
-			       atomic_long_inc_return(&client->writeback_count);
+			       atomic_long_inc_return(&fsc->writeback_count);
 			if (writeback_stat > CONGESTION_ON_THRESH(
-				    client->mount_args->congestion_kb)) {
-				set_bdi_congested(&client->backing_dev_info,
+				    fsc->mount_options->congestion_kb)) {
+				set_bdi_congested(&fsc->backing_dev_info,
 						  BLK_RW_ASYNC);
 			}
 
@@ -846,7 +849,7 @@
 		op->payload_len = cpu_to_le32(len);
 		req->r_request->hdr.data_len = cpu_to_le32(len);
 
-		ceph_osdc_start_request(&client->osdc, req, true);
+		ceph_osdc_start_request(&fsc->client->osdc, req, true);
 		req = NULL;
 
 		/* continue? */
@@ -915,7 +918,7 @@
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	loff_t page_off = pos & PAGE_CACHE_MASK;
 	int pos_in_page = pos & ~PAGE_CACHE_MASK;
 	int end_in_page = pos_in_page + len;
@@ -1053,8 +1056,8 @@
 			  struct page *page, void *fsdata)
 {
 	struct inode *inode = file->f_dentry->d_inode;
-	struct ceph_client *client = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
 	int check_cap = 0;
 
@@ -1123,7 +1126,7 @@
 {
 	struct inode *inode = vma->vm_file->f_dentry->d_inode;
 	struct page *page = vmf->page;
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	loff_t off = page->index << PAGE_CACHE_SHIFT;
 	loff_t size, len;
 	int ret;
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 73c1530..98ab13e 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/fs.h>
 #include <linux/kernel.h>
@@ -9,8 +9,9 @@
 #include <linux/writeback.h>
 
 #include "super.h"
-#include "decode.h"
-#include "messenger.h"
+#include "mds_client.h"
+#include <linux/ceph/decode.h>
+#include <linux/ceph/messenger.h>
 
 /*
  * Capability management
@@ -287,11 +288,11 @@
 	spin_unlock(&mdsc->caps_list_lock);
 }
 
-void ceph_reservation_status(struct ceph_client *client,
+void ceph_reservation_status(struct ceph_fs_client *fsc,
 			     int *total, int *avail, int *used, int *reserved,
 			     int *min)
 {
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 
 	if (total)
 		*total = mdsc->caps_total_count;
@@ -399,7 +400,7 @@
 static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
 			       struct ceph_inode_info *ci)
 {
-	struct ceph_mount_args *ma = mdsc->client->mount_args;
+	struct ceph_mount_options *ma = mdsc->fsc->mount_options;
 
 	ci->i_hold_caps_min = round_jiffies(jiffies +
 					    ma->caps_wanted_delay_min * HZ);
@@ -515,7 +516,7 @@
 		 unsigned seq, unsigned mseq, u64 realmino, int flags,
 		 struct ceph_cap_reservation *caps_reservation)
 {
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_cap *new_cap = NULL;
 	struct ceph_cap *cap;
@@ -873,7 +874,7 @@
 	struct ceph_mds_session *session = cap->session;
 	struct ceph_inode_info *ci = cap->ci;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	int removed = 0;
 
 	dout("__ceph_remove_cap %p from %p\n", cap, &ci->vfs_inode);
@@ -1210,7 +1211,7 @@
 	int mds;
 	struct ceph_cap_snap *capsnap;
 	u32 mseq;
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_mds_session *session = NULL; /* if session != NULL, we hold
 						    session->s_mutex */
 	u64 next_follows = 0;  /* keep track of how far we've gotten through the
@@ -1336,7 +1337,7 @@
 void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 {
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	int was = ci->i_dirty_caps;
 	int dirty = 0;
@@ -1378,7 +1379,7 @@
 static int __mark_caps_flushing(struct inode *inode,
 				 struct ceph_mds_session *session)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int flushing;
 
@@ -1416,17 +1417,6 @@
 /*
  * try to invalidate mapping pages without blocking.
  */
-static int mapping_is_empty(struct address_space *mapping)
-{
-	struct page *page = find_get_page(mapping, 0);
-
-	if (!page)
-		return 1;
-
-	put_page(page);
-	return 0;
-}
-
 static int try_nonblocking_invalidate(struct inode *inode)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1436,7 +1426,7 @@
 	invalidate_mapping_pages(&inode->i_data, 0, -1);
 	spin_lock(&inode->i_lock);
 
-	if (mapping_is_empty(&inode->i_data) &&
+	if (inode->i_data.nrpages == 0 &&
 	    invalidating_gen == ci->i_rdcache_gen) {
 		/* success. */
 		dout("try_nonblocking_invalidate %p success\n", inode);
@@ -1462,8 +1452,8 @@
 void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 		     struct ceph_mds_session *session)
 {
-	struct ceph_client *client = ceph_inode_to_client(&ci->vfs_inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->vfs_inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	struct ceph_cap *cap;
 	int file_wanted, used;
@@ -1533,7 +1523,7 @@
 	 */
 	if ((!is_delayed || mdsc->stopping) &&
 	    ci->i_wrbuffer_ref == 0 &&               /* no dirty pages... */
-	    ci->i_rdcache_gen &&                     /* may have cached pages */
+	    inode->i_data.nrpages &&                 /* have cached pages */
 	    (file_wanted == 0 ||                     /* no open files */
 	     (revoking & (CEPH_CAP_FILE_CACHE|
 			  CEPH_CAP_FILE_LAZYIO))) && /*  or revoking cache */
@@ -1706,7 +1696,7 @@
 static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session,
 			  unsigned *flush_tid)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int unlock_session = session ? 0 : 1;
 	int flushing = 0;
@@ -1872,7 +1862,7 @@
 				       caps_are_flushed(inode, flush_tid));
 	} else {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&inode->i_lock);
 		if (__ceph_caps_dirty(ci))
@@ -2283,7 +2273,8 @@
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int mds = session->s_mds;
-	int seq = le32_to_cpu(grant->seq);
+	unsigned seq = le32_to_cpu(grant->seq);
+	unsigned issue_seq = le32_to_cpu(grant->issue_seq);
 	int newcaps = le32_to_cpu(grant->caps);
 	int issued, implemented, used, wanted, dirty;
 	u64 size = le64_to_cpu(grant->size);
@@ -2295,8 +2286,8 @@
 	int revoked_rdcache = 0;
 	int queue_invalidate = 0;
 
-	dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n",
-	     inode, cap, mds, seq, ceph_cap_string(newcaps));
+	dout("handle_cap_grant inode %p cap %p mds%d seq %u/%u %s\n",
+	     inode, cap, mds, seq, issue_seq, ceph_cap_string(newcaps));
 	dout(" size %llu max_size %llu, i_size %llu\n", size, max_size,
 		inode->i_size);
 
@@ -2392,6 +2383,7 @@
 	}
 
 	cap->seq = seq;
+	cap->issue_seq = issue_seq;
 
 	/* file layout may have changed */
 	ci->i_layout = grant->layout;
@@ -2463,7 +2455,7 @@
 	__releases(inode->i_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	unsigned seq = le32_to_cpu(m->seq);
 	int dirty = le32_to_cpu(m->dirty);
 	int cleaned = 0;
@@ -2711,7 +2703,7 @@
 		      struct ceph_msg *msg)
 {
 	struct ceph_mds_client *mdsc = session->s_mdsc;
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_cap *cap;
 	struct ceph_mds_caps *h;
@@ -2774,15 +2766,7 @@
 		if (op == CEPH_CAP_OP_IMPORT)
 			__queue_cap_release(session, vino.ino, cap_id,
 					    mseq, seq);
-
-		/*
-		 * send any full release message to try to move things
-		 * along for the mds (who clearly thinks we still have this
-		 * cap).
-		 */
-		ceph_add_cap_releases(mdsc, session);
-		ceph_send_cap_releases(mdsc, session);
-		goto done;
+		goto flush_cap_releases;
 	}
 
 	/* these will work even if we don't have a cap yet */
@@ -2810,7 +2794,7 @@
 		dout(" no cap on %p ino %llx.%llx from mds%d\n",
 		     inode, ceph_ino(inode), ceph_snap(inode), mds);
 		spin_unlock(&inode->i_lock);
-		goto done;
+		goto flush_cap_releases;
 	}
 
 	/* note that each of these drops i_lock for us */
@@ -2834,6 +2818,17 @@
 		       ceph_cap_op_name(op));
 	}
 
+	goto done;
+
+flush_cap_releases:
+	/*
+	 * send any full release message to try to move things
+	 * along for the mds (who clearly thinks we still have this
+	 * cap).
+	 */
+	ceph_add_cap_releases(mdsc, session);
+	ceph_send_cap_releases(mdsc, session);
+
 done:
 	mutex_unlock(&session->s_mutex);
 done_unlocked:
diff --git a/fs/ceph/ceph_frag.c b/fs/ceph/ceph_frag.c
index ab6cf35..bdce8b1 100644
--- a/fs/ceph/ceph_frag.c
+++ b/fs/ceph/ceph_frag.c
@@ -1,7 +1,8 @@
 /*
  * Ceph 'frag' type
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
 int ceph_frag_compare(__u32 a, __u32 b)
 {
diff --git a/fs/ceph/ceph_strings.c b/fs/ceph/ceph_strings.c
deleted file mode 100644
index c6179d3..0000000
--- a/fs/ceph/ceph_strings.c
+++ /dev/null
@@ -1,193 +0,0 @@
-/*
- * Ceph string constants
- */
-#include "types.h"
-
-const char *ceph_entity_type_name(int type)
-{
-	switch (type) {
-	case CEPH_ENTITY_TYPE_MDS: return "mds";
-	case CEPH_ENTITY_TYPE_OSD: return "osd";
-	case CEPH_ENTITY_TYPE_MON: return "mon";
-	case CEPH_ENTITY_TYPE_CLIENT: return "client";
-	case CEPH_ENTITY_TYPE_AUTH: return "auth";
-	default: return "unknown";
-	}
-}
-
-const char *ceph_osd_op_name(int op)
-{
-	switch (op) {
-	case CEPH_OSD_OP_READ: return "read";
-	case CEPH_OSD_OP_STAT: return "stat";
-
-	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
-
-	case CEPH_OSD_OP_WRITE: return "write";
-	case CEPH_OSD_OP_DELETE: return "delete";
-	case CEPH_OSD_OP_TRUNCATE: return "truncate";
-	case CEPH_OSD_OP_ZERO: return "zero";
-	case CEPH_OSD_OP_WRITEFULL: return "writefull";
-	case CEPH_OSD_OP_ROLLBACK: return "rollback";
-
-	case CEPH_OSD_OP_APPEND: return "append";
-	case CEPH_OSD_OP_STARTSYNC: return "startsync";
-	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
-	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
-
-	case CEPH_OSD_OP_TMAPUP: return "tmapup";
-	case CEPH_OSD_OP_TMAPGET: return "tmapget";
-	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
-
-	case CEPH_OSD_OP_GETXATTR: return "getxattr";
-	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
-	case CEPH_OSD_OP_SETXATTR: return "setxattr";
-	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
-	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
-	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
-	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
-
-	case CEPH_OSD_OP_PULL: return "pull";
-	case CEPH_OSD_OP_PUSH: return "push";
-	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
-	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
-	case CEPH_OSD_OP_SCRUB: return "scrub";
-
-	case CEPH_OSD_OP_WRLOCK: return "wrlock";
-	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
-	case CEPH_OSD_OP_RDLOCK: return "rdlock";
-	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
-	case CEPH_OSD_OP_UPLOCK: return "uplock";
-	case CEPH_OSD_OP_DNLOCK: return "dnlock";
-
-	case CEPH_OSD_OP_CALL: return "call";
-
-	case CEPH_OSD_OP_PGLS: return "pgls";
-	}
-	return "???";
-}
-
-const char *ceph_mds_state_name(int s)
-{
-	switch (s) {
-		/* down and out */
-	case CEPH_MDS_STATE_DNE:        return "down:dne";
-	case CEPH_MDS_STATE_STOPPED:    return "down:stopped";
-		/* up and out */
-	case CEPH_MDS_STATE_BOOT:       return "up:boot";
-	case CEPH_MDS_STATE_STANDBY:    return "up:standby";
-	case CEPH_MDS_STATE_STANDBY_REPLAY:    return "up:standby-replay";
-	case CEPH_MDS_STATE_CREATING:   return "up:creating";
-	case CEPH_MDS_STATE_STARTING:   return "up:starting";
-		/* up and in */
-	case CEPH_MDS_STATE_REPLAY:     return "up:replay";
-	case CEPH_MDS_STATE_RESOLVE:    return "up:resolve";
-	case CEPH_MDS_STATE_RECONNECT:  return "up:reconnect";
-	case CEPH_MDS_STATE_REJOIN:     return "up:rejoin";
-	case CEPH_MDS_STATE_CLIENTREPLAY: return "up:clientreplay";
-	case CEPH_MDS_STATE_ACTIVE:     return "up:active";
-	case CEPH_MDS_STATE_STOPPING:   return "up:stopping";
-	}
-	return "???";
-}
-
-const char *ceph_session_op_name(int op)
-{
-	switch (op) {
-	case CEPH_SESSION_REQUEST_OPEN: return "request_open";
-	case CEPH_SESSION_OPEN: return "open";
-	case CEPH_SESSION_REQUEST_CLOSE: return "request_close";
-	case CEPH_SESSION_CLOSE: return "close";
-	case CEPH_SESSION_REQUEST_RENEWCAPS: return "request_renewcaps";
-	case CEPH_SESSION_RENEWCAPS: return "renewcaps";
-	case CEPH_SESSION_STALE: return "stale";
-	case CEPH_SESSION_RECALL_STATE: return "recall_state";
-	}
-	return "???";
-}
-
-const char *ceph_mds_op_name(int op)
-{
-	switch (op) {
-	case CEPH_MDS_OP_LOOKUP:  return "lookup";
-	case CEPH_MDS_OP_LOOKUPHASH:  return "lookuphash";
-	case CEPH_MDS_OP_LOOKUPPARENT:  return "lookupparent";
-	case CEPH_MDS_OP_GETATTR:  return "getattr";
-	case CEPH_MDS_OP_SETXATTR: return "setxattr";
-	case CEPH_MDS_OP_SETATTR: return "setattr";
-	case CEPH_MDS_OP_RMXATTR: return "rmxattr";
-	case CEPH_MDS_OP_READDIR: return "readdir";
-	case CEPH_MDS_OP_MKNOD: return "mknod";
-	case CEPH_MDS_OP_LINK: return "link";
-	case CEPH_MDS_OP_UNLINK: return "unlink";
-	case CEPH_MDS_OP_RENAME: return "rename";
-	case CEPH_MDS_OP_MKDIR: return "mkdir";
-	case CEPH_MDS_OP_RMDIR: return "rmdir";
-	case CEPH_MDS_OP_SYMLINK: return "symlink";
-	case CEPH_MDS_OP_CREATE: return "create";
-	case CEPH_MDS_OP_OPEN: return "open";
-	case CEPH_MDS_OP_LOOKUPSNAP: return "lookupsnap";
-	case CEPH_MDS_OP_LSSNAP: return "lssnap";
-	case CEPH_MDS_OP_MKSNAP: return "mksnap";
-	case CEPH_MDS_OP_RMSNAP: return "rmsnap";
-	case CEPH_MDS_OP_SETFILELOCK: return "setfilelock";
-	case CEPH_MDS_OP_GETFILELOCK: return "getfilelock";
-	}
-	return "???";
-}
-
-const char *ceph_cap_op_name(int op)
-{
-	switch (op) {
-	case CEPH_CAP_OP_GRANT: return "grant";
-	case CEPH_CAP_OP_REVOKE: return "revoke";
-	case CEPH_CAP_OP_TRUNC: return "trunc";
-	case CEPH_CAP_OP_EXPORT: return "export";
-	case CEPH_CAP_OP_IMPORT: return "import";
-	case CEPH_CAP_OP_UPDATE: return "update";
-	case CEPH_CAP_OP_DROP: return "drop";
-	case CEPH_CAP_OP_FLUSH: return "flush";
-	case CEPH_CAP_OP_FLUSH_ACK: return "flush_ack";
-	case CEPH_CAP_OP_FLUSHSNAP: return "flushsnap";
-	case CEPH_CAP_OP_FLUSHSNAP_ACK: return "flushsnap_ack";
-	case CEPH_CAP_OP_RELEASE: return "release";
-	case CEPH_CAP_OP_RENEW: return "renew";
-	}
-	return "???";
-}
-
-const char *ceph_lease_op_name(int o)
-{
-	switch (o) {
-	case CEPH_MDS_LEASE_REVOKE: return "revoke";
-	case CEPH_MDS_LEASE_RELEASE: return "release";
-	case CEPH_MDS_LEASE_RENEW: return "renew";
-	case CEPH_MDS_LEASE_REVOKE_ACK: return "revoke_ack";
-	}
-	return "???";
-}
-
-const char *ceph_snap_op_name(int o)
-{
-	switch (o) {
-	case CEPH_SNAP_OP_UPDATE: return "update";
-	case CEPH_SNAP_OP_CREATE: return "create";
-	case CEPH_SNAP_OP_DESTROY: return "destroy";
-	case CEPH_SNAP_OP_SPLIT: return "split";
-	}
-	return "???";
-}
-
-const char *ceph_pool_op_name(int op)
-{
-	switch (op) {
-	case POOL_OP_CREATE: return "create";
-	case POOL_OP_DELETE: return "delete";
-	case POOL_OP_AUID_CHANGE: return "auid change";
-	case POOL_OP_CREATE_SNAP: return "create snap";
-	case POOL_OP_DELETE_SNAP: return "delete snap";
-	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
-	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
-	}
-	return "???";
-}
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 6fd8b20..7ae1b3d 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/device.h>
 #include <linux/slab.h>
@@ -7,143 +7,49 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
+
 #include "super.h"
-#include "mds_client.h"
-#include "mon_client.h"
-#include "auth.h"
 
 #ifdef CONFIG_DEBUG_FS
 
-/*
- * Implement /sys/kernel/debug/ceph fun
- *
- * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
- *      .../osdmap      - current osdmap
- *      .../mdsmap      - current mdsmap
- *      .../monmap      - current monmap
- *      .../osdc        - active osd requests
- *      .../mdsc        - active mds requests
- *      .../monc        - mon client state
- *      .../dentry_lru  - dump contents of dentry lru
- *      .../caps        - expose cap (reservation) stats
- *      .../bdi         - symlink to ../../bdi/something
- */
-
-static struct dentry *ceph_debugfs_dir;
-
-static int monmap_show(struct seq_file *s, void *p)
-{
-	int i;
-	struct ceph_client *client = s->private;
-
-	if (client->monc.monmap == NULL)
-		return 0;
-
-	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
-	for (i = 0; i < client->monc.monmap->num_mon; i++) {
-		struct ceph_entity_inst *inst =
-			&client->monc.monmap->mon_inst[i];
-
-		seq_printf(s, "\t%s%lld\t%s\n",
-			   ENTITY_NAME(inst->name),
-			   pr_addr(&inst->addr.in_addr));
-	}
-	return 0;
-}
+#include "mds_client.h"
 
 static int mdsmap_show(struct seq_file *s, void *p)
 {
 	int i;
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 
-	if (client->mdsc.mdsmap == NULL)
+	if (fsc->mdsc == NULL || fsc->mdsc->mdsmap == NULL)
 		return 0;
-	seq_printf(s, "epoch %d\n", client->mdsc.mdsmap->m_epoch);
-	seq_printf(s, "root %d\n", client->mdsc.mdsmap->m_root);
+	seq_printf(s, "epoch %d\n", fsc->mdsc->mdsmap->m_epoch);
+	seq_printf(s, "root %d\n", fsc->mdsc->mdsmap->m_root);
 	seq_printf(s, "session_timeout %d\n",
-		       client->mdsc.mdsmap->m_session_timeout);
+		       fsc->mdsc->mdsmap->m_session_timeout);
 	seq_printf(s, "session_autoclose %d\n",
-		       client->mdsc.mdsmap->m_session_autoclose);
-	for (i = 0; i < client->mdsc.mdsmap->m_max_mds; i++) {
+		       fsc->mdsc->mdsmap->m_session_autoclose);
+	for (i = 0; i < fsc->mdsc->mdsmap->m_max_mds; i++) {
 		struct ceph_entity_addr *addr =
-			&client->mdsc.mdsmap->m_info[i].addr;
-		int state = client->mdsc.mdsmap->m_info[i].state;
+			&fsc->mdsc->mdsmap->m_info[i].addr;
+		int state = fsc->mdsc->mdsmap->m_info[i].state;
 
-		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i, pr_addr(&addr->in_addr),
+		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i,
+			       ceph_pr_addr(&addr->in_addr),
 			       ceph_mds_state_name(state));
 	}
 	return 0;
 }
 
-static int osdmap_show(struct seq_file *s, void *p)
-{
-	int i;
-	struct ceph_client *client = s->private;
-	struct rb_node *n;
-
-	if (client->osdc.osdmap == NULL)
-		return 0;
-	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
-	seq_printf(s, "flags%s%s\n",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
-		   " NEARFULL" : "",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
-		   " FULL" : "");
-	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
-		struct ceph_pg_pool_info *pool =
-			rb_entry(n, struct ceph_pg_pool_info, node);
-		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
-			   pool->id, pool->v.pg_num, pool->pg_num_mask,
-			   pool->v.lpg_num, pool->lpg_num_mask);
-	}
-	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
-		struct ceph_entity_addr *addr =
-			&client->osdc.osdmap->osd_addr[i];
-		int state = client->osdc.osdmap->osd_state[i];
-		char sb[64];
-
-		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
-			   i, pr_addr(&addr->in_addr),
-			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
-			   ceph_osdmap_state_str(sb, sizeof(sb), state));
-	}
-	return 0;
-}
-
-static int monc_show(struct seq_file *s, void *p)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_client *monc = &client->monc;
-	struct rb_node *rp;
-
-	mutex_lock(&monc->mutex);
-
-	if (monc->have_mdsmap)
-		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
-	if (monc->have_osdmap)
-		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
-	if (monc->want_next_osdmap)
-		seq_printf(s, "want next osdmap\n");
-
-	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
-		__u16 op;
-		req = rb_entry(rp, struct ceph_mon_generic_request, node);
-		op = le16_to_cpu(req->request->hdr.type);
-		if (op == CEPH_MSG_STATFS)
-			seq_printf(s, "%lld statfs\n", req->tid);
-		else
-			seq_printf(s, "%lld unknown\n", req->tid);
-	}
-
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
+/*
+ * mdsc debugfs
+ */
 static int mdsc_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = s->private;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	struct rb_node *rp;
 	int pathlen;
@@ -214,61 +120,12 @@
 	return 0;
 }
 
-static int osdc_show(struct seq_file *s, void *pp)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_osd_client *osdc = &client->osdc;
-	struct rb_node *p;
-
-	mutex_lock(&osdc->request_mutex);
-	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
-		struct ceph_osd_request *req;
-		struct ceph_osd_request_head *head;
-		struct ceph_osd_op *op;
-		int num_ops;
-		int opcode, olen;
-		int i;
-
-		req = rb_entry(p, struct ceph_osd_request, r_node);
-
-		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
-			   req->r_osd ? req->r_osd->o_osd : -1,
-			   le32_to_cpu(req->r_pgid.pool),
-			   le16_to_cpu(req->r_pgid.ps));
-
-		head = req->r_request->front.iov_base;
-		op = (void *)(head + 1);
-
-		num_ops = le16_to_cpu(head->num_ops);
-		olen = le32_to_cpu(head->object_len);
-		seq_printf(s, "%.*s", olen,
-			   (const char *)(head->ops + num_ops));
-
-		if (req->r_reassert_version.epoch)
-			seq_printf(s, "\t%u'%llu",
-			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
-			   le64_to_cpu(req->r_reassert_version.version));
-		else
-			seq_printf(s, "\t");
-
-		for (i = 0; i < num_ops; i++) {
-			opcode = le16_to_cpu(op->op);
-			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
-			op++;
-		}
-
-		seq_printf(s, "\n");
-	}
-	mutex_unlock(&osdc->request_mutex);
-	return 0;
-}
-
 static int caps_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 	int total, avail, used, reserved, min;
 
-	ceph_reservation_status(client, &total, &avail, &used, &reserved, &min);
+	ceph_reservation_status(fsc, &total, &avail, &used, &reserved, &min);
 	seq_printf(s, "total\t\t%d\n"
 		   "avail\t\t%d\n"
 		   "used\t\t%d\n"
@@ -280,8 +137,8 @@
 
 static int dentry_lru_show(struct seq_file *s, void *ptr)
 {
-	struct ceph_client *client = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = s->private;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_dentry_info *di;
 
 	spin_lock(&mdsc->dentry_lru_lock);
@@ -295,199 +152,124 @@
 	return 0;
 }
 
-#define DEFINE_SHOW_FUNC(name)						\
-static int name##_open(struct inode *inode, struct file *file)		\
-{									\
-	struct seq_file *sf;						\
-	int ret;							\
-									\
-	ret = single_open(file, name, NULL);				\
-	sf = file->private_data;					\
-	sf->private = inode->i_private;					\
-	return ret;							\
-}									\
-									\
-static const struct file_operations name##_fops = {			\
-	.open		= name##_open,					\
-	.read		= seq_read,					\
-	.llseek		= seq_lseek,					\
-	.release	= single_release,				\
-};
+CEPH_DEFINE_SHOW_FUNC(mdsmap_show)
+CEPH_DEFINE_SHOW_FUNC(mdsc_show)
+CEPH_DEFINE_SHOW_FUNC(caps_show)
+CEPH_DEFINE_SHOW_FUNC(dentry_lru_show)
 
-DEFINE_SHOW_FUNC(monmap_show)
-DEFINE_SHOW_FUNC(mdsmap_show)
-DEFINE_SHOW_FUNC(osdmap_show)
-DEFINE_SHOW_FUNC(monc_show)
-DEFINE_SHOW_FUNC(mdsc_show)
-DEFINE_SHOW_FUNC(osdc_show)
-DEFINE_SHOW_FUNC(dentry_lru_show)
-DEFINE_SHOW_FUNC(caps_show)
 
+/*
+ * debugfs
+ */
 static int congestion_kb_set(void *data, u64 val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
 
-	if (client)
-		client->mount_args->congestion_kb = (int)val;
-
+	fsc->mount_options->congestion_kb = (int)val;
 	return 0;
 }
 
 static int congestion_kb_get(void *data, u64 *val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
 
-	if (client)
-		*val = (u64)client->mount_args->congestion_kb;
-
+	*val = (u64)fsc->mount_options->congestion_kb;
 	return 0;
 }
 
-
 DEFINE_SIMPLE_ATTRIBUTE(congestion_kb_fops, congestion_kb_get,
 			congestion_kb_set, "%llu\n");
 
-int __init ceph_debugfs_init(void)
+
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
 {
-	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
-	if (!ceph_debugfs_dir)
-		return -ENOMEM;
-	return 0;
+	dout("ceph_fs_debugfs_cleanup\n");
+	debugfs_remove(fsc->debugfs_bdi);
+	debugfs_remove(fsc->debugfs_congestion_kb);
+	debugfs_remove(fsc->debugfs_mdsmap);
+	debugfs_remove(fsc->debugfs_caps);
+	debugfs_remove(fsc->debugfs_mdsc);
+	debugfs_remove(fsc->debugfs_dentry_lru);
 }
 
-void ceph_debugfs_cleanup(void)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
-	debugfs_remove(ceph_debugfs_dir);
-}
+	char name[100];
+	int err = -ENOMEM;
 
-int ceph_debugfs_client_init(struct ceph_client *client)
-{
-	int ret = 0;
-	char name[80];
-
-	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
-		 client->monc.auth->global_id);
-
-	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
-	if (!client->debugfs_dir)
-		goto out;
-
-	client->monc.debugfs_file = debugfs_create_file("monc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &monc_show_fops);
-	if (!client->monc.debugfs_file)
-		goto out;
-
-	client->mdsc.debugfs_file = debugfs_create_file("mdsc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &mdsc_show_fops);
-	if (!client->mdsc.debugfs_file)
-		goto out;
-
-	client->osdc.debugfs_file = debugfs_create_file("osdc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &osdc_show_fops);
-	if (!client->osdc.debugfs_file)
-		goto out;
-
-	client->debugfs_monmap = debugfs_create_file("monmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&monmap_show_fops);
-	if (!client->debugfs_monmap)
-		goto out;
-
-	client->debugfs_mdsmap = debugfs_create_file("mdsmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&mdsmap_show_fops);
-	if (!client->debugfs_mdsmap)
-		goto out;
-
-	client->debugfs_osdmap = debugfs_create_file("osdmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&osdmap_show_fops);
-	if (!client->debugfs_osdmap)
-		goto out;
-
-	client->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
-					0600,
-					client->debugfs_dir,
-					client,
-					&dentry_lru_show_fops);
-	if (!client->debugfs_dentry_lru)
-		goto out;
-
-	client->debugfs_caps = debugfs_create_file("caps",
-						   0400,
-						   client->debugfs_dir,
-						   client,
-						   &caps_show_fops);
-	if (!client->debugfs_caps)
-		goto out;
-
-	client->debugfs_congestion_kb =
+	dout("ceph_fs_debugfs_init\n");
+	fsc->debugfs_congestion_kb =
 		debugfs_create_file("writeback_congestion_kb",
 				    0600,
-				    client->debugfs_dir,
-				    client,
+				    fsc->client->debugfs_dir,
+				    fsc,
 				    &congestion_kb_fops);
-	if (!client->debugfs_congestion_kb)
+	if (!fsc->debugfs_congestion_kb)
 		goto out;
 
-	sprintf(name, "../../bdi/%s", dev_name(client->sb->s_bdi->dev));
-	client->debugfs_bdi = debugfs_create_symlink("bdi", client->debugfs_dir,
-						     name);
+	dout("a\n");
+
+	snprintf(name, sizeof(name), "../../bdi/%s",
+		 dev_name(fsc->backing_dev_info.dev));
+	fsc->debugfs_bdi =
+		debugfs_create_symlink("bdi",
+				       fsc->client->debugfs_dir,
+				       name);
+	if (!fsc->debugfs_bdi)
+		goto out;
+
+	dout("b\n");
+	fsc->debugfs_mdsmap = debugfs_create_file("mdsmap",
+					0600,
+					fsc->client->debugfs_dir,
+					fsc,
+					&mdsmap_show_fops);
+	if (!fsc->debugfs_mdsmap)
+		goto out;
+
+	dout("ca\n");
+	fsc->debugfs_mdsc = debugfs_create_file("mdsc",
+						0600,
+						fsc->client->debugfs_dir,
+						fsc,
+						&mdsc_show_fops);
+	if (!fsc->debugfs_mdsc)
+		goto out;
+
+	dout("da\n");
+	fsc->debugfs_caps = debugfs_create_file("caps",
+						   0400,
+						   fsc->client->debugfs_dir,
+						   fsc,
+						   &caps_show_fops);
+	if (!fsc->debugfs_caps)
+		goto out;
+
+	dout("ea\n");
+	fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
+					0600,
+					fsc->client->debugfs_dir,
+					fsc,
+					&dentry_lru_show_fops);
+	if (!fsc->debugfs_dentry_lru)
+		goto out;
 
 	return 0;
 
 out:
-	ceph_debugfs_client_cleanup(client);
-	return ret;
+	ceph_fs_debugfs_cleanup(fsc);
+	return err;
 }
 
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
-{
-	debugfs_remove(client->debugfs_bdi);
-	debugfs_remove(client->debugfs_caps);
-	debugfs_remove(client->debugfs_dentry_lru);
-	debugfs_remove(client->debugfs_osdmap);
-	debugfs_remove(client->debugfs_mdsmap);
-	debugfs_remove(client->debugfs_monmap);
-	debugfs_remove(client->osdc.debugfs_file);
-	debugfs_remove(client->mdsc.debugfs_file);
-	debugfs_remove(client->monc.debugfs_file);
-	debugfs_remove(client->debugfs_congestion_kb);
-	debugfs_remove(client->debugfs_dir);
-}
 
 #else  /* CONFIG_DEBUG_FS */
 
-int __init ceph_debugfs_init(void)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
 	return 0;
 }
 
-void ceph_debugfs_cleanup(void)
-{
-}
-
-int ceph_debugfs_client_init(struct ceph_client *client)
-{
-	return 0;
-}
-
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
 {
 }
 
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index a1986eb..e0a2dc6 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/spinlock.h>
 #include <linux/fs_struct.h>
@@ -7,6 +7,7 @@
 #include <linux/sched.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * Directory operations: readdir, lookup, create, link, unlink,
@@ -94,10 +95,7 @@
  */
 static int __dcache_readdir(struct file *filp,
 			    void *dirent, filldir_t filldir)
-		__releases(inode->i_lock)
-		__acquires(inode->i_lock)
 {
-	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_file_info *fi = filp->private_data;
 	struct dentry *parent = filp->f_dentry;
 	struct inode *dir = parent->d_inode;
@@ -153,7 +151,6 @@
 
 	atomic_inc(&dentry->d_count);
 	spin_unlock(&dcache_lock);
-	spin_unlock(&inode->i_lock);
 
 	dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, filp->f_pos,
 	     dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
@@ -171,35 +168,30 @@
 		} else {
 			dput(last);
 		}
-		last = NULL;
 	}
-
-	spin_lock(&inode->i_lock);
-	spin_lock(&dcache_lock);
-
 	last = dentry;
 
 	if (err < 0)
-		goto out_unlock;
+		goto out;
 
-	p = p->prev;
 	filp->f_pos++;
 
 	/* make sure a dentry wasn't dropped while we didn't have dcache_lock */
-	if ((ceph_inode(dir)->i_ceph_flags & CEPH_I_COMPLETE))
-		goto more;
-	dout(" lost I_COMPLETE on %p; falling back to mds\n", dir);
-	err = -EAGAIN;
+	if (!ceph_i_test(dir, CEPH_I_COMPLETE)) {
+		dout(" lost I_COMPLETE on %p; falling back to mds\n", dir);
+		err = -EAGAIN;
+		goto out;
+	}
+
+	spin_lock(&dcache_lock);
+	p = p->prev;	/* advance to next dentry */
+	goto more;
 
 out_unlock:
 	spin_unlock(&dcache_lock);
-
-	if (last) {
-		spin_unlock(&inode->i_lock);
+out:
+	if (last)
 		dput(last);
-		spin_lock(&inode->i_lock);
-	}
-
 	return err;
 }
 
@@ -227,15 +219,15 @@
 	struct ceph_file_info *fi = filp->private_data;
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned frag = fpos_frag(filp->f_pos);
 	int off = fpos_off(filp->f_pos);
 	int err;
 	u32 ftype;
 	struct ceph_mds_reply_info_parsed *rinfo;
-	const int max_entries = client->mount_args->max_readdir;
-	const int max_bytes = client->mount_args->max_readdir_bytes;
+	const int max_entries = fsc->mount_options->max_readdir;
+	const int max_bytes = fsc->mount_options->max_readdir_bytes;
 
 	dout("readdir %p filp %p frag %u off %u\n", inode, filp, frag, off);
 	if (fi->at_end)
@@ -267,17 +259,17 @@
 	/* can we use the dcache? */
 	spin_lock(&inode->i_lock);
 	if ((filp->f_pos == 2 || fi->dentry) &&
-	    !ceph_test_opt(client, NOASYNCREADDIR) &&
+	    !ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
 	    ceph_snap(inode) != CEPH_SNAPDIR &&
 	    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
 	    __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
+		spin_unlock(&inode->i_lock);
 		err = __dcache_readdir(filp, dirent, filldir);
-		if (err != -EAGAIN) {
-			spin_unlock(&inode->i_lock);
+		if (err != -EAGAIN)
 			return err;
-		}
+	} else {
+		spin_unlock(&inode->i_lock);
 	}
-	spin_unlock(&inode->i_lock);
 	if (fi->dentry) {
 		err = note_last_dentry(fi, fi->dentry->d_name.name,
 				       fi->dentry->d_name.len);
@@ -487,14 +479,13 @@
 struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 				  struct dentry *dentry, int err)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
 	struct inode *parent = dentry->d_parent->d_inode;
 
 	/* .snap dir? */
 	if (err == -ENOENT &&
-	    ceph_vino(parent).ino != CEPH_INO_ROOT && /* no .snap in root dir */
 	    strcmp(dentry->d_name.name,
-		   client->mount_args->snapdir_name) == 0) {
+		   fsc->mount_options->snapdir_name) == 0) {
 		struct inode *inode = ceph_get_snapdir(parent);
 		dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
 		     dentry, dentry->d_name.len, dentry->d_name.name, inode);
@@ -539,8 +530,8 @@
 static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 				  struct nameidata *nd)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int op;
 	int err;
@@ -572,7 +563,7 @@
 		spin_lock(&dir->i_lock);
 		dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
 		if (strncmp(dentry->d_name.name,
-			    client->mount_args->snapdir_name,
+			    fsc->mount_options->snapdir_name,
 			    dentry->d_name.len) &&
 		    !is_root_ceph_dentry(dir, dentry) &&
 		    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
@@ -629,8 +620,8 @@
 static int ceph_mknod(struct inode *dir, struct dentry *dentry,
 		      int mode, dev_t rdev)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -685,8 +676,8 @@
 static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 			    const char *dest)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -716,8 +707,8 @@
 
 static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
 	int op;
@@ -758,8 +749,8 @@
 static int ceph_link(struct dentry *old_dentry, struct inode *dir,
 		     struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -813,8 +804,8 @@
  */
 static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = dentry->d_inode;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
@@ -854,8 +845,8 @@
 static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
 		       struct inode *new_dir, struct dentry *new_dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(old_dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(old_dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -1076,7 +1067,7 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int left;
 
-	if (!ceph_test_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
+	if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
 		return -EISDIR;
 
 	if (!cf->dir_info) {
@@ -1177,7 +1168,7 @@
 	dout("dentry_lru_add %p %p '%.*s'\n", di, dn,
 	     dn->d_name.len, dn->d_name.name);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_add_tail(&di->lru, &mdsc->dentry_lru);
 		mdsc->num_dentry++;
@@ -1193,7 +1184,7 @@
 	dout("dentry_lru_touch %p %p '%.*s' (offset %lld)\n", di, dn,
 	     dn->d_name.len, dn->d_name.name, di->offset);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_move_tail(&di->lru, &mdsc->dentry_lru);
 		spin_unlock(&mdsc->dentry_lru_lock);
@@ -1208,7 +1199,7 @@
 	dout("dentry_lru_del %p %p '%.*s'\n", di, dn,
 	     dn->d_name.len, dn->d_name.name);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_del_init(&di->lru);
 		mdsc->num_dentry--;
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index 4480cb1..2297d94 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -1,10 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/exportfs.h>
 #include <linux/slab.h>
 #include <asm/unaligned.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * NFS export support
@@ -42,32 +43,37 @@
 static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
 			  int connectable)
 {
+	int type;
 	struct ceph_nfs_fh *fh = (void *)rawfh;
 	struct ceph_nfs_confh *cfh = (void *)rawfh;
 	struct dentry *parent = dentry->d_parent;
 	struct inode *inode = dentry->d_inode;
-	int type;
+	int connected_handle_length = sizeof(*cfh)/4;
+	int handle_length = sizeof(*fh)/4;
 
 	/* don't re-export snaps */
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EINVAL;
 
-	if (*max_len >= sizeof(*cfh)) {
+	if (*max_len >= connected_handle_length) {
 		dout("encode_fh %p connectable\n", dentry);
 		cfh->ino = ceph_ino(dentry->d_inode);
 		cfh->parent_ino = ceph_ino(parent->d_inode);
 		cfh->parent_name_hash = parent->d_name.hash;
-		*max_len = sizeof(*cfh);
+		*max_len = connected_handle_length;
 		type = 2;
-	} else if (*max_len > sizeof(*fh)) {
-		if (connectable)
-			return -ENOSPC;
+	} else if (*max_len >= handle_length) {
+		if (connectable) {
+			*max_len = connected_handle_length;
+			return 255;
+		}
 		dout("encode_fh %p\n", dentry);
 		fh->ino = ceph_ino(dentry->d_inode);
-		*max_len = sizeof(*fh);
+		*max_len = handle_length;
 		type = 1;
 	} else {
-		return -ENOSPC;
+		*max_len = handle_length;
+		return 255;
 	}
 	return type;
 }
@@ -115,7 +121,7 @@
 static struct dentry *__cfh_to_dentry(struct super_block *sb,
 				      struct ceph_nfs_confh *cfh)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
 	struct inode *inode;
 	struct dentry *dentry;
 	struct ceph_vino vino;
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 8c044a4..e77c28c 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1,5 +1,6 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/file.h>
@@ -38,8 +39,8 @@
 static struct ceph_mds_request *
 prepare_open_request(struct super_block *sb, int flags, int create_mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int want_auth = USE_ANY_MDS;
 	int op = (flags & O_CREAT) ? CEPH_MDS_OP_CREATE : CEPH_MDS_OP_OPEN;
@@ -117,8 +118,8 @@
 int ceph_open(struct inode *inode, struct file *file)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	struct ceph_file_info *cf = file->private_data;
 	struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
@@ -216,8 +217,8 @@
 				struct nameidata *nd, int mode,
 				int locked_dir)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct file *file = nd->intent.open.file;
 	struct inode *parent_inode = get_dentry_parent_inode(file->f_dentry);
 	struct ceph_mds_request *req;
@@ -270,163 +271,6 @@
 }
 
 /*
- * build a vector of user pages
- */
-static struct page **get_direct_page_vector(const char __user *data,
-					    int num_pages,
-					    loff_t off, size_t len)
-{
-	struct page **pages;
-	int rc;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-
-	down_read(&current->mm->mmap_sem);
-	rc = get_user_pages(current, current->mm, (unsigned long)data,
-			    num_pages, 0, 0, pages, NULL);
-	up_read(&current->mm->mmap_sem);
-	if (rc < 0)
-		goto fail;
-	return pages;
-
-fail:
-	kfree(pages);
-	return ERR_PTR(rc);
-}
-
-static void put_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		put_page(pages[i]);
-	kfree(pages);
-}
-
-void ceph_release_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		__free_pages(pages[i], 0);
-	kfree(pages);
-}
-
-/*
- * allocate a vector new pages
- */
-static struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
-{
-	struct page **pages;
-	int i;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, flags);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-	for (i = 0; i < num_pages; i++) {
-		pages[i] = __page_cache_alloc(flags);
-		if (pages[i] == NULL) {
-			ceph_release_page_vector(pages, i);
-			return ERR_PTR(-ENOMEM);
-		}
-	}
-	return pages;
-}
-
-/*
- * copy user data into a page vector
- */
-static int copy_user_to_page_vector(struct page **pages,
-				    const char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, PAGE_CACHE_SIZE-po, left);
-		bad = copy_from_user(page_address(pages[i]) + po, data, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		po += l - bad;
-		if (po == PAGE_CACHE_SIZE) {
-			po = 0;
-			i++;
-		}
-	}
-	return len;
-}
-
-/*
- * copy user data from a page vector into a user pointer
- */
-static int copy_page_vector_to_user(struct page **pages, char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, left, PAGE_CACHE_SIZE-po);
-		bad = copy_to_user(data, page_address(pages[i]) + po, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		if (po) {
-			po += l - bad;
-			if (po == PAGE_CACHE_SIZE)
-				po = 0;
-		}
-		i++;
-	}
-	return len;
-}
-
-/*
- * Zero an extent within a page vector.  Offset is relative to the
- * start of the first page.
- */
-static void zero_page_vector_range(int off, int len, struct page **pages)
-{
-	int i = off >> PAGE_CACHE_SHIFT;
-
-	off &= ~PAGE_CACHE_MASK;
-
-	dout("zero_page_vector_page %u~%u\n", off, len);
-
-	/* leading partial page? */
-	if (off) {
-		int end = min((int)PAGE_CACHE_SIZE, off + len);
-		dout("zeroing %d %p head from %d\n", i, pages[i],
-		     (int)off);
-		zero_user_segment(pages[i], off, end);
-		len -= (end - off);
-		i++;
-	}
-	while (len >= PAGE_CACHE_SIZE) {
-		dout("zeroing %d %p len=%d\n", i, pages[i], len);
-		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
-		len -= PAGE_CACHE_SIZE;
-		i++;
-	}
-	/* trailing partial page? */
-	if (len) {
-		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
-		zero_user_segment(pages[i], 0, len);
-	}
-}
-
-
-/*
  * Read a range of bytes striped over one or more objects.  Iterate over
  * objects we stripe over.  (That's not atomic, but good enough for now.)
  *
@@ -438,7 +282,7 @@
 			struct page **pages, int num_pages,
 			int *checkeof)
 {
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	u64 pos, this_len;
 	int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
@@ -459,7 +303,7 @@
 
 more:
 	this_len = left;
-	ret = ceph_osdc_readpages(&client->osdc, ceph_vino(inode),
+	ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
 				  &ci->i_layout, pos, &this_len,
 				  ci->i_truncate_seq,
 				  ci->i_truncate_size,
@@ -477,8 +321,8 @@
 
 		if (read < pos - off) {
 			dout(" zero gap %llu to %llu\n", off + read, pos);
-			zero_page_vector_range(page_off + read,
-					       pos - off - read, pages);
+			ceph_zero_page_vector_range(page_off + read,
+						    pos - off - read, pages);
 		}
 		pos += ret;
 		read = pos - off;
@@ -495,8 +339,8 @@
 		/* was original extent fully inside i_size? */
 		if (pos + left <= inode->i_size) {
 			dout("zero tail\n");
-			zero_page_vector_range(page_off + read, len - read,
-					       pages);
+			ceph_zero_page_vector_range(page_off + read, len - read,
+						    pages);
 			read = len;
 			goto out;
 		}
@@ -531,7 +375,7 @@
 	     (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, off, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, off, len);
 
 		/*
 		 * flush any page cache pages in this range.  this
@@ -552,13 +396,13 @@
 	ret = striped_read(inode, off, len, pages, num_pages, checkeof);
 
 	if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
-		ret = copy_page_vector_to_user(pages, data, off, ret);
+		ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
 	if (ret >= 0)
 		*poff = off + ret;
 
 done:
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else
 		ceph_release_page_vector(pages, num_pages);
 	dout("sync_read result %d\n", ret);
@@ -594,7 +438,7 @@
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_osd_request *req;
 	struct page **pages;
 	int num_pages;
@@ -642,7 +486,7 @@
 	 */
 more:
 	len = left;
-	req = ceph_osdc_new_request(&client->osdc, &ci->i_layout,
+	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
 				    ceph_vino(inode), pos, &len,
 				    CEPH_OSD_OP_WRITE, flags,
 				    ci->i_snap_realm->cached_context,
@@ -655,7 +499,7 @@
 	num_pages = calc_pages_for(pos, len);
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, pos, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, pos, len);
 		if (IS_ERR(pages)) {
 			ret = PTR_ERR(pages);
 			goto out;
@@ -673,7 +517,7 @@
 			ret = PTR_ERR(pages);
 			goto out;
 		}
-		ret = copy_user_to_page_vector(pages, data, pos, len);
+		ret = ceph_copy_user_to_page_vector(pages, data, pos, len);
 		if (ret < 0) {
 			ceph_release_page_vector(pages, num_pages);
 			goto out;
@@ -689,7 +533,7 @@
 	req->r_num_pages = num_pages;
 	req->r_inode = inode;
 
-	ret = ceph_osdc_start_request(&client->osdc, req, false);
+	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!ret) {
 		if (req->r_safe_callback) {
 			/*
@@ -697,15 +541,15 @@
 			 * start_request so that a tid has been assigned.
 			 */
 			spin_lock(&ci->i_unsafe_lock);
-			list_add(&ci->i_unsafe_writes, &req->r_unsafe_item);
+			list_add(&req->r_unsafe_item, &ci->i_unsafe_writes);
 			spin_unlock(&ci->i_unsafe_lock);
 			ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
 		}
-		ret = ceph_osdc_wait_request(&client->osdc, req);
+		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
 	}
 
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else if (file->f_flags & O_SYNC)
 		ceph_release_page_vector(pages, num_pages);
 
@@ -814,7 +658,8 @@
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_sb_to_client(inode->i_sb)->client->osdc;
 	loff_t endoff = pos + iov->iov_len;
 	int want, got = 0;
 	int ret, err;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 62377ec..1d6a45b 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/module.h>
 #include <linux/fs.h>
@@ -13,7 +13,8 @@
 #include <linux/pagevec.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+#include <linux/ceph/decode.h>
 
 /*
  * Ceph inode operations
@@ -384,7 +385,7 @@
 	 */
 	if (ci->i_snap_realm) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+			ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 		struct ceph_snap_realm *realm = ci->i_snap_realm;
 
 		dout(" dropping residual ref to snap realm %p\n", realm);
@@ -685,7 +686,7 @@
 		}
 
 		/* it may be better to set st_size in getattr instead? */
-		if (ceph_test_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
+		if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
 			inode->i_size = ci->i_rbytes;
 		break;
 	default:
@@ -901,7 +902,7 @@
 	struct inode *in = NULL;
 	struct ceph_mds_reply_inode *ininfo;
 	struct ceph_vino vino;
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 	int i = 0;
 	int err = 0;
 
@@ -965,7 +966,7 @@
 	 */
 	if (rinfo->head->is_dentry && !req->r_aborted &&
 	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
-					       client->mount_args->snapdir_name,
+					       fsc->mount_options->snapdir_name,
 					       req->r_dentry->d_name.len))) {
 		/*
 		 * lookup link rename   : null -> possibly existing inode
@@ -1533,7 +1534,7 @@
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	const unsigned int ia_valid = attr->ia_valid;
 	struct ceph_mds_request *req;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(dentry->d_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(dentry->d_sb)->mdsc;
 	int issued;
 	int release = 0, dirtied = 0;
 	int mask = 0;
@@ -1728,8 +1729,8 @@
  */
 int ceph_do_getattr(struct inode *inode, int mask)
 {
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 76e307d..8888c9ba 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -1,8 +1,10 @@
 #include <linux/in.h>
 
-#include "ioctl.h"
 #include "super.h"
-#include "ceph_debug.h"
+#include "mds_client.h"
+#include <linux/ceph/ceph_debug.h>
+
+#include "ioctl.h"
 
 
 /*
@@ -37,7 +39,7 @@
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	struct ceph_ioctl_layout l;
 	int err, i;
@@ -90,6 +92,68 @@
 }
 
 /*
+ * Set a layout policy on a directory inode. All items in the tree
+ * rooted at this inode will inherit this layout on creation,
+ * (It doesn't apply retroactively )
+ * unless a subdirectory has its own layout policy.
+ */
+static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct ceph_mds_request *req;
+	struct ceph_ioctl_layout l;
+	int err, i;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
+
+	/* copy and validate */
+	if (copy_from_user(&l, arg, sizeof(l)))
+		return -EFAULT;
+
+	if ((l.object_size & ~PAGE_MASK) ||
+	    (l.stripe_unit & ~PAGE_MASK) ||
+	    !l.stripe_unit ||
+	    (l.object_size &&
+	        (unsigned)l.object_size % (unsigned)l.stripe_unit))
+		return -EINVAL;
+
+	/* make sure it's a valid data pool */
+	if (l.data_pool > 0) {
+		mutex_lock(&mdsc->mutex);
+		err = -EINVAL;
+		for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
+			if (mdsc->mdsmap->m_data_pg_pools[i] == l.data_pool) {
+				err = 0;
+				break;
+			}
+		mutex_unlock(&mdsc->mutex);
+		if (err)
+			return err;
+	}
+
+	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETDIRLAYOUT,
+				       USE_AUTH_MDS);
+
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+	req->r_inode = igrab(inode);
+
+	req->r_args.setlayout.layout.fl_stripe_unit =
+			cpu_to_le32(l.stripe_unit);
+	req->r_args.setlayout.layout.fl_stripe_count =
+			cpu_to_le32(l.stripe_count);
+	req->r_args.setlayout.layout.fl_object_size =
+			cpu_to_le32(l.object_size);
+	req->r_args.setlayout.layout.fl_pg_pool =
+			cpu_to_le32(l.data_pool);
+	req->r_args.setlayout.layout.fl_pg_preferred =
+			cpu_to_le32(l.preferred_osd);
+
+	err = ceph_mdsc_do_request(mdsc, inode, req);
+	ceph_mdsc_put_request(req);
+	return err;
+}
+
+/*
  * Return object name, size/offset information, and location (OSD
  * number, network address) for a given file offset.
  */
@@ -98,7 +162,8 @@
 	struct ceph_ioctl_dataloc dl;
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_sb_to_client(inode->i_sb)->client->osdc;
 	u64 len = 1, olen;
 	u64 tmp;
 	struct ceph_object_layout ol;
@@ -174,11 +239,15 @@
 	case CEPH_IOC_SET_LAYOUT:
 		return ceph_ioctl_set_layout(file, (void __user *)arg);
 
+	case CEPH_IOC_SET_LAYOUT_POLICY:
+		return ceph_ioctl_set_layout_policy(file, (void __user *)arg);
+
 	case CEPH_IOC_GET_DATALOC:
 		return ceph_ioctl_get_dataloc(file, (void __user *)arg);
 
 	case CEPH_IOC_LAZYIO:
 		return ceph_ioctl_lazyio(file);
 	}
+
 	return -ENOTTY;
 }
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index 88451a3..a6ce54e 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,7 +4,7 @@
 #include <linux/ioctl.h>
 #include <linux/types.h>
 
-#define CEPH_IOCTL_MAGIC 0x97
+#define CEPH_IOCTL_MAGIC 0x98
 
 /* just use u64 to align sanely on all archs */
 struct ceph_ioctl_layout {
@@ -17,6 +17,8 @@
 				   struct ceph_ioctl_layout)
 #define CEPH_IOC_SET_LAYOUT _IOW(CEPH_IOCTL_MAGIC, 2,		\
 				   struct ceph_ioctl_layout)
+#define CEPH_IOC_SET_LAYOUT_POLICY _IOW(CEPH_IOCTL_MAGIC, 5,	\
+				   struct ceph_ioctl_layout)
 
 /*
  * Extract identity, address of the OSD and object storing a given
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index ff4e753..40abde9 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -1,11 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/file.h>
 #include <linux/namei.h>
 
 #include "super.h"
 #include "mds_client.h"
-#include "pagelist.h"
+#include <linux/ceph/pagelist.h>
 
 /**
  * Implement fcntl and flock locking functions.
@@ -16,7 +16,7 @@
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(inode->i_sb)->mdsc;
+		ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -181,8 +181,9 @@
  * Encode the flock and fcntl locks for the given inode into the pagelist.
  * Format is: #fcntl locks, sequential fcntl locks, #flock locks,
  * sequential flock locks.
- * Must be called with BLK already held, and the lock numbers should have
- * been gathered under the same lock holding window.
+ * Must be called with lock_flocks() already held.
+ * If we encounter more of a specific lock type than expected,
+ * we return the value 1.
  */
 int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
 		      int num_fcntl_locks, int num_flock_locks)
@@ -190,6 +191,8 @@
 	struct file_lock *lock;
 	struct ceph_filelock cephlock;
 	int err = 0;
+	int seen_fcntl = 0;
+	int seen_flock = 0;
 
 	dout("encoding %d flock and %d fcntl locks", num_flock_locks,
 	     num_fcntl_locks);
@@ -198,6 +201,11 @@
 		goto fail;
 	for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
 		if (lock->fl_flags & FL_POSIX) {
+			++seen_fcntl;
+			if (seen_fcntl > num_fcntl_locks) {
+				err = -ENOSPC;
+				goto fail;
+			}
 			err = lock_to_ceph_filelock(lock, &cephlock);
 			if (err)
 				goto fail;
@@ -213,6 +221,11 @@
 		goto fail;
 	for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
 		if (lock->fl_flags & FL_FLOCK) {
+			++seen_flock;
+			if (seen_flock > num_flock_locks) {
+				err = -ENOSPC;
+				goto fail;
+			}
 			err = lock_to_ceph_filelock(lock, &cephlock);
 			if (err)
 				goto fail;
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index fad95f8..3142b15 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1,17 +1,21 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/fs.h>
 #include <linux/wait.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
 #include <linux/smp_lock.h>
 
-#include "mds_client.h"
-#include "mon_client.h"
 #include "super.h"
-#include "messenger.h"
-#include "decode.h"
-#include "auth.h"
-#include "pagelist.h"
+#include "mds_client.h"
+
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/pagelist.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 /*
  * A cluster of MDS (metadata server) daemons is responsible for
@@ -286,8 +290,9 @@
 	     atomic_read(&s->s_ref), atomic_read(&s->s_ref)-1);
 	if (atomic_dec_and_test(&s->s_ref)) {
 		if (s->s_authorizer)
-			s->s_mdsc->client->monc.auth->ops->destroy_authorizer(
-				s->s_mdsc->client->monc.auth, s->s_authorizer);
+		     s->s_mdsc->fsc->client->monc.auth->ops->destroy_authorizer(
+			     s->s_mdsc->fsc->client->monc.auth,
+			     s->s_authorizer);
 		kfree(s);
 	}
 }
@@ -344,7 +349,7 @@
 	s->s_seq = 0;
 	mutex_init(&s->s_mutex);
 
-	ceph_con_init(mdsc->client->msgr, &s->s_con);
+	ceph_con_init(mdsc->fsc->client->msgr, &s->s_con);
 	s->s_con.private = s;
 	s->s_con.ops = &mds_con_ops;
 	s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS;
@@ -599,7 +604,7 @@
 	} else if (req->r_dentry) {
 		struct inode *dir = req->r_dentry->d_parent->d_inode;
 
-		if (dir->i_sb != mdsc->client->sb) {
+		if (dir->i_sb != mdsc->fsc->sb) {
 			/* not this fs! */
 			inode = req->r_dentry->d_inode;
 		} else if (ceph_snap(dir) != CEPH_NOSNAP) {
@@ -884,7 +889,7 @@
 	__ceph_remove_cap(cap);
 	if (!__ceph_is_any_real_caps(ci)) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&mdsc->cap_dirty_lock);
 		if (!list_empty(&ci->i_dirty_item)) {
@@ -1146,7 +1151,7 @@
 	struct ceph_msg *msg, *partial = NULL;
 	struct ceph_mds_cap_release *head;
 	int err = -ENOMEM;
-	int extra = mdsc->client->mount_args->cap_release_safety;
+	int extra = mdsc->fsc->mount_options->cap_release_safety;
 	int num;
 
 	dout("add_cap_releases %p mds%d extra %d\n", session, session->s_mds,
@@ -2085,7 +2090,7 @@
 
 	/* insert trace into our cache */
 	mutex_lock(&req->r_fill_mutex);
-	err = ceph_fill_trace(mdsc->client->sb, req, req->r_session);
+	err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
 	if (err == 0) {
 		if (result == 0 && rinfo->dir_nr)
 			ceph_readdir_prepopulate(req, req->r_session);
@@ -2361,19 +2366,35 @@
 
 	if (recon_state->flock) {
 		int num_fcntl_locks, num_flock_locks;
+		struct ceph_pagelist_cursor trunc_point;
 
-		lock_kernel();
-		ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
-		rec.v2.flock_len = (2*sizeof(u32) +
-				    (num_fcntl_locks+num_flock_locks) *
-				    sizeof(struct ceph_filelock));
+		ceph_pagelist_set_cursor(pagelist, &trunc_point);
+		do {
+			lock_flocks();
+			ceph_count_locks(inode, &num_fcntl_locks,
+					 &num_flock_locks);
+			rec.v2.flock_len = (2*sizeof(u32) +
+					    (num_fcntl_locks+num_flock_locks) *
+					    sizeof(struct ceph_filelock));
+			unlock_flocks();
 
-		err = ceph_pagelist_append(pagelist, &rec, reclen);
-		if (!err)
-			err = ceph_encode_locks(inode, pagelist,
-						num_fcntl_locks,
-						num_flock_locks);
-		unlock_kernel();
+			/* pre-alloc pagelist */
+			ceph_pagelist_truncate(pagelist, &trunc_point);
+			err = ceph_pagelist_append(pagelist, &rec, reclen);
+			if (!err)
+				err = ceph_pagelist_reserve(pagelist,
+							    rec.v2.flock_len);
+
+			/* encode locks */
+			if (!err) {
+				lock_flocks();
+				err = ceph_encode_locks(inode,
+							pagelist,
+							num_fcntl_locks,
+							num_flock_locks);
+				unlock_flocks();
+			}
+		} while (err == -ENOSPC);
 	} else {
 		err = ceph_pagelist_append(pagelist, &rec, reclen);
 	}
@@ -2613,7 +2634,7 @@
 			 struct ceph_mds_session *session,
 			 struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_inode_info *ci;
 	struct dentry *parent, *dentry;
@@ -2891,10 +2912,16 @@
 	schedule_delayed(mdsc);
 }
 
+int ceph_mdsc_init(struct ceph_fs_client *fsc)
 
-int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 {
-	mdsc->client = client;
+	struct ceph_mds_client *mdsc;
+
+	mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
+	if (!mdsc)
+		return -ENOMEM;
+	mdsc->fsc = fsc;
+	fsc->mdsc = mdsc;
 	mutex_init(&mdsc->mutex);
 	mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
 	if (mdsc->mdsmap == NULL)
@@ -2927,7 +2954,7 @@
 	INIT_LIST_HEAD(&mdsc->dentry_lru);
 
 	ceph_caps_init(mdsc);
-	ceph_adjust_min_caps(mdsc, client->min_caps);
+	ceph_adjust_min_caps(mdsc, fsc->min_caps);
 
 	return 0;
 }
@@ -2939,7 +2966,7 @@
 static void wait_requests(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_request *req;
-	struct ceph_client *client = mdsc->client;
+	struct ceph_fs_client *fsc = mdsc->fsc;
 
 	mutex_lock(&mdsc->mutex);
 	if (__get_oldest_req(mdsc)) {
@@ -2947,7 +2974,7 @@
 
 		dout("wait_requests waiting for requests\n");
 		wait_for_completion_timeout(&mdsc->safe_umount_waiters,
-				    client->mount_args->mount_timeout * HZ);
+				    fsc->client->options->mount_timeout * HZ);
 
 		/* tear down remaining requests */
 		mutex_lock(&mdsc->mutex);
@@ -3030,7 +3057,7 @@
 {
 	u64 want_tid, want_flush;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return;
 
 	dout("sync\n");
@@ -3053,7 +3080,7 @@
 {
 	int i, n = 0;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return true;
 
 	mutex_lock(&mdsc->mutex);
@@ -3071,8 +3098,8 @@
 {
 	struct ceph_mds_session *session;
 	int i;
-	struct ceph_client *client = mdsc->client;
-	unsigned long timeout = client->mount_args->mount_timeout * HZ;
+	struct ceph_fs_client *fsc = mdsc->fsc;
+	unsigned long timeout = fsc->client->options->mount_timeout * HZ;
 
 	dout("close_sessions\n");
 
@@ -3119,7 +3146,7 @@
 	dout("stopped\n");
 }
 
-void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
+static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
 {
 	dout("stop\n");
 	cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
@@ -3129,6 +3156,15 @@
 	ceph_caps_finalize(mdsc);
 }
 
+void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
+{
+	struct ceph_mds_client *mdsc = fsc->mdsc;
+
+	ceph_mdsc_stop(mdsc);
+	fsc->mdsc = NULL;
+	kfree(mdsc);
+}
+
 
 /*
  * handle mds map update.
@@ -3145,14 +3181,14 @@
 
 	ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
 	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	if (ceph_check_fsid(mdsc->client, &fsid) < 0)
+	if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
 		return;
 	epoch = ceph_decode_32(&p);
 	maplen = ceph_decode_32(&p);
 	dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
 
 	/* do we need it? */
-	ceph_monc_got_mdsmap(&mdsc->client->monc, epoch);
+	ceph_monc_got_mdsmap(&mdsc->fsc->client->monc, epoch);
 	mutex_lock(&mdsc->mutex);
 	if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
 		dout("handle_map epoch %u <= our %u\n",
@@ -3176,7 +3212,7 @@
 	} else {
 		mdsc->mdsmap = newmap;  /* first mds map */
 	}
-	mdsc->client->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
+	mdsc->fsc->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
 
 	__wake_requests(mdsc, &mdsc->waiting_for_map);
 
@@ -3277,7 +3313,7 @@
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 	int ret = 0;
 
 	if (force_new && s->s_authorizer) {
@@ -3311,7 +3347,7 @@
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	return ac->ops->verify_authorizer_reply(ac, s->s_authorizer, len);
 }
@@ -3320,12 +3356,12 @@
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	if (ac->ops->invalidate_authorizer)
 		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
 
-	return ceph_monc_validate_auth(&mdsc->client->monc);
+	return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
 }
 
 static const struct ceph_connection_operations mds_con_ops = {
@@ -3338,7 +3374,4 @@
 	.peer_reset = peer_reset,
 };
 
-
-
-
 /* eof */
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index c98267c..d66d63c 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -8,9 +8,9 @@
 #include <linux/rbtree.h>
 #include <linux/spinlock.h>
 
-#include "types.h"
-#include "messenger.h"
-#include "mdsmap.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/mdsmap.h>
 
 /*
  * Some lock dependencies:
@@ -26,7 +26,7 @@
  *
  */
 
-struct ceph_client;
+struct ceph_fs_client;
 struct ceph_cap;
 
 /*
@@ -230,7 +230,7 @@
  * mds client state
  */
 struct ceph_mds_client {
-	struct ceph_client      *client;
+	struct ceph_fs_client  *fsc;
 	struct mutex            mutex;         /* all nested structures */
 
 	struct ceph_mdsmap      *mdsmap;
@@ -289,11 +289,6 @@
 	int		caps_avail_count;    /* unused, unreserved */
 	int		caps_min_count;      /* keep at least this many
 						(unreserved) */
-
-#ifdef CONFIG_DEBUG_FS
-	struct dentry 	  *debugfs_file;
-#endif
-
 	spinlock_t	  dentry_lru_lock;
 	struct list_head  dentry_lru;
 	int		  num_dentry;
@@ -316,10 +311,9 @@
 extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
 			     struct ceph_msg *msg, int mds);
 
-extern int ceph_mdsc_init(struct ceph_mds_client *mdsc,
-			   struct ceph_client *client);
+extern int ceph_mdsc_init(struct ceph_fs_client *fsc);
 extern void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc);
-extern void ceph_mdsc_stop(struct ceph_mds_client *mdsc);
+extern void ceph_mdsc_destroy(struct ceph_fs_client *fsc);
 
 extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);
 
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 040be6d..73b7d44 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/bug.h>
 #include <linux/err.h>
@@ -6,9 +6,9 @@
 #include <linux/slab.h>
 #include <linux/types.h>
 
-#include "mdsmap.h"
-#include "messenger.h"
-#include "decode.h"
+#include <linux/ceph/mdsmap.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
 
 #include "super.h"
 
@@ -117,7 +117,8 @@
 		}
 
 		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s\n",
-		     i+1, n, global_id, mds, inc, pr_addr(&addr.in_addr),
+		     i+1, n, global_id, mds, inc,
+		     ceph_pr_addr(&addr.in_addr),
 		     ceph_mds_state_name(state));
 		if (mds >= 0 && mds < m->m_max_mds && state > 0) {
 			m->m_info[mds].global_id = global_id;
diff --git a/fs/ceph/pagelist.c b/fs/ceph/pagelist.c
deleted file mode 100644
index 46a368b..0000000
--- a/fs/ceph/pagelist.c
+++ /dev/null
@@ -1,63 +0,0 @@
-
-#include <linux/gfp.h>
-#include <linux/pagemap.h>
-#include <linux/highmem.h>
-
-#include "pagelist.h"
-
-static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
-{
-	struct page *page = list_entry(pl->head.prev, struct page,
-				       lru);
-	kunmap(page);
-}
-
-int ceph_pagelist_release(struct ceph_pagelist *pl)
-{
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-
-	while (!list_empty(&pl->head)) {
-		struct page *page = list_first_entry(&pl->head, struct page,
-						     lru);
-		list_del(&page->lru);
-		__free_page(page);
-	}
-	return 0;
-}
-
-static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
-{
-	struct page *page = __page_cache_alloc(GFP_NOFS);
-	if (!page)
-		return -ENOMEM;
-	pl->room += PAGE_SIZE;
-	list_add_tail(&page->lru, &pl->head);
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-	pl->mapped_tail = kmap(page);
-	return 0;
-}
-
-int ceph_pagelist_append(struct ceph_pagelist *pl, void *buf, size_t len)
-{
-	while (pl->room < len) {
-		size_t bit = pl->room;
-		int ret;
-
-		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
-		       buf, bit);
-		pl->length += bit;
-		pl->room -= bit;
-		buf += bit;
-		len -= bit;
-		ret = ceph_pagelist_addpage(pl);
-		if (ret)
-			return ret;
-	}
-
-	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
-	pl->length += len;
-	pl->room -= len;
-	return 0;
-}
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 190b6c4..39c243a 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -1,10 +1,12 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/sort.h>
 #include <linux/slab.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 /*
  * Snapshots in ceph are driven in large part by cooperation from the
@@ -526,7 +528,7 @@
 			    struct ceph_cap_snap *capsnap)
 {
 	struct inode *inode = &ci->vfs_inode;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 
 	BUG_ON(capsnap->writing);
 	capsnap->size = inode->i_size;
@@ -747,7 +749,7 @@
 		      struct ceph_mds_session *session,
 		      struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	int mds = session->s_mds;
 	u64 split;
 	int op;
diff --git a/fs/ceph/strings.c b/fs/ceph/strings.c
new file mode 100644
index 0000000..cd5097d
--- /dev/null
+++ b/fs/ceph/strings.c
@@ -0,0 +1,117 @@
+/*
+ * Ceph fs string constants
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+
+const char *ceph_mds_state_name(int s)
+{
+	switch (s) {
+		/* down and out */
+	case CEPH_MDS_STATE_DNE:        return "down:dne";
+	case CEPH_MDS_STATE_STOPPED:    return "down:stopped";
+		/* up and out */
+	case CEPH_MDS_STATE_BOOT:       return "up:boot";
+	case CEPH_MDS_STATE_STANDBY:    return "up:standby";
+	case CEPH_MDS_STATE_STANDBY_REPLAY:    return "up:standby-replay";
+	case CEPH_MDS_STATE_CREATING:   return "up:creating";
+	case CEPH_MDS_STATE_STARTING:   return "up:starting";
+		/* up and in */
+	case CEPH_MDS_STATE_REPLAY:     return "up:replay";
+	case CEPH_MDS_STATE_RESOLVE:    return "up:resolve";
+	case CEPH_MDS_STATE_RECONNECT:  return "up:reconnect";
+	case CEPH_MDS_STATE_REJOIN:     return "up:rejoin";
+	case CEPH_MDS_STATE_CLIENTREPLAY: return "up:clientreplay";
+	case CEPH_MDS_STATE_ACTIVE:     return "up:active";
+	case CEPH_MDS_STATE_STOPPING:   return "up:stopping";
+	}
+	return "???";
+}
+
+const char *ceph_session_op_name(int op)
+{
+	switch (op) {
+	case CEPH_SESSION_REQUEST_OPEN: return "request_open";
+	case CEPH_SESSION_OPEN: return "open";
+	case CEPH_SESSION_REQUEST_CLOSE: return "request_close";
+	case CEPH_SESSION_CLOSE: return "close";
+	case CEPH_SESSION_REQUEST_RENEWCAPS: return "request_renewcaps";
+	case CEPH_SESSION_RENEWCAPS: return "renewcaps";
+	case CEPH_SESSION_STALE: return "stale";
+	case CEPH_SESSION_RECALL_STATE: return "recall_state";
+	}
+	return "???";
+}
+
+const char *ceph_mds_op_name(int op)
+{
+	switch (op) {
+	case CEPH_MDS_OP_LOOKUP:  return "lookup";
+	case CEPH_MDS_OP_LOOKUPHASH:  return "lookuphash";
+	case CEPH_MDS_OP_LOOKUPPARENT:  return "lookupparent";
+	case CEPH_MDS_OP_GETATTR:  return "getattr";
+	case CEPH_MDS_OP_SETXATTR: return "setxattr";
+	case CEPH_MDS_OP_SETATTR: return "setattr";
+	case CEPH_MDS_OP_RMXATTR: return "rmxattr";
+	case CEPH_MDS_OP_READDIR: return "readdir";
+	case CEPH_MDS_OP_MKNOD: return "mknod";
+	case CEPH_MDS_OP_LINK: return "link";
+	case CEPH_MDS_OP_UNLINK: return "unlink";
+	case CEPH_MDS_OP_RENAME: return "rename";
+	case CEPH_MDS_OP_MKDIR: return "mkdir";
+	case CEPH_MDS_OP_RMDIR: return "rmdir";
+	case CEPH_MDS_OP_SYMLINK: return "symlink";
+	case CEPH_MDS_OP_CREATE: return "create";
+	case CEPH_MDS_OP_OPEN: return "open";
+	case CEPH_MDS_OP_LOOKUPSNAP: return "lookupsnap";
+	case CEPH_MDS_OP_LSSNAP: return "lssnap";
+	case CEPH_MDS_OP_MKSNAP: return "mksnap";
+	case CEPH_MDS_OP_RMSNAP: return "rmsnap";
+	case CEPH_MDS_OP_SETFILELOCK: return "setfilelock";
+	case CEPH_MDS_OP_GETFILELOCK: return "getfilelock";
+	}
+	return "???";
+}
+
+const char *ceph_cap_op_name(int op)
+{
+	switch (op) {
+	case CEPH_CAP_OP_GRANT: return "grant";
+	case CEPH_CAP_OP_REVOKE: return "revoke";
+	case CEPH_CAP_OP_TRUNC: return "trunc";
+	case CEPH_CAP_OP_EXPORT: return "export";
+	case CEPH_CAP_OP_IMPORT: return "import";
+	case CEPH_CAP_OP_UPDATE: return "update";
+	case CEPH_CAP_OP_DROP: return "drop";
+	case CEPH_CAP_OP_FLUSH: return "flush";
+	case CEPH_CAP_OP_FLUSH_ACK: return "flush_ack";
+	case CEPH_CAP_OP_FLUSHSNAP: return "flushsnap";
+	case CEPH_CAP_OP_FLUSHSNAP_ACK: return "flushsnap_ack";
+	case CEPH_CAP_OP_RELEASE: return "release";
+	case CEPH_CAP_OP_RENEW: return "renew";
+	}
+	return "???";
+}
+
+const char *ceph_lease_op_name(int o)
+{
+	switch (o) {
+	case CEPH_MDS_LEASE_REVOKE: return "revoke";
+	case CEPH_MDS_LEASE_RELEASE: return "release";
+	case CEPH_MDS_LEASE_RENEW: return "renew";
+	case CEPH_MDS_LEASE_REVOKE_ACK: return "revoke_ack";
+	}
+	return "???";
+}
+
+const char *ceph_snap_op_name(int o)
+{
+	switch (o) {
+	case CEPH_SNAP_OP_UPDATE: return "update";
+	case CEPH_SNAP_OP_CREATE: return "create";
+	case CEPH_SNAP_OP_DESTROY: return "destroy";
+	case CEPH_SNAP_OP_SPLIT: return "split";
+	}
+	return "???";
+}
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 9922628..d6e0e04 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -1,5 +1,5 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/backing-dev.h>
 #include <linux/ctype.h>
@@ -15,10 +15,13 @@
 #include <linux/statfs.h>
 #include <linux/string.h>
 
-#include "decode.h"
 #include "super.h"
-#include "mon_client.h"
-#include "auth.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 /*
  * Ceph superblock operations
@@ -26,36 +29,22 @@
  * Handle the basics of mounting, unmounting.
  */
 
-
-/*
- * find filename portion of a path (/foo/bar/baz -> baz)
- */
-const char *ceph_file_part(const char *s, int len)
-{
-	const char *e = s + len;
-
-	while (e != s && *(e-1) != '/')
-		e--;
-	return e;
-}
-
-
 /*
  * super ops
  */
 static void ceph_put_super(struct super_block *s)
 {
-	struct ceph_client *client = ceph_sb_to_client(s);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(s);
 
 	dout("put_super\n");
-	ceph_mdsc_close_sessions(&client->mdsc);
+	ceph_mdsc_close_sessions(fsc->mdsc);
 
 	/*
 	 * ensure we release the bdi before put_anon_super releases
 	 * the device name.
 	 */
-	if (s->s_bdi == &client->backing_dev_info) {
-		bdi_unregister(&client->backing_dev_info);
+	if (s->s_bdi == &fsc->backing_dev_info) {
+		bdi_unregister(&fsc->backing_dev_info);
 		s->s_bdi = NULL;
 	}
 
@@ -64,14 +53,14 @@
 
 static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
-	struct ceph_client *client = ceph_inode_to_client(dentry->d_inode);
-	struct ceph_monmap *monmap = client->monc.monmap;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(dentry->d_inode);
+	struct ceph_monmap *monmap = fsc->client->monc.monmap;
 	struct ceph_statfs st;
 	u64 fsid;
 	int err;
 
 	dout("statfs\n");
-	err = ceph_monc_do_statfs(&client->monc, &st);
+	err = ceph_monc_do_statfs(&fsc->client->monc, &st);
 	if (err < 0)
 		return err;
 
@@ -104,49 +93,237 @@
 
 static int ceph_sync_fs(struct super_block *sb, int wait)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 
 	if (!wait) {
 		dout("sync_fs (non-blocking)\n");
-		ceph_flush_dirty_caps(&client->mdsc);
+		ceph_flush_dirty_caps(fsc->mdsc);
 		dout("sync_fs (non-blocking) done\n");
 		return 0;
 	}
 
 	dout("sync_fs (blocking)\n");
-	ceph_osdc_sync(&ceph_sb_to_client(sb)->osdc);
-	ceph_mdsc_sync(&ceph_sb_to_client(sb)->mdsc);
+	ceph_osdc_sync(&fsc->client->osdc);
+	ceph_mdsc_sync(fsc->mdsc);
 	dout("sync_fs (blocking) done\n");
 	return 0;
 }
 
-static int default_congestion_kb(void)
+/*
+ * mount options
+ */
+enum {
+	Opt_wsize,
+	Opt_rsize,
+	Opt_caps_wanted_delay_min,
+	Opt_caps_wanted_delay_max,
+	Opt_cap_release_safety,
+	Opt_readdir_max_entries,
+	Opt_readdir_max_bytes,
+	Opt_congestion_kb,
+	Opt_last_int,
+	/* int args above */
+	Opt_snapdirname,
+	Opt_last_string,
+	/* string args above */
+	Opt_dirstat,
+	Opt_nodirstat,
+	Opt_rbytes,
+	Opt_norbytes,
+	Opt_noasyncreaddir,
+};
+
+static match_table_t fsopt_tokens = {
+	{Opt_wsize, "wsize=%d"},
+	{Opt_rsize, "rsize=%d"},
+	{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
+	{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
+	{Opt_cap_release_safety, "cap_release_safety=%d"},
+	{Opt_readdir_max_entries, "readdir_max_entries=%d"},
+	{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
+	{Opt_congestion_kb, "write_congestion_kb=%d"},
+	/* int args above */
+	{Opt_snapdirname, "snapdirname=%s"},
+	/* string args above */
+	{Opt_dirstat, "dirstat"},
+	{Opt_nodirstat, "nodirstat"},
+	{Opt_rbytes, "rbytes"},
+	{Opt_norbytes, "norbytes"},
+	{Opt_noasyncreaddir, "noasyncreaddir"},
+	{-1, NULL}
+};
+
+static int parse_fsopt_token(char *c, void *private)
 {
-	int congestion_kb;
+	struct ceph_mount_options *fsopt = private;
+	substring_t argstr[MAX_OPT_ARGS];
+	int token, intval, ret;
 
-	/*
-	 * Copied from NFS
-	 *
-	 * congestion size, scale with available memory.
-	 *
-	 *  64MB:    8192k
-	 * 128MB:   11585k
-	 * 256MB:   16384k
-	 * 512MB:   23170k
-	 *   1GB:   32768k
-	 *   2GB:   46340k
-	 *   4GB:   65536k
-	 *   8GB:   92681k
-	 *  16GB:  131072k
-	 *
-	 * This allows larger machines to have larger/more transfers.
-	 * Limit the default to 256M
-	 */
-	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
-	if (congestion_kb > 256*1024)
-		congestion_kb = 256*1024;
+	token = match_token((char *)c, fsopt_tokens, argstr);
+	if (token < 0)
+		return -EINVAL;
 
-	return congestion_kb;
+	if (token < Opt_last_int) {
+		ret = match_int(&argstr[0], &intval);
+		if (ret < 0) {
+			pr_err("bad mount option arg (not int) "
+			       "at '%s'\n", c);
+			return ret;
+		}
+		dout("got int token %d val %d\n", token, intval);
+	} else if (token > Opt_last_int && token < Opt_last_string) {
+		dout("got string token %d val %s\n", token,
+		     argstr[0].from);
+	} else {
+		dout("got token %d\n", token);
+	}
+
+	switch (token) {
+	case Opt_snapdirname:
+		kfree(fsopt->snapdir_name);
+		fsopt->snapdir_name = kstrndup(argstr[0].from,
+					       argstr[0].to-argstr[0].from,
+					       GFP_KERNEL);
+		if (!fsopt->snapdir_name)
+			return -ENOMEM;
+		break;
+
+		/* misc */
+	case Opt_wsize:
+		fsopt->wsize = intval;
+		break;
+	case Opt_rsize:
+		fsopt->rsize = intval;
+		break;
+	case Opt_caps_wanted_delay_min:
+		fsopt->caps_wanted_delay_min = intval;
+		break;
+	case Opt_caps_wanted_delay_max:
+		fsopt->caps_wanted_delay_max = intval;
+		break;
+	case Opt_readdir_max_entries:
+		fsopt->max_readdir = intval;
+		break;
+	case Opt_readdir_max_bytes:
+		fsopt->max_readdir_bytes = intval;
+		break;
+	case Opt_congestion_kb:
+		fsopt->congestion_kb = intval;
+		break;
+	case Opt_dirstat:
+		fsopt->flags |= CEPH_MOUNT_OPT_DIRSTAT;
+		break;
+	case Opt_nodirstat:
+		fsopt->flags &= ~CEPH_MOUNT_OPT_DIRSTAT;
+		break;
+	case Opt_rbytes:
+		fsopt->flags |= CEPH_MOUNT_OPT_RBYTES;
+		break;
+	case Opt_norbytes:
+		fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
+		break;
+	case Opt_noasyncreaddir:
+		fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
+		break;
+	default:
+		BUG_ON(token);
+	}
+	return 0;
+}
+
+static void destroy_mount_options(struct ceph_mount_options *args)
+{
+	dout("destroy_mount_options %p\n", args);
+	kfree(args->snapdir_name);
+	kfree(args);
+}
+
+static int strcmp_null(const char *s1, const char *s2)
+{
+	if (!s1 && !s2)
+		return 0;
+	if (s1 && !s2)
+		return -1;
+	if (!s1 && s2)
+		return 1;
+	return strcmp(s1, s2);
+}
+
+static int compare_mount_options(struct ceph_mount_options *new_fsopt,
+				 struct ceph_options *new_opt,
+				 struct ceph_fs_client *fsc)
+{
+	struct ceph_mount_options *fsopt1 = new_fsopt;
+	struct ceph_mount_options *fsopt2 = fsc->mount_options;
+	int ofs = offsetof(struct ceph_mount_options, snapdir_name);
+	int ret;
+
+	ret = memcmp(fsopt1, fsopt2, ofs);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(fsopt1->snapdir_name, fsopt2->snapdir_name);
+	if (ret)
+		return ret;
+
+	return ceph_compare_options(new_opt, fsc->client);
+}
+
+static int parse_mount_options(struct ceph_mount_options **pfsopt,
+			       struct ceph_options **popt,
+			       int flags, char *options,
+			       const char *dev_name,
+			       const char **path)
+{
+	struct ceph_mount_options *fsopt;
+	const char *dev_name_end;
+	int err = -ENOMEM;
+
+	fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
+	if (!fsopt)
+		return -ENOMEM;
+
+	dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
+
+        fsopt->sb_flags = flags;
+        fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
+
+        fsopt->rsize = CEPH_MOUNT_RSIZE_DEFAULT;
+        fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
+        fsopt->cap_release_safety = CEPH_CAP_RELEASE_SAFETY_DEFAULT;
+        fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
+        fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
+        fsopt->congestion_kb = default_congestion_kb();
+	
+        /* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
+        err = -EINVAL;
+        if (!dev_name)
+                goto out;
+        *path = strstr(dev_name, ":/");
+        if (*path == NULL) {
+                pr_err("device name is missing path (no :/ in %s)\n",
+                       dev_name);
+                goto out;
+        }
+	dev_name_end = *path;
+	dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
+
+	/* path on server */
+	*path += 2;
+	dout("server path '%s'\n", *path);
+
+	err = ceph_parse_options(popt, options, dev_name, dev_name_end,
+				 parse_fsopt_token, (void *)fsopt);
+	if (err)
+		goto out;
+
+	/* success */
+	*pfsopt = fsopt;
+	return 0;
+
+out:
+	destroy_mount_options(fsopt);
+	return err;
 }
 
 /**
@@ -156,60 +333,176 @@
  */
 static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
 {
-	struct ceph_client *client = ceph_sb_to_client(mnt->mnt_sb);
-	struct ceph_mount_args *args = client->mount_args;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(mnt->mnt_sb);
+	struct ceph_mount_options *fsopt = fsc->mount_options;
+	struct ceph_options *opt = fsc->client->options;
 
-	if (args->flags & CEPH_OPT_FSID)
-		seq_printf(m, ",fsid=%pU", &args->fsid);
-	if (args->flags & CEPH_OPT_NOSHARE)
+	if (opt->flags & CEPH_OPT_FSID)
+		seq_printf(m, ",fsid=%pU", &opt->fsid);
+	if (opt->flags & CEPH_OPT_NOSHARE)
 		seq_puts(m, ",noshare");
-	if (args->flags & CEPH_OPT_DIRSTAT)
-		seq_puts(m, ",dirstat");
-	if ((args->flags & CEPH_OPT_RBYTES) == 0)
-		seq_puts(m, ",norbytes");
-	if (args->flags & CEPH_OPT_NOCRC)
+	if (opt->flags & CEPH_OPT_NOCRC)
 		seq_puts(m, ",nocrc");
-	if (args->flags & CEPH_OPT_NOASYNCREADDIR)
+
+	if (opt->name)
+		seq_printf(m, ",name=%s", opt->name);
+	if (opt->secret)
+		seq_puts(m, ",secret=<hidden>");
+
+	if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
+		seq_printf(m, ",mount_timeout=%d", opt->mount_timeout);
+	if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
+		seq_printf(m, ",osd_idle_ttl=%d", opt->osd_idle_ttl);
+	if (opt->osd_timeout != CEPH_OSD_TIMEOUT_DEFAULT)
+		seq_printf(m, ",osdtimeout=%d", opt->osd_timeout);
+	if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
+		seq_printf(m, ",osdkeepalivetimeout=%d",
+			   opt->osd_keepalive_timeout);
+
+	if (fsopt->flags & CEPH_MOUNT_OPT_DIRSTAT)
+		seq_puts(m, ",dirstat");
+	if ((fsopt->flags & CEPH_MOUNT_OPT_RBYTES) == 0)
+		seq_puts(m, ",norbytes");
+	if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
 		seq_puts(m, ",noasyncreaddir");
 
-	if (args->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
-		seq_printf(m, ",mount_timeout=%d", args->mount_timeout);
-	if (args->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
-		seq_printf(m, ",osd_idle_ttl=%d", args->osd_idle_ttl);
-	if (args->osd_timeout != CEPH_OSD_TIMEOUT_DEFAULT)
-		seq_printf(m, ",osdtimeout=%d", args->osd_timeout);
-	if (args->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
-		seq_printf(m, ",osdkeepalivetimeout=%d",
-			 args->osd_keepalive_timeout);
-	if (args->wsize)
-		seq_printf(m, ",wsize=%d", args->wsize);
-	if (args->rsize != CEPH_MOUNT_RSIZE_DEFAULT)
-		seq_printf(m, ",rsize=%d", args->rsize);
-	if (args->congestion_kb != default_congestion_kb())
-		seq_printf(m, ",write_congestion_kb=%d", args->congestion_kb);
-	if (args->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
+	if (fsopt->wsize)
+		seq_printf(m, ",wsize=%d", fsopt->wsize);
+	if (fsopt->rsize != CEPH_MOUNT_RSIZE_DEFAULT)
+		seq_printf(m, ",rsize=%d", fsopt->rsize);
+	if (fsopt->congestion_kb != default_congestion_kb())
+		seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb);
+	if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
 		seq_printf(m, ",caps_wanted_delay_min=%d",
-			 args->caps_wanted_delay_min);
-	if (args->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
+			 fsopt->caps_wanted_delay_min);
+	if (fsopt->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
 		seq_printf(m, ",caps_wanted_delay_max=%d",
-			   args->caps_wanted_delay_max);
-	if (args->cap_release_safety != CEPH_CAP_RELEASE_SAFETY_DEFAULT)
+			   fsopt->caps_wanted_delay_max);
+	if (fsopt->cap_release_safety != CEPH_CAP_RELEASE_SAFETY_DEFAULT)
 		seq_printf(m, ",cap_release_safety=%d",
-			   args->cap_release_safety);
-	if (args->max_readdir != CEPH_MAX_READDIR_DEFAULT)
-		seq_printf(m, ",readdir_max_entries=%d", args->max_readdir);
-	if (args->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
-		seq_printf(m, ",readdir_max_bytes=%d", args->max_readdir_bytes);
-	if (strcmp(args->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
-		seq_printf(m, ",snapdirname=%s", args->snapdir_name);
-	if (args->name)
-		seq_printf(m, ",name=%s", args->name);
-	if (args->secret)
-		seq_puts(m, ",secret=<hidden>");
+			   fsopt->cap_release_safety);
+	if (fsopt->max_readdir != CEPH_MAX_READDIR_DEFAULT)
+		seq_printf(m, ",readdir_max_entries=%d", fsopt->max_readdir);
+	if (fsopt->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
+		seq_printf(m, ",readdir_max_bytes=%d", fsopt->max_readdir_bytes);
+	if (strcmp(fsopt->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
+		seq_printf(m, ",snapdirname=%s", fsopt->snapdir_name);
 	return 0;
 }
 
 /*
+ * handle any mon messages the standard library doesn't understand.
+ * return error if we don't either.
+ */
+static int extra_mon_dispatch(struct ceph_client *client, struct ceph_msg *msg)
+{
+	struct ceph_fs_client *fsc = client->private;
+	int type = le16_to_cpu(msg->hdr.type);
+
+	switch (type) {
+	case CEPH_MSG_MDS_MAP:
+		ceph_mdsc_handle_map(fsc->mdsc, msg);
+		return 0;
+
+	default:
+		return -1;
+	}
+}
+
+/*
+ * create a new fs client
+ */
+struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
+					struct ceph_options *opt)
+{
+	struct ceph_fs_client *fsc;
+	int err = -ENOMEM;
+
+	fsc = kzalloc(sizeof(*fsc), GFP_KERNEL);
+	if (!fsc)
+		return ERR_PTR(-ENOMEM);
+
+	fsc->client = ceph_create_client(opt, fsc);
+	if (IS_ERR(fsc->client)) {
+		err = PTR_ERR(fsc->client);
+		goto fail;
+	}
+	fsc->client->extra_mon_dispatch = extra_mon_dispatch;
+	fsc->client->supported_features |= CEPH_FEATURE_FLOCK;
+	fsc->client->monc.want_mdsmap = 1;
+
+	fsc->mount_options = fsopt;
+
+	fsc->sb = NULL;
+	fsc->mount_state = CEPH_MOUNT_MOUNTING;
+
+	atomic_long_set(&fsc->writeback_count, 0);
+
+	err = bdi_init(&fsc->backing_dev_info);
+	if (err < 0)
+		goto fail_client;
+
+	err = -ENOMEM;
+	fsc->wb_wq = create_workqueue("ceph-writeback");
+	if (fsc->wb_wq == NULL)
+		goto fail_bdi;
+	fsc->pg_inv_wq = create_singlethread_workqueue("ceph-pg-invalid");
+	if (fsc->pg_inv_wq == NULL)
+		goto fail_wb_wq;
+	fsc->trunc_wq = create_singlethread_workqueue("ceph-trunc");
+	if (fsc->trunc_wq == NULL)
+		goto fail_pg_inv_wq;
+
+	/* set up mempools */
+	err = -ENOMEM;
+	fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
+			      fsc->mount_options->wsize >> PAGE_CACHE_SHIFT);
+	if (!fsc->wb_pagevec_pool)
+		goto fail_trunc_wq;
+
+	/* caps */
+	fsc->min_caps = fsopt->max_readdir;
+
+	return fsc;
+
+fail_trunc_wq:
+	destroy_workqueue(fsc->trunc_wq);
+fail_pg_inv_wq:
+	destroy_workqueue(fsc->pg_inv_wq);
+fail_wb_wq:
+	destroy_workqueue(fsc->wb_wq);
+fail_bdi:
+	bdi_destroy(&fsc->backing_dev_info);
+fail_client:
+	ceph_destroy_client(fsc->client);
+fail:
+	kfree(fsc);
+	return ERR_PTR(err);
+}
+
+void destroy_fs_client(struct ceph_fs_client *fsc)
+{
+	dout("destroy_fs_client %p\n", fsc);
+
+	destroy_workqueue(fsc->wb_wq);
+	destroy_workqueue(fsc->pg_inv_wq);
+	destroy_workqueue(fsc->trunc_wq);
+
+	bdi_destroy(&fsc->backing_dev_info);
+
+	mempool_destroy(fsc->wb_pagevec_pool);
+
+	destroy_mount_options(fsc->mount_options);
+
+	ceph_fs_debugfs_cleanup(fsc);
+
+	ceph_destroy_client(fsc->client);
+
+	kfree(fsc);
+	dout("destroy_fs_client %p done\n", fsc);
+}
+
+/*
  * caches
  */
 struct kmem_cache *ceph_inode_cachep;
@@ -274,12 +567,12 @@
  */
 static void ceph_umount_begin(struct super_block *sb)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 
 	dout("ceph_umount_begin - starting forced umount\n");
-	if (!client)
+	if (!fsc)
 		return;
-	client->mount_state = CEPH_MOUNT_SHUTDOWN;
+	fsc->mount_state = CEPH_MOUNT_SHUTDOWN;
 	return;
 }
 
@@ -294,483 +587,15 @@
 	.umount_begin   = ceph_umount_begin,
 };
 
-
-const char *ceph_msg_type_name(int type)
-{
-	switch (type) {
-	case CEPH_MSG_SHUTDOWN: return "shutdown";
-	case CEPH_MSG_PING: return "ping";
-	case CEPH_MSG_AUTH: return "auth";
-	case CEPH_MSG_AUTH_REPLY: return "auth_reply";
-	case CEPH_MSG_MON_MAP: return "mon_map";
-	case CEPH_MSG_MON_GET_MAP: return "mon_get_map";
-	case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe";
-	case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack";
-	case CEPH_MSG_STATFS: return "statfs";
-	case CEPH_MSG_STATFS_REPLY: return "statfs_reply";
-	case CEPH_MSG_MDS_MAP: return "mds_map";
-	case CEPH_MSG_CLIENT_SESSION: return "client_session";
-	case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect";
-	case CEPH_MSG_CLIENT_REQUEST: return "client_request";
-	case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward";
-	case CEPH_MSG_CLIENT_REPLY: return "client_reply";
-	case CEPH_MSG_CLIENT_CAPS: return "client_caps";
-	case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release";
-	case CEPH_MSG_CLIENT_SNAP: return "client_snap";
-	case CEPH_MSG_CLIENT_LEASE: return "client_lease";
-	case CEPH_MSG_OSD_MAP: return "osd_map";
-	case CEPH_MSG_OSD_OP: return "osd_op";
-	case CEPH_MSG_OSD_OPREPLY: return "osd_opreply";
-	default: return "unknown";
-	}
-}
-
-
-/*
- * mount options
- */
-enum {
-	Opt_wsize,
-	Opt_rsize,
-	Opt_osdtimeout,
-	Opt_osdkeepalivetimeout,
-	Opt_mount_timeout,
-	Opt_osd_idle_ttl,
-	Opt_caps_wanted_delay_min,
-	Opt_caps_wanted_delay_max,
-	Opt_cap_release_safety,
-	Opt_readdir_max_entries,
-	Opt_readdir_max_bytes,
-	Opt_congestion_kb,
-	Opt_last_int,
-	/* int args above */
-	Opt_fsid,
-	Opt_snapdirname,
-	Opt_name,
-	Opt_secret,
-	Opt_last_string,
-	/* string args above */
-	Opt_ip,
-	Opt_noshare,
-	Opt_dirstat,
-	Opt_nodirstat,
-	Opt_rbytes,
-	Opt_norbytes,
-	Opt_nocrc,
-	Opt_noasyncreaddir,
-};
-
-static match_table_t arg_tokens = {
-	{Opt_wsize, "wsize=%d"},
-	{Opt_rsize, "rsize=%d"},
-	{Opt_osdtimeout, "osdtimeout=%d"},
-	{Opt_osdkeepalivetimeout, "osdkeepalive=%d"},
-	{Opt_mount_timeout, "mount_timeout=%d"},
-	{Opt_osd_idle_ttl, "osd_idle_ttl=%d"},
-	{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
-	{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
-	{Opt_cap_release_safety, "cap_release_safety=%d"},
-	{Opt_readdir_max_entries, "readdir_max_entries=%d"},
-	{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
-	{Opt_congestion_kb, "write_congestion_kb=%d"},
-	/* int args above */
-	{Opt_fsid, "fsid=%s"},
-	{Opt_snapdirname, "snapdirname=%s"},
-	{Opt_name, "name=%s"},
-	{Opt_secret, "secret=%s"},
-	/* string args above */
-	{Opt_ip, "ip=%s"},
-	{Opt_noshare, "noshare"},
-	{Opt_dirstat, "dirstat"},
-	{Opt_nodirstat, "nodirstat"},
-	{Opt_rbytes, "rbytes"},
-	{Opt_norbytes, "norbytes"},
-	{Opt_nocrc, "nocrc"},
-	{Opt_noasyncreaddir, "noasyncreaddir"},
-	{-1, NULL}
-};
-
-static int parse_fsid(const char *str, struct ceph_fsid *fsid)
-{
-	int i = 0;
-	char tmp[3];
-	int err = -EINVAL;
-	int d;
-
-	dout("parse_fsid '%s'\n", str);
-	tmp[2] = 0;
-	while (*str && i < 16) {
-		if (ispunct(*str)) {
-			str++;
-			continue;
-		}
-		if (!isxdigit(str[0]) || !isxdigit(str[1]))
-			break;
-		tmp[0] = str[0];
-		tmp[1] = str[1];
-		if (sscanf(tmp, "%x", &d) < 1)
-			break;
-		fsid->fsid[i] = d & 0xff;
-		i++;
-		str += 2;
-	}
-
-	if (i == 16)
-		err = 0;
-	dout("parse_fsid ret %d got fsid %pU", err, fsid);
-	return err;
-}
-
-static struct ceph_mount_args *parse_mount_args(int flags, char *options,
-						const char *dev_name,
-						const char **path)
-{
-	struct ceph_mount_args *args;
-	const char *c;
-	int err = -ENOMEM;
-	substring_t argstr[MAX_OPT_ARGS];
-
-	args = kzalloc(sizeof(*args), GFP_KERNEL);
-	if (!args)
-		return ERR_PTR(-ENOMEM);
-	args->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*args->mon_addr),
-				 GFP_KERNEL);
-	if (!args->mon_addr)
-		goto out;
-
-	dout("parse_mount_args %p, dev_name '%s'\n", args, dev_name);
-
-	/* start with defaults */
-	args->sb_flags = flags;
-	args->flags = CEPH_OPT_DEFAULT;
-	args->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT;
-	args->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT;
-	args->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */
-	args->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT;   /* seconds */
-	args->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
-	args->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
-	args->rsize = CEPH_MOUNT_RSIZE_DEFAULT;
-	args->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
-	args->cap_release_safety = CEPH_CAP_RELEASE_SAFETY_DEFAULT;
-	args->max_readdir = CEPH_MAX_READDIR_DEFAULT;
-	args->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
-	args->congestion_kb = default_congestion_kb();
-
-	/* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
-	err = -EINVAL;
-	if (!dev_name)
-		goto out;
-	*path = strstr(dev_name, ":/");
-	if (*path == NULL) {
-		pr_err("device name is missing path (no :/ in %s)\n",
-		       dev_name);
-		goto out;
-	}
-
-	/* get mon ip(s) */
-	err = ceph_parse_ips(dev_name, *path, args->mon_addr,
-			     CEPH_MAX_MON, &args->num_mon);
-	if (err < 0)
-		goto out;
-
-	/* path on server */
-	*path += 2;
-	dout("server path '%s'\n", *path);
-
-	/* parse mount options */
-	while ((c = strsep(&options, ",")) != NULL) {
-		int token, intval, ret;
-		if (!*c)
-			continue;
-		err = -EINVAL;
-		token = match_token((char *)c, arg_tokens, argstr);
-		if (token < 0) {
-			pr_err("bad mount option at '%s'\n", c);
-			goto out;
-		}
-		if (token < Opt_last_int) {
-			ret = match_int(&argstr[0], &intval);
-			if (ret < 0) {
-				pr_err("bad mount option arg (not int) "
-				       "at '%s'\n", c);
-				continue;
-			}
-			dout("got int token %d val %d\n", token, intval);
-		} else if (token > Opt_last_int && token < Opt_last_string) {
-			dout("got string token %d val %s\n", token,
-			     argstr[0].from);
-		} else {
-			dout("got token %d\n", token);
-		}
-		switch (token) {
-		case Opt_ip:
-			err = ceph_parse_ips(argstr[0].from,
-					     argstr[0].to,
-					     &args->my_addr,
-					     1, NULL);
-			if (err < 0)
-				goto out;
-			args->flags |= CEPH_OPT_MYIP;
-			break;
-
-		case Opt_fsid:
-			err = parse_fsid(argstr[0].from, &args->fsid);
-			if (err == 0)
-				args->flags |= CEPH_OPT_FSID;
-			break;
-		case Opt_snapdirname:
-			kfree(args->snapdir_name);
-			args->snapdir_name = kstrndup(argstr[0].from,
-					      argstr[0].to-argstr[0].from,
-					      GFP_KERNEL);
-			break;
-		case Opt_name:
-			args->name = kstrndup(argstr[0].from,
-					      argstr[0].to-argstr[0].from,
-					      GFP_KERNEL);
-			break;
-		case Opt_secret:
-			args->secret = kstrndup(argstr[0].from,
-						argstr[0].to-argstr[0].from,
-						GFP_KERNEL);
-			break;
-
-			/* misc */
-		case Opt_wsize:
-			args->wsize = intval;
-			break;
-		case Opt_rsize:
-			args->rsize = intval;
-			break;
-		case Opt_osdtimeout:
-			args->osd_timeout = intval;
-			break;
-		case Opt_osdkeepalivetimeout:
-			args->osd_keepalive_timeout = intval;
-			break;
-		case Opt_osd_idle_ttl:
-			args->osd_idle_ttl = intval;
-			break;
-		case Opt_mount_timeout:
-			args->mount_timeout = intval;
-			break;
-		case Opt_caps_wanted_delay_min:
-			args->caps_wanted_delay_min = intval;
-			break;
-		case Opt_caps_wanted_delay_max:
-			args->caps_wanted_delay_max = intval;
-			break;
-		case Opt_readdir_max_entries:
-			args->max_readdir = intval;
-			break;
-		case Opt_readdir_max_bytes:
-			args->max_readdir_bytes = intval;
-			break;
-		case Opt_congestion_kb:
-			args->congestion_kb = intval;
-			break;
-
-		case Opt_noshare:
-			args->flags |= CEPH_OPT_NOSHARE;
-			break;
-
-		case Opt_dirstat:
-			args->flags |= CEPH_OPT_DIRSTAT;
-			break;
-		case Opt_nodirstat:
-			args->flags &= ~CEPH_OPT_DIRSTAT;
-			break;
-		case Opt_rbytes:
-			args->flags |= CEPH_OPT_RBYTES;
-			break;
-		case Opt_norbytes:
-			args->flags &= ~CEPH_OPT_RBYTES;
-			break;
-		case Opt_nocrc:
-			args->flags |= CEPH_OPT_NOCRC;
-			break;
-		case Opt_noasyncreaddir:
-			args->flags |= CEPH_OPT_NOASYNCREADDIR;
-			break;
-
-		default:
-			BUG_ON(token);
-		}
-	}
-	return args;
-
-out:
-	kfree(args->mon_addr);
-	kfree(args);
-	return ERR_PTR(err);
-}
-
-static void destroy_mount_args(struct ceph_mount_args *args)
-{
-	dout("destroy_mount_args %p\n", args);
-	kfree(args->snapdir_name);
-	args->snapdir_name = NULL;
-	kfree(args->name);
-	args->name = NULL;
-	kfree(args->secret);
-	args->secret = NULL;
-	kfree(args);
-}
-
-/*
- * create a fresh client instance
- */
-static struct ceph_client *ceph_create_client(struct ceph_mount_args *args)
-{
-	struct ceph_client *client;
-	int err = -ENOMEM;
-
-	client = kzalloc(sizeof(*client), GFP_KERNEL);
-	if (client == NULL)
-		return ERR_PTR(-ENOMEM);
-
-	mutex_init(&client->mount_mutex);
-
-	init_waitqueue_head(&client->auth_wq);
-
-	client->sb = NULL;
-	client->mount_state = CEPH_MOUNT_MOUNTING;
-	client->mount_args = args;
-
-	client->msgr = NULL;
-
-	client->auth_err = 0;
-	atomic_long_set(&client->writeback_count, 0);
-
-	err = bdi_init(&client->backing_dev_info);
-	if (err < 0)
-		goto fail;
-
-	err = -ENOMEM;
-	client->wb_wq = create_workqueue("ceph-writeback");
-	if (client->wb_wq == NULL)
-		goto fail_bdi;
-	client->pg_inv_wq = create_singlethread_workqueue("ceph-pg-invalid");
-	if (client->pg_inv_wq == NULL)
-		goto fail_wb_wq;
-	client->trunc_wq = create_singlethread_workqueue("ceph-trunc");
-	if (client->trunc_wq == NULL)
-		goto fail_pg_inv_wq;
-
-	/* set up mempools */
-	err = -ENOMEM;
-	client->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
-			      client->mount_args->wsize >> PAGE_CACHE_SHIFT);
-	if (!client->wb_pagevec_pool)
-		goto fail_trunc_wq;
-
-	/* caps */
-	client->min_caps = args->max_readdir;
-
-	/* subsystems */
-	err = ceph_monc_init(&client->monc, client);
-	if (err < 0)
-		goto fail_mempool;
-	err = ceph_osdc_init(&client->osdc, client);
-	if (err < 0)
-		goto fail_monc;
-	err = ceph_mdsc_init(&client->mdsc, client);
-	if (err < 0)
-		goto fail_osdc;
-	return client;
-
-fail_osdc:
-	ceph_osdc_stop(&client->osdc);
-fail_monc:
-	ceph_monc_stop(&client->monc);
-fail_mempool:
-	mempool_destroy(client->wb_pagevec_pool);
-fail_trunc_wq:
-	destroy_workqueue(client->trunc_wq);
-fail_pg_inv_wq:
-	destroy_workqueue(client->pg_inv_wq);
-fail_wb_wq:
-	destroy_workqueue(client->wb_wq);
-fail_bdi:
-	bdi_destroy(&client->backing_dev_info);
-fail:
-	kfree(client);
-	return ERR_PTR(err);
-}
-
-static void ceph_destroy_client(struct ceph_client *client)
-{
-	dout("destroy_client %p\n", client);
-
-	/* unmount */
-	ceph_mdsc_stop(&client->mdsc);
-	ceph_osdc_stop(&client->osdc);
-
-	/*
-	 * make sure mds and osd connections close out before destroying
-	 * the auth module, which is needed to free those connections'
-	 * ceph_authorizers.
-	 */
-	ceph_msgr_flush();
-
-	ceph_monc_stop(&client->monc);
-
-	ceph_debugfs_client_cleanup(client);
-	destroy_workqueue(client->wb_wq);
-	destroy_workqueue(client->pg_inv_wq);
-	destroy_workqueue(client->trunc_wq);
-
-	bdi_destroy(&client->backing_dev_info);
-
-	if (client->msgr)
-		ceph_messenger_destroy(client->msgr);
-	mempool_destroy(client->wb_pagevec_pool);
-
-	destroy_mount_args(client->mount_args);
-
-	kfree(client);
-	dout("destroy_client %p done\n", client);
-}
-
-/*
- * Initially learn our fsid, or verify an fsid matches.
- */
-int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
-{
-	if (client->have_fsid) {
-		if (ceph_fsid_compare(&client->fsid, fsid)) {
-			pr_err("bad fsid, had %pU got %pU",
-			       &client->fsid, fsid);
-			return -1;
-		}
-	} else {
-		pr_info("client%lld fsid %pU\n", client->monc.auth->global_id,
-			fsid);
-		memcpy(&client->fsid, fsid, sizeof(*fsid));
-		ceph_debugfs_client_init(client);
-		client->have_fsid = true;
-	}
-	return 0;
-}
-
-/*
- * true if we have the mon map (and have thus joined the cluster)
- */
-static int have_mon_and_osd_map(struct ceph_client *client)
-{
-	return client->monc.monmap && client->monc.monmap->epoch &&
-	       client->osdc.osdmap && client->osdc.osdmap->epoch;
-}
-
 /*
  * Bootstrap mount by opening the root directory.  Note the mount
  * @started time from caller, and time out if this takes too long.
  */
-static struct dentry *open_root_dentry(struct ceph_client *client,
+static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
 				       const char *path,
 				       unsigned long started)
 {
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req = NULL;
 	int err;
 	struct dentry *root;
@@ -784,14 +609,14 @@
 	req->r_ino1.ino = CEPH_INO_ROOT;
 	req->r_ino1.snap = CEPH_NOSNAP;
 	req->r_started = started;
-	req->r_timeout = client->mount_args->mount_timeout * HZ;
+	req->r_timeout = fsc->client->options->mount_timeout * HZ;
 	req->r_args.getattr.mask = cpu_to_le32(CEPH_STAT_CAP_INODE);
 	req->r_num_caps = 2;
 	err = ceph_mdsc_do_request(mdsc, NULL, req);
 	if (err == 0) {
 		dout("open_root_inode success\n");
 		if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT &&
-		    client->sb->s_root == NULL)
+		    fsc->sb->s_root == NULL)
 			root = d_alloc_root(req->r_target_inode);
 		else
 			root = d_obtain_alias(req->r_target_inode);
@@ -804,105 +629,86 @@
 	return root;
 }
 
+
+
+
 /*
  * mount: join the ceph cluster, and open root directory.
  */
-static int ceph_mount(struct ceph_client *client, struct vfsmount *mnt,
+static int ceph_mount(struct ceph_fs_client *fsc, struct vfsmount *mnt,
 		      const char *path)
 {
-	struct ceph_entity_addr *myaddr = NULL;
 	int err;
-	unsigned long timeout = client->mount_args->mount_timeout * HZ;
 	unsigned long started = jiffies;  /* note the start time */
 	struct dentry *root;
+	int first = 0;   /* first vfsmount for this super_block */
 
 	dout("mount start\n");
-	mutex_lock(&client->mount_mutex);
+	mutex_lock(&fsc->client->mount_mutex);
 
-	/* initialize the messenger */
-	if (client->msgr == NULL) {
-		if (ceph_test_opt(client, MYIP))
-			myaddr = &client->mount_args->my_addr;
-		client->msgr = ceph_messenger_create(myaddr);
-		if (IS_ERR(client->msgr)) {
-			err = PTR_ERR(client->msgr);
-			client->msgr = NULL;
-			goto out;
-		}
-		client->msgr->nocrc = ceph_test_opt(client, NOCRC);
-	}
-
-	/* open session, and wait for mon, mds, and osd maps */
-	err = ceph_monc_open_session(&client->monc);
+	err = __ceph_open_session(fsc->client, started);
 	if (err < 0)
 		goto out;
 
-	while (!have_mon_and_osd_map(client)) {
-		err = -EIO;
-		if (timeout && time_after_eq(jiffies, started + timeout))
-			goto out;
-
-		/* wait */
-		dout("mount waiting for mon_map\n");
-		err = wait_event_interruptible_timeout(client->auth_wq,
-		       have_mon_and_osd_map(client) || (client->auth_err < 0),
-		       timeout);
-		if (err == -EINTR || err == -ERESTARTSYS)
-			goto out;
-		if (client->auth_err < 0) {
-			err = client->auth_err;
-			goto out;
-		}
-	}
-
 	dout("mount opening root\n");
-	root = open_root_dentry(client, "", started);
+	root = open_root_dentry(fsc, "", started);
 	if (IS_ERR(root)) {
 		err = PTR_ERR(root);
 		goto out;
 	}
-	if (client->sb->s_root)
+	if (fsc->sb->s_root) {
 		dput(root);
-	else
-		client->sb->s_root = root;
+	} else {
+		fsc->sb->s_root = root;
+		first = 1;
+
+		err = ceph_fs_debugfs_init(fsc);
+		if (err < 0)
+			goto fail;
+	}
 
 	if (path[0] == 0) {
 		dget(root);
 	} else {
 		dout("mount opening base mountpoint\n");
-		root = open_root_dentry(client, path, started);
+		root = open_root_dentry(fsc, path, started);
 		if (IS_ERR(root)) {
 			err = PTR_ERR(root);
-			dput(client->sb->s_root);
-			client->sb->s_root = NULL;
-			goto out;
+			goto fail;
 		}
 	}
 
 	mnt->mnt_root = root;
-	mnt->mnt_sb = client->sb;
+	mnt->mnt_sb = fsc->sb;
 
-	client->mount_state = CEPH_MOUNT_MOUNTED;
+	fsc->mount_state = CEPH_MOUNT_MOUNTED;
 	dout("mount success\n");
 	err = 0;
 
 out:
-	mutex_unlock(&client->mount_mutex);
+	mutex_unlock(&fsc->client->mount_mutex);
 	return err;
+
+fail:
+	if (first) {
+		dput(fsc->sb->s_root);
+		fsc->sb->s_root = NULL;
+	}
+	goto out;
 }
 
 static int ceph_set_super(struct super_block *s, void *data)
 {
-	struct ceph_client *client = data;
+	struct ceph_fs_client *fsc = data;
 	int ret;
 
 	dout("set_super %p data %p\n", s, data);
 
-	s->s_flags = client->mount_args->sb_flags;
+	s->s_flags = fsc->mount_options->sb_flags;
 	s->s_maxbytes = 1ULL << 40;  /* temp value until we get mdsmap */
 
-	s->s_fs_info = client;
-	client->sb = s;
+	s->s_fs_info = fsc;
+	fsc->sb = s;
 
 	s->s_op = &ceph_super_ops;
 	s->s_export_op = &ceph_export_ops;
@@ -917,7 +723,7 @@
 
 fail:
 	s->s_fs_info = NULL;
-	client->sb = NULL;
+	fsc->sb = NULL;
 	return ret;
 }
 
@@ -926,30 +732,23 @@
  */
 static int ceph_compare_super(struct super_block *sb, void *data)
 {
-	struct ceph_client *new = data;
-	struct ceph_mount_args *args = new->mount_args;
-	struct ceph_client *other = ceph_sb_to_client(sb);
-	int i;
+	struct ceph_fs_client *new = data;
+	struct ceph_mount_options *fsopt = new->mount_options;
+	struct ceph_options *opt = new->client->options;
+	struct ceph_fs_client *other = ceph_sb_to_client(sb);
 
 	dout("ceph_compare_super %p\n", sb);
-	if (args->flags & CEPH_OPT_FSID) {
-		if (ceph_fsid_compare(&args->fsid, &other->fsid)) {
-			dout("fsid doesn't match\n");
-			return 0;
-		}
-	} else {
-		/* do we share (a) monitor? */
-		for (i = 0; i < new->monc.monmap->num_mon; i++)
-			if (ceph_monmap_contains(other->monc.monmap,
-					 &new->monc.monmap->mon_inst[i].addr))
-				break;
-		if (i == new->monc.monmap->num_mon) {
-			dout("mon ip not part of monmap\n");
-			return 0;
-		}
-		dout("mon ip matches existing sb %p\n", sb);
+
+	if (compare_mount_options(fsopt, opt, other)) {
+		dout("monitor(s)/mount options don't match\n");
+		return 0;
 	}
-	if (args->sb_flags != other->mount_args->sb_flags) {
+	if ((opt->flags & CEPH_OPT_FSID) &&
+	    ceph_fsid_compare(&opt->fsid, &other->client->fsid)) {
+		dout("fsid doesn't match\n");
+		return 0;
+	}
+	if (fsopt->sb_flags != other->mount_options->sb_flags) {
 		dout("flags differ\n");
 		return 0;
 	}
@@ -961,19 +760,20 @@
  */
 static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
 
-static int ceph_register_bdi(struct super_block *sb, struct ceph_client *client)
+static int ceph_register_bdi(struct super_block *sb,
+			     struct ceph_fs_client *fsc)
 {
 	int err;
 
 	/* set ra_pages based on rsize mount option? */
-	if (client->mount_args->rsize >= PAGE_CACHE_SIZE)
-		client->backing_dev_info.ra_pages =
-			(client->mount_args->rsize + PAGE_CACHE_SIZE - 1)
+	if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE)
+		fsc->backing_dev_info.ra_pages =
+			(fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1)
 			>> PAGE_SHIFT;
-	err = bdi_register(&client->backing_dev_info, NULL, "ceph-%d",
+	err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%d",
 			   atomic_long_inc_return(&bdi_seq));
 	if (!err)
-		sb->s_bdi = &client->backing_dev_info;
+		sb->s_bdi = &fsc->backing_dev_info;
 	return err;
 }
 
@@ -982,46 +782,52 @@
 		       struct vfsmount *mnt)
 {
 	struct super_block *sb;
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	int err;
 	int (*compare_super)(struct super_block *, void *) = ceph_compare_super;
 	const char *path = NULL;
-	struct ceph_mount_args *args;
+	struct ceph_mount_options *fsopt = NULL;
+	struct ceph_options *opt = NULL;
 
 	dout("ceph_get_sb\n");
-	args = parse_mount_args(flags, data, dev_name, &path);
-	if (IS_ERR(args)) {
-		err = PTR_ERR(args);
+	err = parse_mount_options(&fsopt, &opt, flags, data, dev_name, &path);
+	if (err < 0)
 		goto out_final;
-	}
 
 	/* create client (which we may/may not use) */
-	client = ceph_create_client(args);
-	if (IS_ERR(client)) {
-		err = PTR_ERR(client);
+	fsc = create_fs_client(fsopt, opt);
+	if (IS_ERR(fsc)) {
+		err = PTR_ERR(fsc);
+		kfree(fsopt);
+		kfree(opt);
 		goto out_final;
 	}
 
-	if (client->mount_args->flags & CEPH_OPT_NOSHARE)
+	err = ceph_mdsc_init(fsc);
+	if (err < 0)
+		goto out;
+
+	if (ceph_test_opt(fsc->client, NOSHARE))
 		compare_super = NULL;
-	sb = sget(fs_type, compare_super, ceph_set_super, client);
+	sb = sget(fs_type, compare_super, ceph_set_super, fsc);
 	if (IS_ERR(sb)) {
 		err = PTR_ERR(sb);
 		goto out;
 	}
 
-	if (ceph_sb_to_client(sb) != client) {
-		ceph_destroy_client(client);
-		client = ceph_sb_to_client(sb);
-		dout("get_sb got existing client %p\n", client);
+	if (ceph_sb_to_client(sb) != fsc) {
+		ceph_mdsc_destroy(fsc);
+		destroy_fs_client(fsc);
+		fsc = ceph_sb_to_client(sb);
+		dout("get_sb got existing client %p\n", fsc);
 	} else {
-		dout("get_sb using new client %p\n", client);
-		err = ceph_register_bdi(sb, client);
+		dout("get_sb using new client %p\n", fsc);
+		err = ceph_register_bdi(sb, fsc);
 		if (err < 0)
 			goto out_splat;
 	}
 
-	err = ceph_mount(client, mnt, path);
+	err = ceph_mount(fsc, mnt, path);
 	if (err < 0)
 		goto out_splat;
 	dout("root %p inode %p ino %llx.%llx\n", mnt->mnt_root,
@@ -1029,12 +835,13 @@
 	return 0;
 
 out_splat:
-	ceph_mdsc_close_sessions(&client->mdsc);
+	ceph_mdsc_close_sessions(fsc->mdsc);
 	deactivate_locked_super(sb);
 	goto out_final;
 
 out:
-	ceph_destroy_client(client);
+	ceph_mdsc_destroy(fsc);
+	destroy_fs_client(fsc);
 out_final:
 	dout("ceph_get_sb fail %d\n", err);
 	return err;
@@ -1042,11 +849,12 @@
 
 static void ceph_kill_sb(struct super_block *s)
 {
-	struct ceph_client *client = ceph_sb_to_client(s);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(s);
 	dout("kill_sb %p\n", s);
-	ceph_mdsc_pre_umount(&client->mdsc);
+	ceph_mdsc_pre_umount(fsc->mdsc);
 	kill_anon_super(s);    /* will call put_super after sb is r/o */
-	ceph_destroy_client(client);
+	ceph_mdsc_destroy(fsc);
+	destroy_fs_client(fsc);
 }
 
 static struct file_system_type ceph_fs_type = {
@@ -1062,36 +870,20 @@
 
 static int __init init_ceph(void)
 {
-	int ret = 0;
-
-	ret = ceph_debugfs_init();
-	if (ret < 0)
-		goto out;
-
-	ret = ceph_msgr_init();
-	if (ret < 0)
-		goto out_debugfs;
-
-	ret = init_caches();
+	int ret = init_caches();
 	if (ret)
-		goto out_msgr;
+		goto out;
 
 	ret = register_filesystem(&ceph_fs_type);
 	if (ret)
 		goto out_icache;
 
-	pr_info("loaded (mon/mds/osd proto %d/%d/%d, osdmap %d/%d %d/%d)\n",
-		CEPH_MONC_PROTOCOL, CEPH_MDSC_PROTOCOL, CEPH_OSDC_PROTOCOL,
-		CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT,
-		CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT);
+	pr_info("loaded (mds proto %d)\n", CEPH_MDSC_PROTOCOL);
+
 	return 0;
 
 out_icache:
 	destroy_caches();
-out_msgr:
-	ceph_msgr_exit();
-out_debugfs:
-	ceph_debugfs_cleanup();
 out:
 	return ret;
 }
@@ -1101,8 +893,6 @@
 	dout("exit_ceph\n");
 	unregister_filesystem(&ceph_fs_type);
 	destroy_caches();
-	ceph_msgr_exit();
-	ceph_debugfs_cleanup();
 }
 
 module_init(init_ceph);
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index b87638e..1886294 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -1,7 +1,7 @@
 #ifndef _FS_CEPH_SUPER_H
 #define _FS_CEPH_SUPER_H
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <asm/unaligned.h>
 #include <linux/backing-dev.h>
@@ -14,13 +14,7 @@
 #include <linux/writeback.h>
 #include <linux/slab.h>
 
-#include "types.h"
-#include "messenger.h"
-#include "msgpool.h"
-#include "mon_client.h"
-#include "mds_client.h"
-#include "osd_client.h"
-#include "ceph_fs.h"
+#include <linux/ceph/libceph.h>
 
 /* f_type in struct statfs */
 #define CEPH_SUPER_MAGIC 0x00c36400
@@ -30,42 +24,25 @@
 #define CEPH_BLOCK_SHIFT   20  /* 1 MB */
 #define CEPH_BLOCK         (1 << CEPH_BLOCK_SHIFT)
 
-/*
- * Supported features
- */
-#define CEPH_FEATURE_SUPPORTED CEPH_FEATURE_NOSRCADDR | CEPH_FEATURE_FLOCK
-#define CEPH_FEATURE_REQUIRED  CEPH_FEATURE_NOSRCADDR
+#define CEPH_MOUNT_OPT_DIRSTAT         (1<<4) /* `cat dirname` for stats */
+#define CEPH_MOUNT_OPT_RBYTES          (1<<5) /* dir st_bytes = rbytes */
+#define CEPH_MOUNT_OPT_NOASYNCREADDIR  (1<<7) /* no dcache readdir */
 
-/*
- * mount options
- */
-#define CEPH_OPT_FSID             (1<<0)
-#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
-#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
-#define CEPH_OPT_DIRSTAT          (1<<4) /* funky `cat dirname` for stats */
-#define CEPH_OPT_RBYTES           (1<<5) /* dir st_bytes = rbytes */
-#define CEPH_OPT_NOCRC            (1<<6) /* no data crc on writes */
-#define CEPH_OPT_NOASYNCREADDIR   (1<<7) /* no dcache readdir */
+#define CEPH_MOUNT_OPT_DEFAULT    (CEPH_MOUNT_OPT_RBYTES)
 
-#define CEPH_OPT_DEFAULT   (CEPH_OPT_RBYTES)
+#define ceph_set_mount_opt(fsc, opt) \
+	(fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt;
+#define ceph_test_mount_opt(fsc, opt) \
+	(!!((fsc)->mount_options->flags & CEPH_MOUNT_OPT_##opt))
 
-#define ceph_set_opt(client, opt) \
-	(client)->mount_args->flags |= CEPH_OPT_##opt;
-#define ceph_test_opt(client, opt) \
-	(!!((client)->mount_args->flags & CEPH_OPT_##opt))
+#define CEPH_MAX_READDIR_DEFAULT        1024
+#define CEPH_MAX_READDIR_BYTES_DEFAULT  (512*1024)
+#define CEPH_SNAPDIRNAME_DEFAULT        ".snap"
 
-
-struct ceph_mount_args {
-	int sb_flags;
+struct ceph_mount_options {
 	int flags;
-	struct ceph_fsid fsid;
-	struct ceph_entity_addr my_addr;
-	int num_mon;
-	struct ceph_entity_addr *mon_addr;
-	int mount_timeout;
-	int osd_idle_ttl;
-	int osd_timeout;
-	int osd_keepalive_timeout;
+	int sb_flags;
+
 	int wsize;
 	int rsize;            /* max readahead */
 	int congestion_kb;    /* max writeback in flight */
@@ -73,82 +50,25 @@
 	int cap_release_safety;
 	int max_readdir;       /* max readdir result (entires) */
 	int max_readdir_bytes; /* max readdir result (bytes) */
+
+	/*
+	 * everything above this point can be memcmp'd; everything below
+	 * is handled in compare_mount_options()
+	 */
+
 	char *snapdir_name;   /* default ".snap" */
-	char *name;
-	char *secret;
 };
 
-/*
- * defaults
- */
-#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
-#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
-#define CEPH_OSD_KEEPALIVE_DEFAULT  5
-#define CEPH_OSD_IDLE_TTL_DEFAULT    60
-#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
-#define CEPH_MAX_READDIR_DEFAULT    1024
-#define CEPH_MAX_READDIR_BYTES_DEFAULT    (512*1024)
-
-#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
-#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
-
-#define CEPH_SNAPDIRNAME_DEFAULT ".snap"
-#define CEPH_AUTH_NAME_DEFAULT   "guest"
-/*
- * Delay telling the MDS we no longer want caps, in case we reopen
- * the file.  Delay a minimum amount of time, even if we send a cap
- * message for some other reason.  Otherwise, take the oppotunity to
- * update the mds to avoid sending another message later.
- */
-#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
-#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
-
-#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
-
-/* mount state */
-enum {
-	CEPH_MOUNT_MOUNTING,
-	CEPH_MOUNT_MOUNTED,
-	CEPH_MOUNT_UNMOUNTING,
-	CEPH_MOUNT_UNMOUNTED,
-	CEPH_MOUNT_SHUTDOWN,
-};
-
-/*
- * subtract jiffies
- */
-static inline unsigned long time_sub(unsigned long a, unsigned long b)
-{
-	BUG_ON(time_after(b, a));
-	return (long)a - (long)b;
-}
-
-/*
- * per-filesystem client state
- *
- * possibly shared by multiple mount points, if they are
- * mounting the same ceph filesystem/cluster.
- */
-struct ceph_client {
-	struct ceph_fsid fsid;
-	bool have_fsid;
-
-	struct mutex mount_mutex;       /* serialize mount attempts */
-	struct ceph_mount_args *mount_args;
-
+struct ceph_fs_client {
 	struct super_block *sb;
 
+	struct ceph_mount_options *mount_options;
+	struct ceph_client *client;
+
 	unsigned long mount_state;
-	wait_queue_head_t auth_wq;
-
-	int auth_err;
-
 	int min_caps;                  /* min caps i added */
 
-	struct ceph_messenger *msgr;   /* messenger instance */
-	struct ceph_mon_client monc;
-	struct ceph_mds_client mdsc;
-	struct ceph_osd_client osdc;
+	struct ceph_mds_client *mdsc;
 
 	/* writeback */
 	mempool_t *wb_pagevec_pool;
@@ -160,14 +80,14 @@
 	struct backing_dev_info backing_dev_info;
 
 #ifdef CONFIG_DEBUG_FS
-	struct dentry *debugfs_monmap;
-	struct dentry *debugfs_mdsmap, *debugfs_osdmap;
-	struct dentry *debugfs_dir, *debugfs_dentry_lru, *debugfs_caps;
+	struct dentry *debugfs_dentry_lru, *debugfs_caps;
 	struct dentry *debugfs_congestion_kb;
 	struct dentry *debugfs_bdi;
+	struct dentry *debugfs_mdsc, *debugfs_mdsmap;
 #endif
 };
 
+
 /*
  * File i/o capability.  This tracks shared state with the metadata
  * server that allows us to cache or writeback attributes or to read
@@ -275,6 +195,20 @@
 	int should_free_val;
 };
 
+/*
+ * Ceph dentry state
+ */
+struct ceph_dentry_info {
+	struct ceph_mds_session *lease_session;
+	u32 lease_gen, lease_shared_gen;
+	u32 lease_seq;
+	unsigned long lease_renew_after, lease_renew_from;
+	struct list_head lru;
+	struct dentry *dentry;
+	u64 time;
+	u64 offset;
+};
+
 struct ceph_inode_xattrs_info {
 	/*
 	 * (still encoded) xattr blob. we avoid the overhead of parsing
@@ -296,11 +230,6 @@
 /*
  * Ceph inode.
  */
-#define CEPH_I_COMPLETE  1  /* we have complete directory cached */
-#define CEPH_I_NODELAY   4  /* do not delay cap release */
-#define CEPH_I_FLUSH     8  /* do not delay flush of dirty metadata */
-#define CEPH_I_NOFLUSH  16  /* do not flush dirty caps */
-
 struct ceph_inode_info {
 	struct ceph_vino i_vino;   /* ceph ino + snap */
 
@@ -391,69 +320,9 @@
 	return container_of(inode, struct ceph_inode_info, vfs_inode);
 }
 
-static inline void ceph_i_clear(struct inode *inode, unsigned mask)
+static inline struct ceph_vino ceph_vino(struct inode *inode)
 {
-	struct ceph_inode_info *ci = ceph_inode(inode);
-
-	spin_lock(&inode->i_lock);
-	ci->i_ceph_flags &= ~mask;
-	spin_unlock(&inode->i_lock);
-}
-
-static inline void ceph_i_set(struct inode *inode, unsigned mask)
-{
-	struct ceph_inode_info *ci = ceph_inode(inode);
-
-	spin_lock(&inode->i_lock);
-	ci->i_ceph_flags |= mask;
-	spin_unlock(&inode->i_lock);
-}
-
-static inline bool ceph_i_test(struct inode *inode, unsigned mask)
-{
-	struct ceph_inode_info *ci = ceph_inode(inode);
-	bool r;
-
-	smp_mb();
-	r = (ci->i_ceph_flags & mask) == mask;
-	return r;
-}
-
-
-/* find a specific frag @f */
-extern struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci,
-						u32 f);
-
-/*
- * choose fragment for value @v.  copy frag content to pfrag, if leaf
- * exists
- */
-extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
-			    struct ceph_inode_frag *pfrag,
-			    int *found);
-
-/*
- * Ceph dentry state
- */
-struct ceph_dentry_info {
-	struct ceph_mds_session *lease_session;
-	u32 lease_gen, lease_shared_gen;
-	u32 lease_seq;
-	unsigned long lease_renew_after, lease_renew_from;
-	struct list_head lru;
-	struct dentry *dentry;
-	u64 time;
-	u64 offset;
-};
-
-static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
-{
-	return (struct ceph_dentry_info *)dentry->d_fsdata;
-}
-
-static inline loff_t ceph_make_fpos(unsigned frag, unsigned off)
-{
-	return ((loff_t)frag << 32) | (loff_t)off;
+	return ceph_inode(inode)->i_vino;
 }
 
 /*
@@ -472,18 +341,6 @@
 	return ino;
 }
 
-static inline int ceph_set_ino_cb(struct inode *inode, void *data)
-{
-	ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
-	inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data);
-	return 0;
-}
-
-static inline struct ceph_vino ceph_vino(struct inode *inode)
-{
-	return ceph_inode(inode)->i_vino;
-}
-
 /* for printf-style formatting */
 #define ceph_vinop(i) ceph_inode(i)->i_vino.ino, ceph_inode(i)->i_vino.snap
 
@@ -513,6 +370,73 @@
 
 
 /*
+ * Ceph inode.
+ */
+#define CEPH_I_COMPLETE  1  /* we have complete directory cached */
+#define CEPH_I_NODELAY   4  /* do not delay cap release */
+#define CEPH_I_FLUSH     8  /* do not delay flush of dirty metadata */
+#define CEPH_I_NOFLUSH  16  /* do not flush dirty caps */
+
+static inline void ceph_i_clear(struct inode *inode, unsigned mask)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+
+	spin_lock(&inode->i_lock);
+	ci->i_ceph_flags &= ~mask;
+	spin_unlock(&inode->i_lock);
+}
+
+static inline void ceph_i_set(struct inode *inode, unsigned mask)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+
+	spin_lock(&inode->i_lock);
+	ci->i_ceph_flags |= mask;
+	spin_unlock(&inode->i_lock);
+}
+
+static inline bool ceph_i_test(struct inode *inode, unsigned mask)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+	bool r;
+
+	spin_lock(&inode->i_lock);
+	r = (ci->i_ceph_flags & mask) == mask;
+	spin_unlock(&inode->i_lock);
+	return r;
+}
+
+
+/* find a specific frag @f */
+extern struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci,
+						u32 f);
+
+/*
+ * choose fragment for value @v.  copy frag content to pfrag, if leaf
+ * exists
+ */
+extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
+			    struct ceph_inode_frag *pfrag,
+			    int *found);
+
+static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
+{
+	return (struct ceph_dentry_info *)dentry->d_fsdata;
+}
+
+static inline loff_t ceph_make_fpos(unsigned frag, unsigned off)
+{
+	return ((loff_t)frag << 32) | (loff_t)off;
+}
+
+static inline int ceph_set_ino_cb(struct inode *inode, void *data)
+{
+	ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
+	inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data);
+	return 0;
+}
+
+/*
  * caps helpers
  */
 static inline bool __ceph_is_any_real_caps(struct ceph_inode_info *ci)
@@ -576,18 +500,18 @@
 			     struct ceph_cap_reservation *ctx, int need);
 extern int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
 			       struct ceph_cap_reservation *ctx);
-extern void ceph_reservation_status(struct ceph_client *client,
+extern void ceph_reservation_status(struct ceph_fs_client *client,
 				    int *total, int *avail, int *used,
 				    int *reserved, int *min);
 
-static inline struct ceph_client *ceph_inode_to_client(struct inode *inode)
+static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
 {
-	return (struct ceph_client *)inode->i_sb->s_fs_info;
+	return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
 }
 
-static inline struct ceph_client *ceph_sb_to_client(struct super_block *sb)
+static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
 {
-	return (struct ceph_client *)sb->s_fs_info;
+	return (struct ceph_fs_client *)sb->s_fs_info;
 }
 
 
@@ -617,51 +541,6 @@
 
 
 /*
- * snapshots
- */
-
-/*
- * A "snap context" is the set of existing snapshots when we
- * write data.  It is used by the OSD to guide its COW behavior.
- *
- * The ceph_snap_context is refcounted, and attached to each dirty
- * page, indicating which context the dirty data belonged when it was
- * dirtied.
- */
-struct ceph_snap_context {
-	atomic_t nref;
-	u64 seq;
-	int num_snaps;
-	u64 snaps[];
-};
-
-static inline struct ceph_snap_context *
-ceph_get_snap_context(struct ceph_snap_context *sc)
-{
-	/*
-	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)+1);
-	*/
-	if (sc)
-		atomic_inc(&sc->nref);
-	return sc;
-}
-
-static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
-{
-	if (!sc)
-		return;
-	/*
-	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)-1);
-	*/
-	if (atomic_dec_and_test(&sc->nref)) {
-		/*printk(" deleting snap_context %p\n", sc);*/
-		kfree(sc);
-	}
-}
-
-/*
  * A "snap realm" describes a subset of the file hierarchy sharing
  * the same set of snapshots that apply to it.  The realms themselves
  * are organized into a hierarchy, such that children inherit (some of)
@@ -699,16 +578,33 @@
 	spinlock_t inodes_with_caps_lock;
 };
 
-
-
-/*
- * calculate the number of pages a given length and offset map onto,
- * if we align the data.
- */
-static inline int calc_pages_for(u64 off, u64 len)
+static inline int default_congestion_kb(void)
 {
-	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
-		(off >> PAGE_CACHE_SHIFT);
+	int congestion_kb;
+
+	/*
+	 * Copied from NFS
+	 *
+	 * congestion size, scale with available memory.
+	 *
+	 *  64MB:    8192k
+	 * 128MB:   11585k
+	 * 256MB:   16384k
+	 * 512MB:   23170k
+	 *   1GB:   32768k
+	 *   2GB:   46340k
+	 *   4GB:   65536k
+	 *   8GB:   92681k
+	 *  16GB:  131072k
+	 *
+	 * This allows larger machines to have larger/more transfers.
+	 * Limit the default to 256M
+	 */
+	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	if (congestion_kb > 256*1024)
+		congestion_kb = 256*1024;
+
+	return congestion_kb;
 }
 
 
@@ -741,16 +637,6 @@
 			   ci_item)->writing;
 }
 
-
-/* super.c */
-extern struct kmem_cache *ceph_inode_cachep;
-extern struct kmem_cache *ceph_cap_cachep;
-extern struct kmem_cache *ceph_dentry_cachep;
-extern struct kmem_cache *ceph_file_cachep;
-
-extern const char *ceph_msg_type_name(int type);
-extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
-
 /* inode.c */
 extern const struct inode_operations ceph_file_iops;
 
@@ -857,12 +743,18 @@
 /* file.c */
 extern const struct file_operations ceph_file_fops;
 extern const struct address_space_operations ceph_aops;
+extern int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len);
+extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
 extern int ceph_open(struct inode *inode, struct file *file);
 extern struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
 				       struct nameidata *nd, int mode,
 				       int locked_dir);
 extern int ceph_release(struct inode *inode, struct file *filp);
-extern void ceph_release_page_vector(struct page **pages, int num_pages);
 
 /* dir.c */
 extern const struct file_operations ceph_dir_fops;
@@ -892,12 +784,6 @@
 /* export.c */
 extern const struct export_operations ceph_export_ops;
 
-/* debugfs.c */
-extern int ceph_debugfs_init(void);
-extern void ceph_debugfs_cleanup(void);
-extern int ceph_debugfs_client_init(struct ceph_client *client);
-extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
-
 /* locks.c */
 extern int ceph_lock(struct file *file, int cmd, struct file_lock *fl);
 extern int ceph_flock(struct file *file, int cmd, struct file_lock *fl);
@@ -914,4 +800,8 @@
 	return NULL;
 }
 
+/* debugfs.c */
+extern int ceph_fs_debugfs_init(struct ceph_fs_client *client);
+extern void ceph_fs_debugfs_cleanup(struct ceph_fs_client *client);
+
 #endif /* _FS_CEPH_SUPER_H */
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 9578af6..6e12a6b 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -1,6 +1,9 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
+
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 #include <linux/xattr.h>
 #include <linux/slab.h>
@@ -620,12 +623,12 @@
 static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
 			      const char *value, size_t size, int flags)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
 	struct inode *inode = dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	struct ceph_mds_request *req;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	int err;
 	int i, nr_pages;
 	struct page **pages = NULL;
@@ -713,10 +716,9 @@
 
 	/* preallocate memory for xattr name, value, index node */
 	err = -ENOMEM;
-	newname = kmalloc(name_len + 1, GFP_NOFS);
+	newname = kmemdup(name, name_len + 1, GFP_NOFS);
 	if (!newname)
 		goto out;
-	memcpy(newname, name, name_len + 1);
 
 	if (val_len) {
 		newval = kmalloc(val_len + 1, GFP_NOFS);
@@ -777,8 +779,8 @@
 
 static int ceph_send_removexattr(struct dentry *dentry, const char *name)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = dentry->d_inode;
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	struct ceph_mds_request *req;
diff --git a/fs/exec.c b/fs/exec.c
index 828dd24..6d2b6f9 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -2014,3 +2014,43 @@
 fail:
 	return;
 }
+
+/*
+ * Core dumping helper functions.  These are the only things you should
+ * do on a core-file: use only these functions to write out all the
+ * necessary info.
+ */
+int dump_write(struct file *file, const void *addr, int nr)
+{
+	return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr;
+}
+EXPORT_SYMBOL(dump_write);
+
+int dump_seek(struct file *file, loff_t off)
+{
+	int ret = 1;
+
+	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
+		if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
+			return 0;
+	} else {
+		char *buf = (char *)get_zeroed_page(GFP_KERNEL);
+
+		if (!buf)
+			return 0;
+		while (off > 0) {
+			unsigned long n = off;
+
+			if (n > PAGE_SIZE)
+				n = PAGE_SIZE;
+			if (!dump_write(file, buf, n)) {
+				ret = 0;
+				break;
+			}
+			off -= n;
+		}
+		free_page((unsigned long)buf);
+	}
+	return ret;
+}
+EXPORT_SYMBOL(dump_seek);
diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index eb7368e..3eadd97 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -54,6 +54,9 @@
 	unsigned nr_pages;
 	unsigned long length;
 	loff_t pg_first; /* keep 64bit also in 32-arches */
+	bool read_4_write; /* This means two things: that the read is sync
+			    * And the pages should not be unlocked.
+			    */
 };
 
 static void _pcol_init(struct page_collect *pcol, unsigned expected_pages,
@@ -71,6 +74,7 @@
 	pcol->nr_pages = 0;
 	pcol->length = 0;
 	pcol->pg_first = -1;
+	pcol->read_4_write = false;
 }
 
 static void _pcol_reset(struct page_collect *pcol)
@@ -347,7 +351,8 @@
 		if (PageError(page))
 			ClearPageError(page);
 
-		unlock_page(page);
+		if (!pcol->read_4_write)
+			unlock_page(page);
 		EXOFS_DBGMSG("readpage_strip(0x%lx, 0x%lx) empty page,"
 			     " splitting\n", inode->i_ino, page->index);
 
@@ -428,6 +433,7 @@
 	/* readpage_strip might call read_exec(,is_sync==false) at several
 	 * places but not if we have a single page.
 	 */
+	pcol.read_4_write = is_sync;
 	ret = readpage_strip(&pcol, page);
 	if (ret) {
 		EXOFS_ERR("_readpage => %d\n", ret);
diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig
index cc96655..c465ae0 100644
--- a/fs/gfs2/Kconfig
+++ b/fs/gfs2/Kconfig
@@ -1,6 +1,6 @@
 config GFS2_FS
 	tristate "GFS2 file system support"
-	depends on EXPERIMENTAL && (64BIT || LBDAF)
+	depends on (64BIT || LBDAF)
 	select DLM if GFS2_FS_LOCKING_DLM
 	select CONFIGFS_FS if GFS2_FS_LOCKING_DLM
 	select SYSFS if GFS2_FS_LOCKING_DLM
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 194fe16..6b24afb 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -36,8 +36,8 @@
 #include "glops.h"
 
 
-static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page,
-				   unsigned int from, unsigned int to)
+void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page,
+			    unsigned int from, unsigned int to)
 {
 	struct buffer_head *head = page_buffers(page);
 	unsigned int bsize = head->b_size;
@@ -615,7 +615,7 @@
 	unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
 	int alloc_required;
 	int error = 0;
-	struct gfs2_alloc *al;
+	struct gfs2_alloc *al = NULL;
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
 	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
 	unsigned to = from + len;
@@ -663,6 +663,8 @@
 		rblocks += RES_STATFS + RES_QUOTA;
 	if (&ip->i_inode == sdp->sd_rindex)
 		rblocks += 2 * RES_STATFS;
+	if (alloc_required)
+		rblocks += gfs2_rg_blocks(al);
 
 	error = gfs2_trans_begin(sdp, rblocks,
 				 PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize);
@@ -696,13 +698,11 @@
 
 	page_cache_release(page);
 
-	/*
-	 * XXX(truncate): the call below should probably be replaced with
-	 * a call to the gfs2-specific truncate blocks helper to actually
-	 * release disk blocks..
-	 */
+	gfs2_trans_end(sdp);
 	if (pos + len > ip->i_inode.i_size)
-		truncate_setsize(&ip->i_inode, ip->i_inode.i_size);
+		gfs2_trim_blocks(&ip->i_inode);
+	goto out_trans_fail;
+
 out_endtrans:
 	gfs2_trans_end(sdp);
 out_trans_fail:
@@ -802,10 +802,8 @@
 	page_cache_release(page);
 
 	if (copied) {
-		if (inode->i_size < to) {
+		if (inode->i_size < to)
 			i_size_write(inode, to);
-			ip->i_disksize = inode->i_size;
-		}
 		gfs2_dinode_out(ip, di);
 		mark_inode_dirty(inode);
 	}
@@ -876,8 +874,6 @@
 
 	ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata);
 	if (ret > 0) {
-		if (inode->i_size > ip->i_disksize)
-			ip->i_disksize = inode->i_size;
 		gfs2_dinode_out(ip, dibh->b_data);
 		mark_inode_dirty(inode);
 	}
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 6f48280..5476c06 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -50,7 +50,7 @@
  * @ip: the inode
  * @dibh: the dinode buffer
  * @block: the block number that was allocated
- * @private: any locked page held by the caller process
+ * @page: The (optional) page. This is looked up if @page is NULL
  *
  * Returns: errno
  */
@@ -109,8 +109,7 @@
 /**
  * gfs2_unstuff_dinode - Unstuff a dinode when the data has grown too big
  * @ip: The GFS2 inode to unstuff
- * @unstuffer: the routine that handles unstuffing a non-zero length file
- * @private: private data for the unstuffer
+ * @page: The (optional) page. This is looked up if the @page is NULL
  *
  * This routine unstuffs a dinode and returns it to a "normal" state such
  * that the height can be grown in the traditional way.
@@ -132,7 +131,7 @@
 	if (error)
 		goto out;
 
-	if (ip->i_disksize) {
+	if (i_size_read(&ip->i_inode)) {
 		/* Get a free block, fill it with the stuffed data,
 		   and write it out to disk */
 
@@ -161,7 +160,7 @@
 	di = (struct gfs2_dinode *)dibh->b_data;
 	gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));
 
-	if (ip->i_disksize) {
+	if (i_size_read(&ip->i_inode)) {
 		*(__be64 *)(di + 1) = cpu_to_be64(block);
 		gfs2_add_inode_blocks(&ip->i_inode, 1);
 		di->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
@@ -885,83 +884,14 @@
 }
 
 /**
- * do_grow - Make a file look bigger than it is
- * @ip: the inode
- * @size: the size to set the file to
- *
- * Called with an exclusive lock on @ip.
- *
- * Returns: errno
- */
-
-static int do_grow(struct gfs2_inode *ip, u64 size)
-{
-	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
-	struct gfs2_alloc *al;
-	struct buffer_head *dibh;
-	int error;
-
-	al = gfs2_alloc_get(ip);
-	if (!al)
-		return -ENOMEM;
-
-	error = gfs2_quota_lock_check(ip);
-	if (error)
-		goto out;
-
-	al->al_requested = sdp->sd_max_height + RES_DATA;
-
-	error = gfs2_inplace_reserve(ip);
-	if (error)
-		goto out_gunlock_q;
-
-	error = gfs2_trans_begin(sdp,
-			sdp->sd_max_height + al->al_rgd->rd_length +
-			RES_JDATA + RES_DINODE + RES_STATFS + RES_QUOTA, 0);
-	if (error)
-		goto out_ipres;
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		goto out_end_trans;
-
-	if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) {
-		if (gfs2_is_stuffed(ip)) {
-			error = gfs2_unstuff_dinode(ip, NULL);
-			if (error)
-				goto out_brelse;
-		}
-	}
-
-	ip->i_disksize = size;
-	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
-	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-	gfs2_dinode_out(ip, dibh->b_data);
-
-out_brelse:
-	brelse(dibh);
-out_end_trans:
-	gfs2_trans_end(sdp);
-out_ipres:
-	gfs2_inplace_release(ip);
-out_gunlock_q:
-	gfs2_quota_unlock(ip);
-out:
-	gfs2_alloc_put(ip);
-	return error;
-}
-
-
-/**
  * gfs2_block_truncate_page - Deal with zeroing out data for truncate
  *
  * This is partly borrowed from ext3.
  */
-static int gfs2_block_truncate_page(struct address_space *mapping)
+static int gfs2_block_truncate_page(struct address_space *mapping, loff_t from)
 {
 	struct inode *inode = mapping->host;
 	struct gfs2_inode *ip = GFS2_I(inode);
-	loff_t from = inode->i_size;
 	unsigned long index = from >> PAGE_CACHE_SHIFT;
 	unsigned offset = from & (PAGE_CACHE_SIZE-1);
 	unsigned blocksize, iblock, length, pos;
@@ -1023,9 +953,11 @@
 	return err;
 }
 
-static int trunc_start(struct gfs2_inode *ip, u64 size)
+static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize)
 {
-	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	struct address_space *mapping = inode->i_mapping;
 	struct buffer_head *dibh;
 	int journaled = gfs2_is_jdata(ip);
 	int error;
@@ -1039,31 +971,26 @@
 	if (error)
 		goto out;
 
-	if (gfs2_is_stuffed(ip)) {
-		u64 dsize = size + sizeof(struct gfs2_dinode);
-		ip->i_disksize = size;
-		ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
-		gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-		gfs2_dinode_out(ip, dibh->b_data);
-		if (dsize > dibh->b_size)
-			dsize = dibh->b_size;
-		gfs2_buffer_clear_tail(dibh, dsize);
-		error = 1;
-	} else {
-		if (size & (u64)(sdp->sd_sb.sb_bsize - 1))
-			error = gfs2_block_truncate_page(ip->i_inode.i_mapping);
+	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
 
-		if (!error) {
-			ip->i_disksize = size;
-			ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
-			ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
-			gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-			gfs2_dinode_out(ip, dibh->b_data);
+	if (gfs2_is_stuffed(ip)) {
+		gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + newsize);
+	} else {
+		if (newsize & (u64)(sdp->sd_sb.sb_bsize - 1)) {
+			error = gfs2_block_truncate_page(mapping, newsize);
+			if (error)
+				goto out_brelse;
 		}
+		ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
 	}
 
-	brelse(dibh);
+	i_size_write(inode, newsize);
+	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
+	gfs2_dinode_out(ip, dibh->b_data);
 
+	truncate_pagecache(inode, oldsize, newsize);
+out_brelse:
+	brelse(dibh);
 out:
 	gfs2_trans_end(sdp);
 	return error;
@@ -1123,7 +1050,7 @@
 	if (error)
 		goto out;
 
-	if (!ip->i_disksize) {
+	if (!i_size_read(&ip->i_inode)) {
 		ip->i_height = 0;
 		ip->i_goal = ip->i_no_addr;
 		gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));
@@ -1143,92 +1070,154 @@
 
 /**
  * do_shrink - make a file smaller
- * @ip: the inode
- * @size: the size to make the file
- * @truncator: function to truncate the last partial block
+ * @inode: the inode
+ * @oldsize: the current inode size
+ * @newsize: the size to make the file
  *
- * Called with an exclusive lock on @ip.
+ * Called with an exclusive lock on @inode. The @size must
+ * be equal to or smaller than the current inode size.
  *
  * Returns: errno
  */
 
-static int do_shrink(struct gfs2_inode *ip, u64 size)
+static int do_shrink(struct inode *inode, u64 oldsize, u64 newsize)
 {
+	struct gfs2_inode *ip = GFS2_I(inode);
 	int error;
 
-	error = trunc_start(ip, size);
+	error = trunc_start(inode, oldsize, newsize);
 	if (error < 0)
 		return error;
-	if (error > 0)
+	if (gfs2_is_stuffed(ip))
 		return 0;
 
-	error = trunc_dealloc(ip, size);
-	if (!error)
+	error = trunc_dealloc(ip, newsize);
+	if (error == 0)
 		error = trunc_end(ip);
 
 	return error;
 }
 
-static int do_touch(struct gfs2_inode *ip, u64 size)
+void gfs2_trim_blocks(struct inode *inode)
 {
-	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	u64 size = inode->i_size;
+	int ret;
+
+	ret = do_shrink(inode, size, size);
+	WARN_ON(ret != 0);
+}
+
+/**
+ * do_grow - Touch and update inode size
+ * @inode: The inode
+ * @size: The new size
+ *
+ * This function updates the timestamps on the inode and
+ * may also increase the size of the inode. This function
+ * must not be called with @size any smaller than the current
+ * inode size.
+ *
+ * Although it is not strictly required to unstuff files here,
+ * earlier versions of GFS2 have a bug in the stuffed file reading
+ * code which will result in a buffer overrun if the size is larger
+ * than the max stuffed file size. In order to prevent this from
+ * occuring, such files are unstuffed, but in other cases we can
+ * just update the inode size directly.
+ *
+ * Returns: 0 on success, or -ve on error
+ */
+
+static int do_grow(struct inode *inode, u64 size)
+{
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct buffer_head *dibh;
+	struct gfs2_alloc *al = NULL;
 	int error;
 
-	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
-	if (error)
-		return error;
+	if (gfs2_is_stuffed(ip) &&
+	    (size > (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)))) {
+		al = gfs2_alloc_get(ip);
+		if (al == NULL)
+			return -ENOMEM;
 
-	down_write(&ip->i_rw_mutex);
+		error = gfs2_quota_lock_check(ip);
+		if (error)
+			goto do_grow_alloc_put;
+
+		al->al_requested = 1;
+		error = gfs2_inplace_reserve(ip);
+		if (error)
+			goto do_grow_qunlock;
+	}
+
+	error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT, 0);
+	if (error)
+		goto do_grow_release;
+
+	if (al) {
+		error = gfs2_unstuff_dinode(ip, NULL);
+		if (error)
+			goto do_end_trans;
+	}
 
 	error = gfs2_meta_inode_buffer(ip, &dibh);
 	if (error)
-		goto do_touch_out;
+		goto do_end_trans;
 
+	i_size_write(inode, size);
 	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
 	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
 	gfs2_dinode_out(ip, dibh->b_data);
 	brelse(dibh);
 
-do_touch_out:
-	up_write(&ip->i_rw_mutex);
+do_end_trans:
 	gfs2_trans_end(sdp);
+do_grow_release:
+	if (al) {
+		gfs2_inplace_release(ip);
+do_grow_qunlock:
+		gfs2_quota_unlock(ip);
+do_grow_alloc_put:
+		gfs2_alloc_put(ip);
+	}
 	return error;
 }
 
 /**
- * gfs2_truncatei - make a file a given size
- * @ip: the inode
- * @size: the size to make the file
- * @truncator: function to truncate the last partial block
+ * gfs2_setattr_size - make a file a given size
+ * @inode: the inode
+ * @newsize: the size to make the file
  *
- * The file size can grow, shrink, or stay the same size.
+ * The file size can grow, shrink, or stay the same size. This
+ * is called holding i_mutex and an exclusive glock on the inode
+ * in question.
  *
  * Returns: errno
  */
 
-int gfs2_truncatei(struct gfs2_inode *ip, u64 size)
+int gfs2_setattr_size(struct inode *inode, u64 newsize)
 {
-	int error;
+	int ret;
+	u64 oldsize;
 
-	if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_inode.i_mode)))
-		return -EINVAL;
+	BUG_ON(!S_ISREG(inode->i_mode));
 
-	if (size > ip->i_disksize)
-		error = do_grow(ip, size);
-	else if (size < ip->i_disksize)
-		error = do_shrink(ip, size);
-	else
-		/* update time stamps */
-		error = do_touch(ip, size);
+	ret = inode_newsize_ok(inode, newsize);
+	if (ret)
+		return ret;
 
-	return error;
+	oldsize = inode->i_size;
+	if (newsize >= oldsize)
+		return do_grow(inode, newsize);
+
+	return do_shrink(inode, oldsize, newsize);
 }
 
 int gfs2_truncatei_resume(struct gfs2_inode *ip)
 {
 	int error;
-	error = trunc_dealloc(ip, ip->i_disksize);
+	error = trunc_dealloc(ip, i_size_read(&ip->i_inode));
 	if (!error)
 		error = trunc_end(ip);
 	return error;
@@ -1269,7 +1258,7 @@
 
 	shift = sdp->sd_sb.sb_bsize_shift;
 	BUG_ON(gfs2_is_dir(ip));
-	end_of_file = (ip->i_disksize + sdp->sd_sb.sb_bsize - 1) >> shift;
+	end_of_file = (i_size_read(&ip->i_inode) + sdp->sd_sb.sb_bsize - 1) >> shift;
 	lblock = offset >> shift;
 	lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift;
 	if (lblock_stop > end_of_file)
diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h
index a20a521..42fea03 100644
--- a/fs/gfs2/bmap.h
+++ b/fs/gfs2/bmap.h
@@ -44,14 +44,16 @@
 	}
 }
 
-int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page);
-int gfs2_block_map(struct inode *inode, sector_t lblock, struct buffer_head *bh, int create);
-int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen);
-
-int gfs2_truncatei(struct gfs2_inode *ip, u64 size);
-int gfs2_truncatei_resume(struct gfs2_inode *ip);
-int gfs2_file_dealloc(struct gfs2_inode *ip);
-int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
-			      unsigned int len);
+extern int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page);
+extern int gfs2_block_map(struct inode *inode, sector_t lblock,
+			  struct buffer_head *bh, int create);
+extern int gfs2_extent_map(struct inode *inode, u64 lblock, int *new,
+			   u64 *dblock, unsigned *extlen);
+extern int gfs2_setattr_size(struct inode *inode, u64 size);
+extern void gfs2_trim_blocks(struct inode *inode);
+extern int gfs2_truncatei_resume(struct gfs2_inode *ip);
+extern int gfs2_file_dealloc(struct gfs2_inode *ip);
+extern int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
+				     unsigned int len);
 
 #endif /* __BMAP_DOT_H__ */
diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c
index bb7907b..6798755 100644
--- a/fs/gfs2/dentry.c
+++ b/fs/gfs2/dentry.c
@@ -49,7 +49,7 @@
 		ip = GFS2_I(inode);
 	}
 
-	if (sdp->sd_args.ar_localcaching)
+	if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
 		goto valid;
 
 	had_lock = (gfs2_glock_is_locked_by_me(dip->i_gl) != NULL);
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index b9dd88a..5c356d0 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -79,6 +79,9 @@
 #define gfs2_disk_hash2offset(h) (((u64)(h)) >> 1)
 #define gfs2_dir_offset2hash(p) ((u32)(((u64)(p)) << 1))
 
+struct qstr gfs2_qdot __read_mostly;
+struct qstr gfs2_qdotdot __read_mostly;
+
 typedef int (*leaf_call_t) (struct gfs2_inode *dip, u32 index, u32 len,
 			    u64 leaf_no, void *data);
 typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent,
@@ -127,8 +130,8 @@
 
 	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
 	memcpy(dibh->b_data + offset + sizeof(struct gfs2_dinode), buf, size);
-	if (ip->i_disksize < offset + size)
-		ip->i_disksize = offset + size;
+	if (ip->i_inode.i_size < offset + size)
+		i_size_write(&ip->i_inode, offset + size);
 	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
 	gfs2_dinode_out(ip, dibh->b_data);
 
@@ -225,8 +228,8 @@
 	if (error)
 		return error;
 
-	if (ip->i_disksize < offset + copied)
-		ip->i_disksize = offset + copied;
+	if (ip->i_inode.i_size < offset + copied)
+		i_size_write(&ip->i_inode, offset + copied);
 	ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
 
 	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
@@ -275,12 +278,13 @@
 	unsigned int o;
 	int copied = 0;
 	int error = 0;
+	u64 disksize = i_size_read(&ip->i_inode);
 
-	if (offset >= ip->i_disksize)
+	if (offset >= disksize)
 		return 0;
 
-	if (offset + size > ip->i_disksize)
-		size = ip->i_disksize - offset;
+	if (offset + size > disksize)
+		size = disksize - offset;
 
 	if (!size)
 		return 0;
@@ -727,7 +731,7 @@
 		unsigned hsize = 1 << ip->i_depth;
 		unsigned index;
 		u64 ln;
-		if (hsize * sizeof(u64) != ip->i_disksize) {
+		if (hsize * sizeof(u64) != i_size_read(inode)) {
 			gfs2_consist_inode(ip);
 			return ERR_PTR(-EIO);
 		}
@@ -879,7 +883,7 @@
 	for (x = sdp->sd_hash_ptrs; x--; lp++)
 		*lp = cpu_to_be64(bn);
 
-	dip->i_disksize = sdp->sd_sb.sb_bsize / 2;
+	i_size_write(inode, sdp->sd_sb.sb_bsize / 2);
 	gfs2_add_inode_blocks(&dip->i_inode, 1);
 	dip->i_diskflags |= GFS2_DIF_EXHASH;
 
@@ -1057,11 +1061,12 @@
 	u64 *buf;
 	u64 *from, *to;
 	u64 block;
+	u64 disksize = i_size_read(&dip->i_inode);
 	int x;
 	int error = 0;
 
 	hsize = 1 << dip->i_depth;
-	if (hsize * sizeof(u64) != dip->i_disksize) {
+	if (hsize * sizeof(u64) != disksize) {
 		gfs2_consist_inode(dip);
 		return -EIO;
 	}
@@ -1072,7 +1077,7 @@
 	if (!buf)
 		return -ENOMEM;
 
-	for (block = dip->i_disksize >> sdp->sd_hash_bsize_shift; block--;) {
+	for (block = disksize >> sdp->sd_hash_bsize_shift; block--;) {
 		error = gfs2_dir_read_data(dip, (char *)buf,
 					    block * sdp->sd_hash_bsize,
 					    sdp->sd_hash_bsize, 1);
@@ -1370,7 +1375,7 @@
 	unsigned depth = 0;
 
 	hsize = 1 << dip->i_depth;
-	if (hsize * sizeof(u64) != dip->i_disksize) {
+	if (hsize * sizeof(u64) != i_size_read(inode)) {
 		gfs2_consist_inode(dip);
 		return -EIO;
 	}
@@ -1784,7 +1789,7 @@
 	int error = 0;
 
 	hsize = 1 << dip->i_depth;
-	if (hsize * sizeof(u64) != dip->i_disksize) {
+	if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) {
 		gfs2_consist_inode(dip);
 		return -EIO;
 	}
diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h
index 4f91944..a98f644 100644
--- a/fs/gfs2/dir.h
+++ b/fs/gfs2/dir.h
@@ -17,23 +17,24 @@
 struct gfs2_inode;
 struct gfs2_inum;
 
-struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *filename);
-int gfs2_dir_check(struct inode *dir, const struct qstr *filename,
-		   const struct gfs2_inode *ip);
-int gfs2_dir_add(struct inode *inode, const struct qstr *filename,
-		 const struct gfs2_inode *ip, unsigned int type);
-int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename);
-int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque,
-		  filldir_t filldir);
-int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
-		   const struct gfs2_inode *nip, unsigned int new_type);
+extern struct inode *gfs2_dir_search(struct inode *dir,
+				     const struct qstr *filename);
+extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename,
+			  const struct gfs2_inode *ip);
+extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename,
+			const struct gfs2_inode *ip, unsigned int type);
+extern int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename);
+extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque,
+			 filldir_t filldir);
+extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
+			  const struct gfs2_inode *nip, unsigned int new_type);
 
-int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip);
+extern int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip);
 
-int gfs2_diradd_alloc_required(struct inode *dir,
-			       const struct qstr *filename);
-int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block,
-			    struct buffer_head **bhp);
+extern int gfs2_diradd_alloc_required(struct inode *dir,
+				      const struct qstr *filename);
+extern int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block,
+				   struct buffer_head **bhp);
 
 static inline u32 gfs2_disk_hash(const char *data, int len)
 {
@@ -61,4 +62,7 @@
 	memcpy(dent + 1, name->name, name->len);
 }
 
+extern struct qstr gfs2_qdot;
+extern struct qstr gfs2_qdotdot;
+
 #endif /* __DIR_DOT_H__ */
diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c
index dfe237a..06d5827 100644
--- a/fs/gfs2/export.c
+++ b/fs/gfs2/export.c
@@ -126,16 +126,9 @@
 
 static struct dentry *gfs2_get_parent(struct dentry *child)
 {
-	struct qstr dotdot;
 	struct dentry *dentry;
 
-	/*
-	 * XXX(hch): it would be a good idea to keep this around as a
-	 *	     static variable.
-	 */
-	gfs2_str2qstr(&dotdot, "..");
-
-	dentry = d_obtain_alias(gfs2_lookupi(child->d_inode, &dotdot, 1));
+	dentry = d_obtain_alias(gfs2_lookupi(child->d_inode, &gfs2_qdotdot, 1));
 	if (!IS_ERR(dentry))
 		dentry->d_op = &gfs2_dops;
 	return dentry;
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 4edd662..237ee6a 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -382,8 +382,10 @@
 	rblocks = RES_DINODE + ind_blocks;
 	if (gfs2_is_jdata(ip))
 		rblocks += data_blocks ? data_blocks : 1;
-	if (ind_blocks || data_blocks)
+	if (ind_blocks || data_blocks) {
 		rblocks += RES_STATFS + RES_QUOTA;
+		rblocks += gfs2_rg_blocks(al);
+	}
 	ret = gfs2_trans_begin(sdp, rblocks, 0);
 	if (ret)
 		goto out_trans_fail;
@@ -491,7 +493,7 @@
 			goto fail;
 
 		if (!(file->f_flags & O_LARGEFILE) &&
-		    ip->i_disksize > MAX_NON_LFS) {
+		    i_size_read(inode) > MAX_NON_LFS) {
 			error = -EOVERFLOW;
 			goto fail_gunlock;
 		}
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 9adf8f9..8777885 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -441,6 +441,8 @@
 		else
 			gfs2_glock_put_nolock(gl);
 	}
+	if (held1 && held2 && list_empty(&gl->gl_holders))
+		clear_bit(GLF_QUEUED, &gl->gl_flags);
 
 	gl->gl_state = new_state;
 	gl->gl_tchange = jiffies;
@@ -1012,6 +1014,7 @@
 		if (unlikely((gh->gh_flags & LM_FLAG_PRIORITY) && !insert_pt))
 			insert_pt = &gh2->gh_list;
 	}
+	set_bit(GLF_QUEUED, &gl->gl_flags);
 	if (likely(insert_pt == NULL)) {
 		list_add_tail(&gh->gh_list, &gl->gl_holders);
 		if (unlikely(gh->gh_flags & LM_FLAG_PRIORITY))
@@ -1310,10 +1313,12 @@
 
 	gfs2_glock_hold(gl);
 	holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time;
-	if (time_before(now, holdtime))
-		delay = holdtime - now;
-	if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags))
-		delay = gl->gl_ops->go_min_hold_time;
+	if (test_bit(GLF_QUEUED, &gl->gl_flags)) {
+		if (time_before(now, holdtime))
+			delay = holdtime - now;
+		if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags))
+			delay = gl->gl_ops->go_min_hold_time;
+	}
 
 	spin_lock(&gl->gl_spin);
 	handle_callback(gl, state, delay);
@@ -1512,7 +1517,7 @@
 	spin_unlock(&lru_lock);
 
 	spin_lock(&gl->gl_spin);
-	if (find_first_holder(gl) == NULL && gl->gl_state != LM_ST_UNLOCKED)
+	if (gl->gl_state != LM_ST_UNLOCKED)
 		handle_callback(gl, LM_ST_UNLOCKED, 0);
 	spin_unlock(&gl->gl_spin);
 	gfs2_glock_hold(gl);
@@ -1660,6 +1665,8 @@
 		*p++ = 'I';
 	if (test_bit(GLF_FROZEN, gflags))
 		*p++ = 'F';
+	if (test_bit(GLF_QUEUED, gflags))
+		*p++ = 'q';
 	*p = 0;
 	return buf;
 }
@@ -1776,10 +1783,12 @@
 	}
 #endif
 
-	glock_workqueue = create_workqueue("glock_workqueue");
+	glock_workqueue = alloc_workqueue("glock_workqueue", WQ_RESCUER |
+					  WQ_HIGHPRI | WQ_FREEZEABLE, 0);
 	if (IS_ERR(glock_workqueue))
 		return PTR_ERR(glock_workqueue);
-	gfs2_delete_workqueue = create_workqueue("delete_workqueue");
+	gfs2_delete_workqueue = alloc_workqueue("delete_workqueue", WQ_RESCUER |
+						WQ_FREEZEABLE, 0);
 	if (IS_ERR(gfs2_delete_workqueue)) {
 		destroy_workqueue(glock_workqueue);
 		return PTR_ERR(gfs2_delete_workqueue);
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index 2bda191..db1c26d 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -215,7 +215,7 @@
 void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...);
 
 /**
- * gfs2_glock_nq_init - intialize a holder and enqueue it on a glock
+ * gfs2_glock_nq_init - initialize a holder and enqueue it on a glock
  * @gl: the glock
  * @state: the state we're requesting
  * @flags: the modifier flags
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 49f97d3..0d149dc 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -262,13 +262,12 @@
 	const struct gfs2_inode *ip = gl->gl_object;
 	if (ip == NULL)
 		return 0;
-	gfs2_print_dbg(seq, " I: n:%llu/%llu t:%u f:0x%02lx d:0x%08x s:%llu/%llu\n",
+	gfs2_print_dbg(seq, " I: n:%llu/%llu t:%u f:0x%02lx d:0x%08x s:%llu\n",
 		  (unsigned long long)ip->i_no_formal_ino,
 		  (unsigned long long)ip->i_no_addr,
 		  IF2DT(ip->i_inode.i_mode), ip->i_flags,
 		  (unsigned int)ip->i_diskflags,
-		  (unsigned long long)ip->i_inode.i_size,
-		  (unsigned long long)ip->i_disksize);
+		  (unsigned long long)i_size_read(&ip->i_inode));
 	return 0;
 }
 
@@ -453,7 +452,6 @@
 	[LM_TYPE_META] = &gfs2_meta_glops,
 	[LM_TYPE_INODE] = &gfs2_inode_glops,
 	[LM_TYPE_RGRP] = &gfs2_rgrp_glops,
-	[LM_TYPE_NONDISK] = &gfs2_trans_glops,
 	[LM_TYPE_IOPEN] = &gfs2_iopen_glops,
 	[LM_TYPE_FLOCK] = &gfs2_flock_glops,
 	[LM_TYPE_NONDISK] = &gfs2_nondisk_glops,
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index fdbf4b3..764fbb4 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -196,6 +196,7 @@
 	GLF_REPLY_PENDING		= 9,
 	GLF_INITIAL			= 10,
 	GLF_FROZEN			= 11,
+	GLF_QUEUED			= 12,
 };
 
 struct gfs2_glock {
@@ -267,7 +268,6 @@
 	u64 i_no_formal_ino;
 	u64 i_generation;
 	u64 i_eattr;
-	loff_t i_disksize;
 	unsigned long i_flags;		/* GIF_... */
 	struct gfs2_glock *i_gl; /* Move into i_gh? */
 	struct gfs2_holder i_iopen_gh;
@@ -416,11 +416,8 @@
 	char ar_locktable[GFS2_LOCKNAME_LEN];	/* Name of the Lock Table */
 	char ar_hostdata[GFS2_LOCKNAME_LEN];	/* Host specific data */
 	unsigned int ar_spectator:1;		/* Don't get a journal */
-	unsigned int ar_ignore_local_fs:1;	/* Ignore optimisations */
 	unsigned int ar_localflocks:1;		/* Let the VFS do flock|fcntl */
-	unsigned int ar_localcaching:1;		/* Local caching */
 	unsigned int ar_debug:1;		/* Oops on errors */
-	unsigned int ar_upgrade:1;		/* Upgrade ondisk format */
 	unsigned int ar_posix_acl:1;		/* Enable posix acls */
 	unsigned int ar_quota:2;		/* off/account/on */
 	unsigned int ar_suiddir:1;		/* suiddir support */
@@ -497,7 +494,7 @@
  */
 
 struct lm_lockstruct {
-	unsigned int ls_jid;
+	int ls_jid;
 	unsigned int ls_first;
 	unsigned int ls_first_done;
 	unsigned int ls_nodir;
@@ -572,6 +569,7 @@
 	struct list_head sd_rindex_mru_list;
 	struct gfs2_rgrpd *sd_rindex_forward;
 	unsigned int sd_rgrps;
+	unsigned int sd_max_rg_data;
 
 	/* Journal index stuff */
 
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 08140f1..06370f8 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -359,8 +359,7 @@
 	 * to do that.
 	 */
 	ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink);
-	ip->i_disksize = be64_to_cpu(str->di_size);
-	i_size_write(&ip->i_inode, ip->i_disksize);
+	i_size_write(&ip->i_inode, be64_to_cpu(str->di_size));
 	gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks));
 	atime.tv_sec = be64_to_cpu(str->di_atime);
 	atime.tv_nsec = be32_to_cpu(str->di_atime_nsec);
@@ -1055,7 +1054,7 @@
 	str->di_uid = cpu_to_be32(ip->i_inode.i_uid);
 	str->di_gid = cpu_to_be32(ip->i_inode.i_gid);
 	str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink);
-	str->di_size = cpu_to_be64(ip->i_disksize);
+	str->di_size = cpu_to_be64(i_size_read(&ip->i_inode));
 	str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
 	str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec);
 	str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec);
@@ -1085,8 +1084,8 @@
 	       (unsigned long long)ip->i_no_formal_ino);
 	printk(KERN_INFO "  no_addr = %llu\n",
 	       (unsigned long long)ip->i_no_addr);
-	printk(KERN_INFO "  i_disksize = %llu\n",
-	       (unsigned long long)ip->i_disksize);
+	printk(KERN_INFO "  i_size = %llu\n",
+	       (unsigned long long)i_size_read(&ip->i_inode));
 	printk(KERN_INFO "  blocks = %llu\n",
 	       (unsigned long long)gfs2_get_inode_blocks(&ip->i_inode));
 	printk(KERN_INFO "  i_goal = %llu\n",
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 300ada3..6720d7d 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -19,6 +19,8 @@
 extern int gfs2_internal_read(struct gfs2_inode *ip,
 			      struct file_ra_state *ra_state,
 			      char *buf, loff_t *pos, unsigned size);
+extern void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page,
+				   unsigned int from, unsigned int to);
 extern void gfs2_set_aops(struct inode *inode);
 
 static inline int gfs2_is_stuffed(const struct gfs2_inode *ip)
@@ -80,6 +82,19 @@
 	dent->de_inum.no_addr = cpu_to_be64(ip->i_no_addr);
 }
 
+static inline int gfs2_check_internal_file_size(struct inode *inode,
+						u64 minsize, u64 maxsize)
+{
+	u64 size = i_size_read(inode);
+	if (size < minsize || size > maxsize)
+		goto err;
+	if (size & ((1 << inode->i_blkbits) - 1))
+		goto err;
+	return 0;
+err:
+	gfs2_consist_inode(GFS2_I(inode));
+	return -EIO;
+}
 
 extern void gfs2_set_iop(struct inode *inode);
 extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
index 0e0470e..1c09425 100644
--- a/fs/gfs2/lock_dlm.c
+++ b/fs/gfs2/lock_dlm.c
@@ -42,9 +42,9 @@
 		ret |= LM_OUT_CANCELED;
 		goto out;
 	case -EAGAIN: /* Try lock fails */
+	case -EDEADLK: /* Deadlock detected */
 		goto out;
-	case -EINVAL: /* Invalid */
-	case -ENOMEM: /* Out of memory */
+	case -ETIMEDOUT: /* Canceled due to timeout */
 		ret |= LM_OUT_ERROR;
 		goto out;
 	case 0: /* Success */
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index b1e9630..d7eb1e2 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -24,6 +24,7 @@
 #include "glock.h"
 #include "quota.h"
 #include "recovery.h"
+#include "dir.h"
 
 static struct shrinker qd_shrinker = {
 	.shrink = gfs2_shrink_qd_memory,
@@ -78,6 +79,9 @@
 {
 	int error;
 
+	gfs2_str2qstr(&gfs2_qdot, ".");
+	gfs2_str2qstr(&gfs2_qdotdot, "..");
+
 	error = gfs2_sys_init();
 	if (error)
 		return error;
@@ -140,7 +144,7 @@
 
 	error = -ENOMEM;
 	gfs_recovery_wq = alloc_workqueue("gfs_recovery",
-					  WQ_NON_REENTRANT | WQ_RESCUER, 0);
+					  WQ_RESCUER | WQ_FREEZEABLE, 0);
 	if (!gfs_recovery_wq)
 		goto fail_wq;
 
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 4d4b1e8..aeafc23 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -38,14 +38,6 @@
 #define DO 0
 #define UNDO 1
 
-static const u32 gfs2_old_fs_formats[] = {
-        0
-};
-
-static const u32 gfs2_old_multihost_formats[] = {
-        0
-};
-
 /**
  * gfs2_tune_init - Fill a gfs2_tune structure with default values
  * @gt: tune
@@ -135,8 +127,6 @@
 
 static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent)
 {
-	unsigned int x;
-
 	if (sb->sb_magic != GFS2_MAGIC ||
 	    sb->sb_type != GFS2_METATYPE_SB) {
 		if (!silent)
@@ -150,55 +140,9 @@
 	    sb->sb_multihost_format == GFS2_FORMAT_MULTI)
 		return 0;
 
-	if (sb->sb_fs_format != GFS2_FORMAT_FS) {
-		for (x = 0; gfs2_old_fs_formats[x]; x++)
-			if (gfs2_old_fs_formats[x] == sb->sb_fs_format)
-				break;
+	fs_warn(sdp, "Unknown on-disk format, unable to mount\n");
 
-		if (!gfs2_old_fs_formats[x]) {
-			printk(KERN_WARNING
-			       "GFS2: code version (%u, %u) is incompatible "
-			       "with ondisk format (%u, %u)\n",
-			       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
-			       sb->sb_fs_format, sb->sb_multihost_format);
-			printk(KERN_WARNING
-			       "GFS2: I don't know how to upgrade this FS\n");
-			return -EINVAL;
-		}
-	}
-
-	if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) {
-		for (x = 0; gfs2_old_multihost_formats[x]; x++)
-			if (gfs2_old_multihost_formats[x] ==
-			    sb->sb_multihost_format)
-				break;
-
-		if (!gfs2_old_multihost_formats[x]) {
-			printk(KERN_WARNING
-			       "GFS2: code version (%u, %u) is incompatible "
-			       "with ondisk format (%u, %u)\n",
-			       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
-			       sb->sb_fs_format, sb->sb_multihost_format);
-			printk(KERN_WARNING
-			       "GFS2: I don't know how to upgrade this FS\n");
-			return -EINVAL;
-		}
-	}
-
-	if (!sdp->sd_args.ar_upgrade) {
-		printk(KERN_WARNING
-		       "GFS2: code version (%u, %u) is incompatible "
-		       "with ondisk format (%u, %u)\n",
-		       GFS2_FORMAT_FS, GFS2_FORMAT_MULTI,
-		       sb->sb_fs_format, sb->sb_multihost_format);
-		printk(KERN_INFO
-		       "GFS2: Use the \"upgrade\" mount option to upgrade "
-		       "the FS\n");
-		printk(KERN_INFO "GFS2: See the manual for more details\n");
-		return -EINVAL;
-	}
-
-	return 0;
+	return -EINVAL;
 }
 
 static void end_bio_io_page(struct bio *bio, int error)
@@ -586,7 +530,7 @@
 
 	prev_db = 0;
 
-	for (lb = 0; lb < ip->i_disksize >> sdp->sd_sb.sb_bsize_shift; lb++) {
+	for (lb = 0; lb < i_size_read(jd->jd_inode) >> sdp->sd_sb.sb_bsize_shift; lb++) {
 		bh.b_state = 0;
 		bh.b_blocknr = 0;
 		bh.b_size = 1 << ip->i_inode.i_blkbits;
@@ -1022,7 +966,6 @@
 	if (!strcmp("lock_nolock", proto)) {
 		lm = &nolock_ops;
 		sdp->sd_args.ar_localflocks = 1;
-		sdp->sd_args.ar_localcaching = 1;
 #ifdef CONFIG_GFS2_FS_LOCKING_DLM
 	} else if (!strcmp("lock_dlm", proto)) {
 		lm = &gfs2_dlm_ops;
@@ -1113,8 +1056,6 @@
 
 static int wait_on_journal(struct gfs2_sbd *sdp)
 {
-	if (sdp->sd_args.ar_spectator)
-		return 0;
 	if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
 		return 0;
 
@@ -1217,6 +1158,20 @@
 	if (error)
 		goto fail_sb;
 
+	/*
+	 * If user space has failed to join the cluster or some similar
+	 * failure has occurred, then the journal id will contain a
+	 * negative (error) number. This will then be returned to the
+	 * caller (of the mount syscall). We do this even for spectator
+	 * mounts (which just write a jid of 0 to indicate "ok" even though
+	 * the jid is unused in the spectator case)
+	 */
+	if (sdp->sd_lockstruct.ls_jid < 0) {
+		error = sdp->sd_lockstruct.ls_jid;
+		sdp->sd_lockstruct.ls_jid = 0;
+		goto fail_sb;
+	}
+
 	error = init_inodes(sdp, DO);
 	if (error)
 		goto fail_sb;
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 1009be2..0534510 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -18,6 +18,8 @@
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
 #include <linux/fiemap.h>
+#include <linux/swap.h>
+#include <linux/falloc.h>
 #include <asm/uaccess.h>
 
 #include "gfs2.h"
@@ -217,7 +219,7 @@
 			goto out_gunlock_q;
 
 		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 al->al_rgd->rd_length +
+					 gfs2_rg_blocks(al) +
 					 2 * RES_DINODE + RES_STATFS +
 					 RES_QUOTA, 0);
 		if (error)
@@ -406,7 +408,6 @@
 
 	ip = ghs[1].gh_gl->gl_object;
 
-	ip->i_disksize = size;
 	i_size_write(inode, size);
 
 	error = gfs2_meta_inode_buffer(ip, &dibh);
@@ -461,7 +462,7 @@
 	ip = ghs[1].gh_gl->gl_object;
 
 	ip->i_inode.i_nlink = 2;
-	ip->i_disksize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode);
+	i_size_write(inode, sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode));
 	ip->i_diskflags |= GFS2_DIF_JDATA;
 	ip->i_entries = 2;
 
@@ -470,18 +471,15 @@
 	if (!gfs2_assert_withdraw(sdp, !error)) {
 		struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data;
 		struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1);
-		struct qstr str;
 
-		gfs2_str2qstr(&str, ".");
 		gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-		gfs2_qstr2dirent(&str, GFS2_DIRENT_SIZE(str.len), dent);
+		gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent);
 		dent->de_inum = di->di_num; /* already GFS2 endian */
 		dent->de_type = cpu_to_be16(DT_DIR);
 		di->di_entries = cpu_to_be32(1);
 
-		gfs2_str2qstr(&str, "..");
 		dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1));
-		gfs2_qstr2dirent(&str, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent);
+		gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent);
 
 		gfs2_inum_out(dip, dent);
 		dent->de_type = cpu_to_be16(DT_DIR);
@@ -522,7 +520,6 @@
 static int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
 		       struct gfs2_inode *ip)
 {
-	struct qstr dotname;
 	int error;
 
 	if (ip->i_entries != 2) {
@@ -539,13 +536,11 @@
 	if (error)
 		return error;
 
-	gfs2_str2qstr(&dotname, ".");
-	error = gfs2_dir_del(ip, &dotname);
+	error = gfs2_dir_del(ip, &gfs2_qdot);
 	if (error)
 		return error;
 
-	gfs2_str2qstr(&dotname, "..");
-	error = gfs2_dir_del(ip, &dotname);
+	error = gfs2_dir_del(ip, &gfs2_qdotdot);
 	if (error)
 		return error;
 
@@ -694,11 +689,8 @@
 	struct inode *dir = &to->i_inode;
 	struct super_block *sb = dir->i_sb;
 	struct inode *tmp;
-	struct qstr dotdot;
 	int error = 0;
 
-	gfs2_str2qstr(&dotdot, "..");
-
 	igrab(dir);
 
 	for (;;) {
@@ -711,7 +703,7 @@
 			break;
 		}
 
-		tmp = gfs2_lookupi(dir, &dotdot, 1);
+		tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1);
 		if (IS_ERR(tmp)) {
 			error = PTR_ERR(tmp);
 			break;
@@ -744,7 +736,7 @@
 	struct gfs2_inode *ip = GFS2_I(odentry->d_inode);
 	struct gfs2_inode *nip = NULL;
 	struct gfs2_sbd *sdp = GFS2_SB(odir);
-	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, };
+	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh;
 	struct gfs2_rgrpd *nrgd;
 	unsigned int num_gh;
 	int dir_rename = 0;
@@ -758,6 +750,9 @@
 			return 0;
 	}
 
+	error = gfs2_rindex_hold(sdp, &ri_gh);
+	if (error)
+		return error;
 
 	if (odip != ndip) {
 		error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE,
@@ -887,12 +882,12 @@
 
 		al->al_requested = sdp->sd_max_dirres;
 
-		error = gfs2_inplace_reserve(ndip);
+		error = gfs2_inplace_reserve_ri(ndip);
 		if (error)
 			goto out_gunlock_q;
 
 		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 al->al_rgd->rd_length +
+					 gfs2_rg_blocks(al) +
 					 4 * RES_DINODE + 4 * RES_LEAF +
 					 RES_STATFS + RES_QUOTA + 4, 0);
 		if (error)
@@ -920,9 +915,6 @@
 	}
 
 	if (dir_rename) {
-		struct qstr name;
-		gfs2_str2qstr(&name, "..");
-
 		error = gfs2_change_nlink(ndip, +1);
 		if (error)
 			goto out_end_trans;
@@ -930,7 +922,7 @@
 		if (error)
 			goto out_end_trans;
 
-		error = gfs2_dir_mvino(ip, &name, ndip, DT_DIR);
+		error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
 		if (error)
 			goto out_end_trans;
 	} else {
@@ -972,6 +964,7 @@
 	if (r_gh.gh_gl)
 		gfs2_glock_dq_uninit(&r_gh);
 out:
+	gfs2_glock_dq_uninit(&ri_gh);
 	return error;
 }
 
@@ -990,7 +983,7 @@
 	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
 	struct gfs2_holder i_gh;
 	struct buffer_head *dibh;
-	unsigned int x;
+	unsigned int x, size;
 	char *buf;
 	int error;
 
@@ -1002,7 +995,8 @@
 		return NULL;
 	}
 
-	if (!ip->i_disksize) {
+	size = (unsigned int)i_size_read(&ip->i_inode);
+	if (size == 0) {
 		gfs2_consist_inode(ip);
 		buf = ERR_PTR(-EIO);
 		goto out;
@@ -1014,7 +1008,7 @@
 		goto out;
 	}
 
-	x = ip->i_disksize + 1;
+	x = size + 1;
 	buf = kmalloc(x, GFP_NOFS);
 	if (!buf)
 		buf = ERR_PTR(-ENOMEM);
@@ -1071,30 +1065,6 @@
 	return error;
 }
 
-/*
- * XXX(truncate): the truncate_setsize calls should be moved to the end.
- */
-static int setattr_size(struct inode *inode, struct iattr *attr)
-{
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
-	int error;
-
-	if (attr->ia_size != ip->i_disksize) {
-		error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks);
-		if (error)
-			return error;
-		truncate_setsize(inode, attr->ia_size);
-		gfs2_trans_end(sdp);
-	}
-
-	error = gfs2_truncatei(ip, attr->ia_size);
-	if (error && (inode->i_size != ip->i_disksize))
-		i_size_write(inode, ip->i_disksize);
-
-	return error;
-}
-
 static int setattr_chown(struct inode *inode, struct iattr *attr)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
@@ -1195,7 +1165,7 @@
 		goto out;
 
 	if (attr->ia_valid & ATTR_SIZE)
-		error = setattr_size(inode, attr);
+		error = gfs2_setattr_size(inode, attr->ia_size);
 	else if (attr->ia_valid & (ATTR_UID | ATTR_GID))
 		error = setattr_chown(inode, attr);
 	else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode))
@@ -1301,6 +1271,257 @@
 	return ret;
 }
 
+static void empty_write_end(struct page *page, unsigned from,
+			   unsigned to)
+{
+	struct gfs2_inode *ip = GFS2_I(page->mapping->host);
+
+	page_zero_new_buffers(page, from, to);
+	flush_dcache_page(page);
+	mark_page_accessed(page);
+
+	if (!gfs2_is_writeback(ip))
+		gfs2_page_add_databufs(ip, page, from, to);
+
+	block_commit_write(page, from, to);
+}
+
+
+static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
+{
+	unsigned start, end, next;
+	struct buffer_head *bh, *head;
+	int error;
+
+	if (!page_has_buffers(page)) {
+		error = block_prepare_write(page, from, to, gfs2_block_map);
+		if (unlikely(error))
+			return error;
+
+		empty_write_end(page, from, to);
+		return 0;
+	}
+
+	bh = head = page_buffers(page);
+	next = end = 0;
+	while (next < from) {
+		next += bh->b_size;
+		bh = bh->b_this_page;
+	}
+	start = next;
+	do {
+		next += bh->b_size;
+		if (buffer_mapped(bh)) {
+			if (end) {
+				error = block_prepare_write(page, start, end,
+							    gfs2_block_map);
+				if (unlikely(error))
+					return error;
+				empty_write_end(page, start, end);
+				end = 0;
+			}
+			start = next;
+		}
+		else
+			end = next;
+		bh = bh->b_this_page;
+	} while (next < to);
+
+	if (end) {
+		error = block_prepare_write(page, start, end, gfs2_block_map);
+		if (unlikely(error))
+			return error;
+		empty_write_end(page, start, end);
+	}
+
+	return 0;
+}
+
+static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
+			   int mode)
+{
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct buffer_head *dibh;
+	int error;
+	u64 start = offset >> PAGE_CACHE_SHIFT;
+	unsigned int start_offset = offset & ~PAGE_CACHE_MASK;
+	u64 end = (offset + len - 1) >> PAGE_CACHE_SHIFT;
+	pgoff_t curr;
+	struct page *page;
+	unsigned int end_offset = (offset + len) & ~PAGE_CACHE_MASK;
+	unsigned int from, to;
+
+	if (!end_offset)
+		end_offset = PAGE_CACHE_SIZE;
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (unlikely(error))
+		goto out;
+
+	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+
+	if (gfs2_is_stuffed(ip)) {
+		error = gfs2_unstuff_dinode(ip, NULL);
+		if (unlikely(error))
+			goto out;
+	}
+
+	curr = start;
+	offset = start << PAGE_CACHE_SHIFT;
+	from = start_offset;
+	to = PAGE_CACHE_SIZE;
+	while (curr <= end) {
+		page = grab_cache_page_write_begin(inode->i_mapping, curr,
+						   AOP_FLAG_NOFS);
+		if (unlikely(!page)) {
+			error = -ENOMEM;
+			goto out;
+		}
+
+		if (curr == end)
+			to = end_offset;
+		error = write_empty_blocks(page, from, to);
+		if (!error && offset + to > inode->i_size &&
+		    !(mode & FALLOC_FL_KEEP_SIZE)) {
+			i_size_write(inode, offset + to);
+		}
+		unlock_page(page);
+		page_cache_release(page);
+		if (error)
+			goto out;
+		curr++;
+		offset += PAGE_CACHE_SIZE;
+		from = 0;
+	}
+
+	gfs2_dinode_out(ip, dibh->b_data);
+	mark_inode_dirty(inode);
+
+	brelse(dibh);
+
+out:
+	return error;
+}
+
+static void calc_max_reserv(struct gfs2_inode *ip, loff_t max, loff_t *len,
+			    unsigned int *data_blocks, unsigned int *ind_blocks)
+{
+	const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	unsigned int max_blocks = ip->i_alloc->al_rgd->rd_free_clone;
+	unsigned int tmp, max_data = max_blocks - 3 * (sdp->sd_max_height - 1);
+
+	for (tmp = max_data; tmp > sdp->sd_diptrs;) {
+		tmp = DIV_ROUND_UP(tmp, sdp->sd_inptrs);
+		max_data -= tmp;
+	}
+	/* This calculation isn't the exact reverse of gfs2_write_calc_reserve,
+	   so it might end up with fewer data blocks */
+	if (max_data <= *data_blocks)
+		return;
+	*data_blocks = max_data;
+	*ind_blocks = max_blocks - max_data;
+	*len = ((loff_t)max_data - 3) << sdp->sd_sb.sb_bsize_shift;
+	if (*len > max) {
+		*len = max;
+		gfs2_write_calc_reserv(ip, max, data_blocks, ind_blocks);
+	}
+}
+
+static long gfs2_fallocate(struct inode *inode, int mode, loff_t offset,
+			   loff_t len)
+{
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	struct gfs2_inode *ip = GFS2_I(inode);
+	unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
+	loff_t bytes, max_bytes;
+	struct gfs2_alloc *al;
+	int error;
+	loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift;
+	next = (next + 1) << sdp->sd_sb.sb_bsize_shift;
+
+	offset = (offset >> sdp->sd_sb.sb_bsize_shift) <<
+		 sdp->sd_sb.sb_bsize_shift;
+
+	len = next - offset;
+	bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2;
+	if (!bytes)
+		bytes = UINT_MAX;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh);
+	error = gfs2_glock_nq(&ip->i_gh);
+	if (unlikely(error))
+		goto out_uninit;
+
+	if (!gfs2_write_alloc_required(ip, offset, len))
+		goto out_unlock;
+
+	while (len > 0) {
+		if (len < bytes)
+			bytes = len;
+		al = gfs2_alloc_get(ip);
+		if (!al) {
+			error = -ENOMEM;
+			goto out_unlock;
+		}
+
+		error = gfs2_quota_lock_check(ip);
+		if (error)
+			goto out_alloc_put;
+
+retry:
+		gfs2_write_calc_reserv(ip, bytes, &data_blocks, &ind_blocks);
+
+		al->al_requested = data_blocks + ind_blocks;
+		error = gfs2_inplace_reserve(ip);
+		if (error) {
+			if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) {
+				bytes >>= 1;
+				goto retry;
+			}
+			goto out_qunlock;
+		}
+		max_bytes = bytes;
+		calc_max_reserv(ip, len, &max_bytes, &data_blocks, &ind_blocks);
+		al->al_requested = data_blocks + ind_blocks;
+
+		rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA +
+			  RES_RG_HDR + gfs2_rg_blocks(al);
+		if (gfs2_is_jdata(ip))
+			rblocks += data_blocks ? data_blocks : 1;
+
+		error = gfs2_trans_begin(sdp, rblocks,
+					 PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize);
+		if (error)
+			goto out_trans_fail;
+
+		error = fallocate_chunk(inode, offset, max_bytes, mode);
+		gfs2_trans_end(sdp);
+
+		if (error)
+			goto out_trans_fail;
+
+		len -= max_bytes;
+		offset += max_bytes;
+		gfs2_inplace_release(ip);
+		gfs2_quota_unlock(ip);
+		gfs2_alloc_put(ip);
+	}
+	goto out_unlock;
+
+out_trans_fail:
+	gfs2_inplace_release(ip);
+out_qunlock:
+	gfs2_quota_unlock(ip);
+out_alloc_put:
+	gfs2_alloc_put(ip);
+out_unlock:
+	gfs2_glock_dq(&ip->i_gh);
+out_uninit:
+	gfs2_holder_uninit(&ip->i_gh);
+	return error;
+}
+
+
 static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		       u64 start, u64 len)
 {
@@ -1351,6 +1572,7 @@
 	.getxattr = gfs2_getxattr,
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
+	.fallocate = gfs2_fallocate,
 	.fiemap = gfs2_fiemap,
 };
 
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 1bc6b56..58a9b99 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -735,10 +735,8 @@
 		goto out;
 
 	size = loc + sizeof(struct gfs2_quota);
-	if (size > inode->i_size) {
-		ip->i_disksize = size;
+	if (size > inode->i_size)
 		i_size_write(inode, size);
-	}
 	inode->i_mtime = inode->i_atime = CURRENT_TIME;
 	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
 	gfs2_dinode_out(ip, dibh->b_data);
@@ -817,7 +815,7 @@
 		goto out_alloc;
 
 	if (nalloc)
-		blocks += al->al_rgd->rd_length + nalloc * ind_blocks + RES_STATFS;
+		blocks += gfs2_rg_blocks(al) + nalloc * ind_blocks + RES_STATFS;
 
 	error = gfs2_trans_begin(sdp, blocks, 0);
 	if (error)
@@ -1190,18 +1188,17 @@
 int gfs2_quota_init(struct gfs2_sbd *sdp)
 {
 	struct gfs2_inode *ip = GFS2_I(sdp->sd_qc_inode);
-	unsigned int blocks = ip->i_disksize >> sdp->sd_sb.sb_bsize_shift;
+	u64 size = i_size_read(sdp->sd_qc_inode);
+	unsigned int blocks = size >> sdp->sd_sb.sb_bsize_shift;
 	unsigned int x, slot = 0;
 	unsigned int found = 0;
 	u64 dblock;
 	u32 extlen = 0;
 	int error;
 
-	if (!ip->i_disksize || ip->i_disksize > (64 << 20) ||
-	    ip->i_disksize & (sdp->sd_sb.sb_bsize - 1)) {
-		gfs2_consist_inode(ip);
+	if (gfs2_check_internal_file_size(sdp->sd_qc_inode, 1, 64 << 20))
 		return -EIO;
-	}
+
 	sdp->sd_quota_slots = blocks * sdp->sd_qc_per_block;
 	sdp->sd_quota_chunks = DIV_ROUND_UP(sdp->sd_quota_slots, 8 * PAGE_SIZE);
 
@@ -1589,6 +1586,7 @@
 		error = gfs2_inplace_reserve(ip);
 		if (error)
 			goto out_alloc;
+		blocks += gfs2_rg_blocks(al);
 	}
 
 	error = gfs2_trans_begin(sdp, blocks + RES_DINODE + 1, 0);
diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c
index f7f89a9..f2a02ed 100644
--- a/fs/gfs2/recovery.c
+++ b/fs/gfs2/recovery.c
@@ -455,11 +455,13 @@
 	int ro = 0;
 	unsigned int pass;
 	int error;
+	int jlocked = 0;
 
-	if (jd->jd_jid != sdp->sd_lockstruct.ls_jid) {
+	if (sdp->sd_args.ar_spectator ||
+	    (jd->jd_jid != sdp->sd_lockstruct.ls_jid)) {
 		fs_info(sdp, "jid=%u: Trying to acquire journal lock...\n",
 			jd->jd_jid);
-
+		jlocked = 1;
 		/* Acquire the journal lock so we can do recovery */
 
 		error = gfs2_glock_nq_num(sdp, jd->jd_jid, &gfs2_journal_glops,
@@ -554,13 +556,12 @@
 			jd->jd_jid, t);
 	}
 
-	if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
-		gfs2_glock_dq_uninit(&ji_gh);
-
 	gfs2_recovery_done(sdp, jd->jd_jid, LM_RD_SUCCESS);
 
-	if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
+	if (jlocked) {
+		gfs2_glock_dq_uninit(&ji_gh);
 		gfs2_glock_dq_uninit(&j_gh);
+	}
 
 	fs_info(sdp, "jid=%u: Done\n", jd->jd_jid);
 	goto done;
@@ -568,7 +569,7 @@
 fail_gunlock_tr:
 	gfs2_glock_dq_uninit(&t_gh);
 fail_gunlock_ji:
-	if (jd->jd_jid != sdp->sd_lockstruct.ls_jid) {
+	if (jlocked) {
 		gfs2_glock_dq_uninit(&ji_gh);
 fail_gunlock_j:
 		gfs2_glock_dq_uninit(&j_gh);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 171a744..fb67f59 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -500,7 +500,7 @@
 	for (rgrps = 0;; rgrps++) {
 		loff_t pos = rgrps * sizeof(struct gfs2_rindex);
 
-		if (pos + sizeof(struct gfs2_rindex) >= ip->i_disksize)
+		if (pos + sizeof(struct gfs2_rindex) >= i_size_read(inode))
 			break;
 		error = gfs2_internal_read(ip, &ra_state, buf, &pos,
 					   sizeof(struct gfs2_rindex));
@@ -588,7 +588,9 @@
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct inode *inode = &ip->i_inode;
 	struct file_ra_state ra_state;
-	u64 rgrp_count = ip->i_disksize;
+	u64 rgrp_count = i_size_read(inode);
+	struct gfs2_rgrpd *rgd;
+	unsigned int max_data = 0;
 	int error;
 
 	do_div(rgrp_count, sizeof(struct gfs2_rindex));
@@ -603,6 +605,10 @@
 		}
 	}
 
+	list_for_each_entry(rgd, &sdp->sd_rindex_list, rd_list)
+		if (rgd->rd_data > max_data)
+			max_data = rgd->rd_data;
+	sdp->sd_max_rg_data = max_data;
 	sdp->sd_rindex_uptodate = 1;
 	return 0;
 }
@@ -622,13 +628,15 @@
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct inode *inode = &ip->i_inode;
 	struct file_ra_state ra_state;
+	struct gfs2_rgrpd *rgd;
+	unsigned int max_data = 0;
 	int error;
 
 	file_ra_state_init(&ra_state, inode->i_mapping);
 	for (sdp->sd_rgrps = 0;; sdp->sd_rgrps++) {
 		/* Ignore partials */
 		if ((sdp->sd_rgrps + 1) * sizeof(struct gfs2_rindex) >
-		    ip->i_disksize)
+		    i_size_read(inode))
 			break;
 		error = read_rindex_entry(ip, &ra_state);
 		if (error) {
@@ -636,6 +644,10 @@
 			return error;
 		}
 	}
+	list_for_each_entry(rgd, &sdp->sd_rindex_list, rd_list)
+		if (rgd->rd_data > max_data)
+			max_data = rgd->rd_data;
+	sdp->sd_max_rg_data = max_data;
 
 	sdp->sd_rindex_uptodate = 1;
 	return 0;
@@ -1188,7 +1200,8 @@
  * Returns: errno
  */
 
-int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file, unsigned int line)
+int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex,
+			   char *file, unsigned int line)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct gfs2_alloc *al = ip->i_alloc;
@@ -1199,12 +1212,15 @@
 		return -EINVAL;
 
 try_again:
-	/* We need to hold the rindex unless the inode we're using is
-	   the rindex itself, in which case it's already held. */
-	if (ip != GFS2_I(sdp->sd_rindex))
-		error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
-	else if (!sdp->sd_rgrps) /* We may not have the rindex read in, so: */
-		error = gfs2_ri_update_special(ip);
+	if (hold_rindex) {
+		/* We need to hold the rindex unless the inode we're using is
+		   the rindex itself, in which case it's already held. */
+		if (ip != GFS2_I(sdp->sd_rindex))
+			error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
+		else if (!sdp->sd_rgrps) /* We may not have the rindex read
+					    in, so: */
+			error = gfs2_ri_update_special(ip);
+	}
 
 	if (error)
 		return error;
@@ -1215,7 +1231,7 @@
 	   try to free it, and try the allocation again. */
 	error = get_local_rgrp(ip, &unlinked, &last_unlinked);
 	if (error) {
-		if (ip != GFS2_I(sdp->sd_rindex))
+		if (hold_rindex && ip != GFS2_I(sdp->sd_rindex))
 			gfs2_glock_dq_uninit(&al->al_ri_gh);
 		if (error != -EAGAIN)
 			return error;
@@ -1257,7 +1273,7 @@
 	al->al_rgd = NULL;
 	if (al->al_rgd_gh.gh_gl)
 		gfs2_glock_dq_uninit(&al->al_rgd_gh);
-	if (ip != GFS2_I(sdp->sd_rindex))
+	if (ip != GFS2_I(sdp->sd_rindex) && al->al_ri_gh.gh_gl)
 		gfs2_glock_dq_uninit(&al->al_ri_gh);
 }
 
@@ -1496,11 +1512,19 @@
 	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct buffer_head *dibh;
 	struct gfs2_alloc *al = ip->i_alloc;
-	struct gfs2_rgrpd *rgd = al->al_rgd;
+	struct gfs2_rgrpd *rgd;
 	u32 goal, blk;
 	u64 block;
 	int error;
 
+	/* Only happens if there is a bug in gfs2, return something distinctive
+	 * to ensure that it is noticed.
+	 */
+	if (al == NULL)
+		return -ECANCELED;
+
+	rgd = al->al_rgd;
+
 	if (rgrp_contains_block(rgd, ip->i_goal))
 		goal = ip->i_goal - rgd->rd_data0;
 	else
diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h
index f07119d..0e35c04 100644
--- a/fs/gfs2/rgrp.h
+++ b/fs/gfs2/rgrp.h
@@ -39,10 +39,12 @@
 	ip->i_alloc = NULL;
 }
 
-extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file,
-				  unsigned int line);
+extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex,
+				  char *file, unsigned int line);
 #define gfs2_inplace_reserve(ip) \
-gfs2_inplace_reserve_i((ip), __FILE__, __LINE__)
+	gfs2_inplace_reserve_i((ip), 1, __FILE__, __LINE__)
+#define gfs2_inplace_reserve_ri(ip) \
+	gfs2_inplace_reserve_i((ip), 0, __FILE__, __LINE__)
 
 extern void gfs2_inplace_release(struct gfs2_inode *ip);
 
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 77cb9f8..047d117 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -85,6 +85,7 @@
 	{Opt_locktable, "locktable=%s"},
 	{Opt_hostdata, "hostdata=%s"},
 	{Opt_spectator, "spectator"},
+	{Opt_spectator, "norecovery"},
 	{Opt_ignore_local_fs, "ignore_local_fs"},
 	{Opt_localflocks, "localflocks"},
 	{Opt_localcaching, "localcaching"},
@@ -159,13 +160,13 @@
 			args->ar_spectator = 1;
 			break;
 		case Opt_ignore_local_fs:
-			args->ar_ignore_local_fs = 1;
+			/* Retained for backwards compat only */
 			break;
 		case Opt_localflocks:
 			args->ar_localflocks = 1;
 			break;
 		case Opt_localcaching:
-			args->ar_localcaching = 1;
+			/* Retained for backwards compat only */
 			break;
 		case Opt_debug:
 			if (args->ar_errors == GFS2_ERRORS_PANIC) {
@@ -179,7 +180,7 @@
 			args->ar_debug = 0;
 			break;
 		case Opt_upgrade:
-			args->ar_upgrade = 1;
+			/* Retained for backwards compat only */
 			break;
 		case Opt_acl:
 			args->ar_posix_acl = 1;
@@ -342,15 +343,14 @@
 {
 	struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
 	struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
+	u64 size = i_size_read(jd->jd_inode);
 
-	if (ip->i_disksize < (8 << 20) || ip->i_disksize > (1 << 30) ||
-	    (ip->i_disksize & (sdp->sd_sb.sb_bsize - 1))) {
-		gfs2_consist_inode(ip);
+	if (gfs2_check_internal_file_size(jd->jd_inode, 8 << 20, 1 << 30))
 		return -EIO;
-	}
-	jd->jd_blocks = ip->i_disksize >> sdp->sd_sb.sb_bsize_shift;
 
-	if (gfs2_write_alloc_required(ip, 0, ip->i_disksize)) {
+	jd->jd_blocks = size >> sdp->sd_sb.sb_bsize_shift;
+
+	if (gfs2_write_alloc_required(ip, 0, size)) {
 		gfs2_consist_inode(ip);
 		return -EIO;
 	}
@@ -1129,9 +1129,7 @@
 
 	/* Some flags must not be changed */
 	if (args_neq(&args, &sdp->sd_args, spectator) ||
-	    args_neq(&args, &sdp->sd_args, ignore_local_fs) ||
 	    args_neq(&args, &sdp->sd_args, localflocks) ||
-	    args_neq(&args, &sdp->sd_args, localcaching) ||
 	    args_neq(&args, &sdp->sd_args, meta))
 		return -EINVAL;
 
@@ -1234,16 +1232,10 @@
 		seq_printf(s, ",hostdata=%s", args->ar_hostdata);
 	if (args->ar_spectator)
 		seq_printf(s, ",spectator");
-	if (args->ar_ignore_local_fs)
-		seq_printf(s, ",ignore_local_fs");
 	if (args->ar_localflocks)
 		seq_printf(s, ",localflocks");
-	if (args->ar_localcaching)
-		seq_printf(s, ",localcaching");
 	if (args->ar_debug)
 		seq_printf(s, ",debug");
-	if (args->ar_upgrade)
-		seq_printf(s, ",upgrade");
 	if (args->ar_posix_acl)
 		seq_printf(s, ",acl");
 	if (args->ar_quota != GFS2_QUOTA_DEFAULT) {
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index ccacffd..748ccb5 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -230,7 +230,10 @@
 
 	if (gltype > LM_TYPE_JOURNAL)
 		return -EINVAL;
-	glops = gfs2_glops_list[gltype];
+	if (gltype == LM_TYPE_NONDISK && glnum == GFS2_TRANS_LOCK)
+		glops = &gfs2_trans_glops;
+	else
+		glops = gfs2_glops_list[gltype];
 	if (glops == NULL)
 		return -EINVAL;
 	if (!test_and_set_bit(SDF_DEMOTE, &sdp->sd_flags))
@@ -399,31 +402,32 @@
 
 static ssize_t jid_show(struct gfs2_sbd *sdp, char *buf)
 {
-	return sprintf(buf, "%u\n", sdp->sd_lockstruct.ls_jid);
+	return sprintf(buf, "%d\n", sdp->sd_lockstruct.ls_jid);
 }
 
 static ssize_t jid_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
 {
-        unsigned jid;
+        int jid;
 	int rv;
 
-	rv = sscanf(buf, "%u", &jid);
+	rv = sscanf(buf, "%d", &jid);
 	if (rv != 1)
 		return -EINVAL;
 
 	spin_lock(&sdp->sd_jindex_spin);
 	rv = -EINVAL;
-	if (sdp->sd_args.ar_spectator)
-		goto out;
 	if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL)
 		goto out;
 	rv = -EBUSY;
-	if (test_and_clear_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
+	if (test_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0)
 		goto out;
+	rv = 0;
+	if (sdp->sd_args.ar_spectator && jid > 0)
+		rv = jid = -EINVAL;
 	sdp->sd_lockstruct.ls_jid = jid;
+	clear_bit(SDF_NOJOURNALID, &sdp->sd_flags);
 	smp_mb__after_clear_bit();
 	wake_up_bit(&sdp->sd_flags, SDF_NOJOURNALID);
-	rv = 0;
 out:
 	spin_unlock(&sdp->sd_jindex_spin);
 	return rv ? rv : len;
@@ -617,7 +621,7 @@
 	add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
 	add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
 	if (!test_bit(SDF_NOJOURNALID, &sdp->sd_flags))
-		add_uevent_var(env, "JOURNALID=%u", sdp->sd_lockstruct.ls_jid);
+		add_uevent_var(env, "JOURNALID=%d", sdp->sd_lockstruct.ls_jid);
 	if (gfs2_uuid_valid(uuid))
 		add_uevent_var(env, "UUID=%pUB", uuid);
 	return 0;
diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h
index 148d55c..cedb0bb 100644
--- a/fs/gfs2/trace_gfs2.h
+++ b/fs/gfs2/trace_gfs2.h
@@ -39,7 +39,8 @@
 	{(1UL << GLF_INVALIDATE_IN_PROGRESS),	"i" },		\
 	{(1UL << GLF_REPLY_PENDING),		"r" },		\
 	{(1UL << GLF_INITIAL),			"I" },		\
-	{(1UL << GLF_FROZEN),			"F" })
+	{(1UL << GLF_FROZEN),			"F" },		\
+	{(1UL << GLF_QUEUED),			"q" })
 
 #ifndef NUMPTY
 #define NUMPTY
diff --git a/fs/gfs2/trans.h b/fs/gfs2/trans.h
index edf9d4b..fb56b78 100644
--- a/fs/gfs2/trans.h
+++ b/fs/gfs2/trans.h
@@ -20,11 +20,20 @@
 #define RES_JDATA	1
 #define RES_DATA	1
 #define RES_LEAF	1
+#define RES_RG_HDR	1
 #define RES_RG_BIT	2
 #define RES_EATTR	1
 #define RES_STATFS	1
 #define RES_QUOTA	2
 
+/* reserve either the number of blocks to be allocated plus the rg header
+ * block, or all of the blocks in the rg, whichever is smaller */
+static inline unsigned int gfs2_rg_blocks(const struct gfs2_alloc *al)
+{
+	return (al->al_requested < al->al_rgd->rd_length)?
+	       al->al_requested + 1 : al->al_rgd->rd_length;
+}
+
 int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks,
 		     unsigned int revokes);
 
diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index 776af6e..30b58f0 100644
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -734,7 +734,7 @@
 		goto out_gunlock_q;
 
 	error = gfs2_trans_begin(GFS2_SB(&ip->i_inode),
-				 blks + al->al_rgd->rd_length +
+				 blks + gfs2_rg_blocks(al) +
 				 RES_DINODE + RES_STATFS + RES_QUOTA, 0);
 	if (error)
 		goto out_ipres;
diff --git a/fs/hfs/bfind.c b/fs/hfs/bfind.c
index 4129cdb..571abe9 100644
--- a/fs/hfs/bfind.c
+++ b/fs/hfs/bfind.c
@@ -23,7 +23,7 @@
 	fd->search_key = ptr;
 	fd->key = ptr + tree->max_key_len + 2;
 	dprint(DBG_BNODE_REFS, "find_init: %d (%p)\n", tree->cnid, __builtin_return_address(0));
-	down(&tree->tree_lock);
+	mutex_lock(&tree->tree_lock);
 	return 0;
 }
 
@@ -32,7 +32,7 @@
 	hfs_bnode_put(fd->bnode);
 	kfree(fd->search_key);
 	dprint(DBG_BNODE_REFS, "find_exit: %d (%p)\n", fd->tree->cnid, __builtin_return_address(0));
-	up(&fd->tree->tree_lock);
+	mutex_unlock(&fd->tree->tree_lock);
 	fd->tree = NULL;
 }
 
diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c
index 38a0a99..3ebc437 100644
--- a/fs/hfs/btree.c
+++ b/fs/hfs/btree.c
@@ -27,7 +27,7 @@
 	if (!tree)
 		return NULL;
 
-	init_MUTEX(&tree->tree_lock);
+	mutex_init(&tree->tree_lock);
 	spin_lock_init(&tree->hash_lock);
 	/* Set the correct compare function */
 	tree->sb = sb;
diff --git a/fs/hfs/btree.h b/fs/hfs/btree.h
index cc51905..2a1d712 100644
--- a/fs/hfs/btree.h
+++ b/fs/hfs/btree.h
@@ -33,7 +33,7 @@
 	unsigned int depth;
 
 	//unsigned int map1_size, map_size;
-	struct semaphore tree_lock;
+	struct mutex tree_lock;
 
 	unsigned int pages_per_bnode;
 	spinlock_t hash_lock;
diff --git a/fs/hfsplus/bfind.c b/fs/hfsplus/bfind.c
index 5007a41..d182438 100644
--- a/fs/hfsplus/bfind.c
+++ b/fs/hfsplus/bfind.c
@@ -23,7 +23,7 @@
 	fd->search_key = ptr;
 	fd->key = ptr + tree->max_key_len + 2;
 	dprint(DBG_BNODE_REFS, "find_init: %d (%p)\n", tree->cnid, __builtin_return_address(0));
-	down(&tree->tree_lock);
+	mutex_lock(&tree->tree_lock);
 	return 0;
 }
 
@@ -32,7 +32,7 @@
 	hfs_bnode_put(fd->bnode);
 	kfree(fd->search_key);
 	dprint(DBG_BNODE_REFS, "find_exit: %d (%p)\n", fd->tree->cnid, __builtin_return_address(0));
-	up(&fd->tree->tree_lock);
+	mutex_unlock(&fd->tree->tree_lock);
 	fd->tree = NULL;
 }
 
@@ -52,6 +52,10 @@
 		rec = (e + b) / 2;
 		len = hfs_brec_lenoff(bnode, rec, &off);
 		keylen = hfs_brec_keylen(bnode, rec);
+		if (keylen == 0) {
+			res = -EINVAL;
+			goto fail;
+		}
 		hfs_bnode_read(bnode, fd->key, off, keylen);
 		cmpval = bnode->tree->keycmp(fd->key, fd->search_key);
 		if (!cmpval) {
@@ -67,6 +71,10 @@
 	if (rec != e && e >= 0) {
 		len = hfs_brec_lenoff(bnode, e, &off);
 		keylen = hfs_brec_keylen(bnode, e);
+		if (keylen == 0) {
+			res = -EINVAL;
+			goto fail;
+		}
 		hfs_bnode_read(bnode, fd->key, off, keylen);
 	}
 done:
@@ -75,6 +83,7 @@
 	fd->keylength = keylen;
 	fd->entryoffset = off + keylen;
 	fd->entrylength = len - keylen;
+fail:
 	return res;
 }
 
@@ -198,6 +207,10 @@
 
 	len = hfs_brec_lenoff(bnode, fd->record, &off);
 	keylen = hfs_brec_keylen(bnode, fd->record);
+	if (keylen == 0) {
+		res = -EINVAL;
+		goto out;
+	}
 	fd->keyoffset = off;
 	fd->keylength = keylen;
 	fd->entryoffset = off + keylen;
diff --git a/fs/hfsplus/bitmap.c b/fs/hfsplus/bitmap.c
index ea30afc..ad57f59 100644
--- a/fs/hfsplus/bitmap.c
+++ b/fs/hfsplus/bitmap.c
@@ -17,6 +17,7 @@
 
 int hfsplus_block_allocate(struct super_block *sb, u32 size, u32 offset, u32 *max)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	struct page *page;
 	struct address_space *mapping;
 	__be32 *pptr, *curr, *end;
@@ -29,8 +30,8 @@
 		return size;
 
 	dprint(DBG_BITMAP, "block_allocate: %u,%u,%u\n", size, offset, len);
-	mutex_lock(&HFSPLUS_SB(sb).alloc_file->i_mutex);
-	mapping = HFSPLUS_SB(sb).alloc_file->i_mapping;
+	mutex_lock(&sbi->alloc_mutex);
+	mapping = sbi->alloc_file->i_mapping;
 	page = read_mapping_page(mapping, offset / PAGE_CACHE_BITS, NULL);
 	if (IS_ERR(page)) {
 		start = size;
@@ -150,16 +151,17 @@
 	set_page_dirty(page);
 	kunmap(page);
 	*max = offset + (curr - pptr) * 32 + i - start;
-	HFSPLUS_SB(sb).free_blocks -= *max;
+	sbi->free_blocks -= *max;
 	sb->s_dirt = 1;
 	dprint(DBG_BITMAP, "-> %u,%u\n", start, *max);
 out:
-	mutex_unlock(&HFSPLUS_SB(sb).alloc_file->i_mutex);
+	mutex_unlock(&sbi->alloc_mutex);
 	return start;
 }
 
 int hfsplus_block_free(struct super_block *sb, u32 offset, u32 count)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	struct page *page;
 	struct address_space *mapping;
 	__be32 *pptr, *curr, *end;
@@ -172,11 +174,11 @@
 
 	dprint(DBG_BITMAP, "block_free: %u,%u\n", offset, count);
 	/* are all of the bits in range? */
-	if ((offset + count) > HFSPLUS_SB(sb).total_blocks)
+	if ((offset + count) > sbi->total_blocks)
 		return -2;
 
-	mutex_lock(&HFSPLUS_SB(sb).alloc_file->i_mutex);
-	mapping = HFSPLUS_SB(sb).alloc_file->i_mapping;
+	mutex_lock(&sbi->alloc_mutex);
+	mapping = sbi->alloc_file->i_mapping;
 	pnr = offset / PAGE_CACHE_BITS;
 	page = read_mapping_page(mapping, pnr, NULL);
 	pptr = kmap(page);
@@ -224,9 +226,9 @@
 out:
 	set_page_dirty(page);
 	kunmap(page);
-	HFSPLUS_SB(sb).free_blocks += len;
+	sbi->free_blocks += len;
 	sb->s_dirt = 1;
-	mutex_unlock(&HFSPLUS_SB(sb).alloc_file->i_mutex);
+	mutex_unlock(&sbi->alloc_mutex);
 
 	return 0;
 }
diff --git a/fs/hfsplus/brec.c b/fs/hfsplus/brec.c
index c88e5d7..2f39d05 100644
--- a/fs/hfsplus/brec.c
+++ b/fs/hfsplus/brec.c
@@ -42,10 +42,13 @@
 		recoff = hfs_bnode_read_u16(node, node->tree->node_size - (rec + 1) * 2);
 		if (!recoff)
 			return 0;
-		if (node->tree->attributes & HFS_TREE_BIGKEYS)
-			retval = hfs_bnode_read_u16(node, recoff) + 2;
-		else
-			retval = (hfs_bnode_read_u8(node, recoff) | 1) + 1;
+
+		retval = hfs_bnode_read_u16(node, recoff) + 2;
+		if (retval > node->tree->max_key_len + 2) {
+			printk(KERN_ERR "hfs: keylen %d too large\n",
+				retval);
+			retval = 0;
+		}
 	}
 	return retval;
 }
@@ -216,7 +219,7 @@
 static struct hfs_bnode *hfs_bnode_split(struct hfs_find_data *fd)
 {
 	struct hfs_btree *tree;
-	struct hfs_bnode *node, *new_node;
+	struct hfs_bnode *node, *new_node, *next_node;
 	struct hfs_bnode_desc node_desc;
 	int num_recs, new_rec_off, new_off, old_rec_off;
 	int data_start, data_end, size;
@@ -235,6 +238,17 @@
 	new_node->type = node->type;
 	new_node->height = node->height;
 
+	if (node->next)
+		next_node = hfs_bnode_find(tree, node->next);
+	else
+		next_node = NULL;
+
+	if (IS_ERR(next_node)) {
+		hfs_bnode_put(node);
+		hfs_bnode_put(new_node);
+		return next_node;
+	}
+
 	size = tree->node_size / 2 - node->num_recs * 2 - 14;
 	old_rec_off = tree->node_size - 4;
 	num_recs = 1;
@@ -248,6 +262,8 @@
 		/* panic? */
 		hfs_bnode_put(node);
 		hfs_bnode_put(new_node);
+		if (next_node)
+			hfs_bnode_put(next_node);
 		return ERR_PTR(-ENOSPC);
 	}
 
@@ -302,8 +318,7 @@
 	hfs_bnode_write(node, &node_desc, 0, sizeof(node_desc));
 
 	/* update next bnode header */
-	if (new_node->next) {
-		struct hfs_bnode *next_node = hfs_bnode_find(tree, new_node->next);
+	if (next_node) {
 		next_node->prev = new_node->this;
 		hfs_bnode_read(next_node, &node_desc, 0, sizeof(node_desc));
 		node_desc.prev = cpu_to_be32(next_node->prev);
diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c
index e49fcee..22e4d4e 100644
--- a/fs/hfsplus/btree.c
+++ b/fs/hfsplus/btree.c
@@ -30,7 +30,7 @@
 	if (!tree)
 		return NULL;
 
-	init_MUTEX(&tree->tree_lock);
+	mutex_init(&tree->tree_lock);
 	spin_lock_init(&tree->hash_lock);
 	tree->sb = sb;
 	tree->cnid = id;
@@ -39,10 +39,16 @@
 		goto free_tree;
 	tree->inode = inode;
 
+	if (!HFSPLUS_I(tree->inode)->first_blocks) {
+		printk(KERN_ERR
+		       "hfs: invalid btree extent records (0 size).\n");
+		goto free_inode;
+	}
+
 	mapping = tree->inode->i_mapping;
 	page = read_mapping_page(mapping, 0, NULL);
 	if (IS_ERR(page))
-		goto free_tree;
+		goto free_inode;
 
 	/* Load the header */
 	head = (struct hfs_btree_header_rec *)(kmap(page) + sizeof(struct hfs_bnode_desc));
@@ -57,27 +63,56 @@
 	tree->max_key_len = be16_to_cpu(head->max_key_len);
 	tree->depth = be16_to_cpu(head->depth);
 
-	/* Set the correct compare function */
-	if (id == HFSPLUS_EXT_CNID) {
+	/* Verify the tree and set the correct compare function */
+	switch (id) {
+	case HFSPLUS_EXT_CNID:
+		if (tree->max_key_len != HFSPLUS_EXT_KEYLEN - sizeof(u16)) {
+			printk(KERN_ERR "hfs: invalid extent max_key_len %d\n",
+				tree->max_key_len);
+			goto fail_page;
+		}
+		if (tree->attributes & HFS_TREE_VARIDXKEYS) {
+			printk(KERN_ERR "hfs: invalid extent btree flag\n");
+			goto fail_page;
+		}
+
 		tree->keycmp = hfsplus_ext_cmp_key;
-	} else if (id == HFSPLUS_CAT_CNID) {
-		if ((HFSPLUS_SB(sb).flags & HFSPLUS_SB_HFSX) &&
+		break;
+	case HFSPLUS_CAT_CNID:
+		if (tree->max_key_len != HFSPLUS_CAT_KEYLEN - sizeof(u16)) {
+			printk(KERN_ERR "hfs: invalid catalog max_key_len %d\n",
+				tree->max_key_len);
+			goto fail_page;
+		}
+		if (!(tree->attributes & HFS_TREE_VARIDXKEYS)) {
+			printk(KERN_ERR "hfs: invalid catalog btree flag\n");
+			goto fail_page;
+		}
+
+		if (test_bit(HFSPLUS_SB_HFSX, &HFSPLUS_SB(sb)->flags) &&
 		    (head->key_type == HFSPLUS_KEY_BINARY))
 			tree->keycmp = hfsplus_cat_bin_cmp_key;
 		else {
 			tree->keycmp = hfsplus_cat_case_cmp_key;
-			HFSPLUS_SB(sb).flags |= HFSPLUS_SB_CASEFOLD;
+			set_bit(HFSPLUS_SB_CASEFOLD, &HFSPLUS_SB(sb)->flags);
 		}
-	} else {
+		break;
+	default:
 		printk(KERN_ERR "hfs: unknown B*Tree requested\n");
 		goto fail_page;
 	}
 
+	if (!(tree->attributes & HFS_TREE_BIGKEYS)) {
+		printk(KERN_ERR "hfs: invalid btree flag\n");
+		goto fail_page;
+	}
+
 	size = tree->node_size;
 	if (!is_power_of_2(size))
 		goto fail_page;
 	if (!tree->node_count)
 		goto fail_page;
+
 	tree->node_size_shift = ffs(size) - 1;
 
 	tree->pages_per_bnode = (tree->node_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
@@ -87,10 +122,11 @@
 	return tree;
 
  fail_page:
-	tree->inode->i_mapping->a_ops = &hfsplus_aops;
 	page_cache_release(page);
- free_tree:
+ free_inode:
+	tree->inode->i_mapping->a_ops = &hfsplus_aops;
 	iput(tree->inode);
+ free_tree:
 	kfree(tree);
 	return NULL;
 }
@@ -192,17 +228,18 @@
 
 	while (!tree->free_nodes) {
 		struct inode *inode = tree->inode;
+		struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 		u32 count;
 		int res;
 
 		res = hfsplus_file_extend(inode);
 		if (res)
 			return ERR_PTR(res);
-		HFSPLUS_I(inode).phys_size = inode->i_size =
-				(loff_t)HFSPLUS_I(inode).alloc_blocks <<
-				HFSPLUS_SB(tree->sb).alloc_blksz_shift;
-		HFSPLUS_I(inode).fs_blocks = HFSPLUS_I(inode).alloc_blocks <<
-					     HFSPLUS_SB(tree->sb).fs_shift;
+		hip->phys_size = inode->i_size =
+			(loff_t)hip->alloc_blocks <<
+				HFSPLUS_SB(tree->sb)->alloc_blksz_shift;
+		hip->fs_blocks =
+			hip->alloc_blocks << HFSPLUS_SB(tree->sb)->fs_shift;
 		inode_set_bytes(inode, inode->i_size);
 		count = inode->i_size >> tree->node_size_shift;
 		tree->free_nodes = count - tree->node_count;
diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c
index f6874ac..8af45fc 100644
--- a/fs/hfsplus/catalog.c
+++ b/fs/hfsplus/catalog.c
@@ -67,7 +67,7 @@
 	key->key_len = cpu_to_be16(6 + ustrlen);
 }
 
-static void hfsplus_set_perms(struct inode *inode, struct hfsplus_perm *perms)
+void hfsplus_cat_set_perms(struct inode *inode, struct hfsplus_perm *perms)
 {
 	if (inode->i_flags & S_IMMUTABLE)
 		perms->rootflags |= HFSPLUS_FLG_IMMUTABLE;
@@ -77,15 +77,24 @@
 		perms->rootflags |= HFSPLUS_FLG_APPEND;
 	else
 		perms->rootflags &= ~HFSPLUS_FLG_APPEND;
-	HFSPLUS_I(inode).rootflags = perms->rootflags;
-	HFSPLUS_I(inode).userflags = perms->userflags;
+
+	perms->userflags = HFSPLUS_I(inode)->userflags;
 	perms->mode = cpu_to_be16(inode->i_mode);
 	perms->owner = cpu_to_be32(inode->i_uid);
 	perms->group = cpu_to_be32(inode->i_gid);
+
+	if (S_ISREG(inode->i_mode))
+		perms->dev = cpu_to_be32(inode->i_nlink);
+	else if (S_ISBLK(inode->i_mode) || S_ISCHR(inode->i_mode))
+		perms->dev = cpu_to_be32(inode->i_rdev);
+	else
+		perms->dev = 0;
 }
 
 static int hfsplus_cat_build_record(hfsplus_cat_entry *entry, u32 cnid, struct inode *inode)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(inode->i_sb);
+
 	if (S_ISDIR(inode->i_mode)) {
 		struct hfsplus_cat_folder *folder;
 
@@ -93,13 +102,13 @@
 		memset(folder, 0, sizeof(*folder));
 		folder->type = cpu_to_be16(HFSPLUS_FOLDER);
 		folder->id = cpu_to_be32(inode->i_ino);
-		HFSPLUS_I(inode).create_date =
+		HFSPLUS_I(inode)->create_date =
 			folder->create_date =
 			folder->content_mod_date =
 			folder->attribute_mod_date =
 			folder->access_date = hfsp_now2mt();
-		hfsplus_set_perms(inode, &folder->permissions);
-		if (inode == HFSPLUS_SB(inode->i_sb).hidden_dir)
+		hfsplus_cat_set_perms(inode, &folder->permissions);
+		if (inode == sbi->hidden_dir)
 			/* invisible and namelocked */
 			folder->user_info.frFlags = cpu_to_be16(0x5000);
 		return sizeof(*folder);
@@ -111,19 +120,19 @@
 		file->type = cpu_to_be16(HFSPLUS_FILE);
 		file->flags = cpu_to_be16(HFSPLUS_FILE_THREAD_EXISTS);
 		file->id = cpu_to_be32(cnid);
-		HFSPLUS_I(inode).create_date =
+		HFSPLUS_I(inode)->create_date =
 			file->create_date =
 			file->content_mod_date =
 			file->attribute_mod_date =
 			file->access_date = hfsp_now2mt();
 		if (cnid == inode->i_ino) {
-			hfsplus_set_perms(inode, &file->permissions);
+			hfsplus_cat_set_perms(inode, &file->permissions);
 			if (S_ISLNK(inode->i_mode)) {
 				file->user_info.fdType = cpu_to_be32(HFSP_SYMLINK_TYPE);
 				file->user_info.fdCreator = cpu_to_be32(HFSP_SYMLINK_CREATOR);
 			} else {
-				file->user_info.fdType = cpu_to_be32(HFSPLUS_SB(inode->i_sb).type);
-				file->user_info.fdCreator = cpu_to_be32(HFSPLUS_SB(inode->i_sb).creator);
+				file->user_info.fdType = cpu_to_be32(sbi->type);
+				file->user_info.fdCreator = cpu_to_be32(sbi->creator);
 			}
 			if ((file->permissions.rootflags | file->permissions.userflags) & HFSPLUS_FLG_IMMUTABLE)
 				file->flags |= cpu_to_be16(HFSPLUS_FILE_LOCKED);
@@ -131,8 +140,8 @@
 			file->user_info.fdType = cpu_to_be32(HFSP_HARDLINK_TYPE);
 			file->user_info.fdCreator = cpu_to_be32(HFSP_HFSPLUS_CREATOR);
 			file->user_info.fdFlags = cpu_to_be16(0x100);
-			file->create_date = HFSPLUS_I(HFSPLUS_SB(inode->i_sb).hidden_dir).create_date;
-			file->permissions.dev = cpu_to_be32(HFSPLUS_I(inode).dev);
+			file->create_date = HFSPLUS_I(sbi->hidden_dir)->create_date;
+			file->permissions.dev = cpu_to_be32(HFSPLUS_I(inode)->linkid);
 		}
 		return sizeof(*file);
 	}
@@ -180,15 +189,14 @@
 
 int hfsplus_create_cat(u32 cnid, struct inode *dir, struct qstr *str, struct inode *inode)
 {
+	struct super_block *sb = dir->i_sb;
 	struct hfs_find_data fd;
-	struct super_block *sb;
 	hfsplus_cat_entry entry;
 	int entry_size;
 	int err;
 
 	dprint(DBG_CAT_MOD, "create_cat: %s,%u(%d)\n", str->name, cnid, inode->i_nlink);
-	sb = dir->i_sb;
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
 
 	hfsplus_cat_build_key(sb, fd.search_key, cnid, NULL);
 	entry_size = hfsplus_fill_cat_thread(sb, &entry, S_ISDIR(inode->i_mode) ?
@@ -234,7 +242,7 @@
 
 int hfsplus_delete_cat(u32 cnid, struct inode *dir, struct qstr *str)
 {
-	struct super_block *sb;
+	struct super_block *sb = dir->i_sb;
 	struct hfs_find_data fd;
 	struct hfsplus_fork_raw fork;
 	struct list_head *pos;
@@ -242,8 +250,7 @@
 	u16 type;
 
 	dprint(DBG_CAT_MOD, "delete_cat: %s,%u\n", str ? str->name : NULL, cnid);
-	sb = dir->i_sb;
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
 
 	if (!str) {
 		int len;
@@ -279,7 +286,7 @@
 		hfsplus_free_fork(sb, cnid, &fork, HFSPLUS_TYPE_RSRC);
 	}
 
-	list_for_each(pos, &HFSPLUS_I(dir).open_dir_list) {
+	list_for_each(pos, &HFSPLUS_I(dir)->open_dir_list) {
 		struct hfsplus_readdir_data *rd =
 			list_entry(pos, struct hfsplus_readdir_data, list);
 		if (fd.tree->keycmp(fd.search_key, (void *)&rd->key) < 0)
@@ -312,7 +319,7 @@
 		       struct inode *src_dir, struct qstr *src_name,
 		       struct inode *dst_dir, struct qstr *dst_name)
 {
-	struct super_block *sb;
+	struct super_block *sb = src_dir->i_sb;
 	struct hfs_find_data src_fd, dst_fd;
 	hfsplus_cat_entry entry;
 	int entry_size, type;
@@ -320,8 +327,7 @@
 
 	dprint(DBG_CAT_MOD, "rename_cat: %u - %lu,%s - %lu,%s\n", cnid, src_dir->i_ino, src_name->name,
 		dst_dir->i_ino, dst_name->name);
-	sb = src_dir->i_sb;
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &src_fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &src_fd);
 	dst_fd = src_fd;
 
 	/* find the old dir entry and read the data */
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 764fd1b..d236d85 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -39,7 +39,7 @@
 
 	dentry->d_op = &hfsplus_dentry_operations;
 	dentry->d_fsdata = NULL;
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
 	hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, &dentry->d_name);
 again:
 	err = hfs_brec_read(&fd, &entry, sizeof(entry));
@@ -68,9 +68,9 @@
 		cnid = be32_to_cpu(entry.file.id);
 		if (entry.file.user_info.fdType == cpu_to_be32(HFSP_HARDLINK_TYPE) &&
 		    entry.file.user_info.fdCreator == cpu_to_be32(HFSP_HFSPLUS_CREATOR) &&
-		    (entry.file.create_date == HFSPLUS_I(HFSPLUS_SB(sb).hidden_dir).create_date ||
-		     entry.file.create_date == HFSPLUS_I(sb->s_root->d_inode).create_date) &&
-		    HFSPLUS_SB(sb).hidden_dir) {
+		    (entry.file.create_date == HFSPLUS_I(HFSPLUS_SB(sb)->hidden_dir)->create_date ||
+		     entry.file.create_date == HFSPLUS_I(sb->s_root->d_inode)->create_date) &&
+		    HFSPLUS_SB(sb)->hidden_dir) {
 			struct qstr str;
 			char name[32];
 
@@ -86,7 +86,8 @@
 				linkid = be32_to_cpu(entry.file.permissions.dev);
 				str.len = sprintf(name, "iNode%d", linkid);
 				str.name = name;
-				hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_SB(sb).hidden_dir->i_ino, &str);
+				hfsplus_cat_build_key(sb, fd.search_key,
+					HFSPLUS_SB(sb)->hidden_dir->i_ino, &str);
 				goto again;
 			}
 		} else if (!dentry->d_fsdata)
@@ -101,7 +102,7 @@
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 	if (S_ISREG(inode->i_mode))
-		HFSPLUS_I(inode).dev = linkid;
+		HFSPLUS_I(inode)->linkid = linkid;
 out:
 	d_add(dentry, inode);
 	return NULL;
@@ -124,7 +125,7 @@
 	if (filp->f_pos >= inode->i_size)
 		return 0;
 
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
 	hfsplus_cat_build_key(sb, fd.search_key, inode->i_ino, NULL);
 	err = hfs_brec_find(&fd);
 	if (err)
@@ -180,8 +181,9 @@
 				err = -EIO;
 				goto out;
 			}
-			if (HFSPLUS_SB(sb).hidden_dir &&
-			    HFSPLUS_SB(sb).hidden_dir->i_ino == be32_to_cpu(entry.folder.id))
+			if (HFSPLUS_SB(sb)->hidden_dir &&
+			    HFSPLUS_SB(sb)->hidden_dir->i_ino ==
+					be32_to_cpu(entry.folder.id))
 				goto next;
 			if (filldir(dirent, strbuf, len, filp->f_pos,
 				    be32_to_cpu(entry.folder.id), DT_DIR))
@@ -217,7 +219,7 @@
 		}
 		filp->private_data = rd;
 		rd->file = filp;
-		list_add(&rd->list, &HFSPLUS_I(inode).open_dir_list);
+		list_add(&rd->list, &HFSPLUS_I(inode)->open_dir_list);
 	}
 	memcpy(&rd->key, fd.key, sizeof(struct hfsplus_cat_key));
 out:
@@ -229,38 +231,18 @@
 {
 	struct hfsplus_readdir_data *rd = file->private_data;
 	if (rd) {
+		mutex_lock(&inode->i_mutex);
 		list_del(&rd->list);
+		mutex_unlock(&inode->i_mutex);
 		kfree(rd);
 	}
 	return 0;
 }
 
-static int hfsplus_create(struct inode *dir, struct dentry *dentry, int mode,
-			  struct nameidata *nd)
-{
-	struct inode *inode;
-	int res;
-
-	inode = hfsplus_new_inode(dir->i_sb, mode);
-	if (!inode)
-		return -ENOSPC;
-
-	res = hfsplus_create_cat(inode->i_ino, dir, &dentry->d_name, inode);
-	if (res) {
-		inode->i_nlink = 0;
-		hfsplus_delete_inode(inode);
-		iput(inode);
-		return res;
-	}
-	hfsplus_instantiate(dentry, inode, inode->i_ino);
-	mark_inode_dirty(inode);
-	return 0;
-}
-
 static int hfsplus_link(struct dentry *src_dentry, struct inode *dst_dir,
 			struct dentry *dst_dentry)
 {
-	struct super_block *sb = dst_dir->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(dst_dir->i_sb);
 	struct inode *inode = src_dentry->d_inode;
 	struct inode *src_dir = src_dentry->d_parent->d_inode;
 	struct qstr str;
@@ -270,7 +252,10 @@
 
 	if (HFSPLUS_IS_RSRC(inode))
 		return -EPERM;
+	if (!S_ISREG(inode->i_mode))
+		return -EPERM;
 
+	mutex_lock(&sbi->vh_mutex);
 	if (inode->i_ino == (u32)(unsigned long)src_dentry->d_fsdata) {
 		for (;;) {
 			get_random_bytes(&id, sizeof(cnid));
@@ -279,40 +264,41 @@
 			str.len = sprintf(name, "iNode%d", id);
 			res = hfsplus_rename_cat(inode->i_ino,
 						 src_dir, &src_dentry->d_name,
-						 HFSPLUS_SB(sb).hidden_dir, &str);
+						 sbi->hidden_dir, &str);
 			if (!res)
 				break;
 			if (res != -EEXIST)
-				return res;
+				goto out;
 		}
-		HFSPLUS_I(inode).dev = id;
-		cnid = HFSPLUS_SB(sb).next_cnid++;
+		HFSPLUS_I(inode)->linkid = id;
+		cnid = sbi->next_cnid++;
 		src_dentry->d_fsdata = (void *)(unsigned long)cnid;
 		res = hfsplus_create_cat(cnid, src_dir, &src_dentry->d_name, inode);
 		if (res)
 			/* panic? */
-			return res;
-		HFSPLUS_SB(sb).file_count++;
+			goto out;
+		sbi->file_count++;
 	}
-	cnid = HFSPLUS_SB(sb).next_cnid++;
+	cnid = sbi->next_cnid++;
 	res = hfsplus_create_cat(cnid, dst_dir, &dst_dentry->d_name, inode);
 	if (res)
-		return res;
+		goto out;
 
 	inc_nlink(inode);
 	hfsplus_instantiate(dst_dentry, inode, cnid);
 	atomic_inc(&inode->i_count);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	mark_inode_dirty(inode);
-	HFSPLUS_SB(sb).file_count++;
-	sb->s_dirt = 1;
-
-	return 0;
+	sbi->file_count++;
+	dst_dir->i_sb->s_dirt = 1;
+out:
+	mutex_unlock(&sbi->vh_mutex);
+	return res;
 }
 
 static int hfsplus_unlink(struct inode *dir, struct dentry *dentry)
 {
-	struct super_block *sb = dir->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb);
 	struct inode *inode = dentry->d_inode;
 	struct qstr str;
 	char name[32];
@@ -322,21 +308,22 @@
 	if (HFSPLUS_IS_RSRC(inode))
 		return -EPERM;
 
+	mutex_lock(&sbi->vh_mutex);
 	cnid = (u32)(unsigned long)dentry->d_fsdata;
 	if (inode->i_ino == cnid &&
-	    atomic_read(&HFSPLUS_I(inode).opencnt)) {
+	    atomic_read(&HFSPLUS_I(inode)->opencnt)) {
 		str.name = name;
 		str.len = sprintf(name, "temp%lu", inode->i_ino);
 		res = hfsplus_rename_cat(inode->i_ino,
 					 dir, &dentry->d_name,
-					 HFSPLUS_SB(sb).hidden_dir, &str);
+					 sbi->hidden_dir, &str);
 		if (!res)
 			inode->i_flags |= S_DEAD;
-		return res;
+		goto out;
 	}
 	res = hfsplus_delete_cat(cnid, dir, &dentry->d_name);
 	if (res)
-		return res;
+		goto out;
 
 	if (inode->i_nlink > 0)
 		drop_nlink(inode);
@@ -344,10 +331,10 @@
 		clear_nlink(inode);
 	if (!inode->i_nlink) {
 		if (inode->i_ino != cnid) {
-			HFSPLUS_SB(sb).file_count--;
-			if (!atomic_read(&HFSPLUS_I(inode).opencnt)) {
+			sbi->file_count--;
+			if (!atomic_read(&HFSPLUS_I(inode)->opencnt)) {
 				res = hfsplus_delete_cat(inode->i_ino,
-							 HFSPLUS_SB(sb).hidden_dir,
+							 sbi->hidden_dir,
 							 NULL);
 				if (!res)
 					hfsplus_delete_inode(inode);
@@ -356,107 +343,108 @@
 		} else
 			hfsplus_delete_inode(inode);
 	} else
-		HFSPLUS_SB(sb).file_count--;
+		sbi->file_count--;
 	inode->i_ctime = CURRENT_TIME_SEC;
 	mark_inode_dirty(inode);
-
+out:
+	mutex_unlock(&sbi->vh_mutex);
 	return res;
 }
 
-static int hfsplus_mkdir(struct inode *dir, struct dentry *dentry, int mode)
-{
-	struct inode *inode;
-	int res;
-
-	inode = hfsplus_new_inode(dir->i_sb, S_IFDIR | mode);
-	if (!inode)
-		return -ENOSPC;
-
-	res = hfsplus_create_cat(inode->i_ino, dir, &dentry->d_name, inode);
-	if (res) {
-		inode->i_nlink = 0;
-		hfsplus_delete_inode(inode);
-		iput(inode);
-		return res;
-	}
-	hfsplus_instantiate(dentry, inode, inode->i_ino);
-	mark_inode_dirty(inode);
-	return 0;
-}
-
 static int hfsplus_rmdir(struct inode *dir, struct dentry *dentry)
 {
-	struct inode *inode;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb);
+	struct inode *inode = dentry->d_inode;
 	int res;
 
-	inode = dentry->d_inode;
 	if (inode->i_size != 2)
 		return -ENOTEMPTY;
+
+	mutex_lock(&sbi->vh_mutex);
 	res = hfsplus_delete_cat(inode->i_ino, dir, &dentry->d_name);
 	if (res)
-		return res;
+		goto out;
 	clear_nlink(inode);
 	inode->i_ctime = CURRENT_TIME_SEC;
 	hfsplus_delete_inode(inode);
 	mark_inode_dirty(inode);
-	return 0;
+out:
+	mutex_unlock(&sbi->vh_mutex);
+	return res;
 }
 
 static int hfsplus_symlink(struct inode *dir, struct dentry *dentry,
 			   const char *symname)
 {
-	struct super_block *sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb);
 	struct inode *inode;
-	int res;
+	int res = -ENOSPC;
 
-	sb = dir->i_sb;
-	inode = hfsplus_new_inode(sb, S_IFLNK | S_IRWXUGO);
+	mutex_lock(&sbi->vh_mutex);
+	inode = hfsplus_new_inode(dir->i_sb, S_IFLNK | S_IRWXUGO);
 	if (!inode)
-		return -ENOSPC;
+		goto out;
 
 	res = page_symlink(inode, symname, strlen(symname) + 1);
-	if (res) {
-		inode->i_nlink = 0;
-		hfsplus_delete_inode(inode);
-		iput(inode);
-		return res;
-	}
+	if (res)
+		goto out_err;
 
-	mark_inode_dirty(inode);
 	res = hfsplus_create_cat(inode->i_ino, dir, &dentry->d_name, inode);
+	if (res)
+		goto out_err;
 
-	if (!res) {
-		hfsplus_instantiate(dentry, inode, inode->i_ino);
-		mark_inode_dirty(inode);
-	}
+	hfsplus_instantiate(dentry, inode, inode->i_ino);
+	mark_inode_dirty(inode);
+	goto out;
 
+out_err:
+	inode->i_nlink = 0;
+	hfsplus_delete_inode(inode);
+	iput(inode);
+out:
+	mutex_unlock(&sbi->vh_mutex);
 	return res;
 }
 
 static int hfsplus_mknod(struct inode *dir, struct dentry *dentry,
 			 int mode, dev_t rdev)
 {
-	struct super_block *sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb);
 	struct inode *inode;
-	int res;
+	int res = -ENOSPC;
 
-	sb = dir->i_sb;
-	inode = hfsplus_new_inode(sb, mode);
+	mutex_lock(&sbi->vh_mutex);
+	inode = hfsplus_new_inode(dir->i_sb, mode);
 	if (!inode)
-		return -ENOSPC;
+		goto out;
+
+	if (S_ISBLK(mode) || S_ISCHR(mode) || S_ISFIFO(mode) || S_ISSOCK(mode))
+		init_special_inode(inode, mode, rdev);
 
 	res = hfsplus_create_cat(inode->i_ino, dir, &dentry->d_name, inode);
 	if (res) {
 		inode->i_nlink = 0;
 		hfsplus_delete_inode(inode);
 		iput(inode);
-		return res;
+		goto out;
 	}
-	init_special_inode(inode, mode, rdev);
+
 	hfsplus_instantiate(dentry, inode, inode->i_ino);
 	mark_inode_dirty(inode);
+out:
+	mutex_unlock(&sbi->vh_mutex);
+	return res;
+}
 
-	return 0;
+static int hfsplus_create(struct inode *dir, struct dentry *dentry, int mode,
+			  struct nameidata *nd)
+{
+	return hfsplus_mknod(dir, dentry, mode, 0);
+}
+
+static int hfsplus_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+	return hfsplus_mknod(dir, dentry, mode | S_IFDIR, 0);
 }
 
 static int hfsplus_rename(struct inode *old_dir, struct dentry *old_dentry,
@@ -466,7 +454,10 @@
 
 	/* Unlink destination if it already exists */
 	if (new_dentry->d_inode) {
-		res = hfsplus_unlink(new_dir, new_dentry);
+		if (S_ISDIR(new_dentry->d_inode->i_mode))
+			res = hfsplus_rmdir(new_dir, new_dentry);
+		else
+			res = hfsplus_unlink(new_dir, new_dentry);
 		if (res)
 			return res;
 	}
diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index 0022eec..0c9cb18 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -85,33 +85,47 @@
 
 static void __hfsplus_ext_write_extent(struct inode *inode, struct hfs_find_data *fd)
 {
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	int res;
 
-	hfsplus_ext_build_key(fd->search_key, inode->i_ino, HFSPLUS_I(inode).cached_start,
-			      HFSPLUS_IS_RSRC(inode) ?  HFSPLUS_TYPE_RSRC : HFSPLUS_TYPE_DATA);
+	WARN_ON(!mutex_is_locked(&hip->extents_lock));
+
+	hfsplus_ext_build_key(fd->search_key, inode->i_ino, hip->cached_start,
+			      HFSPLUS_IS_RSRC(inode) ?
+				HFSPLUS_TYPE_RSRC : HFSPLUS_TYPE_DATA);
+
 	res = hfs_brec_find(fd);
-	if (HFSPLUS_I(inode).flags & HFSPLUS_FLG_EXT_NEW) {
+	if (hip->flags & HFSPLUS_FLG_EXT_NEW) {
 		if (res != -ENOENT)
 			return;
-		hfs_brec_insert(fd, HFSPLUS_I(inode).cached_extents, sizeof(hfsplus_extent_rec));
-		HFSPLUS_I(inode).flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
+		hfs_brec_insert(fd, hip->cached_extents,
+				sizeof(hfsplus_extent_rec));
+		hip->flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
 	} else {
 		if (res)
 			return;
-		hfs_bnode_write(fd->bnode, HFSPLUS_I(inode).cached_extents, fd->entryoffset, fd->entrylength);
-		HFSPLUS_I(inode).flags &= ~HFSPLUS_FLG_EXT_DIRTY;
+		hfs_bnode_write(fd->bnode, hip->cached_extents,
+				fd->entryoffset, fd->entrylength);
+		hip->flags &= ~HFSPLUS_FLG_EXT_DIRTY;
+	}
+}
+
+static void hfsplus_ext_write_extent_locked(struct inode *inode)
+{
+	if (HFSPLUS_I(inode)->flags & HFSPLUS_FLG_EXT_DIRTY) {
+		struct hfs_find_data fd;
+
+		hfs_find_init(HFSPLUS_SB(inode->i_sb)->ext_tree, &fd);
+		__hfsplus_ext_write_extent(inode, &fd);
+		hfs_find_exit(&fd);
 	}
 }
 
 void hfsplus_ext_write_extent(struct inode *inode)
 {
-	if (HFSPLUS_I(inode).flags & HFSPLUS_FLG_EXT_DIRTY) {
-		struct hfs_find_data fd;
-
-		hfs_find_init(HFSPLUS_SB(inode->i_sb).ext_tree, &fd);
-		__hfsplus_ext_write_extent(inode, &fd);
-		hfs_find_exit(&fd);
-	}
+	mutex_lock(&HFSPLUS_I(inode)->extents_lock);
+	hfsplus_ext_write_extent_locked(inode);
+	mutex_unlock(&HFSPLUS_I(inode)->extents_lock);
 }
 
 static inline int __hfsplus_ext_read_extent(struct hfs_find_data *fd,
@@ -136,33 +150,39 @@
 
 static inline int __hfsplus_ext_cache_extent(struct hfs_find_data *fd, struct inode *inode, u32 block)
 {
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	int res;
 
-	if (HFSPLUS_I(inode).flags & HFSPLUS_FLG_EXT_DIRTY)
+	WARN_ON(!mutex_is_locked(&hip->extents_lock));
+
+	if (hip->flags & HFSPLUS_FLG_EXT_DIRTY)
 		__hfsplus_ext_write_extent(inode, fd);
 
-	res = __hfsplus_ext_read_extent(fd, HFSPLUS_I(inode).cached_extents, inode->i_ino,
-					block, HFSPLUS_IS_RSRC(inode) ? HFSPLUS_TYPE_RSRC : HFSPLUS_TYPE_DATA);
+	res = __hfsplus_ext_read_extent(fd, hip->cached_extents, inode->i_ino,
+					block, HFSPLUS_IS_RSRC(inode) ?
+						HFSPLUS_TYPE_RSRC :
+						HFSPLUS_TYPE_DATA);
 	if (!res) {
-		HFSPLUS_I(inode).cached_start = be32_to_cpu(fd->key->ext.start_block);
-		HFSPLUS_I(inode).cached_blocks = hfsplus_ext_block_count(HFSPLUS_I(inode).cached_extents);
+		hip->cached_start = be32_to_cpu(fd->key->ext.start_block);
+		hip->cached_blocks = hfsplus_ext_block_count(hip->cached_extents);
 	} else {
-		HFSPLUS_I(inode).cached_start = HFSPLUS_I(inode).cached_blocks = 0;
-		HFSPLUS_I(inode).flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
+		hip->cached_start = hip->cached_blocks = 0;
+		hip->flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
 	}
 	return res;
 }
 
 static int hfsplus_ext_read_extent(struct inode *inode, u32 block)
 {
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	struct hfs_find_data fd;
 	int res;
 
-	if (block >= HFSPLUS_I(inode).cached_start &&
-	    block < HFSPLUS_I(inode).cached_start + HFSPLUS_I(inode).cached_blocks)
+	if (block >= hip->cached_start &&
+	    block < hip->cached_start + hip->cached_blocks)
 		return 0;
 
-	hfs_find_init(HFSPLUS_SB(inode->i_sb).ext_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(inode->i_sb)->ext_tree, &fd);
 	res = __hfsplus_ext_cache_extent(&fd, inode, block);
 	hfs_find_exit(&fd);
 	return res;
@@ -172,21 +192,21 @@
 int hfsplus_get_block(struct inode *inode, sector_t iblock,
 		      struct buffer_head *bh_result, int create)
 {
-	struct super_block *sb;
+	struct super_block *sb = inode->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	int res = -EIO;
 	u32 ablock, dblock, mask;
 	int shift;
 
-	sb = inode->i_sb;
-
 	/* Convert inode block to disk allocation block */
-	shift = HFSPLUS_SB(sb).alloc_blksz_shift - sb->s_blocksize_bits;
-	ablock = iblock >> HFSPLUS_SB(sb).fs_shift;
+	shift = sbi->alloc_blksz_shift - sb->s_blocksize_bits;
+	ablock = iblock >> sbi->fs_shift;
 
-	if (iblock >= HFSPLUS_I(inode).fs_blocks) {
-		if (iblock > HFSPLUS_I(inode).fs_blocks || !create)
+	if (iblock >= hip->fs_blocks) {
+		if (iblock > hip->fs_blocks || !create)
 			return -EIO;
-		if (ablock >= HFSPLUS_I(inode).alloc_blocks) {
+		if (ablock >= hip->alloc_blocks) {
 			res = hfsplus_file_extend(inode);
 			if (res)
 				return res;
@@ -194,33 +214,33 @@
 	} else
 		create = 0;
 
-	if (ablock < HFSPLUS_I(inode).first_blocks) {
-		dblock = hfsplus_ext_find_block(HFSPLUS_I(inode).first_extents, ablock);
+	if (ablock < hip->first_blocks) {
+		dblock = hfsplus_ext_find_block(hip->first_extents, ablock);
 		goto done;
 	}
 
 	if (inode->i_ino == HFSPLUS_EXT_CNID)
 		return -EIO;
 
-	mutex_lock(&HFSPLUS_I(inode).extents_lock);
+	mutex_lock(&hip->extents_lock);
 	res = hfsplus_ext_read_extent(inode, ablock);
 	if (!res) {
-		dblock = hfsplus_ext_find_block(HFSPLUS_I(inode).cached_extents, ablock -
-					     HFSPLUS_I(inode).cached_start);
+		dblock = hfsplus_ext_find_block(hip->cached_extents,
+						ablock - hip->cached_start);
 	} else {
-		mutex_unlock(&HFSPLUS_I(inode).extents_lock);
+		mutex_unlock(&hip->extents_lock);
 		return -EIO;
 	}
-	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&hip->extents_lock);
 
 done:
 	dprint(DBG_EXTENT, "get_block(%lu): %llu - %u\n", inode->i_ino, (long long)iblock, dblock);
-	mask = (1 << HFSPLUS_SB(sb).fs_shift) - 1;
-	map_bh(bh_result, sb, (dblock << HFSPLUS_SB(sb).fs_shift) + HFSPLUS_SB(sb).blockoffset + (iblock & mask));
+	mask = (1 << sbi->fs_shift) - 1;
+	map_bh(bh_result, sb, (dblock << sbi->fs_shift) + sbi->blockoffset + (iblock & mask));
 	if (create) {
 		set_buffer_new(bh_result);
-		HFSPLUS_I(inode).phys_size += sb->s_blocksize;
-		HFSPLUS_I(inode).fs_blocks++;
+		hip->phys_size += sb->s_blocksize;
+		hip->fs_blocks++;
 		inode_add_bytes(inode, sb->s_blocksize);
 		mark_inode_dirty(inode);
 	}
@@ -327,7 +347,7 @@
 	if (total_blocks == blocks)
 		return 0;
 
-	hfs_find_init(HFSPLUS_SB(sb).ext_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->ext_tree, &fd);
 	do {
 		res = __hfsplus_ext_read_extent(&fd, ext_entry, cnid,
 						total_blocks, type);
@@ -348,29 +368,33 @@
 int hfsplus_file_extend(struct inode *inode)
 {
 	struct super_block *sb = inode->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	u32 start, len, goal;
 	int res;
 
-	if (HFSPLUS_SB(sb).alloc_file->i_size * 8 < HFSPLUS_SB(sb).total_blocks - HFSPLUS_SB(sb).free_blocks + 8) {
+	if (sbi->alloc_file->i_size * 8 <
+	    sbi->total_blocks - sbi->free_blocks + 8) {
 		// extend alloc file
-		printk(KERN_ERR "hfs: extend alloc file! (%Lu,%u,%u)\n", HFSPLUS_SB(sb).alloc_file->i_size * 8,
-			HFSPLUS_SB(sb).total_blocks, HFSPLUS_SB(sb).free_blocks);
+		printk(KERN_ERR "hfs: extend alloc file! (%Lu,%u,%u)\n",
+				sbi->alloc_file->i_size * 8,
+				sbi->total_blocks, sbi->free_blocks);
 		return -ENOSPC;
 	}
 
-	mutex_lock(&HFSPLUS_I(inode).extents_lock);
-	if (HFSPLUS_I(inode).alloc_blocks == HFSPLUS_I(inode).first_blocks)
-		goal = hfsplus_ext_lastblock(HFSPLUS_I(inode).first_extents);
+	mutex_lock(&hip->extents_lock);
+	if (hip->alloc_blocks == hip->first_blocks)
+		goal = hfsplus_ext_lastblock(hip->first_extents);
 	else {
-		res = hfsplus_ext_read_extent(inode, HFSPLUS_I(inode).alloc_blocks);
+		res = hfsplus_ext_read_extent(inode, hip->alloc_blocks);
 		if (res)
 			goto out;
-		goal = hfsplus_ext_lastblock(HFSPLUS_I(inode).cached_extents);
+		goal = hfsplus_ext_lastblock(hip->cached_extents);
 	}
 
-	len = HFSPLUS_I(inode).clump_blocks;
-	start = hfsplus_block_allocate(sb, HFSPLUS_SB(sb).total_blocks, goal, &len);
-	if (start >= HFSPLUS_SB(sb).total_blocks) {
+	len = hip->clump_blocks;
+	start = hfsplus_block_allocate(sb, sbi->total_blocks, goal, &len);
+	if (start >= sbi->total_blocks) {
 		start = hfsplus_block_allocate(sb, goal, 0, &len);
 		if (start >= goal) {
 			res = -ENOSPC;
@@ -379,56 +403,56 @@
 	}
 
 	dprint(DBG_EXTENT, "extend %lu: %u,%u\n", inode->i_ino, start, len);
-	if (HFSPLUS_I(inode).alloc_blocks <= HFSPLUS_I(inode).first_blocks) {
-		if (!HFSPLUS_I(inode).first_blocks) {
+
+	if (hip->alloc_blocks <= hip->first_blocks) {
+		if (!hip->first_blocks) {
 			dprint(DBG_EXTENT, "first extents\n");
 			/* no extents yet */
-			HFSPLUS_I(inode).first_extents[0].start_block = cpu_to_be32(start);
-			HFSPLUS_I(inode).first_extents[0].block_count = cpu_to_be32(len);
+			hip->first_extents[0].start_block = cpu_to_be32(start);
+			hip->first_extents[0].block_count = cpu_to_be32(len);
 			res = 0;
 		} else {
 			/* try to append to extents in inode */
-			res = hfsplus_add_extent(HFSPLUS_I(inode).first_extents,
-						 HFSPLUS_I(inode).alloc_blocks,
+			res = hfsplus_add_extent(hip->first_extents,
+						 hip->alloc_blocks,
 						 start, len);
 			if (res == -ENOSPC)
 				goto insert_extent;
 		}
 		if (!res) {
-			hfsplus_dump_extent(HFSPLUS_I(inode).first_extents);
-			HFSPLUS_I(inode).first_blocks += len;
+			hfsplus_dump_extent(hip->first_extents);
+			hip->first_blocks += len;
 		}
 	} else {
-		res = hfsplus_add_extent(HFSPLUS_I(inode).cached_extents,
-					 HFSPLUS_I(inode).alloc_blocks -
-					 HFSPLUS_I(inode).cached_start,
+		res = hfsplus_add_extent(hip->cached_extents,
+					 hip->alloc_blocks - hip->cached_start,
 					 start, len);
 		if (!res) {
-			hfsplus_dump_extent(HFSPLUS_I(inode).cached_extents);
-			HFSPLUS_I(inode).flags |= HFSPLUS_FLG_EXT_DIRTY;
-			HFSPLUS_I(inode).cached_blocks += len;
+			hfsplus_dump_extent(hip->cached_extents);
+			hip->flags |= HFSPLUS_FLG_EXT_DIRTY;
+			hip->cached_blocks += len;
 		} else if (res == -ENOSPC)
 			goto insert_extent;
 	}
 out:
-	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&hip->extents_lock);
 	if (!res) {
-		HFSPLUS_I(inode).alloc_blocks += len;
+		hip->alloc_blocks += len;
 		mark_inode_dirty(inode);
 	}
 	return res;
 
 insert_extent:
 	dprint(DBG_EXTENT, "insert new extent\n");
-	hfsplus_ext_write_extent(inode);
+	hfsplus_ext_write_extent_locked(inode);
 
-	memset(HFSPLUS_I(inode).cached_extents, 0, sizeof(hfsplus_extent_rec));
-	HFSPLUS_I(inode).cached_extents[0].start_block = cpu_to_be32(start);
-	HFSPLUS_I(inode).cached_extents[0].block_count = cpu_to_be32(len);
-	hfsplus_dump_extent(HFSPLUS_I(inode).cached_extents);
-	HFSPLUS_I(inode).flags |= HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW;
-	HFSPLUS_I(inode).cached_start = HFSPLUS_I(inode).alloc_blocks;
-	HFSPLUS_I(inode).cached_blocks = len;
+	memset(hip->cached_extents, 0, sizeof(hfsplus_extent_rec));
+	hip->cached_extents[0].start_block = cpu_to_be32(start);
+	hip->cached_extents[0].block_count = cpu_to_be32(len);
+	hfsplus_dump_extent(hip->cached_extents);
+	hip->flags |= HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW;
+	hip->cached_start = hip->alloc_blocks;
+	hip->cached_blocks = len;
 
 	res = 0;
 	goto out;
@@ -437,13 +461,15 @@
 void hfsplus_file_truncate(struct inode *inode)
 {
 	struct super_block *sb = inode->i_sb;
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	struct hfs_find_data fd;
 	u32 alloc_cnt, blk_cnt, start;
 	int res;
 
-	dprint(DBG_INODE, "truncate: %lu, %Lu -> %Lu\n", inode->i_ino,
-	       (long long)HFSPLUS_I(inode).phys_size, inode->i_size);
-	if (inode->i_size > HFSPLUS_I(inode).phys_size) {
+	dprint(DBG_INODE, "truncate: %lu, %Lu -> %Lu\n",
+		inode->i_ino, (long long)hip->phys_size, inode->i_size);
+
+	if (inode->i_size > hip->phys_size) {
 		struct address_space *mapping = inode->i_mapping;
 		struct page *page;
 		void *fsdata;
@@ -460,47 +486,48 @@
 			return;
 		mark_inode_dirty(inode);
 		return;
-	} else if (inode->i_size == HFSPLUS_I(inode).phys_size)
+	} else if (inode->i_size == hip->phys_size)
 		return;
 
-	blk_cnt = (inode->i_size + HFSPLUS_SB(sb).alloc_blksz - 1) >> HFSPLUS_SB(sb).alloc_blksz_shift;
-	alloc_cnt = HFSPLUS_I(inode).alloc_blocks;
+	blk_cnt = (inode->i_size + HFSPLUS_SB(sb)->alloc_blksz - 1) >>
+			HFSPLUS_SB(sb)->alloc_blksz_shift;
+	alloc_cnt = hip->alloc_blocks;
 	if (blk_cnt == alloc_cnt)
 		goto out;
 
-	mutex_lock(&HFSPLUS_I(inode).extents_lock);
-	hfs_find_init(HFSPLUS_SB(sb).ext_tree, &fd);
+	mutex_lock(&hip->extents_lock);
+	hfs_find_init(HFSPLUS_SB(sb)->ext_tree, &fd);
 	while (1) {
-		if (alloc_cnt == HFSPLUS_I(inode).first_blocks) {
-			hfsplus_free_extents(sb, HFSPLUS_I(inode).first_extents,
+		if (alloc_cnt == hip->first_blocks) {
+			hfsplus_free_extents(sb, hip->first_extents,
 					     alloc_cnt, alloc_cnt - blk_cnt);
-			hfsplus_dump_extent(HFSPLUS_I(inode).first_extents);
-			HFSPLUS_I(inode).first_blocks = blk_cnt;
+			hfsplus_dump_extent(hip->first_extents);
+			hip->first_blocks = blk_cnt;
 			break;
 		}
 		res = __hfsplus_ext_cache_extent(&fd, inode, alloc_cnt);
 		if (res)
 			break;
-		start = HFSPLUS_I(inode).cached_start;
-		hfsplus_free_extents(sb, HFSPLUS_I(inode).cached_extents,
+		start = hip->cached_start;
+		hfsplus_free_extents(sb, hip->cached_extents,
 				     alloc_cnt - start, alloc_cnt - blk_cnt);
-		hfsplus_dump_extent(HFSPLUS_I(inode).cached_extents);
+		hfsplus_dump_extent(hip->cached_extents);
 		if (blk_cnt > start) {
-			HFSPLUS_I(inode).flags |= HFSPLUS_FLG_EXT_DIRTY;
+			hip->flags |= HFSPLUS_FLG_EXT_DIRTY;
 			break;
 		}
 		alloc_cnt = start;
-		HFSPLUS_I(inode).cached_start = HFSPLUS_I(inode).cached_blocks = 0;
-		HFSPLUS_I(inode).flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
+		hip->cached_start = hip->cached_blocks = 0;
+		hip->flags &= ~(HFSPLUS_FLG_EXT_DIRTY | HFSPLUS_FLG_EXT_NEW);
 		hfs_brec_remove(&fd);
 	}
 	hfs_find_exit(&fd);
-	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&hip->extents_lock);
 
-	HFSPLUS_I(inode).alloc_blocks = blk_cnt;
+	hip->alloc_blocks = blk_cnt;
 out:
-	HFSPLUS_I(inode).phys_size = inode->i_size;
-	HFSPLUS_I(inode).fs_blocks = (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
-	inode_set_bytes(inode, HFSPLUS_I(inode).fs_blocks << sb->s_blocksize_bits);
+	hip->phys_size = inode->i_size;
+	hip->fs_blocks = (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
+	inode_set_bytes(inode, hip->fs_blocks << sb->s_blocksize_bits);
 	mark_inode_dirty(inode);
 }
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index dc856be..cb3653e 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -62,7 +62,7 @@
 	unsigned int depth;
 
 	//unsigned int map1_size, map_size;
-	struct semaphore tree_lock;
+	struct mutex tree_lock;
 
 	unsigned int pages_per_bnode;
 	spinlock_t hash_lock;
@@ -121,16 +121,21 @@
 	u32 sect_count;
 	int fs_shift;
 
-	/* Stuff in host order from Vol Header */
+	/* immutable data from the volume header */
 	u32 alloc_blksz;
 	int alloc_blksz_shift;
 	u32 total_blocks;
+	u32 data_clump_blocks, rsrc_clump_blocks;
+
+	/* mutable data from the volume header, protected by alloc_mutex */
 	u32 free_blocks;
-	u32 next_alloc;
+	struct mutex alloc_mutex;
+
+	/* mutable data from the volume header, protected by vh_mutex */
 	u32 next_cnid;
 	u32 file_count;
 	u32 folder_count;
-	u32 data_clump_blocks, rsrc_clump_blocks;
+	struct mutex vh_mutex;
 
 	/* Config options */
 	u32 creator;
@@ -143,40 +148,50 @@
 	int part, session;
 
 	unsigned long flags;
-
-	struct hlist_head rsrc_inodes;
 };
 
-#define HFSPLUS_SB_WRITEBACKUP	0x0001
-#define HFSPLUS_SB_NODECOMPOSE	0x0002
-#define HFSPLUS_SB_FORCE	0x0004
-#define HFSPLUS_SB_HFSX		0x0008
-#define HFSPLUS_SB_CASEFOLD	0x0010
+#define HFSPLUS_SB_WRITEBACKUP	0
+#define HFSPLUS_SB_NODECOMPOSE	1
+#define HFSPLUS_SB_FORCE	2
+#define HFSPLUS_SB_HFSX		3
+#define HFSPLUS_SB_CASEFOLD	4
 
 
 struct hfsplus_inode_info {
-	struct mutex extents_lock;
-	u32 clump_blocks, alloc_blocks;
-	sector_t fs_blocks;
-	/* Allocation extents from catalog record or volume header */
-	hfsplus_extent_rec first_extents;
-	u32 first_blocks;
-	hfsplus_extent_rec cached_extents;
-	u32 cached_start, cached_blocks;
 	atomic_t opencnt;
 
-	struct inode *rsrc_inode;
+	/*
+	 * Extent allocation information, protected by extents_lock.
+	 */
+	u32 first_blocks;
+	u32 clump_blocks;
+	u32 alloc_blocks;
+	u32 cached_start;
+	u32 cached_blocks;
+	hfsplus_extent_rec first_extents;
+	hfsplus_extent_rec cached_extents;
 	unsigned long flags;
+	struct mutex extents_lock;
 
+	/*
+	 * Immutable data.
+	 */
+	struct inode *rsrc_inode;
 	__be32 create_date;
-	/* Device number in hfsplus_permissions in catalog */
-	u32 dev;
-	/* BSD system and user file flags */
-	u8 rootflags;
-	u8 userflags;
 
+	/*
+	 * Protected by sbi->vh_mutex.
+	 */
+	u32 linkid;
+
+	/*
+	 * Protected by i_mutex.
+	 */
+	sector_t fs_blocks;
+	u8 userflags;		/* BSD user file flags */
 	struct list_head open_dir_list;
 	loff_t phys_size;
+
 	struct inode vfs_inode;
 };
 
@@ -184,8 +199,8 @@
 #define HFSPLUS_FLG_EXT_DIRTY	0x0002
 #define HFSPLUS_FLG_EXT_NEW	0x0004
 
-#define HFSPLUS_IS_DATA(inode)   (!(HFSPLUS_I(inode).flags & HFSPLUS_FLG_RSRC))
-#define HFSPLUS_IS_RSRC(inode)   (HFSPLUS_I(inode).flags & HFSPLUS_FLG_RSRC)
+#define HFSPLUS_IS_DATA(inode)   (!(HFSPLUS_I(inode)->flags & HFSPLUS_FLG_RSRC))
+#define HFSPLUS_IS_RSRC(inode)   (HFSPLUS_I(inode)->flags & HFSPLUS_FLG_RSRC)
 
 struct hfs_find_data {
 	/* filled by caller */
@@ -311,6 +326,7 @@
 int hfsplus_delete_cat(u32, struct inode *, struct qstr *);
 int hfsplus_rename_cat(u32, struct inode *, struct qstr *,
 		       struct inode *, struct qstr *);
+void hfsplus_cat_set_perms(struct inode *inode, struct hfsplus_perm *perms);
 
 /* dir.c */
 extern const struct inode_operations hfsplus_dir_inode_operations;
@@ -372,26 +388,15 @@
 int hfs_part_find(struct super_block *, sector_t *, sector_t *);
 
 /* access macros */
-/*
 static inline struct hfsplus_sb_info *HFSPLUS_SB(struct super_block *sb)
 {
 	return sb->s_fs_info;
 }
+
 static inline struct hfsplus_inode_info *HFSPLUS_I(struct inode *inode)
 {
 	return list_entry(inode, struct hfsplus_inode_info, vfs_inode);
 }
-*/
-#define HFSPLUS_SB(super)	(*(struct hfsplus_sb_info *)(super)->s_fs_info)
-#define HFSPLUS_I(inode)	(*list_entry(inode, struct hfsplus_inode_info, vfs_inode))
-
-#if 1
-#define hfsplus_kmap(p)		({ struct page *__p = (p); kmap(__p); })
-#define hfsplus_kunmap(p)	({ struct page *__p = (p); kunmap(__p); __p; })
-#else
-#define hfsplus_kmap(p)		kmap(p)
-#define hfsplus_kunmap(p)	kunmap(p)
-#endif
 
 #define sb_bread512(sb, sec, data) ({			\
 	struct buffer_head *__bh;			\
@@ -419,6 +424,4 @@
 #define hfsp_ut2mt(t)		__hfsp_ut2mt((t).tv_sec)
 #define hfsp_now2mt()		__hfsp_ut2mt(get_seconds())
 
-#define kdev_t_to_nr(x)		(x)
-
 #endif
diff --git a/fs/hfsplus/hfsplus_raw.h b/fs/hfsplus/hfsplus_raw.h
index fe99fe8..6892899 100644
--- a/fs/hfsplus/hfsplus_raw.h
+++ b/fs/hfsplus/hfsplus_raw.h
@@ -200,6 +200,7 @@
 	struct hfsplus_unistr name;
 } __packed;
 
+#define HFSPLUS_CAT_KEYLEN	(sizeof(struct hfsplus_cat_key))
 
 /* Structs from hfs.h */
 struct hfsp_point {
@@ -323,7 +324,7 @@
 	__be32 start_block;
 } __packed;
 
-#define HFSPLUS_EXT_KEYLEN 12
+#define HFSPLUS_EXT_KEYLEN	sizeof(struct hfsplus_ext_key)
 
 /* HFS+ generic BTree key */
 typedef union {
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index c5a979d..7844928 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -36,7 +36,7 @@
 	*pagep = NULL;
 	ret = cont_write_begin(file, mapping, pos, len, flags, pagep, fsdata,
 				hfsplus_get_block,
-				&HFSPLUS_I(mapping->host).phys_size);
+				&HFSPLUS_I(mapping->host)->phys_size);
 	if (unlikely(ret)) {
 		loff_t isize = mapping->host->i_size;
 		if (pos + len > isize)
@@ -62,13 +62,13 @@
 
 	switch (inode->i_ino) {
 	case HFSPLUS_EXT_CNID:
-		tree = HFSPLUS_SB(sb).ext_tree;
+		tree = HFSPLUS_SB(sb)->ext_tree;
 		break;
 	case HFSPLUS_CAT_CNID:
-		tree = HFSPLUS_SB(sb).cat_tree;
+		tree = HFSPLUS_SB(sb)->cat_tree;
 		break;
 	case HFSPLUS_ATTR_CNID:
-		tree = HFSPLUS_SB(sb).attr_tree;
+		tree = HFSPLUS_SB(sb)->attr_tree;
 		break;
 	default:
 		BUG();
@@ -172,12 +172,13 @@
 	struct hfs_find_data fd;
 	struct super_block *sb = dir->i_sb;
 	struct inode *inode = NULL;
+	struct hfsplus_inode_info *hip;
 	int err;
 
 	if (HFSPLUS_IS_RSRC(dir) || strcmp(dentry->d_name.name, "rsrc"))
 		goto out;
 
-	inode = HFSPLUS_I(dir).rsrc_inode;
+	inode = HFSPLUS_I(dir)->rsrc_inode;
 	if (inode)
 		goto out;
 
@@ -185,12 +186,13 @@
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 
+	hip = HFSPLUS_I(inode);
 	inode->i_ino = dir->i_ino;
-	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	mutex_init(&HFSPLUS_I(inode).extents_lock);
-	HFSPLUS_I(inode).flags = HFSPLUS_FLG_RSRC;
+	INIT_LIST_HEAD(&hip->open_dir_list);
+	mutex_init(&hip->extents_lock);
+	hip->flags = HFSPLUS_FLG_RSRC;
 
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
 	err = hfsplus_find_cat(sb, dir->i_ino, &fd);
 	if (!err)
 		err = hfsplus_cat_read_inode(inode, &fd);
@@ -199,10 +201,18 @@
 		iput(inode);
 		return ERR_PTR(err);
 	}
-	HFSPLUS_I(inode).rsrc_inode = dir;
-	HFSPLUS_I(dir).rsrc_inode = inode;
+	hip->rsrc_inode = dir;
+	HFSPLUS_I(dir)->rsrc_inode = inode;
 	igrab(dir);
-	hlist_add_head(&inode->i_hash, &HFSPLUS_SB(sb).rsrc_inodes);
+
+	/*
+	 * __mark_inode_dirty expects inodes to be hashed.  Since we don't
+	 * want resource fork inodes in the regular inode space, we make them
+	 * appear hashed, but do not put on any lists.  hlist_del()
+	 * will work fine and require no locking.
+	 */
+	inode->i_hash.pprev = &inode->i_hash.next;
+
 	mark_inode_dirty(inode);
 out:
 	d_add(dentry, inode);
@@ -211,30 +221,27 @@
 
 static void hfsplus_get_perms(struct inode *inode, struct hfsplus_perm *perms, int dir)
 {
-	struct super_block *sb = inode->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(inode->i_sb);
 	u16 mode;
 
 	mode = be16_to_cpu(perms->mode);
 
 	inode->i_uid = be32_to_cpu(perms->owner);
 	if (!inode->i_uid && !mode)
-		inode->i_uid = HFSPLUS_SB(sb).uid;
+		inode->i_uid = sbi->uid;
 
 	inode->i_gid = be32_to_cpu(perms->group);
 	if (!inode->i_gid && !mode)
-		inode->i_gid = HFSPLUS_SB(sb).gid;
+		inode->i_gid = sbi->gid;
 
 	if (dir) {
-		mode = mode ? (mode & S_IALLUGO) :
-			(S_IRWXUGO & ~(HFSPLUS_SB(sb).umask));
+		mode = mode ? (mode & S_IALLUGO) : (S_IRWXUGO & ~(sbi->umask));
 		mode |= S_IFDIR;
 	} else if (!mode)
-		mode = S_IFREG | ((S_IRUGO|S_IWUGO) &
-			~(HFSPLUS_SB(sb).umask));
+		mode = S_IFREG | ((S_IRUGO|S_IWUGO) & ~(sbi->umask));
 	inode->i_mode = mode;
 
-	HFSPLUS_I(inode).rootflags = perms->rootflags;
-	HFSPLUS_I(inode).userflags = perms->userflags;
+	HFSPLUS_I(inode)->userflags = perms->userflags;
 	if (perms->rootflags & HFSPLUS_FLG_IMMUTABLE)
 		inode->i_flags |= S_IMMUTABLE;
 	else
@@ -245,30 +252,13 @@
 		inode->i_flags &= ~S_APPEND;
 }
 
-static void hfsplus_set_perms(struct inode *inode, struct hfsplus_perm *perms)
-{
-	if (inode->i_flags & S_IMMUTABLE)
-		perms->rootflags |= HFSPLUS_FLG_IMMUTABLE;
-	else
-		perms->rootflags &= ~HFSPLUS_FLG_IMMUTABLE;
-	if (inode->i_flags & S_APPEND)
-		perms->rootflags |= HFSPLUS_FLG_APPEND;
-	else
-		perms->rootflags &= ~HFSPLUS_FLG_APPEND;
-	perms->userflags = HFSPLUS_I(inode).userflags;
-	perms->mode = cpu_to_be16(inode->i_mode);
-	perms->owner = cpu_to_be32(inode->i_uid);
-	perms->group = cpu_to_be32(inode->i_gid);
-	perms->dev = cpu_to_be32(HFSPLUS_I(inode).dev);
-}
-
 static int hfsplus_file_open(struct inode *inode, struct file *file)
 {
 	if (HFSPLUS_IS_RSRC(inode))
-		inode = HFSPLUS_I(inode).rsrc_inode;
+		inode = HFSPLUS_I(inode)->rsrc_inode;
 	if (!(file->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
 		return -EOVERFLOW;
-	atomic_inc(&HFSPLUS_I(inode).opencnt);
+	atomic_inc(&HFSPLUS_I(inode)->opencnt);
 	return 0;
 }
 
@@ -277,12 +267,13 @@
 	struct super_block *sb = inode->i_sb;
 
 	if (HFSPLUS_IS_RSRC(inode))
-		inode = HFSPLUS_I(inode).rsrc_inode;
-	if (atomic_dec_and_test(&HFSPLUS_I(inode).opencnt)) {
+		inode = HFSPLUS_I(inode)->rsrc_inode;
+	if (atomic_dec_and_test(&HFSPLUS_I(inode)->opencnt)) {
 		mutex_lock(&inode->i_mutex);
 		hfsplus_file_truncate(inode);
 		if (inode->i_flags & S_DEAD) {
-			hfsplus_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL);
+			hfsplus_delete_cat(inode->i_ino,
+					   HFSPLUS_SB(sb)->hidden_dir, NULL);
 			hfsplus_delete_inode(inode);
 		}
 		mutex_unlock(&inode->i_mutex);
@@ -361,47 +352,52 @@
 
 struct inode *hfsplus_new_inode(struct super_block *sb, int mode)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	struct inode *inode = new_inode(sb);
+	struct hfsplus_inode_info *hip;
+
 	if (!inode)
 		return NULL;
 
-	inode->i_ino = HFSPLUS_SB(sb).next_cnid++;
+	inode->i_ino = sbi->next_cnid++;
 	inode->i_mode = mode;
 	inode->i_uid = current_fsuid();
 	inode->i_gid = current_fsgid();
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	mutex_init(&HFSPLUS_I(inode).extents_lock);
-	atomic_set(&HFSPLUS_I(inode).opencnt, 0);
-	HFSPLUS_I(inode).flags = 0;
-	memset(HFSPLUS_I(inode).first_extents, 0, sizeof(hfsplus_extent_rec));
-	memset(HFSPLUS_I(inode).cached_extents, 0, sizeof(hfsplus_extent_rec));
-	HFSPLUS_I(inode).alloc_blocks = 0;
-	HFSPLUS_I(inode).first_blocks = 0;
-	HFSPLUS_I(inode).cached_start = 0;
-	HFSPLUS_I(inode).cached_blocks = 0;
-	HFSPLUS_I(inode).phys_size = 0;
-	HFSPLUS_I(inode).fs_blocks = 0;
-	HFSPLUS_I(inode).rsrc_inode = NULL;
+
+	hip = HFSPLUS_I(inode);
+	INIT_LIST_HEAD(&hip->open_dir_list);
+	mutex_init(&hip->extents_lock);
+	atomic_set(&hip->opencnt, 0);
+	hip->flags = 0;
+	memset(hip->first_extents, 0, sizeof(hfsplus_extent_rec));
+	memset(hip->cached_extents, 0, sizeof(hfsplus_extent_rec));
+	hip->alloc_blocks = 0;
+	hip->first_blocks = 0;
+	hip->cached_start = 0;
+	hip->cached_blocks = 0;
+	hip->phys_size = 0;
+	hip->fs_blocks = 0;
+	hip->rsrc_inode = NULL;
 	if (S_ISDIR(inode->i_mode)) {
 		inode->i_size = 2;
-		HFSPLUS_SB(sb).folder_count++;
+		sbi->folder_count++;
 		inode->i_op = &hfsplus_dir_inode_operations;
 		inode->i_fop = &hfsplus_dir_operations;
 	} else if (S_ISREG(inode->i_mode)) {
-		HFSPLUS_SB(sb).file_count++;
+		sbi->file_count++;
 		inode->i_op = &hfsplus_file_inode_operations;
 		inode->i_fop = &hfsplus_file_operations;
 		inode->i_mapping->a_ops = &hfsplus_aops;
-		HFSPLUS_I(inode).clump_blocks = HFSPLUS_SB(sb).data_clump_blocks;
+		hip->clump_blocks = sbi->data_clump_blocks;
 	} else if (S_ISLNK(inode->i_mode)) {
-		HFSPLUS_SB(sb).file_count++;
+		sbi->file_count++;
 		inode->i_op = &page_symlink_inode_operations;
 		inode->i_mapping->a_ops = &hfsplus_aops;
-		HFSPLUS_I(inode).clump_blocks = 1;
+		hip->clump_blocks = 1;
 	} else
-		HFSPLUS_SB(sb).file_count++;
+		sbi->file_count++;
 	insert_inode_hash(inode);
 	mark_inode_dirty(inode);
 	sb->s_dirt = 1;
@@ -414,11 +410,11 @@
 	struct super_block *sb = inode->i_sb;
 
 	if (S_ISDIR(inode->i_mode)) {
-		HFSPLUS_SB(sb).folder_count--;
+		HFSPLUS_SB(sb)->folder_count--;
 		sb->s_dirt = 1;
 		return;
 	}
-	HFSPLUS_SB(sb).file_count--;
+	HFSPLUS_SB(sb)->file_count--;
 	if (S_ISREG(inode->i_mode)) {
 		if (!inode->i_nlink) {
 			inode->i_size = 0;
@@ -434,34 +430,39 @@
 void hfsplus_inode_read_fork(struct inode *inode, struct hfsplus_fork_raw *fork)
 {
 	struct super_block *sb = inode->i_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
 	u32 count;
 	int i;
 
-	memcpy(&HFSPLUS_I(inode).first_extents, &fork->extents,
-	       sizeof(hfsplus_extent_rec));
+	memcpy(&hip->first_extents, &fork->extents, sizeof(hfsplus_extent_rec));
 	for (count = 0, i = 0; i < 8; i++)
 		count += be32_to_cpu(fork->extents[i].block_count);
-	HFSPLUS_I(inode).first_blocks = count;
-	memset(HFSPLUS_I(inode).cached_extents, 0, sizeof(hfsplus_extent_rec));
-	HFSPLUS_I(inode).cached_start = 0;
-	HFSPLUS_I(inode).cached_blocks = 0;
+	hip->first_blocks = count;
+	memset(hip->cached_extents, 0, sizeof(hfsplus_extent_rec));
+	hip->cached_start = 0;
+	hip->cached_blocks = 0;
 
-	HFSPLUS_I(inode).alloc_blocks = be32_to_cpu(fork->total_blocks);
-	inode->i_size = HFSPLUS_I(inode).phys_size = be64_to_cpu(fork->total_size);
-	HFSPLUS_I(inode).fs_blocks = (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
-	inode_set_bytes(inode, HFSPLUS_I(inode).fs_blocks << sb->s_blocksize_bits);
-	HFSPLUS_I(inode).clump_blocks = be32_to_cpu(fork->clump_size) >> HFSPLUS_SB(sb).alloc_blksz_shift;
-	if (!HFSPLUS_I(inode).clump_blocks)
-		HFSPLUS_I(inode).clump_blocks = HFSPLUS_IS_RSRC(inode) ? HFSPLUS_SB(sb).rsrc_clump_blocks :
-				HFSPLUS_SB(sb).data_clump_blocks;
+	hip->alloc_blocks = be32_to_cpu(fork->total_blocks);
+	hip->phys_size = inode->i_size = be64_to_cpu(fork->total_size);
+	hip->fs_blocks =
+		(inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
+	inode_set_bytes(inode, hip->fs_blocks << sb->s_blocksize_bits);
+	hip->clump_blocks =
+		be32_to_cpu(fork->clump_size) >> sbi->alloc_blksz_shift;
+	if (!hip->clump_blocks) {
+		hip->clump_blocks = HFSPLUS_IS_RSRC(inode) ?
+			sbi->rsrc_clump_blocks :
+			sbi->data_clump_blocks;
+	}
 }
 
 void hfsplus_inode_write_fork(struct inode *inode, struct hfsplus_fork_raw *fork)
 {
-	memcpy(&fork->extents, &HFSPLUS_I(inode).first_extents,
+	memcpy(&fork->extents, &HFSPLUS_I(inode)->first_extents,
 	       sizeof(hfsplus_extent_rec));
 	fork->total_size = cpu_to_be64(inode->i_size);
-	fork->total_blocks = cpu_to_be32(HFSPLUS_I(inode).alloc_blocks);
+	fork->total_blocks = cpu_to_be32(HFSPLUS_I(inode)->alloc_blocks);
 }
 
 int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
@@ -472,7 +473,7 @@
 
 	type = hfs_bnode_read_u16(fd->bnode, fd->entryoffset);
 
-	HFSPLUS_I(inode).dev = 0;
+	HFSPLUS_I(inode)->linkid = 0;
 	if (type == HFSPLUS_FOLDER) {
 		struct hfsplus_cat_folder *folder = &entry.folder;
 
@@ -486,8 +487,8 @@
 		inode->i_atime = hfsp_mt2ut(folder->access_date);
 		inode->i_mtime = hfsp_mt2ut(folder->content_mod_date);
 		inode->i_ctime = hfsp_mt2ut(folder->attribute_mod_date);
-		HFSPLUS_I(inode).create_date = folder->create_date;
-		HFSPLUS_I(inode).fs_blocks = 0;
+		HFSPLUS_I(inode)->create_date = folder->create_date;
+		HFSPLUS_I(inode)->fs_blocks = 0;
 		inode->i_op = &hfsplus_dir_inode_operations;
 		inode->i_fop = &hfsplus_dir_operations;
 	} else if (type == HFSPLUS_FILE) {
@@ -518,7 +519,7 @@
 		inode->i_atime = hfsp_mt2ut(file->access_date);
 		inode->i_mtime = hfsp_mt2ut(file->content_mod_date);
 		inode->i_ctime = hfsp_mt2ut(file->attribute_mod_date);
-		HFSPLUS_I(inode).create_date = file->create_date;
+		HFSPLUS_I(inode)->create_date = file->create_date;
 	} else {
 		printk(KERN_ERR "hfs: bad catalog entry used to create inode\n");
 		res = -EIO;
@@ -533,12 +534,12 @@
 	hfsplus_cat_entry entry;
 
 	if (HFSPLUS_IS_RSRC(inode))
-		main_inode = HFSPLUS_I(inode).rsrc_inode;
+		main_inode = HFSPLUS_I(inode)->rsrc_inode;
 
 	if (!main_inode->i_nlink)
 		return 0;
 
-	if (hfs_find_init(HFSPLUS_SB(main_inode->i_sb).cat_tree, &fd))
+	if (hfs_find_init(HFSPLUS_SB(main_inode->i_sb)->cat_tree, &fd))
 		/* panic? */
 		return -EIO;
 
@@ -554,7 +555,7 @@
 		hfs_bnode_read(fd.bnode, &entry, fd.entryoffset,
 					sizeof(struct hfsplus_cat_folder));
 		/* simple node checks? */
-		hfsplus_set_perms(inode, &folder->permissions);
+		hfsplus_cat_set_perms(inode, &folder->permissions);
 		folder->access_date = hfsp_ut2mt(inode->i_atime);
 		folder->content_mod_date = hfsp_ut2mt(inode->i_mtime);
 		folder->attribute_mod_date = hfsp_ut2mt(inode->i_ctime);
@@ -576,11 +577,7 @@
 		hfs_bnode_read(fd.bnode, &entry, fd.entryoffset,
 					sizeof(struct hfsplus_cat_file));
 		hfsplus_inode_write_fork(inode, &file->data_fork);
-		if (S_ISREG(inode->i_mode))
-			HFSPLUS_I(inode).dev = inode->i_nlink;
-		if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))
-			HFSPLUS_I(inode).dev = kdev_t_to_nr(inode->i_rdev);
-		hfsplus_set_perms(inode, &file->permissions);
+		hfsplus_cat_set_perms(inode, &file->permissions);
 		if ((file->permissions.rootflags | file->permissions.userflags) & HFSPLUS_FLG_IMMUTABLE)
 			file->flags |= cpu_to_be16(HFSPLUS_FILE_LOCKED);
 		else
diff --git a/fs/hfsplus/ioctl.c b/fs/hfsplus/ioctl.c
index ac405f0..5b4667e 100644
--- a/fs/hfsplus/ioctl.c
+++ b/fs/hfsplus/ioctl.c
@@ -17,83 +17,98 @@
 #include <linux/mount.h>
 #include <linux/sched.h>
 #include <linux/xattr.h>
-#include <linux/smp_lock.h>
 #include <asm/uaccess.h>
 #include "hfsplus_fs.h"
 
-long hfsplus_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+static int hfsplus_ioctl_getflags(struct file *file, int __user *user_flags)
 {
-	struct inode *inode = filp->f_path.dentry->d_inode;
-	unsigned int flags;
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
+	unsigned int flags = 0;
 
-	lock_kernel();
+	if (inode->i_flags & S_IMMUTABLE)
+		flags |= FS_IMMUTABLE_FL;
+	if (inode->i_flags |= S_APPEND)
+		flags |= FS_APPEND_FL;
+	if (hip->userflags & HFSPLUS_FLG_NODUMP)
+		flags |= FS_NODUMP_FL;
+
+	return put_user(flags, user_flags);
+}
+
+static int hfsplus_ioctl_setflags(struct file *file, int __user *user_flags)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct hfsplus_inode_info *hip = HFSPLUS_I(inode);
+	unsigned int flags;
+	int err = 0;
+
+	err = mnt_want_write(file->f_path.mnt);
+	if (err)
+		goto out;
+
+	if (!is_owner_or_cap(inode)) {
+		err = -EACCES;
+		goto out_drop_write;
+	}
+
+	if (get_user(flags, user_flags)) {
+		err = -EFAULT;
+		goto out_drop_write;
+	}
+
+	mutex_lock(&inode->i_mutex);
+
+	if ((flags & (FS_IMMUTABLE_FL|FS_APPEND_FL)) ||
+	    inode->i_flags & (S_IMMUTABLE|S_APPEND)) {
+		if (!capable(CAP_LINUX_IMMUTABLE)) {
+			err = -EPERM;
+			goto out_unlock_inode;
+		}
+	}
+
+	/* don't silently ignore unsupported ext2 flags */
+	if (flags & ~(FS_IMMUTABLE_FL|FS_APPEND_FL|FS_NODUMP_FL)) {
+		err = -EOPNOTSUPP;
+		goto out_unlock_inode;
+	}
+
+	if (flags & FS_IMMUTABLE_FL)
+		inode->i_flags |= S_IMMUTABLE;
+	else
+		inode->i_flags &= ~S_IMMUTABLE;
+
+	if (flags & FS_APPEND_FL)
+		inode->i_flags |= S_APPEND;
+	else
+		inode->i_flags &= ~S_APPEND;
+
+	if (flags & FS_NODUMP_FL)
+		hip->userflags |= HFSPLUS_FLG_NODUMP;
+	else
+		hip->userflags &= ~HFSPLUS_FLG_NODUMP;
+
+	inode->i_ctime = CURRENT_TIME_SEC;
+	mark_inode_dirty(inode);
+
+out_unlock_inode:
+	mutex_lock(&inode->i_mutex);
+out_drop_write:
+	mnt_drop_write(file->f_path.mnt);
+out:
+	return err;
+}
+
+long hfsplus_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+
 	switch (cmd) {
 	case HFSPLUS_IOC_EXT2_GETFLAGS:
-		flags = 0;
-		if (HFSPLUS_I(inode).rootflags & HFSPLUS_FLG_IMMUTABLE)
-			flags |= FS_IMMUTABLE_FL; /* EXT2_IMMUTABLE_FL */
-		if (HFSPLUS_I(inode).rootflags & HFSPLUS_FLG_APPEND)
-			flags |= FS_APPEND_FL; /* EXT2_APPEND_FL */
-		if (HFSPLUS_I(inode).userflags & HFSPLUS_FLG_NODUMP)
-			flags |= FS_NODUMP_FL; /* EXT2_NODUMP_FL */
-		return put_user(flags, (int __user *)arg);
-	case HFSPLUS_IOC_EXT2_SETFLAGS: {
-		int err = 0;
-		err = mnt_want_write(filp->f_path.mnt);
-		if (err) {
-			unlock_kernel();
-			return err;
-		}
-
-		if (!is_owner_or_cap(inode)) {
-			err = -EACCES;
-			goto setflags_out;
-		}
-		if (get_user(flags, (int __user *)arg)) {
-			err = -EFAULT;
-			goto setflags_out;
-		}
-		if (flags & (FS_IMMUTABLE_FL|FS_APPEND_FL) ||
-		    HFSPLUS_I(inode).rootflags & (HFSPLUS_FLG_IMMUTABLE|HFSPLUS_FLG_APPEND)) {
-			if (!capable(CAP_LINUX_IMMUTABLE)) {
-				err = -EPERM;
-				goto setflags_out;
-			}
-		}
-
-		/* don't silently ignore unsupported ext2 flags */
-		if (flags & ~(FS_IMMUTABLE_FL|FS_APPEND_FL|FS_NODUMP_FL)) {
-			err = -EOPNOTSUPP;
-			goto setflags_out;
-		}
-		if (flags & FS_IMMUTABLE_FL) { /* EXT2_IMMUTABLE_FL */
-			inode->i_flags |= S_IMMUTABLE;
-			HFSPLUS_I(inode).rootflags |= HFSPLUS_FLG_IMMUTABLE;
-		} else {
-			inode->i_flags &= ~S_IMMUTABLE;
-			HFSPLUS_I(inode).rootflags &= ~HFSPLUS_FLG_IMMUTABLE;
-		}
-		if (flags & FS_APPEND_FL) { /* EXT2_APPEND_FL */
-			inode->i_flags |= S_APPEND;
-			HFSPLUS_I(inode).rootflags |= HFSPLUS_FLG_APPEND;
-		} else {
-			inode->i_flags &= ~S_APPEND;
-			HFSPLUS_I(inode).rootflags &= ~HFSPLUS_FLG_APPEND;
-		}
-		if (flags & FS_NODUMP_FL) /* EXT2_NODUMP_FL */
-			HFSPLUS_I(inode).userflags |= HFSPLUS_FLG_NODUMP;
-		else
-			HFSPLUS_I(inode).userflags &= ~HFSPLUS_FLG_NODUMP;
-
-		inode->i_ctime = CURRENT_TIME_SEC;
-		mark_inode_dirty(inode);
-setflags_out:
-		mnt_drop_write(filp->f_path.mnt);
-		unlock_kernel();
-		return err;
-	}
+		return hfsplus_ioctl_getflags(file, argp);
+	case HFSPLUS_IOC_EXT2_SETFLAGS:
+		return hfsplus_ioctl_setflags(file, argp);
 	default:
-		unlock_kernel();
 		return -ENOTTY;
 	}
 }
@@ -110,7 +125,7 @@
 	if (!S_ISREG(inode->i_mode) || HFSPLUS_IS_RSRC(inode))
 		return -EOPNOTSUPP;
 
-	res = hfs_find_init(HFSPLUS_SB(inode->i_sb).cat_tree, &fd);
+	res = hfs_find_init(HFSPLUS_SB(inode->i_sb)->cat_tree, &fd);
 	if (res)
 		return res;
 	res = hfsplus_find_cat(inode->i_sb, inode->i_ino, &fd);
@@ -153,7 +168,7 @@
 		return -EOPNOTSUPP;
 
 	if (size) {
-		res = hfs_find_init(HFSPLUS_SB(inode->i_sb).cat_tree, &fd);
+		res = hfs_find_init(HFSPLUS_SB(inode->i_sb)->cat_tree, &fd);
 		if (res)
 			return res;
 		res = hfsplus_find_cat(inode->i_sb, inode->i_ino, &fd);
@@ -177,7 +192,7 @@
 		} else
 			res = size ? -ERANGE : 4;
 	} else
-		res = -ENODATA;
+		res = -EOPNOTSUPP;
 out:
 	if (size)
 		hfs_find_exit(&fd);
diff --git a/fs/hfsplus/options.c b/fs/hfsplus/options.c
index 572628b..f9ab276 100644
--- a/fs/hfsplus/options.c
+++ b/fs/hfsplus/options.c
@@ -143,13 +143,13 @@
 			kfree(p);
 			break;
 		case opt_decompose:
-			sbi->flags &= ~HFSPLUS_SB_NODECOMPOSE;
+			clear_bit(HFSPLUS_SB_NODECOMPOSE, &sbi->flags);
 			break;
 		case opt_nodecompose:
-			sbi->flags |= HFSPLUS_SB_NODECOMPOSE;
+			set_bit(HFSPLUS_SB_NODECOMPOSE, &sbi->flags);
 			break;
 		case opt_force:
-			sbi->flags |= HFSPLUS_SB_FORCE;
+			set_bit(HFSPLUS_SB_FORCE, &sbi->flags);
 			break;
 		default:
 			return 0;
@@ -171,7 +171,7 @@
 
 int hfsplus_show_options(struct seq_file *seq, struct vfsmount *mnt)
 {
-	struct hfsplus_sb_info *sbi = &HFSPLUS_SB(mnt->mnt_sb);
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(mnt->mnt_sb);
 
 	if (sbi->creator != HFSPLUS_DEF_CR_TYPE)
 		seq_printf(seq, ",creator=%.4s", (char *)&sbi->creator);
@@ -184,7 +184,7 @@
 		seq_printf(seq, ",session=%u", sbi->session);
 	if (sbi->nls)
 		seq_printf(seq, ",nls=%s", sbi->nls->charset);
-	if (sbi->flags & HFSPLUS_SB_NODECOMPOSE)
+	if (test_bit(HFSPLUS_SB_NODECOMPOSE, &sbi->flags))
 		seq_printf(seq, ",nodecompose");
 	return 0;
 }
diff --git a/fs/hfsplus/part_tbl.c b/fs/hfsplus/part_tbl.c
index 1528a6f..208b16c 100644
--- a/fs/hfsplus/part_tbl.c
+++ b/fs/hfsplus/part_tbl.c
@@ -74,6 +74,7 @@
 int hfs_part_find(struct super_block *sb,
 		  sector_t *part_start, sector_t *part_size)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	struct buffer_head *bh;
 	__be16 *data;
 	int i, size, res;
@@ -95,7 +96,7 @@
 		for (i = 0; i < size; p++, i++) {
 			if (p->pdStart && p->pdSize &&
 			    p->pdFSID == cpu_to_be32(0x54465331)/*"TFS1"*/ &&
-			    (HFSPLUS_SB(sb).part < 0 || HFSPLUS_SB(sb).part == i)) {
+			    (sbi->part < 0 || sbi->part == i)) {
 				*part_start += be32_to_cpu(p->pdStart);
 				*part_size = be32_to_cpu(p->pdSize);
 				res = 0;
@@ -111,7 +112,7 @@
 		size = be32_to_cpu(pm->pmMapBlkCnt);
 		for (i = 0; i < size;) {
 			if (!memcmp(pm->pmPartType,"Apple_HFS", 9) &&
-			    (HFSPLUS_SB(sb).part < 0 || HFSPLUS_SB(sb).part == i)) {
+			    (sbi->part < 0 || sbi->part == i)) {
 				*part_start += be32_to_cpu(pm->pmPyPartStart);
 				*part_size = be32_to_cpu(pm->pmPartBlkCnt);
 				res = 0;
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 3b55c05..9a88d75 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -12,7 +12,6 @@
 #include <linux/pagemap.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
-#include <linux/smp_lock.h>
 #include <linux/vfs.h>
 #include <linux/nls.h>
 
@@ -21,40 +20,11 @@
 
 #include "hfsplus_fs.h"
 
-struct inode *hfsplus_iget(struct super_block *sb, unsigned long ino)
+static int hfsplus_system_read_inode(struct inode *inode)
 {
-	struct hfs_find_data fd;
-	struct hfsplus_vh *vhdr;
-	struct inode *inode;
-	long err = -EIO;
+	struct hfsplus_vh *vhdr = HFSPLUS_SB(inode->i_sb)->s_vhdr;
 
-	inode = iget_locked(sb, ino);
-	if (!inode)
-		return ERR_PTR(-ENOMEM);
-	if (!(inode->i_state & I_NEW))
-		return inode;
-
-	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	mutex_init(&HFSPLUS_I(inode).extents_lock);
-	HFSPLUS_I(inode).flags = 0;
-	HFSPLUS_I(inode).rsrc_inode = NULL;
-	atomic_set(&HFSPLUS_I(inode).opencnt, 0);
-
-	if (inode->i_ino >= HFSPLUS_FIRSTUSER_CNID) {
-	read_inode:
-		hfs_find_init(HFSPLUS_SB(inode->i_sb).cat_tree, &fd);
-		err = hfsplus_find_cat(inode->i_sb, inode->i_ino, &fd);
-		if (!err)
-			err = hfsplus_cat_read_inode(inode, &fd);
-		hfs_find_exit(&fd);
-		if (err)
-			goto bad_inode;
-		goto done;
-	}
-	vhdr = HFSPLUS_SB(inode->i_sb).s_vhdr;
-	switch(inode->i_ino) {
-	case HFSPLUS_ROOT_CNID:
-		goto read_inode;
+	switch (inode->i_ino) {
 	case HFSPLUS_EXT_CNID:
 		hfsplus_inode_read_fork(inode, &vhdr->ext_file);
 		inode->i_mapping->a_ops = &hfsplus_btree_aops;
@@ -75,74 +45,101 @@
 		inode->i_mapping->a_ops = &hfsplus_btree_aops;
 		break;
 	default:
-		goto bad_inode;
+		return -EIO;
 	}
 
-done:
+	return 0;
+}
+
+struct inode *hfsplus_iget(struct super_block *sb, unsigned long ino)
+{
+	struct hfs_find_data fd;
+	struct inode *inode;
+	int err;
+
+	inode = iget_locked(sb, ino);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+	if (!(inode->i_state & I_NEW))
+		return inode;
+
+	INIT_LIST_HEAD(&HFSPLUS_I(inode)->open_dir_list);
+	mutex_init(&HFSPLUS_I(inode)->extents_lock);
+	HFSPLUS_I(inode)->flags = 0;
+	HFSPLUS_I(inode)->rsrc_inode = NULL;
+	atomic_set(&HFSPLUS_I(inode)->opencnt, 0);
+
+	if (inode->i_ino >= HFSPLUS_FIRSTUSER_CNID ||
+	    inode->i_ino == HFSPLUS_ROOT_CNID) {
+		hfs_find_init(HFSPLUS_SB(inode->i_sb)->cat_tree, &fd);
+		err = hfsplus_find_cat(inode->i_sb, inode->i_ino, &fd);
+		if (!err)
+			err = hfsplus_cat_read_inode(inode, &fd);
+		hfs_find_exit(&fd);
+	} else {
+		err = hfsplus_system_read_inode(inode);
+	}
+
+	if (err) {
+		iget_failed(inode);
+		return ERR_PTR(err);
+	}
+
 	unlock_new_inode(inode);
 	return inode;
+}
 
-bad_inode:
-	iget_failed(inode);
-	return ERR_PTR(err);
+static int hfsplus_system_write_inode(struct inode *inode)
+{
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(inode->i_sb);
+	struct hfsplus_vh *vhdr = sbi->s_vhdr;
+	struct hfsplus_fork_raw *fork;
+	struct hfs_btree *tree = NULL;
+
+	switch (inode->i_ino) {
+	case HFSPLUS_EXT_CNID:
+		fork = &vhdr->ext_file;
+		tree = sbi->ext_tree;
+		break;
+	case HFSPLUS_CAT_CNID:
+		fork = &vhdr->cat_file;
+		tree = sbi->cat_tree;
+		break;
+	case HFSPLUS_ALLOC_CNID:
+		fork = &vhdr->alloc_file;
+		break;
+	case HFSPLUS_START_CNID:
+		fork = &vhdr->start_file;
+		break;
+	case HFSPLUS_ATTR_CNID:
+		fork = &vhdr->attr_file;
+		tree = sbi->attr_tree;
+	default:
+		return -EIO;
+	}
+
+	if (fork->total_size != cpu_to_be64(inode->i_size)) {
+		set_bit(HFSPLUS_SB_WRITEBACKUP, &sbi->flags);
+		inode->i_sb->s_dirt = 1;
+	}
+	hfsplus_inode_write_fork(inode, fork);
+	if (tree)
+		hfs_btree_write(tree);
+	return 0;
 }
 
 static int hfsplus_write_inode(struct inode *inode,
 		struct writeback_control *wbc)
 {
-	struct hfsplus_vh *vhdr;
-	int ret = 0;
-
 	dprint(DBG_INODE, "hfsplus_write_inode: %lu\n", inode->i_ino);
+
 	hfsplus_ext_write_extent(inode);
-	if (inode->i_ino >= HFSPLUS_FIRSTUSER_CNID) {
+
+	if (inode->i_ino >= HFSPLUS_FIRSTUSER_CNID ||
+	    inode->i_ino == HFSPLUS_ROOT_CNID)
 		return hfsplus_cat_write_inode(inode);
-	}
-	vhdr = HFSPLUS_SB(inode->i_sb).s_vhdr;
-	switch (inode->i_ino) {
-	case HFSPLUS_ROOT_CNID:
-		ret = hfsplus_cat_write_inode(inode);
-		break;
-	case HFSPLUS_EXT_CNID:
-		if (vhdr->ext_file.total_size != cpu_to_be64(inode->i_size)) {
-			HFSPLUS_SB(inode->i_sb).flags |= HFSPLUS_SB_WRITEBACKUP;
-			inode->i_sb->s_dirt = 1;
-		}
-		hfsplus_inode_write_fork(inode, &vhdr->ext_file);
-		hfs_btree_write(HFSPLUS_SB(inode->i_sb).ext_tree);
-		break;
-	case HFSPLUS_CAT_CNID:
-		if (vhdr->cat_file.total_size != cpu_to_be64(inode->i_size)) {
-			HFSPLUS_SB(inode->i_sb).flags |= HFSPLUS_SB_WRITEBACKUP;
-			inode->i_sb->s_dirt = 1;
-		}
-		hfsplus_inode_write_fork(inode, &vhdr->cat_file);
-		hfs_btree_write(HFSPLUS_SB(inode->i_sb).cat_tree);
-		break;
-	case HFSPLUS_ALLOC_CNID:
-		if (vhdr->alloc_file.total_size != cpu_to_be64(inode->i_size)) {
-			HFSPLUS_SB(inode->i_sb).flags |= HFSPLUS_SB_WRITEBACKUP;
-			inode->i_sb->s_dirt = 1;
-		}
-		hfsplus_inode_write_fork(inode, &vhdr->alloc_file);
-		break;
-	case HFSPLUS_START_CNID:
-		if (vhdr->start_file.total_size != cpu_to_be64(inode->i_size)) {
-			HFSPLUS_SB(inode->i_sb).flags |= HFSPLUS_SB_WRITEBACKUP;
-			inode->i_sb->s_dirt = 1;
-		}
-		hfsplus_inode_write_fork(inode, &vhdr->start_file);
-		break;
-	case HFSPLUS_ATTR_CNID:
-		if (vhdr->attr_file.total_size != cpu_to_be64(inode->i_size)) {
-			HFSPLUS_SB(inode->i_sb).flags |= HFSPLUS_SB_WRITEBACKUP;
-			inode->i_sb->s_dirt = 1;
-		}
-		hfsplus_inode_write_fork(inode, &vhdr->attr_file);
-		hfs_btree_write(HFSPLUS_SB(inode->i_sb).attr_tree);
-		break;
-	}
-	return ret;
+	else
+		return hfsplus_system_write_inode(inode);
 }
 
 static void hfsplus_evict_inode(struct inode *inode)
@@ -151,51 +148,53 @@
 	truncate_inode_pages(&inode->i_data, 0);
 	end_writeback(inode);
 	if (HFSPLUS_IS_RSRC(inode)) {
-		HFSPLUS_I(HFSPLUS_I(inode).rsrc_inode).rsrc_inode = NULL;
-		iput(HFSPLUS_I(inode).rsrc_inode);
+		HFSPLUS_I(HFSPLUS_I(inode)->rsrc_inode)->rsrc_inode = NULL;
+		iput(HFSPLUS_I(inode)->rsrc_inode);
 	}
 }
 
 int hfsplus_sync_fs(struct super_block *sb, int wait)
 {
-	struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+	struct hfsplus_vh *vhdr = sbi->s_vhdr;
 
 	dprint(DBG_SUPER, "hfsplus_write_super\n");
 
-	lock_super(sb);
+	mutex_lock(&sbi->vh_mutex);
+	mutex_lock(&sbi->alloc_mutex);
 	sb->s_dirt = 0;
 
-	vhdr->free_blocks = cpu_to_be32(HFSPLUS_SB(sb).free_blocks);
-	vhdr->next_alloc = cpu_to_be32(HFSPLUS_SB(sb).next_alloc);
-	vhdr->next_cnid = cpu_to_be32(HFSPLUS_SB(sb).next_cnid);
-	vhdr->folder_count = cpu_to_be32(HFSPLUS_SB(sb).folder_count);
-	vhdr->file_count = cpu_to_be32(HFSPLUS_SB(sb).file_count);
+	vhdr->free_blocks = cpu_to_be32(sbi->free_blocks);
+	vhdr->next_cnid = cpu_to_be32(sbi->next_cnid);
+	vhdr->folder_count = cpu_to_be32(sbi->folder_count);
+	vhdr->file_count = cpu_to_be32(sbi->file_count);
 
-	mark_buffer_dirty(HFSPLUS_SB(sb).s_vhbh);
-	if (HFSPLUS_SB(sb).flags & HFSPLUS_SB_WRITEBACKUP) {
-		if (HFSPLUS_SB(sb).sect_count) {
+	mark_buffer_dirty(sbi->s_vhbh);
+	if (test_and_clear_bit(HFSPLUS_SB_WRITEBACKUP, &sbi->flags)) {
+		if (sbi->sect_count) {
 			struct buffer_head *bh;
 			u32 block, offset;
 
-			block = HFSPLUS_SB(sb).blockoffset;
-			block += (HFSPLUS_SB(sb).sect_count - 2) >> (sb->s_blocksize_bits - 9);
-			offset = ((HFSPLUS_SB(sb).sect_count - 2) << 9) & (sb->s_blocksize - 1);
-			printk(KERN_DEBUG "hfs: backup: %u,%u,%u,%u\n", HFSPLUS_SB(sb).blockoffset,
-				HFSPLUS_SB(sb).sect_count, block, offset);
+			block = sbi->blockoffset;
+			block += (sbi->sect_count - 2) >> (sb->s_blocksize_bits - 9);
+			offset = ((sbi->sect_count - 2) << 9) & (sb->s_blocksize - 1);
+			printk(KERN_DEBUG "hfs: backup: %u,%u,%u,%u\n",
+					  sbi->blockoffset, sbi->sect_count,
+					  block, offset);
 			bh = sb_bread(sb, block);
 			if (bh) {
 				vhdr = (struct hfsplus_vh *)(bh->b_data + offset);
 				if (be16_to_cpu(vhdr->signature) == HFSPLUS_VOLHEAD_SIG) {
-					memcpy(vhdr, HFSPLUS_SB(sb).s_vhdr, sizeof(*vhdr));
+					memcpy(vhdr, sbi->s_vhdr, sizeof(*vhdr));
 					mark_buffer_dirty(bh);
 					brelse(bh);
 				} else
 					printk(KERN_WARNING "hfs: backup not found!\n");
 			}
 		}
-		HFSPLUS_SB(sb).flags &= ~HFSPLUS_SB_WRITEBACKUP;
 	}
-	unlock_super(sb);
+	mutex_unlock(&sbi->alloc_mutex);
+	mutex_unlock(&sbi->vh_mutex);
 	return 0;
 }
 
@@ -209,48 +208,48 @@
 
 static void hfsplus_put_super(struct super_block *sb)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
+
 	dprint(DBG_SUPER, "hfsplus_put_super\n");
+
 	if (!sb->s_fs_info)
 		return;
 
-	lock_kernel();
-
 	if (sb->s_dirt)
 		hfsplus_write_super(sb);
-	if (!(sb->s_flags & MS_RDONLY) && HFSPLUS_SB(sb).s_vhdr) {
-		struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr;
+	if (!(sb->s_flags & MS_RDONLY) && sbi->s_vhdr) {
+		struct hfsplus_vh *vhdr = sbi->s_vhdr;
 
 		vhdr->modify_date = hfsp_now2mt();
 		vhdr->attributes |= cpu_to_be32(HFSPLUS_VOL_UNMNT);
 		vhdr->attributes &= cpu_to_be32(~HFSPLUS_VOL_INCNSTNT);
-		mark_buffer_dirty(HFSPLUS_SB(sb).s_vhbh);
-		sync_dirty_buffer(HFSPLUS_SB(sb).s_vhbh);
+		mark_buffer_dirty(sbi->s_vhbh);
+		sync_dirty_buffer(sbi->s_vhbh);
 	}
 
-	hfs_btree_close(HFSPLUS_SB(sb).cat_tree);
-	hfs_btree_close(HFSPLUS_SB(sb).ext_tree);
-	iput(HFSPLUS_SB(sb).alloc_file);
-	iput(HFSPLUS_SB(sb).hidden_dir);
-	brelse(HFSPLUS_SB(sb).s_vhbh);
-	unload_nls(HFSPLUS_SB(sb).nls);
+	hfs_btree_close(sbi->cat_tree);
+	hfs_btree_close(sbi->ext_tree);
+	iput(sbi->alloc_file);
+	iput(sbi->hidden_dir);
+	brelse(sbi->s_vhbh);
+	unload_nls(sbi->nls);
 	kfree(sb->s_fs_info);
 	sb->s_fs_info = NULL;
-
-	unlock_kernel();
 }
 
 static int hfsplus_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
 
 	buf->f_type = HFSPLUS_SUPER_MAGIC;
 	buf->f_bsize = sb->s_blocksize;
-	buf->f_blocks = HFSPLUS_SB(sb).total_blocks << HFSPLUS_SB(sb).fs_shift;
-	buf->f_bfree = HFSPLUS_SB(sb).free_blocks << HFSPLUS_SB(sb).fs_shift;
+	buf->f_blocks = sbi->total_blocks << sbi->fs_shift;
+	buf->f_bfree = sbi->free_blocks << sbi->fs_shift;
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = 0xFFFFFFFF;
-	buf->f_ffree = 0xFFFFFFFF - HFSPLUS_SB(sb).next_cnid;
+	buf->f_ffree = 0xFFFFFFFF - sbi->next_cnid;
 	buf->f_fsid.val[0] = (u32)id;
 	buf->f_fsid.val[1] = (u32)(id >> 32);
 	buf->f_namelen = HFSPLUS_MAX_STRLEN;
@@ -263,11 +262,11 @@
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
 		return 0;
 	if (!(*flags & MS_RDONLY)) {
-		struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr;
+		struct hfsplus_vh *vhdr = HFSPLUS_SB(sb)->s_vhdr;
 		struct hfsplus_sb_info sbi;
 
 		memset(&sbi, 0, sizeof(struct hfsplus_sb_info));
-		sbi.nls = HFSPLUS_SB(sb).nls;
+		sbi.nls = HFSPLUS_SB(sb)->nls;
 		if (!hfsplus_parse_options(data, &sbi))
 			return -EINVAL;
 
@@ -276,7 +275,7 @@
 			       "running fsck.hfsplus is recommended.  leaving read-only.\n");
 			sb->s_flags |= MS_RDONLY;
 			*flags |= MS_RDONLY;
-		} else if (sbi.flags & HFSPLUS_SB_FORCE) {
+		} else if (test_bit(HFSPLUS_SB_FORCE, &sbi.flags)) {
 			/* nothing */
 		} else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) {
 			printk(KERN_WARNING "hfs: filesystem is marked locked, leaving read-only.\n");
@@ -320,7 +319,8 @@
 		return -ENOMEM;
 
 	sb->s_fs_info = sbi;
-	INIT_HLIST_HEAD(&sbi->rsrc_inodes);
+	mutex_init(&sbi->alloc_mutex);
+	mutex_init(&sbi->vh_mutex);
 	hfsplus_fill_defaults(sbi);
 	if (!hfsplus_parse_options(data, sbi)) {
 		printk(KERN_ERR "hfs: unable to parse mount options\n");
@@ -344,7 +344,7 @@
 		err = -EINVAL;
 		goto cleanup;
 	}
-	vhdr = HFSPLUS_SB(sb).s_vhdr;
+	vhdr = sbi->s_vhdr;
 
 	/* Copy parts of the volume header into the superblock */
 	sb->s_magic = HFSPLUS_VOLHEAD_SIG;
@@ -353,18 +353,19 @@
 		printk(KERN_ERR "hfs: wrong filesystem version\n");
 		goto cleanup;
 	}
-	HFSPLUS_SB(sb).total_blocks = be32_to_cpu(vhdr->total_blocks);
-	HFSPLUS_SB(sb).free_blocks = be32_to_cpu(vhdr->free_blocks);
-	HFSPLUS_SB(sb).next_alloc = be32_to_cpu(vhdr->next_alloc);
-	HFSPLUS_SB(sb).next_cnid = be32_to_cpu(vhdr->next_cnid);
-	HFSPLUS_SB(sb).file_count = be32_to_cpu(vhdr->file_count);
-	HFSPLUS_SB(sb).folder_count = be32_to_cpu(vhdr->folder_count);
-	HFSPLUS_SB(sb).data_clump_blocks = be32_to_cpu(vhdr->data_clump_sz) >> HFSPLUS_SB(sb).alloc_blksz_shift;
-	if (!HFSPLUS_SB(sb).data_clump_blocks)
-		HFSPLUS_SB(sb).data_clump_blocks = 1;
-	HFSPLUS_SB(sb).rsrc_clump_blocks = be32_to_cpu(vhdr->rsrc_clump_sz) >> HFSPLUS_SB(sb).alloc_blksz_shift;
-	if (!HFSPLUS_SB(sb).rsrc_clump_blocks)
-		HFSPLUS_SB(sb).rsrc_clump_blocks = 1;
+	sbi->total_blocks = be32_to_cpu(vhdr->total_blocks);
+	sbi->free_blocks = be32_to_cpu(vhdr->free_blocks);
+	sbi->next_cnid = be32_to_cpu(vhdr->next_cnid);
+	sbi->file_count = be32_to_cpu(vhdr->file_count);
+	sbi->folder_count = be32_to_cpu(vhdr->folder_count);
+	sbi->data_clump_blocks =
+		be32_to_cpu(vhdr->data_clump_sz) >> sbi->alloc_blksz_shift;
+	if (!sbi->data_clump_blocks)
+		sbi->data_clump_blocks = 1;
+	sbi->rsrc_clump_blocks =
+		be32_to_cpu(vhdr->rsrc_clump_sz) >> sbi->alloc_blksz_shift;
+	if (!sbi->rsrc_clump_blocks)
+		sbi->rsrc_clump_blocks = 1;
 
 	/* Set up operations so we can load metadata */
 	sb->s_op = &hfsplus_sops;
@@ -374,7 +375,7 @@
 		printk(KERN_WARNING "hfs: Filesystem was not cleanly unmounted, "
 		       "running fsck.hfsplus is recommended.  mounting read-only.\n");
 		sb->s_flags |= MS_RDONLY;
-	} else if (sbi->flags & HFSPLUS_SB_FORCE) {
+	} else if (test_and_clear_bit(HFSPLUS_SB_FORCE, &sbi->flags)) {
 		/* nothing */
 	} else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) {
 		printk(KERN_WARNING "hfs: Filesystem is marked locked, mounting read-only.\n");
@@ -384,16 +385,15 @@
 		       "use the force option at your own risk, mounting read-only.\n");
 		sb->s_flags |= MS_RDONLY;
 	}
-	sbi->flags &= ~HFSPLUS_SB_FORCE;
 
 	/* Load metadata objects (B*Trees) */
-	HFSPLUS_SB(sb).ext_tree = hfs_btree_open(sb, HFSPLUS_EXT_CNID);
-	if (!HFSPLUS_SB(sb).ext_tree) {
+	sbi->ext_tree = hfs_btree_open(sb, HFSPLUS_EXT_CNID);
+	if (!sbi->ext_tree) {
 		printk(KERN_ERR "hfs: failed to load extents file\n");
 		goto cleanup;
 	}
-	HFSPLUS_SB(sb).cat_tree = hfs_btree_open(sb, HFSPLUS_CAT_CNID);
-	if (!HFSPLUS_SB(sb).cat_tree) {
+	sbi->cat_tree = hfs_btree_open(sb, HFSPLUS_CAT_CNID);
+	if (!sbi->cat_tree) {
 		printk(KERN_ERR "hfs: failed to load catalog file\n");
 		goto cleanup;
 	}
@@ -404,7 +404,7 @@
 		err = PTR_ERR(inode);
 		goto cleanup;
 	}
-	HFSPLUS_SB(sb).alloc_file = inode;
+	sbi->alloc_file = inode;
 
 	/* Load the root directory */
 	root = hfsplus_iget(sb, HFSPLUS_ROOT_CNID);
@@ -423,7 +423,7 @@
 
 	str.len = sizeof(HFSP_HIDDENDIR_NAME) - 1;
 	str.name = HFSP_HIDDENDIR_NAME;
-	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
+	hfs_find_init(sbi->cat_tree, &fd);
 	hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_ROOT_CNID, &str);
 	if (!hfs_brec_read(&fd, &entry, sizeof(entry))) {
 		hfs_find_exit(&fd);
@@ -434,7 +434,7 @@
 			err = PTR_ERR(inode);
 			goto cleanup;
 		}
-		HFSPLUS_SB(sb).hidden_dir = inode;
+		sbi->hidden_dir = inode;
 	} else
 		hfs_find_exit(&fd);
 
@@ -449,15 +449,19 @@
 	be32_add_cpu(&vhdr->write_count, 1);
 	vhdr->attributes &= cpu_to_be32(~HFSPLUS_VOL_UNMNT);
 	vhdr->attributes |= cpu_to_be32(HFSPLUS_VOL_INCNSTNT);
-	mark_buffer_dirty(HFSPLUS_SB(sb).s_vhbh);
-	sync_dirty_buffer(HFSPLUS_SB(sb).s_vhbh);
+	mark_buffer_dirty(sbi->s_vhbh);
+	sync_dirty_buffer(sbi->s_vhbh);
 
-	if (!HFSPLUS_SB(sb).hidden_dir) {
+	if (!sbi->hidden_dir) {
 		printk(KERN_DEBUG "hfs: create hidden dir...\n");
-		HFSPLUS_SB(sb).hidden_dir = hfsplus_new_inode(sb, S_IFDIR);
-		hfsplus_create_cat(HFSPLUS_SB(sb).hidden_dir->i_ino, sb->s_root->d_inode,
-				   &str, HFSPLUS_SB(sb).hidden_dir);
-		mark_inode_dirty(HFSPLUS_SB(sb).hidden_dir);
+
+		mutex_lock(&sbi->vh_mutex);
+		sbi->hidden_dir = hfsplus_new_inode(sb, S_IFDIR);
+		hfsplus_create_cat(sbi->hidden_dir->i_ino, sb->s_root->d_inode,
+				   &str, sbi->hidden_dir);
+		mutex_unlock(&sbi->vh_mutex);
+
+		mark_inode_dirty(sbi->hidden_dir);
 	}
 out:
 	unload_nls(sbi->nls);
@@ -486,7 +490,7 @@
 
 static void hfsplus_destroy_inode(struct inode *inode)
 {
-	kmem_cache_free(hfsplus_inode_cachep, &HFSPLUS_I(inode));
+	kmem_cache_free(hfsplus_inode_cachep, HFSPLUS_I(inode));
 }
 
 #define HFSPLUS_INODE_SIZE	sizeof(struct hfsplus_inode_info)
diff --git a/fs/hfsplus/unicode.c b/fs/hfsplus/unicode.c
index 628ccf6..b66d67d 100644
--- a/fs/hfsplus/unicode.c
+++ b/fs/hfsplus/unicode.c
@@ -121,7 +121,7 @@
 int hfsplus_uni2asc(struct super_block *sb, const struct hfsplus_unistr *ustr, char *astr, int *len_p)
 {
 	const hfsplus_unichr *ip;
-	struct nls_table *nls = HFSPLUS_SB(sb).nls;
+	struct nls_table *nls = HFSPLUS_SB(sb)->nls;
 	u8 *op;
 	u16 cc, c0, c1;
 	u16 *ce1, *ce2;
@@ -132,7 +132,7 @@
 	ustrlen = be16_to_cpu(ustr->length);
 	len = *len_p;
 	ce1 = NULL;
-	compose = !(HFSPLUS_SB(sb).flags & HFSPLUS_SB_NODECOMPOSE);
+	compose = !test_bit(HFSPLUS_SB_NODECOMPOSE, &HFSPLUS_SB(sb)->flags);
 
 	while (ustrlen > 0) {
 		c0 = be16_to_cpu(*ip++);
@@ -246,7 +246,7 @@
 static inline int asc2unichar(struct super_block *sb, const char *astr, int len,
 			      wchar_t *uc)
 {
-	int size = HFSPLUS_SB(sb).nls->char2uni(astr, len, uc);
+	int size = HFSPLUS_SB(sb)->nls->char2uni(astr, len, uc);
 	if (size <= 0) {
 		*uc = '?';
 		size = 1;
@@ -293,7 +293,7 @@
 	u16 *dstr, outlen = 0;
 	wchar_t c;
 
-	decompose = !(HFSPLUS_SB(sb).flags & HFSPLUS_SB_NODECOMPOSE);
+	decompose = !test_bit(HFSPLUS_SB_NODECOMPOSE, &HFSPLUS_SB(sb)->flags);
 	while (outlen < HFSPLUS_MAX_STRLEN && len > 0) {
 		size = asc2unichar(sb, astr, len, &c);
 
@@ -330,8 +330,8 @@
 	wchar_t c;
 	u16 c2;
 
-	casefold = (HFSPLUS_SB(sb).flags & HFSPLUS_SB_CASEFOLD);
-	decompose = !(HFSPLUS_SB(sb).flags & HFSPLUS_SB_NODECOMPOSE);
+	casefold = test_bit(HFSPLUS_SB_CASEFOLD, &HFSPLUS_SB(sb)->flags);
+	decompose = !test_bit(HFSPLUS_SB_NODECOMPOSE, &HFSPLUS_SB(sb)->flags);
 	hash = init_name_hash();
 	astr = str->name;
 	len = str->len;
@@ -373,8 +373,8 @@
 	u16 c1, c2;
 	wchar_t c;
 
-	casefold = (HFSPLUS_SB(sb).flags & HFSPLUS_SB_CASEFOLD);
-	decompose = !(HFSPLUS_SB(sb).flags & HFSPLUS_SB_NODECOMPOSE);
+	casefold = test_bit(HFSPLUS_SB_CASEFOLD, &HFSPLUS_SB(sb)->flags);
+	decompose = !test_bit(HFSPLUS_SB_NODECOMPOSE, &HFSPLUS_SB(sb)->flags);
 	astr1 = s1->name;
 	len1 = s1->len;
 	astr2 = s2->name;
diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c
index bed78ac8..8972c20 100644
--- a/fs/hfsplus/wrapper.c
+++ b/fs/hfsplus/wrapper.c
@@ -65,8 +65,8 @@
 	*start = 0;
 	*size = sb->s_bdev->bd_inode->i_size >> 9;
 
-	if (HFSPLUS_SB(sb).session >= 0) {
-		te.cdte_track = HFSPLUS_SB(sb).session;
+	if (HFSPLUS_SB(sb)->session >= 0) {
+		te.cdte_track = HFSPLUS_SB(sb)->session;
 		te.cdte_format = CDROM_LBA;
 		res = ioctl_by_bdev(sb->s_bdev, CDROMREADTOCENTRY, (unsigned long)&te);
 		if (!res && (te.cdte_ctrl & CDROM_DATA_TRACK) == 4) {
@@ -87,6 +87,7 @@
 /* Takes in super block, returns true if good data read */
 int hfsplus_read_wrapper(struct super_block *sb)
 {
+	struct hfsplus_sb_info *sbi = HFSPLUS_SB(sb);
 	struct buffer_head *bh;
 	struct hfsplus_vh *vhdr;
 	struct hfsplus_wd wd;
@@ -122,7 +123,7 @@
 		if (vhdr->signature == cpu_to_be16(HFSPLUS_VOLHEAD_SIG))
 			break;
 		if (vhdr->signature == cpu_to_be16(HFSPLUS_VOLHEAD_SIGX)) {
-			HFSPLUS_SB(sb).flags |= HFSPLUS_SB_HFSX;
+			set_bit(HFSPLUS_SB_HFSX, &sbi->flags);
 			break;
 		}
 		brelse(bh);
@@ -143,11 +144,11 @@
 	if (blocksize < HFSPLUS_SECTOR_SIZE ||
 	    ((blocksize - 1) & blocksize))
 		return -EINVAL;
-	HFSPLUS_SB(sb).alloc_blksz = blocksize;
-	HFSPLUS_SB(sb).alloc_blksz_shift = 0;
+	sbi->alloc_blksz = blocksize;
+	sbi->alloc_blksz_shift = 0;
 	while ((blocksize >>= 1) != 0)
-		HFSPLUS_SB(sb).alloc_blksz_shift++;
-	blocksize = min(HFSPLUS_SB(sb).alloc_blksz, (u32)PAGE_SIZE);
+		sbi->alloc_blksz_shift++;
+	blocksize = min(sbi->alloc_blksz, (u32)PAGE_SIZE);
 
 	/* align block size to block offset */
 	while (part_start & ((blocksize >> HFSPLUS_SECTOR_SHIFT) - 1))
@@ -158,23 +159,26 @@
 		return -EINVAL;
 	}
 
-	HFSPLUS_SB(sb).blockoffset = part_start >>
-			(sb->s_blocksize_bits - HFSPLUS_SECTOR_SHIFT);
-	HFSPLUS_SB(sb).sect_count = part_size;
-	HFSPLUS_SB(sb).fs_shift = HFSPLUS_SB(sb).alloc_blksz_shift -
-			sb->s_blocksize_bits;
+	sbi->blockoffset =
+		part_start >> (sb->s_blocksize_bits - HFSPLUS_SECTOR_SHIFT);
+	sbi->sect_count = part_size;
+	sbi->fs_shift = sbi->alloc_blksz_shift - sb->s_blocksize_bits;
 
 	bh = sb_bread512(sb, part_start + HFSPLUS_VOLHEAD_SECTOR, vhdr);
 	if (!bh)
 		return -EIO;
 
 	/* should still be the same... */
-	if (vhdr->signature != (HFSPLUS_SB(sb).flags & HFSPLUS_SB_HFSX ?
-				cpu_to_be16(HFSPLUS_VOLHEAD_SIGX) :
-				cpu_to_be16(HFSPLUS_VOLHEAD_SIG)))
-		goto error;
-	HFSPLUS_SB(sb).s_vhbh = bh;
-	HFSPLUS_SB(sb).s_vhdr = vhdr;
+	if (test_bit(HFSPLUS_SB_HFSX, &sbi->flags)) {
+		if (vhdr->signature != cpu_to_be16(HFSPLUS_VOLHEAD_SIGX))
+			goto error;
+	} else {
+		if (vhdr->signature != cpu_to_be16(HFSPLUS_VOLHEAD_SIG))
+			goto error;
+	}
+
+	sbi->s_vhbh = bh;
+	sbi->s_vhdr = vhdr;
 
 	return 0;
  error:
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
index cdfb8c6..c16f8d8 100644
--- a/fs/nfsd/nfsfh.h
+++ b/fs/nfsd/nfsfh.h
@@ -196,8 +196,6 @@
 static inline void
 fh_unlock(struct svc_fh *fhp)
 {
-	BUG_ON(!fhp->fh_dentry);
-
 	if (fhp->fh_locked) {
 		fill_post_wcc(fhp);
 		mutex_unlock(&fhp->fh_dentry->d_inode->i_mutex);
diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig
index 22c629e..b388443 100644
--- a/fs/notify/Kconfig
+++ b/fs/notify/Kconfig
@@ -3,4 +3,4 @@
 
 source "fs/notify/dnotify/Kconfig"
 source "fs/notify/inotify/Kconfig"
-source "fs/notify/fanotify/Kconfig"
+#source "fs/notify/fanotify/Kconfig"
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index d59c4a6..81976ff 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -668,14 +668,11 @@
 	xfs_perag_put(pag);
 }
 
-void
-__xfs_inode_clear_reclaim_tag(
-	xfs_mount_t	*mp,
+STATIC void
+__xfs_inode_clear_reclaim(
 	xfs_perag_t	*pag,
 	xfs_inode_t	*ip)
 {
-	radix_tree_tag_clear(&pag->pag_ici_root,
-			XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
 	pag->pag_ici_reclaimable--;
 	if (!pag->pag_ici_reclaimable) {
 		/* clear the reclaim tag from the perag radix tree */
@@ -689,6 +686,17 @@
 	}
 }
 
+void
+__xfs_inode_clear_reclaim_tag(
+	xfs_mount_t	*mp,
+	xfs_perag_t	*pag,
+	xfs_inode_t	*ip)
+{
+	radix_tree_tag_clear(&pag->pag_ici_root,
+			XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+	__xfs_inode_clear_reclaim(pag, ip);
+}
+
 /*
  * Inodes in different states need to be treated differently, and the return
  * value of xfs_iflush is not sufficient to get this right. The following table
@@ -838,6 +846,7 @@
 	if (!radix_tree_delete(&pag->pag_ici_root,
 				XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino)))
 		ASSERT(0);
+	__xfs_inode_clear_reclaim(pag, ip);
 	write_unlock(&pag->pag_ici_lock);
 
 	/*
diff --git a/include/asm-generic/hardirq.h b/include/asm-generic/hardirq.h
index c0771aa..04d0a97 100644
--- a/include/asm-generic/hardirq.h
+++ b/include/asm-generic/hardirq.h
@@ -9,6 +9,7 @@
 } ____cacheline_aligned irq_cpustat_t;
 
 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
+#include <linux/irq.h>
 
 #ifndef ack_bad_irq
 static inline void ack_bad_irq(unsigned int irq)
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index e2bd73e..f4d4120 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -129,6 +129,10 @@
 #define move_pte(pte, prot, old_addr, new_addr)	(pte)
 #endif
 
+#ifndef flush_tlb_fix_spurious_fault
+#define flush_tlb_fix_spurious_fault(vma, address) flush_tlb_page(vma, address)
+#endif
+
 #ifndef pgprot_noncached
 #define pgprot_noncached(prot)	(prot)
 #endif
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 8a92a17..f4229fb 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -220,6 +220,8 @@
 									\
 	BUG_TABLE							\
 									\
+	JUMP_TABLE							\
+									\
 	/* PCI quirks */						\
 	.pci_fixup        : AT(ADDR(.pci_fixup) - LOAD_OFFSET) {	\
 		VMLINUX_SYMBOL(__start_pci_fixups_early) = .;		\
@@ -563,6 +565,14 @@
 #define BUG_TABLE
 #endif
 
+#define JUMP_TABLE							\
+	. = ALIGN(8);							\
+	__jump_table : AT(ADDR(__jump_table) - LOAD_OFFSET) {		\
+		VMLINUX_SYMBOL(__start___jump_table) = .;		\
+		*(__jump_table)						\
+		VMLINUX_SYMBOL(__stop___jump_table) = .;		\
+	}
+
 #ifdef CONFIG_PM_TRACE
 #define TRACEDATA							\
 	. = ALIGN(4);							\
@@ -677,7 +687,9 @@
 				- LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__per_cpu_start) = .;			\
 		*(.data..percpu..first)					\
+		. = ALIGN(PAGE_SIZE);					\
 		*(.data..percpu..page_aligned)				\
+		*(.data..percpu..readmostly)				\
 		*(.data..percpu)					\
 		*(.data..percpu..shared_aligned)			\
 		VMLINUX_SYMBOL(__per_cpu_end) = .;			\
@@ -703,7 +715,9 @@
 		VMLINUX_SYMBOL(__per_cpu_load) = .;			\
 		VMLINUX_SYMBOL(__per_cpu_start) = .;			\
 		*(.data..percpu..first)					\
+		. = ALIGN(PAGE_SIZE);					\
 		*(.data..percpu..page_aligned)				\
+		*(.data..percpu..readmostly)				\
 		*(.data..percpu)					\
 		*(.data..percpu..shared_aligned)			\
 		VMLINUX_SYMBOL(__per_cpu_end) = .;			\
diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
index 267a86c..2040e6c 100644
--- a/include/drm/ttm/ttm_bo_api.h
+++ b/include/drm/ttm/ttm_bo_api.h
@@ -246,9 +246,11 @@
 
 	atomic_t reserved;
 
-
 	/**
 	 * Members protected by the bo::lock
+	 * In addition, setting sync_obj to anything else
+	 * than NULL requires bo::reserved to be held. This allows for
+	 * checking NULL while reserved but not holding bo::lock.
 	 */
 
 	void *sync_obj_arg;
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 626b629..4e8ea8c 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -118,7 +118,6 @@
 header-y += ext2_fs.h
 header-y += fadvise.h
 header-y += falloc.h
-header-y += fanotify.h
 header-y += fb.h
 header-y += fcntl.h
 header-y += fd.h
diff --git a/include/linux/acpi_pmtmr.h b/include/linux/acpi_pmtmr.h
index 7e3d285..1d0ef1a 100644
--- a/include/linux/acpi_pmtmr.h
+++ b/include/linux/acpi_pmtmr.h
@@ -25,8 +25,6 @@
 	return acpi_pm_read_verified() & ACPI_PM_MASK;
 }
 
-extern void pmtimer_wait(unsigned);
-
 #else
 
 static inline u32 acpi_pm_read_early(void)
diff --git a/include/linux/amba/pl022.h b/include/linux/amba/pl022.h
index abf26cc..4ce98f5 100644
--- a/include/linux/amba/pl022.h
+++ b/include/linux/amba/pl022.h
@@ -228,6 +228,7 @@
 };
 
 
+struct dma_chan;
 /**
  * struct pl022_ssp_master - device.platform_data for SPI controller devices.
  * @num_chipselect: chipselects are used to distinguish individual
@@ -235,11 +236,16 @@
  *     each slave has a chipselect signal, but it's common that not
  *     every chipselect is connected to a slave.
  * @enable_dma: if true enables DMA driven transfers.
+ * @dma_rx_param: parameter to locate an RX DMA channel.
+ * @dma_tx_param: parameter to locate a TX DMA channel.
  */
 struct pl022_ssp_controller {
 	u16 bus_id;
 	u8 num_chipselect;
 	u8 enable_dma:1;
+	bool (*dma_filter)(struct dma_chan *chan, void *filter_param);
+	void *dma_rx_param;
+	void *dma_tx_param;
 };
 
 /**
@@ -270,20 +276,13 @@
  * @dma_config: DMA configuration for SSP controller and peripheral
  */
 struct pl022_config_chip {
-	struct device *dev;
-	enum ssp_loopback lbm;
 	enum ssp_interface iface;
 	enum ssp_hierarchy hierarchy;
 	bool slave_tx_disable;
 	struct ssp_clock_params clk_freq;
-	enum ssp_rx_endian endian_rx;
-	enum ssp_tx_endian endian_tx;
-	enum ssp_data_size data_size;
 	enum ssp_mode com_mode;
 	enum ssp_rx_level_trig rx_lev_trig;
 	enum ssp_tx_level_trig tx_lev_trig;
-	enum ssp_spi_clk_phase clk_phase;
-	enum ssp_spi_clk_pol clk_pol;
 	enum ssp_microwire_ctrl_len ctrl_len;
 	enum ssp_microwire_wait_state wait_state;
 	enum ssp_duplex duplex;
diff --git a/fs/ceph/auth.h b/include/linux/ceph/auth.h
similarity index 97%
rename from fs/ceph/auth.h
rename to include/linux/ceph/auth.h
index d38a2fb..7fff521 100644
--- a/fs/ceph/auth.h
+++ b/include/linux/ceph/auth.h
@@ -1,8 +1,8 @@
 #ifndef _FS_CEPH_AUTH_H
 #define _FS_CEPH_AUTH_H
 
-#include "types.h"
-#include "buffer.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/buffer.h>
 
 /*
  * Abstract interface for communicating with the authenticate module.
diff --git a/fs/ceph/buffer.h b/include/linux/ceph/buffer.h
similarity index 100%
rename from fs/ceph/buffer.h
rename to include/linux/ceph/buffer.h
diff --git a/fs/ceph/ceph_debug.h b/include/linux/ceph/ceph_debug.h
similarity index 86%
rename from fs/ceph/ceph_debug.h
rename to include/linux/ceph/ceph_debug.h
index 1818c23..aa2e191 100644
--- a/fs/ceph/ceph_debug.h
+++ b/include/linux/ceph/ceph_debug.h
@@ -3,7 +3,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#ifdef CONFIG_CEPH_FS_PRETTYDEBUG
+#ifdef CONFIG_CEPH_LIB_PRETTYDEBUG
 
 /*
  * wrap pr_debug to include a filename:lineno prefix on each line.
@@ -14,7 +14,8 @@
 # if defined(DEBUG) || defined(CONFIG_DYNAMIC_DEBUG)
 extern const char *ceph_file_part(const char *s, int len);
 #  define dout(fmt, ...)						\
-	pr_debug(" %12.12s:%-4d : " fmt,				\
+	pr_debug("%.*s %12.12s:%-4d : " fmt,				\
+		 8 - (int)sizeof(KBUILD_MODNAME), "    ",		\
 		 ceph_file_part(__FILE__, sizeof(__FILE__)),		\
 		 __LINE__, ##__VA_ARGS__)
 # else
diff --git a/fs/ceph/ceph_frag.h b/include/linux/ceph/ceph_frag.h
similarity index 100%
rename from fs/ceph/ceph_frag.h
rename to include/linux/ceph/ceph_frag.h
diff --git a/fs/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h
similarity index 99%
rename from fs/ceph/ceph_fs.h
rename to include/linux/ceph/ceph_fs.h
index d5619ac..c3c74ae 100644
--- a/fs/ceph/ceph_fs.h
+++ b/include/linux/ceph/ceph_fs.h
@@ -299,6 +299,7 @@
 	CEPH_MDS_OP_SETATTR    = 0x01108,
 	CEPH_MDS_OP_SETFILELOCK= 0x01109,
 	CEPH_MDS_OP_GETFILELOCK= 0x00110,
+	CEPH_MDS_OP_SETDIRLAYOUT=0x0110a,
 
 	CEPH_MDS_OP_MKNOD      = 0x01201,
 	CEPH_MDS_OP_LINK       = 0x01202,
diff --git a/fs/ceph/ceph_hash.h b/include/linux/ceph/ceph_hash.h
similarity index 100%
rename from fs/ceph/ceph_hash.h
rename to include/linux/ceph/ceph_hash.h
diff --git a/include/linux/ceph/debugfs.h b/include/linux/ceph/debugfs.h
new file mode 100644
index 0000000..2a79702
--- /dev/null
+++ b/include/linux/ceph/debugfs.h
@@ -0,0 +1,33 @@
+#ifndef _FS_CEPH_DEBUGFS_H
+#define _FS_CEPH_DEBUGFS_H
+
+#include "ceph_debug.h"
+#include "types.h"
+
+#define CEPH_DEFINE_SHOW_FUNC(name)					\
+static int name##_open(struct inode *inode, struct file *file)		\
+{									\
+	struct seq_file *sf;						\
+	int ret;							\
+									\
+	ret = single_open(file, name, NULL);				\
+	sf = file->private_data;					\
+	sf->private = inode->i_private;					\
+	return ret;							\
+}									\
+									\
+static const struct file_operations name##_fops = {			\
+	.open		= name##_open,					\
+	.read		= seq_read,					\
+	.llseek		= seq_lseek,					\
+	.release	= single_release,				\
+};
+
+/* debugfs.c */
+extern int ceph_debugfs_init(void);
+extern void ceph_debugfs_cleanup(void);
+extern int ceph_debugfs_client_init(struct ceph_client *client);
+extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
+
+#endif
+
diff --git a/fs/ceph/decode.h b/include/linux/ceph/decode.h
similarity index 96%
rename from fs/ceph/decode.h
rename to include/linux/ceph/decode.h
index 3d25415..c5b6939 100644
--- a/fs/ceph/decode.h
+++ b/include/linux/ceph/decode.h
@@ -191,6 +191,11 @@
 		ceph_encode_need(p, end, n, bad);		\
 		ceph_encode_copy(p, pv, n);			\
 	} while (0)
+#define ceph_encode_string_safe(p, end, s, n, bad)		\
+	do {							\
+		ceph_encode_need(p, end, n, bad);		\
+		ceph_encode_string(p, end, s, n);		\
+	} while (0)
 
 
 #endif
diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h
new file mode 100644
index 0000000..f22b2e9
--- /dev/null
+++ b/include/linux/ceph/libceph.h
@@ -0,0 +1,249 @@
+#ifndef _FS_CEPH_LIBCEPH_H
+#define _FS_CEPH_LIBCEPH_H
+
+#include "ceph_debug.h"
+
+#include <asm/unaligned.h>
+#include <linux/backing-dev.h>
+#include <linux/completion.h>
+#include <linux/exportfs.h>
+#include <linux/fs.h>
+#include <linux/mempool.h>
+#include <linux/pagemap.h>
+#include <linux/wait.h>
+#include <linux/writeback.h>
+#include <linux/slab.h>
+
+#include "types.h"
+#include "messenger.h"
+#include "msgpool.h"
+#include "mon_client.h"
+#include "osd_client.h"
+#include "ceph_fs.h"
+
+/*
+ * Supported features
+ */
+#define CEPH_FEATURE_SUPPORTED_DEFAULT CEPH_FEATURE_NOSRCADDR
+#define CEPH_FEATURE_REQUIRED_DEFAULT  CEPH_FEATURE_NOSRCADDR
+
+/*
+ * mount options
+ */
+#define CEPH_OPT_FSID             (1<<0)
+#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
+#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
+#define CEPH_OPT_NOCRC            (1<<3) /* no data crc on writes */
+
+#define CEPH_OPT_DEFAULT   (0);
+
+#define ceph_set_opt(client, opt) \
+	(client)->options->flags |= CEPH_OPT_##opt;
+#define ceph_test_opt(client, opt) \
+	(!!((client)->options->flags & CEPH_OPT_##opt))
+
+struct ceph_options {
+	int flags;
+	struct ceph_fsid fsid;
+	struct ceph_entity_addr my_addr;
+	int mount_timeout;
+	int osd_idle_ttl;
+	int osd_timeout;
+	int osd_keepalive_timeout;
+
+	/*
+	 * any type that can't be simply compared or doesn't need need
+	 * to be compared should go beyond this point,
+	 * ceph_compare_options() should be updated accordingly
+	 */
+
+	struct ceph_entity_addr *mon_addr; /* should be the first
+					      pointer type of args */
+	int num_mon;
+	char *name;
+	char *secret;
+};
+
+/*
+ * defaults
+ */
+#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
+#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
+#define CEPH_OSD_KEEPALIVE_DEFAULT  5
+#define CEPH_OSD_IDLE_TTL_DEFAULT    60
+#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
+
+#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
+#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
+
+#define CEPH_AUTH_NAME_DEFAULT   "guest"
+
+/*
+ * Delay telling the MDS we no longer want caps, in case we reopen
+ * the file.  Delay a minimum amount of time, even if we send a cap
+ * message for some other reason.  Otherwise, take the oppotunity to
+ * update the mds to avoid sending another message later.
+ */
+#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
+#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
+
+#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
+
+/* mount state */
+enum {
+	CEPH_MOUNT_MOUNTING,
+	CEPH_MOUNT_MOUNTED,
+	CEPH_MOUNT_UNMOUNTING,
+	CEPH_MOUNT_UNMOUNTED,
+	CEPH_MOUNT_SHUTDOWN,
+};
+
+/*
+ * subtract jiffies
+ */
+static inline unsigned long time_sub(unsigned long a, unsigned long b)
+{
+	BUG_ON(time_after(b, a));
+	return (long)a - (long)b;
+}
+
+struct ceph_mds_client;
+
+/*
+ * per client state
+ *
+ * possibly shared by multiple mount points, if they are
+ * mounting the same ceph filesystem/cluster.
+ */
+struct ceph_client {
+	struct ceph_fsid fsid;
+	bool have_fsid;
+
+	void *private;
+
+	struct ceph_options *options;
+
+	struct mutex mount_mutex;      /* serialize mount attempts */
+	wait_queue_head_t auth_wq;
+	int auth_err;
+
+	int (*extra_mon_dispatch)(struct ceph_client *, struct ceph_msg *);
+
+	u32 supported_features;
+	u32 required_features;
+
+	struct ceph_messenger *msgr;   /* messenger instance */
+	struct ceph_mon_client monc;
+	struct ceph_osd_client osdc;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *debugfs_dir;
+	struct dentry *debugfs_monmap;
+	struct dentry *debugfs_osdmap;
+#endif
+};
+
+
+
+/*
+ * snapshots
+ */
+
+/*
+ * A "snap context" is the set of existing snapshots when we
+ * write data.  It is used by the OSD to guide its COW behavior.
+ *
+ * The ceph_snap_context is refcounted, and attached to each dirty
+ * page, indicating which context the dirty data belonged when it was
+ * dirtied.
+ */
+struct ceph_snap_context {
+	atomic_t nref;
+	u64 seq;
+	int num_snaps;
+	u64 snaps[];
+};
+
+static inline struct ceph_snap_context *
+ceph_get_snap_context(struct ceph_snap_context *sc)
+{
+	/*
+	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)+1);
+	*/
+	if (sc)
+		atomic_inc(&sc->nref);
+	return sc;
+}
+
+static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
+{
+	if (!sc)
+		return;
+	/*
+	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)-1);
+	*/
+	if (atomic_dec_and_test(&sc->nref)) {
+		/*printk(" deleting snap_context %p\n", sc);*/
+		kfree(sc);
+	}
+}
+
+/*
+ * calculate the number of pages a given length and offset map onto,
+ * if we align the data.
+ */
+static inline int calc_pages_for(u64 off, u64 len)
+{
+	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
+		(off >> PAGE_CACHE_SHIFT);
+}
+
+/* ceph_common.c */
+extern const char *ceph_msg_type_name(int type);
+extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
+extern struct kmem_cache *ceph_inode_cachep;
+extern struct kmem_cache *ceph_cap_cachep;
+extern struct kmem_cache *ceph_dentry_cachep;
+extern struct kmem_cache *ceph_file_cachep;
+
+extern int ceph_parse_options(struct ceph_options **popt, char *options,
+			      const char *dev_name, const char *dev_name_end,
+			      int (*parse_extra_token)(char *c, void *private),
+			      void *private);
+extern void ceph_destroy_options(struct ceph_options *opt);
+extern int ceph_compare_options(struct ceph_options *new_opt,
+				struct ceph_client *client);
+extern struct ceph_client *ceph_create_client(struct ceph_options *opt,
+					      void *private);
+extern u64 ceph_client_id(struct ceph_client *client);
+extern void ceph_destroy_client(struct ceph_client *client);
+extern int __ceph_open_session(struct ceph_client *client,
+			       unsigned long started);
+extern int ceph_open_session(struct ceph_client *client);
+
+/* pagevec.c */
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+
+extern struct page **ceph_get_direct_page_vector(const char __user *data,
+					    int num_pages,
+					    loff_t off, size_t len);
+extern void ceph_put_page_vector(struct page **pages, int num_pages);
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
+extern int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len);
+extern int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_page_vector_to_user(struct page **pages, char __user *data,
+				    loff_t off, size_t len);
+extern void ceph_zero_page_vector_range(int off, int len, struct page **pages);
+
+
+#endif /* _FS_CEPH_SUPER_H */
diff --git a/fs/ceph/mdsmap.h b/include/linux/ceph/mdsmap.h
similarity index 100%
rename from fs/ceph/mdsmap.h
rename to include/linux/ceph/mdsmap.h
diff --git a/fs/ceph/messenger.h b/include/linux/ceph/messenger.h
similarity index 95%
rename from fs/ceph/messenger.h
rename to include/linux/ceph/messenger.h
index 76fbc95..5956d62 100644
--- a/fs/ceph/messenger.h
+++ b/include/linux/ceph/messenger.h
@@ -65,6 +65,9 @@
 	 */
 	u32 global_seq;
 	spinlock_t global_seq_lock;
+
+	u32 supported_features;
+	u32 required_features;
 };
 
 /*
@@ -82,6 +85,10 @@
 	struct ceph_pagelist *pagelist; /* instead of pages */
 	struct list_head list_head;
 	struct kref kref;
+	struct bio  *bio;		/* instead of pages/pagelist */
+	struct bio  *bio_iter;		/* bio iterator */
+	int bio_seg;			/* current bio segment */
+	struct ceph_pagelist *trail;	/* the trailing part of the data */
 	bool front_is_vmalloc;
 	bool more_to_follow;
 	bool needs_out_seq;
@@ -205,7 +212,7 @@
 };
 
 
-extern const char *pr_addr(const struct sockaddr_storage *ss);
+extern const char *ceph_pr_addr(const struct sockaddr_storage *ss);
 extern int ceph_parse_ips(const char *c, const char *end,
 			  struct ceph_entity_addr *addr,
 			  int max_count, int *count);
@@ -216,7 +223,8 @@
 extern void ceph_msgr_flush(void);
 
 extern struct ceph_messenger *ceph_messenger_create(
-	struct ceph_entity_addr *myaddr);
+	struct ceph_entity_addr *myaddr,
+	u32 features, u32 required);
 extern void ceph_messenger_destroy(struct ceph_messenger *);
 
 extern void ceph_con_init(struct ceph_messenger *msgr,
diff --git a/fs/ceph/mon_client.h b/include/linux/ceph/mon_client.h
similarity index 99%
rename from fs/ceph/mon_client.h
rename to include/linux/ceph/mon_client.h
index 8e396f2..545f859 100644
--- a/fs/ceph/mon_client.h
+++ b/include/linux/ceph/mon_client.h
@@ -79,6 +79,7 @@
 	u64 last_tid;
 
 	/* mds/osd map */
+	int want_mdsmap;
 	int want_next_osdmap; /* 1 = want, 2 = want+asked */
 	u32 have_osdmap, have_mdsmap;
 
diff --git a/fs/ceph/msgpool.h b/include/linux/ceph/msgpool.h
similarity index 100%
rename from fs/ceph/msgpool.h
rename to include/linux/ceph/msgpool.h
diff --git a/fs/ceph/msgr.h b/include/linux/ceph/msgr.h
similarity index 100%
rename from fs/ceph/msgr.h
rename to include/linux/ceph/msgr.h
diff --git a/fs/ceph/osd_client.h b/include/linux/ceph/osd_client.h
similarity index 75%
rename from fs/ceph/osd_client.h
rename to include/linux/ceph/osd_client.h
index ce77698..6c91fb0 100644
--- a/fs/ceph/osd_client.h
+++ b/include/linux/ceph/osd_client.h
@@ -15,6 +15,7 @@
 struct ceph_osd_request;
 struct ceph_osd_client;
 struct ceph_authorizer;
+struct ceph_pagelist;
 
 /*
  * completion callback for async writepages
@@ -68,6 +69,7 @@
 	struct list_head  r_unsafe_item;
 
 	struct inode *r_inode;         	      /* for use by callbacks */
+	void *r_priv;			      /* ditto */
 
 	char              r_oid[40];          /* object name */
 	int               r_oid_len;
@@ -80,6 +82,11 @@
 	struct page     **r_pages;            /* pages for data payload */
 	int               r_pages_from_pool;
 	int               r_own_pages;        /* if true, i own page list */
+#ifdef CONFIG_BLOCK
+	struct bio       *r_bio;	      /* instead of pages */
+#endif
+
+	struct ceph_pagelist *r_trail;	      /* trailing part of the data */
 };
 
 struct ceph_osd_client {
@@ -110,6 +117,42 @@
 	struct ceph_msgpool	msgpool_op_reply;
 };
 
+struct ceph_osd_req_op {
+	u16 op;           /* CEPH_OSD_OP_* */
+	u32 flags;        /* CEPH_OSD_FLAG_* */
+	union {
+		struct {
+			u64 offset, length;
+			u64 truncate_size;
+			u32 truncate_seq;
+		} extent;
+		struct {
+			const char *name;
+			u32 name_len;
+			const char  *val;
+			u32 value_len;
+			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
+			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
+		} xattr;
+		struct {
+			const char *class_name;
+			__u8 class_len;
+			const char *method_name;
+			__u8 method_len;
+			__u8 argc;
+			const char *indata;
+			u32 indata_len;
+		} cls;
+		struct {
+			u64 cookie, count;
+		} pgls;
+	        struct {
+		        u64 snapid;
+	        } snap;
+	};
+	u32 payload_len;
+};
+
 extern int ceph_osdc_init(struct ceph_osd_client *osdc,
 			  struct ceph_client *client);
 extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
@@ -119,6 +162,30 @@
 extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
 				 struct ceph_msg *msg);
 
+extern void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op);
+
+extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio);
+
+extern void ceph_osdc_build_request(struct ceph_osd_request *req,
+				    u64 off, u64 *plen,
+				    struct ceph_osd_req_op *src_ops,
+				    struct ceph_snap_context *snapc,
+				    struct timespec *mtime,
+				    const char *oid,
+				    int oid_len);
+
 extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
 				      struct ceph_file_layout *layout,
 				      struct ceph_vino vino,
diff --git a/fs/ceph/osdmap.h b/include/linux/ceph/osdmap.h
similarity index 96%
rename from fs/ceph/osdmap.h
rename to include/linux/ceph/osdmap.h
index 970b547..ba4c205 100644
--- a/fs/ceph/osdmap.h
+++ b/include/linux/ceph/osdmap.h
@@ -4,7 +4,7 @@
 #include <linux/rbtree.h>
 #include "types.h"
 #include "ceph_fs.h"
-#include "crush/crush.h"
+#include <linux/crush/crush.h>
 
 /*
  * The osd map describes the current membership of the osd cluster and
@@ -125,4 +125,6 @@
 extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
 				struct ceph_pg pgid);
 
+extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name);
+
 #endif
diff --git a/fs/ceph/pagelist.h b/include/linux/ceph/pagelist.h
similarity index 62%
rename from fs/ceph/pagelist.h
rename to include/linux/ceph/pagelist.h
index e8a4187..9660d6b 100644
--- a/fs/ceph/pagelist.h
+++ b/include/linux/ceph/pagelist.h
@@ -8,6 +8,14 @@
 	void *mapped_tail;
 	size_t length;
 	size_t room;
+	struct list_head free_list;
+	size_t num_pages_free;
+};
+
+struct ceph_pagelist_cursor {
+	struct ceph_pagelist *pl;   /* pagelist, for error checking */
+	struct list_head *page_lru; /* page in list */
+	size_t room;		    /* room remaining to reset to */
 };
 
 static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
@@ -16,10 +24,23 @@
 	pl->mapped_tail = NULL;
 	pl->length = 0;
 	pl->room = 0;
+	INIT_LIST_HEAD(&pl->free_list);
+	pl->num_pages_free = 0;
 }
+
 extern int ceph_pagelist_release(struct ceph_pagelist *pl);
 
-extern int ceph_pagelist_append(struct ceph_pagelist *pl, void *d, size_t l);
+extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l);
+
+extern int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space);
+
+extern int ceph_pagelist_free_reserve(struct ceph_pagelist *pl);
+
+extern void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+				     struct ceph_pagelist_cursor *c);
+
+extern int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+				  struct ceph_pagelist_cursor *c);
 
 static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v)
 {
diff --git a/fs/ceph/rados.h b/include/linux/ceph/rados.h
similarity index 100%
rename from fs/ceph/rados.h
rename to include/linux/ceph/rados.h
diff --git a/fs/ceph/types.h b/include/linux/ceph/types.h
similarity index 100%
rename from fs/ceph/types.h
rename to include/linux/ceph/types.h
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 0c99102..709dfb9 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -75,7 +75,7 @@
 
 	unsigned long flags;
 	/* ID for this css, if possible */
-	struct css_id *id;
+	struct css_id __rcu *id;
 };
 
 /* bits in struct cgroup_subsys_state flags field */
@@ -205,7 +205,7 @@
 	struct list_head children;	/* my children */
 
 	struct cgroup *parent;		/* my parent */
-	struct dentry *dentry;	  	/* cgroup fs entry, RCU protected */
+	struct dentry __rcu *dentry;	/* cgroup fs entry, RCU protected */
 
 	/* Private pointers for each registered subsystem */
 	struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index c1a62c5..320d6c9 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -16,7 +16,11 @@
 # define __release(x)	__context__(x,-1)
 # define __cond_lock(x,c)	((c) ? ({ __acquire(x); 1; }) : 0)
 # define __percpu	__attribute__((noderef, address_space(3)))
+#ifdef CONFIG_SPARSE_RCU_POINTER
+# define __rcu		__attribute__((noderef, address_space(4)))
+#else
 # define __rcu
+#endif
 extern void __chk_user_ptr(const volatile void __user *);
 extern void __chk_io_ptr(const volatile void __iomem *);
 #else
diff --git a/include/linux/coredump.h b/include/linux/coredump.h
index 8ba66a9..ba4b85a 100644
--- a/include/linux/coredump.h
+++ b/include/linux/coredump.h
@@ -9,37 +9,7 @@
  * These are the only things you should do on a core-file: use only these
  * functions to write out all the necessary info.
  */
-static inline int dump_write(struct file *file, const void *addr, int nr)
-{
-	return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
-
-static inline int dump_seek(struct file *file, loff_t off)
-{
-	int ret = 1;
-
-	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
-		if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
-			return 0;
-	} else {
-		char *buf = (char *)get_zeroed_page(GFP_KERNEL);
-
-		if (!buf)
-			return 0;
-		while (off > 0) {
-			unsigned long n = off;
-
-			if (n > PAGE_SIZE)
-				n = PAGE_SIZE;
-			if (!dump_write(file, buf, n)) {
-				ret = 0;
-				break;
-			}
-			off -= n;
-		}
-		free_page((unsigned long)buf);
-	}
-	return ret;
-}
+extern int dump_write(struct file *file, const void *addr, int nr);
+extern int dump_seek(struct file *file, loff_t off);
 
 #endif /* _LINUX_COREDUMP_H */
diff --git a/include/linux/cred.h b/include/linux/cred.h
index 4d2c395..4aaeab3 100644
--- a/include/linux/cred.h
+++ b/include/linux/cred.h
@@ -84,7 +84,7 @@
 	atomic_t	usage;
 	pid_t		tgid;			/* thread group process ID */
 	spinlock_t	lock;
-	struct key	*session_keyring;	/* keyring inherited over fork */
+	struct key __rcu *session_keyring;	/* keyring inherited over fork */
 	struct key	*process_keyring;	/* keyring private to this process */
 	struct rcu_head	rcu;			/* RCU deletion hook */
 };
diff --git a/fs/ceph/crush/crush.h b/include/linux/crush/crush.h
similarity index 100%
rename from fs/ceph/crush/crush.h
rename to include/linux/crush/crush.h
diff --git a/fs/ceph/crush/hash.h b/include/linux/crush/hash.h
similarity index 100%
rename from fs/ceph/crush/hash.h
rename to include/linux/crush/hash.h
diff --git a/fs/ceph/crush/mapper.h b/include/linux/crush/mapper.h
similarity index 100%
rename from fs/ceph/crush/mapper.h
rename to include/linux/crush/mapper.h
diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h
index 29b3ce3..2833452 100644
--- a/include/linux/debug_locks.h
+++ b/include/linux/debug_locks.h
@@ -49,7 +49,6 @@
 
 #ifdef CONFIG_LOCKDEP
 extern void debug_show_all_locks(void);
-extern void __debug_show_held_locks(struct task_struct *task);
 extern void debug_show_held_locks(struct task_struct *task);
 extern void debug_check_no_locks_freed(const void *from, unsigned long len);
 extern void debug_check_no_locks_held(struct task_struct *task);
@@ -58,10 +57,6 @@
 {
 }
 
-static inline void __debug_show_held_locks(struct task_struct *task)
-{
-}
-
 static inline void debug_show_held_locks(struct task_struct *task)
 {
 }
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index d7cecc9..a7d9dc2 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -57,15 +57,15 @@
 extern int dmar_dev_scope_init(void);
 
 /* Intel IOMMU detection */
-extern void detect_intel_iommu(void);
+extern int detect_intel_iommu(void);
 extern int enable_drhd_fault_handling(void);
 
 extern int parse_ioapics_under_ir(void);
 extern int alloc_iommu(struct dmar_drhd_unit *);
 #else
-static inline void detect_intel_iommu(void)
+static inline int detect_intel_iommu(void)
 {
-	return;
+	return -ENODEV;
 }
 
 static inline int dmar_table_init(void)
@@ -106,6 +106,7 @@
 		__u64 high;
 	};
 };
+
 #ifdef CONFIG_INTR_REMAP
 extern int intr_remapping_enabled;
 extern int intr_remapping_supported(void);
@@ -119,11 +120,8 @@
 extern int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index,
    			u16 sub_handle);
 extern int map_irq_to_irte_handle(int irq, u16 *sub_handle);
-extern int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index);
-extern int flush_irte(int irq);
 extern int free_irte(int irq);
 
-extern int irq_remapped(int irq);
 extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev);
 extern struct intel_iommu *map_ioapic_to_ir(int apic);
 extern struct intel_iommu *map_hpet_to_ir(u8 id);
@@ -177,7 +175,6 @@
 	return 0;
 }
 
-#define irq_remapped(irq)		(0)
 #define enable_intr_remapping(mode)	(-1)
 #define disable_intr_remapping()	(0)
 #define reenable_intr_remapping(mode)	(0)
@@ -187,8 +184,9 @@
 /* Can't use the common MSI interrupt functions
  * since DMAR is not a pci device
  */
-extern void dmar_msi_unmask(unsigned int irq);
-extern void dmar_msi_mask(unsigned int irq);
+struct irq_data;
+extern void dmar_msi_unmask(struct irq_data *data);
+extern void dmar_msi_mask(struct irq_data *data);
 extern void dmar_msi_read(int irq, struct msi_msg *msg);
 extern void dmar_msi_write(int irq, struct msi_msg *msg);
 extern int dmar_set_interrupt(struct intel_iommu *iommu);
diff --git a/include/linux/dynamic_debug.h b/include/linux/dynamic_debug.h
index 52c0da4..bef3cda 100644
--- a/include/linux/dynamic_debug.h
+++ b/include/linux/dynamic_debug.h
@@ -1,6 +1,8 @@
 #ifndef _DYNAMIC_DEBUG_H
 #define _DYNAMIC_DEBUG_H
 
+#include <linux/jump_label.h>
+
 /* dynamic_printk_enabled, and dynamic_printk_enabled2 are bitmasks in which
  * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They
  * use independent hash functions, to reduce the chance of false positives.
@@ -22,8 +24,6 @@
 	const char *function;
 	const char *filename;
 	const char *format;
-	char primary_hash;
-	char secondary_hash;
 	unsigned int lineno:24;
 	/*
  	 * The flags field controls the behaviour at the callsite.
@@ -33,6 +33,7 @@
 #define _DPRINTK_FLAGS_PRINT   (1<<0)  /* printk() a message using the format */
 #define _DPRINTK_FLAGS_DEFAULT 0
 	unsigned int flags:8;
+	char enabled;
 } __attribute__((aligned(8)));
 
 
@@ -42,33 +43,35 @@
 #if defined(CONFIG_DYNAMIC_DEBUG)
 extern int ddebug_remove_module(const char *mod_name);
 
-#define __dynamic_dbg_enabled(dd)  ({	     \
-	int __ret = 0;							     \
-	if (unlikely((dynamic_debug_enabled & (1LL << DEBUG_HASH)) &&	     \
-			(dynamic_debug_enabled2 & (1LL << DEBUG_HASH2))))   \
-				if (unlikely(dd.flags))			     \
-					__ret = 1;			     \
-	__ret; })
-
 #define dynamic_pr_debug(fmt, ...) do {					\
+	__label__ do_printk;						\
+	__label__ out;							\
 	static struct _ddebug descriptor				\
 	__used								\
 	__attribute__((section("__verbose"), aligned(8))) =		\
-	{ KBUILD_MODNAME, __func__, __FILE__, fmt, DEBUG_HASH,	\
-		DEBUG_HASH2, __LINE__, _DPRINTK_FLAGS_DEFAULT };	\
-	if (__dynamic_dbg_enabled(descriptor))				\
-		printk(KERN_DEBUG pr_fmt(fmt),	##__VA_ARGS__);		\
+	{ KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,		\
+		_DPRINTK_FLAGS_DEFAULT };				\
+	JUMP_LABEL(&descriptor.enabled, do_printk);			\
+	goto out;							\
+do_printk:								\
+	printk(KERN_DEBUG pr_fmt(fmt),	##__VA_ARGS__);			\
+out:	;								\
 	} while (0)
 
 
 #define dynamic_dev_dbg(dev, fmt, ...) do {				\
+	__label__ do_printk;						\
+	__label__ out;							\
 	static struct _ddebug descriptor				\
 	__used								\
 	__attribute__((section("__verbose"), aligned(8))) =		\
-	{ KBUILD_MODNAME, __func__, __FILE__, fmt, DEBUG_HASH,	\
-		DEBUG_HASH2, __LINE__, _DPRINTK_FLAGS_DEFAULT };	\
-	if (__dynamic_dbg_enabled(descriptor))				\
-		dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);	\
+	{ KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,		\
+		_DPRINTK_FLAGS_DEFAULT };				\
+	JUMP_LABEL(&descriptor.enabled, do_printk);			\
+	goto out;							\
+do_printk:								\
+	dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);		\
+out:	;								\
 	} while (0)
 
 #else
diff --git a/include/linux/edac.h b/include/linux/edac.h
index 7cf92e8..36c6644 100644
--- a/include/linux/edac.h
+++ b/include/linux/edac.h
@@ -13,6 +13,7 @@
 #define _LINUX_EDAC_H_
 
 #include <asm/atomic.h>
+#include <linux/sysdev.h>
 
 #define EDAC_OPSTATE_INVAL	-1
 #define EDAC_OPSTATE_POLL	0
@@ -22,9 +23,12 @@
 extern int edac_op_state;
 extern int edac_err_assert;
 extern atomic_t edac_handlers;
+extern struct sysdev_class edac_class;
 
 extern int edac_handler_set(void);
 extern void edac_atomic_assert_error(void);
+extern struct sysdev_class *edac_get_sysfs_class(void);
+extern void edac_put_sysfs_class(void);
 
 static inline void opstate_init(void)
 {
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 926b503..4fd978e 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -93,6 +93,7 @@
 	struct elevator_type *elevator_type;
 	struct mutex sysfs_lock;
 	struct hlist_head *hash;
+	unsigned int registered:1;
 };
 
 /*
diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h
index f59ed29..133c0ba 100644
--- a/include/linux/fdtable.h
+++ b/include/linux/fdtable.h
@@ -31,7 +31,7 @@
 
 struct fdtable {
 	unsigned int max_fds;
-	struct file ** fd;      /* current fd array */
+	struct file __rcu **fd;      /* current fd array */
 	fd_set *close_on_exec;
 	fd_set *open_fds;
 	struct rcu_head rcu;
@@ -46,7 +46,7 @@
    * read mostly part
    */
 	atomic_t count;
-	struct fdtable *fdt;
+	struct fdtable __rcu *fdt;
 	struct fdtable fdtab;
   /*
    * written part on a separate cache line in SMP
@@ -55,7 +55,7 @@
 	int next_fd;
 	struct embedded_fd_set close_on_exec_init;
 	struct embedded_fd_set open_fds_init;
-	struct file * fd_array[NR_OPEN_DEFAULT];
+	struct file __rcu * fd_array[NR_OPEN_DEFAULT];
 };
 
 #define rcu_dereference_check_fdtable(files, fdtfd) \
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 63d069b..3168dcf 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1384,7 +1384,7 @@
 	 * Saved mount options for lazy filesystems using
 	 * generic_show_options()
 	 */
-	char *s_options;
+	char __rcu *s_options;
 };
 
 extern struct timespec current_fs_time(struct super_block *sb);
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 02b8b24..8beabb9 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -191,8 +191,8 @@
 	unsigned int		flags;
 
 #ifdef CONFIG_PERF_EVENTS
-	int			perf_refcount;
-	struct hlist_head	*perf_events;
+	int				perf_refcount;
+	struct hlist_head __percpu	*perf_events;
 #endif
 };
 
@@ -252,8 +252,8 @@
 
 extern int  perf_trace_init(struct perf_event *event);
 extern void perf_trace_destroy(struct perf_event *event);
-extern int  perf_trace_enable(struct perf_event *event);
-extern void perf_trace_disable(struct perf_event *event);
+extern int  perf_trace_add(struct perf_event *event, int flags);
+extern void perf_trace_del(struct perf_event *event, int flags);
 extern int  ftrace_profile_set_filter(struct perf_event *event, int event_id,
 				     char *filter_str);
 extern void ftrace_profile_free_filter(struct perf_event *event);
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 5f2f4c4..af3f06b 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -129,8 +129,8 @@
 struct disk_part_tbl {
 	struct rcu_head rcu_head;
 	int len;
-	struct hd_struct *last_lookup;
-	struct hd_struct *part[];
+	struct hd_struct __rcu *last_lookup;
+	struct hd_struct __rcu *part[];
 };
 
 struct gendisk {
@@ -149,7 +149,7 @@
 	 * non-critical accesses use RCU.  Always access through
 	 * helpers.
 	 */
-	struct disk_part_tbl *part_tbl;
+	struct disk_part_tbl __rcu *part_tbl;
 	struct hd_struct part0;
 
 	const struct block_device_operations *fops;
diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 7dfdc06..8a389b6 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -63,6 +63,8 @@
 #define HARDIRQ_OFFSET	(1UL << HARDIRQ_SHIFT)
 #define NMI_OFFSET	(1UL << NMI_SHIFT)
 
+#define SOFTIRQ_DISABLE_OFFSET	(2 * SOFTIRQ_OFFSET)
+
 #ifndef PREEMPT_ACTIVE
 #define PREEMPT_ACTIVE_BITS	1
 #define PREEMPT_ACTIVE_SHIFT	(NMI_SHIFT + NMI_BITS)
@@ -81,10 +83,13 @@
 /*
  * Are we doing bottom half or hardware interrupt processing?
  * Are we in a softirq context? Interrupt context?
+ * in_softirq - Are we currently processing softirq or have bh disabled?
+ * in_serving_softirq - Are we currently processing softirq?
  */
 #define in_irq()		(hardirq_count())
 #define in_softirq()		(softirq_count())
 #define in_interrupt()		(irq_count())
+#define in_serving_softirq()	(softirq_count() & SOFTIRQ_OFFSET)
 
 /*
  * Are we in NMI context?
@@ -131,14 +136,16 @@
 
 struct task_struct;
 
-#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+#if !defined(CONFIG_VIRT_CPU_ACCOUNTING) && !defined(CONFIG_IRQ_TIME_ACCOUNTING)
 static inline void account_system_vtime(struct task_struct *tsk)
 {
 }
+#else
+extern void account_system_vtime(struct task_struct *tsk);
 #endif
 
 #if defined(CONFIG_NO_HZ)
-#if defined(CONFIG_TINY_RCU)
+#if defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 extern void rcu_enter_nohz(void);
 extern void rcu_exit_nohz(void);
 
diff --git a/include/linux/htirq.h b/include/linux/htirq.h
index c96ea46..70a1dbb 100644
--- a/include/linux/htirq.h
+++ b/include/linux/htirq.h
@@ -9,8 +9,9 @@
 /* Helper functions.. */
 void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg);
 void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg);
-void mask_ht_irq(unsigned int irq);
-void unmask_ht_irq(unsigned int irq);
+struct irq_data;
+void mask_ht_irq(struct irq_data *data);
+void unmask_ht_irq(struct irq_data *data);
 
 /* The arch hook for getting things started */
 int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev);
diff --git a/include/linux/idr.h b/include/linux/idr.h
index e968db7..cdb715e 100644
--- a/include/linux/idr.h
+++ b/include/linux/idr.h
@@ -50,14 +50,14 @@
 
 struct idr_layer {
 	unsigned long		 bitmap; /* A zero bit means "space here" */
-	struct idr_layer	*ary[1<<IDR_BITS];
+	struct idr_layer __rcu	*ary[1<<IDR_BITS];
 	int			 count;	 /* When zero, we can release it */
 	int			 layer;	 /* distance from leaf */
 	struct rcu_head		 rcu_head;
 };
 
 struct idr {
-	struct idr_layer *top;
+	struct idr_layer __rcu *top;
 	struct idr_layer *id_free;
 	int		  layers; /* only valid without concurrent changes */
 	int		  id_free_cnt;
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 1f43fa5..2fea6c8 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -82,11 +82,17 @@
 # define CAP_INIT_BSET  CAP_FULL_SET
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
+#define INIT_TASK_RCU_TREE_PREEMPT()					\
+	.rcu_blocked_node = NULL,
+#else
+#define INIT_TASK_RCU_TREE_PREEMPT(tsk)
+#endif
+#ifdef CONFIG_PREEMPT_RCU
 #define INIT_TASK_RCU_PREEMPT(tsk)					\
 	.rcu_read_lock_nesting = 0,					\
 	.rcu_read_unlock_special = 0,					\
-	.rcu_blocked_node = NULL,					\
-	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),
+	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),		\
+	INIT_TASK_RCU_TREE_PREEMPT()
 #else
 #define INIT_TASK_RCU_PREEMPT(tsk)
 #endif
@@ -137,8 +143,8 @@
 	.children	= LIST_HEAD_INIT(tsk.children),			\
 	.sibling	= LIST_HEAD_INIT(tsk.sibling),			\
 	.group_leader	= &tsk,						\
-	.real_cred	= &init_cred,					\
-	.cred		= &init_cred,					\
+	RCU_INIT_POINTER(.real_cred, &init_cred),			\
+	RCU_INIT_POINTER(.cred, &init_cred),				\
 	.cred_guard_mutex =						\
 		 __MUTEX_INITIALIZER(tsk.cred_guard_mutex),		\
 	.comm		= "swapper",					\
diff --git a/include/linux/input.h b/include/linux/input.h
index 896a922..d6ae176 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -1196,7 +1196,7 @@
 	int (*flush)(struct input_dev *dev, struct file *file);
 	int (*event)(struct input_dev *dev, unsigned int type, unsigned int code, int value);
 
-	struct input_handle *grab;
+	struct input_handle __rcu *grab;
 
 	spinlock_t event_lock;
 	struct mutex mutex;
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a0384a4..4143285 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -18,6 +18,7 @@
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
 #include <asm/system.h>
+#include <trace/events/irq.h>
 
 /*
  * These correspond to the IORESOURCE_IRQ_* defines in
@@ -407,7 +408,12 @@
 asmlinkage void __do_softirq(void);
 extern void open_softirq(int nr, void (*action)(struct softirq_action *));
 extern void softirq_init(void);
-#define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
+static inline void __raise_softirq_irqoff(unsigned int nr)
+{
+	trace_softirq_raise((struct softirq_action *)(unsigned long)nr, NULL);
+	or_softirq_pending(1UL << nr);
+}
+
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
 extern void wakeup_softirqd(void);
@@ -641,11 +647,8 @@
 struct seq_file;
 int show_interrupts(struct seq_file *p, void *v);
 
-struct irq_desc;
-
 extern int early_irq_init(void);
 extern int arch_probe_nr_irqs(void);
 extern int arch_early_irq_init(void);
-extern int arch_init_chip_data(struct irq_desc *desc, int node);
 
 #endif
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index 64d5291..3e70b21 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -53,7 +53,7 @@
 
 	struct radix_tree_root radix_root;
 	struct hlist_head cic_list;
-	void *ioc_data;
+	void __rcu *ioc_data;
 };
 
 static inline struct io_context *ioc_task_link(struct io_context *ioc)
diff --git a/include/linux/irq.h b/include/linux/irq.h
index c03243a..e963911 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -72,6 +72,10 @@
 #define IRQ_ONESHOT		0x08000000	/* IRQ is not unmasked after hardirq */
 #define IRQ_NESTED_THREAD	0x10000000	/* IRQ is nested into another, no own handler thread */
 
+#define IRQF_MODIFY_MASK	\
+	(IRQ_TYPE_SENSE_MASK | IRQ_NOPROBE | IRQ_NOREQUEST | \
+	 IRQ_NOAUTOEN | IRQ_MOVE_PCNTXT | IRQ_LEVEL)
+
 #ifdef CONFIG_IRQ_PER_CPU
 # define CHECK_IRQ_PER_CPU(var) ((var) & IRQ_PER_CPU)
 # define IRQ_NO_BALANCING_MASK	(IRQ_PER_CPU | IRQ_NO_BALANCING)
@@ -80,36 +84,77 @@
 # define IRQ_NO_BALANCING_MASK	IRQ_NO_BALANCING
 #endif
 
-struct proc_dir_entry;
 struct msi_desc;
 
 /**
+ * struct irq_data - per irq and irq chip data passed down to chip functions
+ * @irq:		interrupt number
+ * @node:		node index useful for balancing
+ * @chip:		low level interrupt hardware access
+ * @handler_data:	per-IRQ data for the irq_chip methods
+ * @chip_data:		platform-specific per-chip private data for the chip
+ *			methods, to allow shared chip implementations
+ * @msi_desc:		MSI descriptor
+ * @affinity:		IRQ affinity on SMP
+ *
+ * The fields here need to overlay the ones in irq_desc until we
+ * cleaned up the direct references and switched everything over to
+ * irq_data.
+ */
+struct irq_data {
+	unsigned int		irq;
+	unsigned int		node;
+	struct irq_chip		*chip;
+	void			*handler_data;
+	void			*chip_data;
+	struct msi_desc		*msi_desc;
+#ifdef CONFIG_SMP
+	cpumask_var_t		affinity;
+#endif
+};
+
+/**
  * struct irq_chip - hardware interrupt chip descriptor
  *
  * @name:		name for /proc/interrupts
- * @startup:		start up the interrupt (defaults to ->enable if NULL)
- * @shutdown:		shut down the interrupt (defaults to ->disable if NULL)
- * @enable:		enable the interrupt (defaults to chip->unmask if NULL)
- * @disable:		disable the interrupt
- * @ack:		start of a new interrupt
- * @mask:		mask an interrupt source
- * @mask_ack:		ack and mask an interrupt source
- * @unmask:		unmask an interrupt source
- * @eoi:		end of interrupt - chip level
- * @end:		end of interrupt - flow level
- * @set_affinity:	set the CPU affinity on SMP machines
- * @retrigger:		resend an IRQ to the CPU
- * @set_type:		set the flow type (IRQ_TYPE_LEVEL/etc.) of an IRQ
- * @set_wake:		enable/disable power-management wake-on of an IRQ
+ * @startup:		deprecated, replaced by irq_startup
+ * @shutdown:		deprecated, replaced by irq_shutdown
+ * @enable:		deprecated, replaced by irq_enable
+ * @disable:		deprecated, replaced by irq_disable
+ * @ack:		deprecated, replaced by irq_ack
+ * @mask:		deprecated, replaced by irq_mask
+ * @mask_ack:		deprecated, replaced by irq_mask_ack
+ * @unmask:		deprecated, replaced by irq_unmask
+ * @eoi:		deprecated, replaced by irq_eoi
+ * @end:		deprecated, will go away with __do_IRQ()
+ * @set_affinity:	deprecated, replaced by irq_set_affinity
+ * @retrigger:		deprecated, replaced by irq_retrigger
+ * @set_type:		deprecated, replaced by irq_set_type
+ * @set_wake:		deprecated, replaced by irq_wake
+ * @bus_lock:		deprecated, replaced by irq_bus_lock
+ * @bus_sync_unlock:	deprecated, replaced by irq_bus_sync_unlock
  *
- * @bus_lock:		function to lock access to slow bus (i2c) chips
- * @bus_sync_unlock:	function to sync and unlock slow bus (i2c) chips
+ * @irq_startup:	start up the interrupt (defaults to ->enable if NULL)
+ * @irq_shutdown:	shut down the interrupt (defaults to ->disable if NULL)
+ * @irq_enable:		enable the interrupt (defaults to chip->unmask if NULL)
+ * @irq_disable:	disable the interrupt
+ * @irq_ack:		start of a new interrupt
+ * @irq_mask:		mask an interrupt source
+ * @irq_mask_ack:	ack and mask an interrupt source
+ * @irq_unmask:		unmask an interrupt source
+ * @irq_eoi:		end of interrupt
+ * @irq_set_affinity:	set the CPU affinity on SMP machines
+ * @irq_retrigger:	resend an IRQ to the CPU
+ * @irq_set_type:	set the flow type (IRQ_TYPE_LEVEL/etc.) of an IRQ
+ * @irq_set_wake:	enable/disable power-management wake-on of an IRQ
+ * @irq_bus_lock:	function to lock access to slow bus (i2c) chips
+ * @irq_bus_sync_unlock:function to sync and unlock slow bus (i2c) chips
  *
  * @release:		release function solely used by UML
- * @typename:		obsoleted by name, kept as migration helper
  */
 struct irq_chip {
 	const char	*name;
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	unsigned int	(*startup)(unsigned int irq);
 	void		(*shutdown)(unsigned int irq);
 	void		(*enable)(unsigned int irq);
@@ -130,154 +175,66 @@
 
 	void		(*bus_lock)(unsigned int irq);
 	void		(*bus_sync_unlock)(unsigned int irq);
+#endif
+	unsigned int	(*irq_startup)(struct irq_data *data);
+	void		(*irq_shutdown)(struct irq_data *data);
+	void		(*irq_enable)(struct irq_data *data);
+	void		(*irq_disable)(struct irq_data *data);
+
+	void		(*irq_ack)(struct irq_data *data);
+	void		(*irq_mask)(struct irq_data *data);
+	void		(*irq_mask_ack)(struct irq_data *data);
+	void		(*irq_unmask)(struct irq_data *data);
+	void		(*irq_eoi)(struct irq_data *data);
+
+	int		(*irq_set_affinity)(struct irq_data *data, const struct cpumask *dest, bool force);
+	int		(*irq_retrigger)(struct irq_data *data);
+	int		(*irq_set_type)(struct irq_data *data, unsigned int flow_type);
+	int		(*irq_set_wake)(struct irq_data *data, unsigned int on);
+
+	void		(*irq_bus_lock)(struct irq_data *data);
+	void		(*irq_bus_sync_unlock)(struct irq_data *data);
 
 	/* Currently used only by UML, might disappear one day.*/
 #ifdef CONFIG_IRQ_RELEASE_METHOD
 	void		(*release)(unsigned int irq, void *dev_id);
 #endif
-	/*
-	 * For compatibility, ->typename is copied into ->name.
-	 * Will disappear.
-	 */
-	const char	*typename;
 };
 
-struct timer_rand_state;
-struct irq_2_iommu;
-/**
- * struct irq_desc - interrupt descriptor
- * @irq:		interrupt number for this descriptor
- * @timer_rand_state:	pointer to timer rand state struct
- * @kstat_irqs:		irq stats per cpu
- * @irq_2_iommu:	iommu with this irq
- * @handle_irq:		highlevel irq-events handler [if NULL, __do_IRQ()]
- * @chip:		low level interrupt hardware access
- * @msi_desc:		MSI descriptor
- * @handler_data:	per-IRQ data for the irq_chip methods
- * @chip_data:		platform-specific per-chip private data for the chip
- *			methods, to allow shared chip implementations
- * @action:		the irq action chain
- * @status:		status information
- * @depth:		disable-depth, for nested irq_disable() calls
- * @wake_depth:		enable depth, for multiple set_irq_wake() callers
- * @irq_count:		stats field to detect stalled irqs
- * @last_unhandled:	aging timer for unhandled count
- * @irqs_unhandled:	stats field for spurious unhandled interrupts
- * @lock:		locking for SMP
- * @affinity:		IRQ affinity on SMP
- * @node:		node index useful for balancing
- * @pending_mask:	pending rebalanced interrupts
- * @threads_active:	number of irqaction threads currently running
- * @wait_for_threads:	wait queue for sync_irq to wait for threaded handlers
- * @dir:		/proc/irq/ procfs entry
- * @name:		flow handler name for /proc/interrupts output
- */
-struct irq_desc {
-	unsigned int		irq;
-	struct timer_rand_state *timer_rand_state;
-	unsigned int            *kstat_irqs;
-#ifdef CONFIG_INTR_REMAP
-	struct irq_2_iommu      *irq_2_iommu;
-#endif
-	irq_flow_handler_t	handle_irq;
-	struct irq_chip		*chip;
-	struct msi_desc		*msi_desc;
-	void			*handler_data;
-	void			*chip_data;
-	struct irqaction	*action;	/* IRQ action list */
-	unsigned int		status;		/* IRQ status */
-
-	unsigned int		depth;		/* nested irq disables */
-	unsigned int		wake_depth;	/* nested wake enables */
-	unsigned int		irq_count;	/* For detecting broken IRQs */
-	unsigned long		last_unhandled;	/* Aging timer for unhandled count */
-	unsigned int		irqs_unhandled;
-	raw_spinlock_t		lock;
-#ifdef CONFIG_SMP
-	cpumask_var_t		affinity;
-	const struct cpumask	*affinity_hint;
-	unsigned int		node;
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_var_t		pending_mask;
-#endif
-#endif
-	atomic_t		threads_active;
-	wait_queue_head_t       wait_for_threads;
-#ifdef CONFIG_PROC_FS
-	struct proc_dir_entry	*dir;
-#endif
-	const char		*name;
-} ____cacheline_internodealigned_in_smp;
-
-extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
-					struct irq_desc *desc, int node);
-extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
-
-#ifndef CONFIG_SPARSE_IRQ
-extern struct irq_desc irq_desc[NR_IRQS];
-#endif
-
-#ifdef CONFIG_NUMA_IRQ_DESC
-extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int node);
-#else
-static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
-{
-	return desc;
-}
-#endif
-
-extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
+/* This include will go away once we isolated irq_desc usage to core code */
+#include <linux/irqdesc.h>
 
 /*
  * Pick up the arch-dependent methods:
  */
 #include <asm/hw_irq.h>
 
+#ifndef NR_IRQS_LEGACY
+# define NR_IRQS_LEGACY 0
+#endif
+
+#ifndef ARCH_IRQ_INIT_FLAGS
+# define ARCH_IRQ_INIT_FLAGS	0
+#endif
+
+#define IRQ_DEFAULT_INIT_FLAGS	(IRQ_DISABLED | ARCH_IRQ_INIT_FLAGS)
+
+struct irqaction;
 extern int setup_irq(unsigned int irq, struct irqaction *new);
 extern void remove_irq(unsigned int irq, struct irqaction *act);
 
 #ifdef CONFIG_GENERIC_HARDIRQS
 
-#ifdef CONFIG_SMP
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-
+#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_PENDING_IRQ)
 void move_native_irq(int irq);
 void move_masked_irq(int irq);
-
-#else /* CONFIG_GENERIC_PENDING_IRQ */
-
-static inline void move_irq(int irq)
-{
-}
-
-static inline void move_native_irq(int irq)
-{
-}
-
-static inline void move_masked_irq(int irq)
-{
-}
-
-#endif /* CONFIG_GENERIC_PENDING_IRQ */
-
-#else /* CONFIG_SMP */
-
-#define move_native_irq(x)
-#define move_masked_irq(x)
-
-#endif /* CONFIG_SMP */
+#else
+static inline void move_native_irq(int irq) { }
+static inline void move_masked_irq(int irq) { }
+#endif
 
 extern int no_irq_affinity;
 
-static inline int irq_balancing_disabled(unsigned int irq)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-	return desc->status & IRQ_NO_BALANCING_MASK;
-}
-
 /* Handle irq action chains: */
 extern irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action);
 
@@ -293,42 +250,10 @@
 extern void handle_bad_irq(unsigned int irq, struct irq_desc *desc);
 extern void handle_nested_irq(unsigned int irq);
 
-/*
- * Monolithic do_IRQ implementation.
- */
-#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
-extern unsigned int __do_IRQ(unsigned int irq);
-#endif
-
-/*
- * Architectures call this to let the generic IRQ layer
- * handle an interrupt. If the descriptor is attached to an
- * irqchip-style controller then we call the ->handle_irq() handler,
- * and it calls __do_IRQ() if it's attached to an irqtype-style controller.
- */
-static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
-	desc->handle_irq(irq, desc);
-#else
-	if (likely(desc->handle_irq))
-		desc->handle_irq(irq, desc);
-	else
-		__do_IRQ(irq);
-#endif
-}
-
-static inline void generic_handle_irq(unsigned int irq)
-{
-	generic_handle_irq_desc(irq, irq_to_desc(irq));
-}
-
 /* Handling of unhandled and spurious interrupts: */
 extern void note_interrupt(unsigned int irq, struct irq_desc *desc,
 			   irqreturn_t action_ret);
 
-/* Resending of interrupts :*/
-void check_irq_resend(struct irq_desc *desc, unsigned int irq);
 
 /* Enable/disable irq debugging output: */
 extern int noirqdebug_setup(char *str);
@@ -351,16 +276,6 @@
 __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 		  const char *name);
 
-/* caller has locked the irq_desc and both params are valid */
-static inline void __set_irq_handler_unlocked(int irq,
-					      irq_flow_handler_t handler)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-	desc->handle_irq = handler;
-}
-
 /*
  * Set a highlevel flow handler for a given IRQ:
  */
@@ -384,26 +299,42 @@
 
 extern void set_irq_nested_thread(unsigned int irq, int nest);
 
-extern void set_irq_noprobe(unsigned int irq);
-extern void set_irq_probe(unsigned int irq);
+void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set);
+
+static inline void irq_set_status_flags(unsigned int irq, unsigned long set)
+{
+	irq_modify_status(irq, 0, set);
+}
+
+static inline void irq_clear_status_flags(unsigned int irq, unsigned long clr)
+{
+	irq_modify_status(irq, clr, 0);
+}
+
+static inline void set_irq_noprobe(unsigned int irq)
+{
+	irq_modify_status(irq, 0, IRQ_NOPROBE);
+}
+
+static inline void set_irq_probe(unsigned int irq)
+{
+	irq_modify_status(irq, IRQ_NOPROBE, 0);
+}
 
 /* Handle dynamic irq creation and destruction */
 extern unsigned int create_irq_nr(unsigned int irq_want, int node);
 extern int create_irq(void);
 extern void destroy_irq(unsigned int irq);
 
-/* Test to see if a driver has successfully requested an irq */
-static inline int irq_has_action(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	return desc->action != NULL;
-}
-
-/* Dynamic irq helper functions */
-extern void dynamic_irq_init(unsigned int irq);
-void dynamic_irq_init_keep_chip_data(unsigned int irq);
+/*
+ * Dynamic irq helper functions. Obsolete. Use irq_alloc_desc* and
+ * irq_free_desc instead.
+ */
 extern void dynamic_irq_cleanup(unsigned int irq);
-void dynamic_irq_cleanup_keep_chip_data(unsigned int irq);
+static inline void dynamic_irq_init(unsigned int irq)
+{
+	dynamic_irq_cleanup(irq);
+}
 
 /* Set/get chip/data for an IRQ: */
 extern int set_irq_chip(unsigned int irq, struct irq_chip *chip);
@@ -411,114 +342,78 @@
 extern int set_irq_chip_data(unsigned int irq, void *data);
 extern int set_irq_type(unsigned int irq, unsigned int type);
 extern int set_irq_msi(unsigned int irq, struct msi_desc *entry);
+extern struct irq_data *irq_get_irq_data(unsigned int irq);
 
-#define get_irq_chip(irq)	(irq_to_desc(irq)->chip)
-#define get_irq_chip_data(irq)	(irq_to_desc(irq)->chip_data)
-#define get_irq_data(irq)	(irq_to_desc(irq)->handler_data)
-#define get_irq_msi(irq)	(irq_to_desc(irq)->msi_desc)
+static inline struct irq_chip *get_irq_chip(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->chip : NULL;
+}
 
-#define get_irq_desc_chip(desc)		((desc)->chip)
-#define get_irq_desc_chip_data(desc)	((desc)->chip_data)
-#define get_irq_desc_data(desc)		((desc)->handler_data)
-#define get_irq_desc_msi(desc)		((desc)->msi_desc)
+static inline struct irq_chip *irq_data_get_irq_chip(struct irq_data *d)
+{
+	return d->chip;
+}
+
+static inline void *get_irq_chip_data(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->chip_data : NULL;
+}
+
+static inline void *irq_data_get_irq_chip_data(struct irq_data *d)
+{
+	return d->chip_data;
+}
+
+static inline void *get_irq_data(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->handler_data : NULL;
+}
+
+static inline void *irq_data_get_irq_data(struct irq_data *d)
+{
+	return d->handler_data;
+}
+
+static inline struct msi_desc *get_irq_msi(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->msi_desc : NULL;
+}
+
+static inline struct msi_desc *irq_data_get_msi(struct irq_data *d)
+{
+	return d->msi_desc;
+}
+
+int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node);
+void irq_free_descs(unsigned int irq, unsigned int cnt);
+int irq_reserve_irqs(unsigned int from, unsigned int cnt);
+
+static inline int irq_alloc_desc(int node)
+{
+	return irq_alloc_descs(-1, 0, 1, node);
+}
+
+static inline int irq_alloc_desc_at(unsigned int at, int node)
+{
+	return irq_alloc_descs(at, at, 1, node);
+}
+
+static inline int irq_alloc_desc_from(unsigned int from, int node)
+{
+	return irq_alloc_descs(-1, from, 1, node);
+}
+
+static inline void irq_free_desc(unsigned int irq)
+{
+	irq_free_descs(irq, 1);
+}
 
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
 #endif /* !CONFIG_S390 */
 
-#ifdef CONFIG_SMP
-/**
- * alloc_desc_masks - allocate cpumasks for irq_desc
- * @desc:	pointer to irq_desc struct
- * @node:	node which will be handling the cpumasks
- * @boot:	true if need bootmem
- *
- * Allocates affinity and pending_mask cpumask if required.
- * Returns true if successful (or not required).
- */
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-							bool boot)
-{
-	gfp_t gfp = GFP_ATOMIC;
-
-	if (boot)
-		gfp = GFP_NOWAIT;
-
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	if (!alloc_cpumask_var_node(&desc->affinity, gfp, node))
-		return false;
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	if (!alloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
-		free_cpumask_var(desc->affinity);
-		return false;
-	}
-#endif
-#endif
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-	cpumask_setall(desc->affinity);
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_clear(desc->pending_mask);
-#endif
-}
-
-/**
- * init_copy_desc_masks - copy cpumasks for irq_desc
- * @old_desc:	pointer to old irq_desc struct
- * @new_desc:	pointer to new irq_desc struct
- *
- * Insures affinity and pending_masks are copied to new irq_desc.
- * If !CONFIG_CPUMASKS_OFFSTACK the cpumasks are embedded in the
- * irq_desc struct so the copy is redundant.
- */
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	cpumask_copy(new_desc->affinity, old_desc->affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_copy(new_desc->pending_mask, old_desc->pending_mask);
-#endif
-#endif
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-	free_cpumask_var(old_desc->affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	free_cpumask_var(old_desc->pending_mask);
-#endif
-}
-
-#else /* !CONFIG_SMP */
-
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-								bool boot)
-{
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-}
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-}
-#endif	/* CONFIG_SMP */
-
 #endif /* _LINUX_IRQ_H */
diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h
new file mode 100644
index 0000000..4fa09d4
--- /dev/null
+++ b/include/linux/irq_work.h
@@ -0,0 +1,20 @@
+#ifndef _LINUX_IRQ_WORK_H
+#define _LINUX_IRQ_WORK_H
+
+struct irq_work {
+	struct irq_work *next;
+	void (*func)(struct irq_work *);
+};
+
+static inline
+void init_irq_work(struct irq_work *entry, void (*func)(struct irq_work *))
+{
+	entry->next = NULL;
+	entry->func = func;
+}
+
+bool irq_work_queue(struct irq_work *entry);
+void irq_work_run(void);
+void irq_work_sync(struct irq_work *entry);
+
+#endif /* _LINUX_IRQ_WORK_H */
diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
new file mode 100644
index 0000000..979c68c
--- /dev/null
+++ b/include/linux/irqdesc.h
@@ -0,0 +1,159 @@
+#ifndef _LINUX_IRQDESC_H
+#define _LINUX_IRQDESC_H
+
+/*
+ * Core internal functions to deal with irq descriptors
+ *
+ * This include will move to kernel/irq once we cleaned up the tree.
+ * For now it's included from <linux/irq.h>
+ */
+
+struct proc_dir_entry;
+struct timer_rand_state;
+/**
+ * struct irq_desc - interrupt descriptor
+ * @irq_data:		per irq and chip data passed down to chip functions
+ * @timer_rand_state:	pointer to timer rand state struct
+ * @kstat_irqs:		irq stats per cpu
+ * @handle_irq:		highlevel irq-events handler [if NULL, __do_IRQ()]
+ * @action:		the irq action chain
+ * @status:		status information
+ * @depth:		disable-depth, for nested irq_disable() calls
+ * @wake_depth:		enable depth, for multiple set_irq_wake() callers
+ * @irq_count:		stats field to detect stalled irqs
+ * @last_unhandled:	aging timer for unhandled count
+ * @irqs_unhandled:	stats field for spurious unhandled interrupts
+ * @lock:		locking for SMP
+ * @pending_mask:	pending rebalanced interrupts
+ * @threads_active:	number of irqaction threads currently running
+ * @wait_for_threads:	wait queue for sync_irq to wait for threaded handlers
+ * @dir:		/proc/irq/ procfs entry
+ * @name:		flow handler name for /proc/interrupts output
+ */
+struct irq_desc {
+
+#ifdef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+	struct irq_data		irq_data;
+#else
+	/*
+	 * This union will go away, once we fixed the direct access to
+	 * irq_desc all over the place. The direct fields are a 1:1
+	 * overlay of irq_data.
+	 */
+	union {
+		struct irq_data		irq_data;
+		struct {
+			unsigned int		irq;
+			unsigned int		node;
+			struct irq_chip		*chip;
+			void			*handler_data;
+			void			*chip_data;
+			struct msi_desc		*msi_desc;
+#ifdef CONFIG_SMP
+			cpumask_var_t		affinity;
+#endif
+		};
+	};
+#endif
+
+	struct timer_rand_state *timer_rand_state;
+	unsigned int		*kstat_irqs;
+	irq_flow_handler_t	handle_irq;
+	struct irqaction	*action;	/* IRQ action list */
+	unsigned int		status;		/* IRQ status */
+
+	unsigned int		depth;		/* nested irq disables */
+	unsigned int		wake_depth;	/* nested wake enables */
+	unsigned int		irq_count;	/* For detecting broken IRQs */
+	unsigned long		last_unhandled;	/* Aging timer for unhandled count */
+	unsigned int		irqs_unhandled;
+	raw_spinlock_t		lock;
+#ifdef CONFIG_SMP
+	const struct cpumask	*affinity_hint;
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_var_t		pending_mask;
+#endif
+#endif
+	atomic_t		threads_active;
+	wait_queue_head_t       wait_for_threads;
+#ifdef CONFIG_PROC_FS
+	struct proc_dir_entry	*dir;
+#endif
+	const char		*name;
+} ____cacheline_internodealigned_in_smp;
+
+#ifndef CONFIG_SPARSE_IRQ
+extern struct irq_desc irq_desc[NR_IRQS];
+#endif
+
+/* Will be removed once the last users in power and sh are gone */
+extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
+static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
+{
+	return desc;
+}
+
+#ifdef CONFIG_GENERIC_HARDIRQS
+
+#define get_irq_desc_chip(desc)		((desc)->irq_data.chip)
+#define get_irq_desc_chip_data(desc)	((desc)->irq_data.chip_data)
+#define get_irq_desc_data(desc)		((desc)->irq_data.handler_data)
+#define get_irq_desc_msi(desc)		((desc)->irq_data.msi_desc)
+
+/*
+ * Monolithic do_IRQ implementation.
+ */
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+extern unsigned int __do_IRQ(unsigned int irq);
+#endif
+
+/*
+ * Architectures call this to let the generic IRQ layer
+ * handle an interrupt. If the descriptor is attached to an
+ * irqchip-style controller then we call the ->handle_irq() handler,
+ * and it calls __do_IRQ() if it's attached to an irqtype-style controller.
+ */
+static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *desc)
+{
+#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+	desc->handle_irq(irq, desc);
+#else
+	if (likely(desc->handle_irq))
+		desc->handle_irq(irq, desc);
+	else
+		__do_IRQ(irq);
+#endif
+}
+
+static inline void generic_handle_irq(unsigned int irq)
+{
+	generic_handle_irq_desc(irq, irq_to_desc(irq));
+}
+
+/* Test to see if a driver has successfully requested an irq */
+static inline int irq_has_action(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	return desc->action != NULL;
+}
+
+static inline int irq_balancing_disabled(unsigned int irq)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+	return desc->status & IRQ_NO_BALANCING_MASK;
+}
+
+/* caller has locked the irq_desc and both params are valid */
+static inline void __set_irq_handler_unlocked(int irq,
+					      irq_flow_handler_t handler)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+	desc->handle_irq = handler;
+}
+#endif
+
+#endif
diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index 7bf89bc..05aa8c23 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -25,6 +25,7 @@
 
 extern int nr_irqs;
 extern struct irq_desc *irq_to_desc(unsigned int irq);
+unsigned int irq_get_next_irq(unsigned int offset);
 
 # define for_each_irq_desc(irq, desc)					\
 	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs;		\
@@ -47,6 +48,10 @@
 #define irq_node(irq)	0
 #endif
 
+# define for_each_active_irq(irq)			\
+	for (irq = irq_get_next_irq(0); irq < nr_irqs;	\
+	     irq = irq_get_next_irq(irq + 1))
+
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
 #define for_each_irq_nr(irq)                   \
diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
new file mode 100644
index 0000000..b67cb18
--- /dev/null
+++ b/include/linux/jump_label.h
@@ -0,0 +1,74 @@
+#ifndef _LINUX_JUMP_LABEL_H
+#define _LINUX_JUMP_LABEL_H
+
+#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_HAVE_ARCH_JUMP_LABEL)
+# include <asm/jump_label.h>
+# define HAVE_JUMP_LABEL
+#endif
+
+enum jump_label_type {
+	JUMP_LABEL_ENABLE,
+	JUMP_LABEL_DISABLE
+};
+
+struct module;
+
+#ifdef HAVE_JUMP_LABEL
+
+extern struct jump_entry __start___jump_table[];
+extern struct jump_entry __stop___jump_table[];
+
+extern void arch_jump_label_transform(struct jump_entry *entry,
+				 enum jump_label_type type);
+extern void arch_jump_label_text_poke_early(jump_label_t addr);
+extern void jump_label_update(unsigned long key, enum jump_label_type type);
+extern void jump_label_apply_nops(struct module *mod);
+extern int jump_label_text_reserved(void *start, void *end);
+
+#define jump_label_enable(key) \
+	jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE);
+
+#define jump_label_disable(key) \
+	jump_label_update((unsigned long)key, JUMP_LABEL_DISABLE);
+
+#else
+
+#define JUMP_LABEL(key, label)			\
+do {						\
+	if (unlikely(*key))			\
+		goto label;			\
+} while (0)
+
+#define jump_label_enable(cond_var)	\
+do {					\
+       *(cond_var) = 1;			\
+} while (0)
+
+#define jump_label_disable(cond_var)	\
+do {					\
+       *(cond_var) = 0;			\
+} while (0)
+
+static inline int jump_label_apply_nops(struct module *mod)
+{
+	return 0;
+}
+
+static inline int jump_label_text_reserved(void *start, void *end)
+{
+	return 0;
+}
+
+#endif
+
+#define COND_STMT(key, stmt)					\
+do {								\
+	__label__ jl_enabled;					\
+	JUMP_LABEL(key, jl_enabled);				\
+	if (0) {						\
+jl_enabled:							\
+		stmt;						\
+	}							\
+} while (0)
+
+#endif
diff --git a/include/linux/jump_label_ref.h b/include/linux/jump_label_ref.h
new file mode 100644
index 0000000..e5d012a
--- /dev/null
+++ b/include/linux/jump_label_ref.h
@@ -0,0 +1,44 @@
+#ifndef _LINUX_JUMP_LABEL_REF_H
+#define _LINUX_JUMP_LABEL_REF_H
+
+#include <linux/jump_label.h>
+#include <asm/atomic.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+static inline void jump_label_inc(atomic_t *key)
+{
+	if (atomic_add_return(1, key) == 1)
+		jump_label_enable(key);
+}
+
+static inline void jump_label_dec(atomic_t *key)
+{
+	if (atomic_dec_and_test(key))
+		jump_label_disable(key);
+}
+
+#else /* !HAVE_JUMP_LABEL */
+
+static inline void jump_label_inc(atomic_t *key)
+{
+	atomic_inc(key);
+}
+
+static inline void jump_label_dec(atomic_t *key)
+{
+	atomic_dec(key);
+}
+
+#undef JUMP_LABEL
+#define JUMP_LABEL(key, label)						\
+do {									\
+	if (unlikely(__builtin_choose_expr(				\
+	      __builtin_types_compatible_p(typeof(key), atomic_t *),	\
+	      atomic_read((atomic_t *)(key)), *(key))))			\
+		goto label;						\
+} while (0)
+
+#endif /* HAVE_JUMP_LABEL */
+
+#endif /* _LINUX_JUMP_LABEL_REF_H */
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b0a35e..1759ba5 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -58,7 +58,18 @@
 
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
 #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
-#define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
+#define roundup(x, y) (					\
+{							\
+	typeof(y) __y = y;				\
+	(((x) + (__y - 1)) / __y) * __y;		\
+}							\
+)
+#define rounddown(x, y) (				\
+{							\
+	typeof(x) __x = (x);				\
+	__x - (__x % (y));				\
+}							\
+)
 #define DIV_ROUND_CLOSEST(x, divisor)(			\
 {							\
 	typeof(divisor) __divisor = divisor;		\
diff --git a/include/linux/key.h b/include/linux/key.h
index cd50dfa..3db0adc 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -178,8 +178,9 @@
 	 */
 	union {
 		unsigned long		value;
+		void __rcu		*rcudata;
 		void			*data;
-		struct keyring_list	*subscriptions;
+		struct keyring_list __rcu *subscriptions;
 	} payload;
 };
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c13cc48..ac740b2 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -205,7 +205,7 @@
 
 	struct mutex irq_lock;
 #ifdef CONFIG_HAVE_KVM_IRQCHIP
-	struct kvm_irq_routing_table *irq_routing;
+	struct kvm_irq_routing_table __rcu *irq_routing;
 	struct hlist_head mask_notifier_list;
 	struct hlist_head irq_ack_notifier_list;
 #endif
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 06aed83..71c09b26 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -32,6 +32,17 @@
 #define MAX_LOCKDEP_SUBCLASSES		8UL
 
 /*
+ * NR_LOCKDEP_CACHING_CLASSES ... Number of classes
+ * cached in the instance of lockdep_map
+ *
+ * Currently main class (subclass == 0) and signle depth subclass
+ * are cached in lockdep_map. This optimization is mainly targeting
+ * on rq->lock. double_rq_lock() acquires this highly competitive with
+ * single depth.
+ */
+#define NR_LOCKDEP_CACHING_CLASSES	2
+
+/*
  * Lock-classes are keyed via unique addresses, by embedding the
  * lockclass-key into the kernel (or module) .data section. (For
  * static locks we use the lock address itself as the key.)
@@ -138,7 +149,7 @@
  */
 struct lockdep_map {
 	struct lock_class_key		*key;
-	struct lock_class		*class_cache;
+	struct lock_class		*class_cache[NR_LOCKDEP_CACHING_CLASSES];
 	const char			*name;
 #ifdef CONFIG_LOCK_STAT
 	int				cpu;
@@ -424,14 +435,6 @@
 
 #endif /* CONFIG_LOCKDEP */
 
-#ifdef CONFIG_GENERIC_HARDIRQS
-extern void early_init_irq_lock_class(void);
-#else
-static inline void early_init_irq_lock_class(void)
-{
-}
-#endif
-
 #ifdef CONFIG_TRACE_IRQFLAGS
 extern void early_boot_irqs_off(void);
 extern void early_boot_irqs_on(void);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index ee7e258..cb57d65 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -299,7 +299,7 @@
 	 * new_owner->mm == mm
 	 * new_owner->alloc_lock is held
 	 */
-	struct task_struct *owner;
+	struct task_struct __rcu *owner;
 #endif
 
 #ifdef CONFIG_PROC_FS
diff --git a/include/linux/module.h b/include/linux/module.h
index aace066..b29e745 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -350,7 +350,10 @@
 	struct tracepoint *tracepoints;
 	unsigned int num_tracepoints;
 #endif
-
+#ifdef HAVE_JUMP_LABEL
+	struct jump_entry *jump_entries;
+	unsigned int num_jump_entries;
+#endif
 #ifdef CONFIG_TRACING
 	const char **trace_bprintk_fmt_start;
 	unsigned int num_trace_bprintk_fmt;
diff --git a/include/linux/msi.h b/include/linux/msi.h
index 91b05c1..05acced 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -10,12 +10,13 @@
 };
 
 /* Helper functions */
-struct irq_desc;
-extern void mask_msi_irq(unsigned int irq);
-extern void unmask_msi_irq(unsigned int irq);
-extern void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
-extern void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
-extern void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
+struct irq_data;
+struct msi_desc;
+extern void mask_msi_irq(struct irq_data *data);
+extern void unmask_msi_irq(struct irq_data *data);
+extern void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
+extern void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
+extern void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
 extern void read_msi_msg(unsigned int irq, struct msi_msg *msg);
 extern void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg);
 extern void write_msi_msg(unsigned int irq, struct msi_msg *msg);
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 9ed534c..70cd060 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -39,8 +39,9 @@
 	CTA_TUPLE_MASTER,
 	CTA_NAT_SEQ_ADJ_ORIG,
 	CTA_NAT_SEQ_ADJ_REPLY,
-	CTA_SECMARK,
+	CTA_SECMARK,		/* obsolete */
 	CTA_ZONE,
+	CTA_SECCTX,
 	__CTA_MAX
 };
 #define CTA_MAX (__CTA_MAX - 1)
@@ -172,4 +173,11 @@
 };
 #define CTA_HELP_MAX (__CTA_HELP_MAX - 1)
 
+enum ctattr_secctx {
+	CTA_SECCTX_UNSPEC,
+	CTA_SECCTX_NAME,
+	__CTA_SECCTX_MAX
+};
+#define CTA_SECCTX_MAX (__CTA_SECCTX_MAX - 1)
+
 #endif /* _IPCONNTRACK_NETLINK_H */
diff --git a/include/linux/netfilter/xt_SECMARK.h b/include/linux/netfilter/xt_SECMARK.h
index 6fcd344..989092b 100644
--- a/include/linux/netfilter/xt_SECMARK.h
+++ b/include/linux/netfilter/xt_SECMARK.h
@@ -11,18 +11,12 @@
  * packets are being marked for.
  */
 #define SECMARK_MODE_SEL	0x01		/* SELinux */
-#define SECMARK_SELCTX_MAX	256
-
-struct xt_secmark_target_selinux_info {
-	__u32 selsid;
-	char selctx[SECMARK_SELCTX_MAX];
-};
+#define SECMARK_SECCTX_MAX	256
 
 struct xt_secmark_target_info {
 	__u8 mode;
-	union {
-		struct xt_secmark_target_selinux_info sel;
-	} u;
+	__u32 secid;
+	char secctx[SECMARK_SECCTX_MAX];
 };
 
 #endif /*_XT_SECMARK_H_target */
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 508f8cf..d0edf7d 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -185,7 +185,7 @@
 	struct nfs4_cached_acl	*nfs4_acl;
         /* NFSv4 state */
 	struct list_head	open_states;
-	struct nfs_delegation	*delegation;
+	struct nfs_delegation __rcu *delegation;
 	fmode_t			 delegation_state;
 	struct rw_semaphore	rwsem;
 #endif /* CONFIG_NFS_V4*/
diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index b2f1a4d..2026f9e 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -49,28 +49,28 @@
 
 struct notifier_block {
 	int (*notifier_call)(struct notifier_block *, unsigned long, void *);
-	struct notifier_block *next;
+	struct notifier_block __rcu *next;
 	int priority;
 };
 
 struct atomic_notifier_head {
 	spinlock_t lock;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct blocking_notifier_head {
 	struct rw_semaphore rwsem;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct raw_notifier_head {
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct srcu_notifier_head {
 	struct mutex mutex;
 	struct srcu_struct srcu;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 #define ATOMIC_INIT_NOTIFIER_HEAD(name) do {	\
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index 5171639..32fb812 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -15,6 +15,7 @@
 
 #include <linux/types.h>
 #include <linux/spinlock.h>
+#include <linux/init.h>
 #include <asm/atomic.h>
  
 /* Each escaped entry is prefixed by ESCAPE_CODE
@@ -185,4 +186,10 @@
 int oprofile_add_data64(struct op_entry *entry, u64 val);
 int oprofile_write_commit(struct op_entry *entry);
 
+#ifdef CONFIG_PERF_EVENTS
+int __init oprofile_perf_init(struct oprofile_operations *ops);
+void oprofile_perf_exit(void);
+char *op_name_from_perf_id(void);
+#endif /* CONFIG_PERF_EVENTS */
+
 #endif /* OPROFILE_H */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 570fdde..2615c37 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -517,6 +517,7 @@
 #define PCI_DEVICE_ID_AMD_11H_NB_DRAM	0x1302
 #define PCI_DEVICE_ID_AMD_11H_NB_MISC	0x1303
 #define PCI_DEVICE_ID_AMD_11H_NB_LINK	0x1304
+#define PCI_DEVICE_ID_AMD_15H_NB_MISC	0x1603
 #define PCI_DEVICE_ID_AMD_LANCE		0x2000
 #define PCI_DEVICE_ID_AMD_LANCE_HOME	0x2001
 #define PCI_DEVICE_ID_AMD_SCSI		0x2020
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index ce2dc65..27ef6b1 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -139,6 +139,15 @@
 	__aligned(PAGE_SIZE)
 
 /*
+ * Declaration/definition used for per-CPU variables that must be read mostly.
+ */
+#define DECLARE_PER_CPU_READ_MOSTLY(type, name)			\
+	DECLARE_PER_CPU_SECTION(type, name, "..readmostly")
+
+#define DEFINE_PER_CPU_READ_MOSTLY(type, name)				\
+	DEFINE_PER_CPU_SECTION(type, name, "..readmostly")
+
+/*
  * Intermodule exports for per-CPU variables.  sparse forgets about
  * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to
  * noop if __CHECKER__.
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 49466b1..0eb5083 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -39,6 +39,15 @@
 	preempt_enable();				\
 } while (0)
 
+#define get_cpu_ptr(var) ({				\
+	preempt_disable();				\
+	this_cpu_ptr(var); })
+
+#define put_cpu_ptr(var) do {				\
+	(void)(var);					\
+	preempt_enable();				\
+} while (0)
+
 #ifdef CONFIG_SMP
 
 /* minimum unit size, also is the maximum supported allocation size */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 716f99b..057bf22 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -486,6 +486,8 @@
 #include <linux/workqueue.h>
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
+#include <linux/irq_work.h>
+#include <linux/jump_label_ref.h>
 #include <asm/atomic.h>
 #include <asm/local.h>
 
@@ -529,16 +531,22 @@
 			int		last_cpu;
 		};
 		struct { /* software */
-			s64		remaining;
 			struct hrtimer	hrtimer;
 		};
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 		struct { /* breakpoint */
 			struct arch_hw_breakpoint	info;
 			struct list_head		bp_list;
+			/*
+			 * Crufty hack to avoid the chicken and egg
+			 * problem hw_breakpoint has with context
+			 * creation and event initalization.
+			 */
+			struct task_struct		*bp_target;
 		};
 #endif
 	};
+	int				state;
 	local64_t			prev_count;
 	u64				sample_period;
 	u64				last_period;
@@ -550,6 +558,13 @@
 #endif
 };
 
+/*
+ * hw_perf_event::state flags
+ */
+#define PERF_HES_STOPPED	0x01 /* the counter is stopped */
+#define PERF_HES_UPTODATE	0x02 /* event->count up-to-date */
+#define PERF_HES_ARCH		0x04
+
 struct perf_event;
 
 /*
@@ -561,36 +576,70 @@
  * struct pmu - generic performance monitoring unit
  */
 struct pmu {
-	int (*enable)			(struct perf_event *event);
-	void (*disable)			(struct perf_event *event);
-	int (*start)			(struct perf_event *event);
-	void (*stop)			(struct perf_event *event);
+	struct list_head		entry;
+
+	int * __percpu			pmu_disable_count;
+	struct perf_cpu_context * __percpu pmu_cpu_context;
+	int				task_ctx_nr;
+
+	/*
+	 * Fully disable/enable this PMU, can be used to protect from the PMI
+	 * as well as for lazy/batch writing of the MSRs.
+	 */
+	void (*pmu_enable)		(struct pmu *pmu); /* optional */
+	void (*pmu_disable)		(struct pmu *pmu); /* optional */
+
+	/*
+	 * Try and initialize the event for this PMU.
+	 * Should return -ENOENT when the @event doesn't match this PMU.
+	 */
+	int (*event_init)		(struct perf_event *event);
+
+#define PERF_EF_START	0x01		/* start the counter when adding    */
+#define PERF_EF_RELOAD	0x02		/* reload the counter when starting */
+#define PERF_EF_UPDATE	0x04		/* update the counter when stopping */
+
+	/*
+	 * Adds/Removes a counter to/from the PMU, can be done inside
+	 * a transaction, see the ->*_txn() methods.
+	 */
+	int  (*add)			(struct perf_event *event, int flags);
+	void (*del)			(struct perf_event *event, int flags);
+
+	/*
+	 * Starts/Stops a counter present on the PMU. The PMI handler
+	 * should stop the counter when perf_event_overflow() returns
+	 * !0. ->start() will be used to continue.
+	 */
+	void (*start)			(struct perf_event *event, int flags);
+	void (*stop)			(struct perf_event *event, int flags);
+
+	/*
+	 * Updates the counter value of the event.
+	 */
 	void (*read)			(struct perf_event *event);
-	void (*unthrottle)		(struct perf_event *event);
 
 	/*
-	 * Group events scheduling is treated as a transaction, add group
-	 * events as a whole and perform one schedulability test. If the test
-	 * fails, roll back the whole group
+	 * Group events scheduling is treated as a transaction, add
+	 * group events as a whole and perform one schedulability test.
+	 * If the test fails, roll back the whole group
+	 *
+	 * Start the transaction, after this ->add() doesn't need to
+	 * do schedulability tests.
 	 */
-
+	void (*start_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * Start the transaction, after this ->enable() doesn't need
-	 * to do schedulability tests.
-	 */
-	void (*start_txn)	(const struct pmu *pmu);
-	/*
-	 * If ->start_txn() disabled the ->enable() schedulability test
+	 * If ->start_txn() disabled the ->add() schedulability test
 	 * then ->commit_txn() is required to perform one. On success
 	 * the transaction is closed. On error the transaction is kept
 	 * open until ->cancel_txn() is called.
 	 */
-	int  (*commit_txn)	(const struct pmu *pmu);
+	int  (*commit_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * Will cancel the transaction, assumes ->disable() is called for
-	 * each successfull ->enable() during the transaction.
+	 * Will cancel the transaction, assumes ->del() is called
+	 * for each successfull ->add() during the transaction.
 	 */
-	void (*cancel_txn)	(const struct pmu *pmu);
+	void (*cancel_txn)	(struct pmu *pmu); /* optional */
 };
 
 /**
@@ -631,11 +680,6 @@
 	void				*data_pages[0];
 };
 
-struct perf_pending_entry {
-	struct perf_pending_entry *next;
-	void (*func)(struct perf_pending_entry *);
-};
-
 struct perf_sample_data;
 
 typedef void (*perf_overflow_handler_t)(struct perf_event *, int,
@@ -656,6 +700,7 @@
 
 #define PERF_ATTACH_CONTEXT	0x01
 #define PERF_ATTACH_GROUP	0x02
+#define PERF_ATTACH_TASK	0x04
 
 /**
  * struct perf_event - performance event kernel representation:
@@ -669,7 +714,7 @@
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
-	const struct pmu		*pmu;
+	struct pmu			*pmu;
 
 	enum perf_event_active_state	state;
 	unsigned int			attach_state;
@@ -743,7 +788,7 @@
 	int				pending_wakeup;
 	int				pending_kill;
 	int				pending_disable;
-	struct perf_pending_entry	pending;
+	struct irq_work			pending;
 
 	atomic_t			event_limit;
 
@@ -763,12 +808,19 @@
 #endif /* CONFIG_PERF_EVENTS */
 };
 
+enum perf_event_context_type {
+	task_context,
+	cpu_context,
+};
+
 /**
  * struct perf_event_context - event context structure
  *
  * Used as a container for task events and CPU events as well:
  */
 struct perf_event_context {
+	enum perf_event_context_type	type;
+	struct pmu			*pmu;
 	/*
 	 * Protect the states of the events in the list,
 	 * nr_active, and the list:
@@ -808,6 +860,12 @@
 	struct rcu_head			rcu_head;
 };
 
+/*
+ * Number of contexts where an event can trigger:
+ * 	task, softirq, hardirq, nmi.
+ */
+#define PERF_NR_CONTEXTS	4
+
 /**
  * struct perf_event_cpu_context - per cpu event context structure
  */
@@ -815,18 +873,9 @@
 	struct perf_event_context	ctx;
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
-	int				max_pertask;
 	int				exclusive;
-	struct swevent_hlist		*swevent_hlist;
-	struct mutex			hlist_mutex;
-	int				hlist_refcount;
-
-	/*
-	 * Recursion avoidance:
-	 *
-	 * task, softirq, irq, nmi context
-	 */
-	int				recursion[4];
+	struct list_head		rotation_list;
+	int				jiffies_interval;
 };
 
 struct perf_output_handle {
@@ -842,26 +891,34 @@
 
 #ifdef CONFIG_PERF_EVENTS
 
-/*
- * Set by architecture code:
- */
-extern int perf_max_events;
+extern int perf_pmu_register(struct pmu *pmu);
+extern void perf_pmu_unregister(struct pmu *pmu);
 
-extern const struct pmu *hw_perf_event_init(struct perf_event *event);
+extern int perf_num_counters(void);
+extern const char *perf_pmu_name(void);
+extern void __perf_event_task_sched_in(struct task_struct *task);
+extern void __perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
 
-extern void perf_event_task_sched_in(struct task_struct *task);
-extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
-extern void perf_event_task_tick(struct task_struct *task);
+extern atomic_t perf_task_events;
+
+static inline void perf_event_task_sched_in(struct task_struct *task)
+{
+	COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
+}
+
+static inline
+void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
+{
+	COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
+}
+
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
-extern void set_perf_event_pending(void);
-extern void perf_event_do_pending(void);
+extern void perf_event_delayed_put(struct task_struct *task);
 extern void perf_event_print_debug(void);
-extern void __perf_disable(void);
-extern bool __perf_enable(void);
-extern void perf_disable(void);
-extern void perf_enable(void);
+extern void perf_pmu_disable(struct pmu *pmu);
+extern void perf_pmu_enable(struct pmu *pmu);
 extern int perf_event_task_disable(void);
 extern int perf_event_task_enable(void);
 extern void perf_event_update_userpage(struct perf_event *event);
@@ -869,7 +926,7 @@
 extern struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr,
 				int cpu,
-				pid_t pid,
+				struct task_struct *task,
 				perf_overflow_handler_t callback);
 extern u64 perf_event_read_value(struct perf_event *event,
 				 u64 *enabled, u64 *running);
@@ -920,14 +977,7 @@
  */
 static inline int is_software_event(struct perf_event *event)
 {
-	switch (event->attr.type) {
-	case PERF_TYPE_SOFTWARE:
-	case PERF_TYPE_TRACEPOINT:
-	/* for now the breakpoint stuff also works as software event */
-	case PERF_TYPE_BREAKPOINT:
-		return 1;
-	}
-	return 0;
+	return event->pmu->task_ctx_nr == perf_sw_context;
 }
 
 extern atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
@@ -954,18 +1004,20 @@
 	perf_arch_fetch_caller_regs(regs, CALLER_ADDR0);
 }
 
-static inline void
+static __always_inline void
 perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr)
 {
-	if (atomic_read(&perf_swevent_enabled[event_id])) {
-		struct pt_regs hot_regs;
+	struct pt_regs hot_regs;
 
-		if (!regs) {
-			perf_fetch_caller_regs(&hot_regs);
-			regs = &hot_regs;
-		}
-		__perf_sw_event(event_id, nr, nmi, regs, addr);
+	JUMP_LABEL(&perf_swevent_enabled[event_id], have_event);
+	return;
+
+have_event:
+	if (!regs) {
+		perf_fetch_caller_regs(&hot_regs);
+		regs = &hot_regs;
 	}
+	__perf_sw_event(event_id, nr, nmi, regs, addr);
 }
 
 extern void perf_event_mmap(struct vm_area_struct *vma);
@@ -976,7 +1028,21 @@
 extern void perf_event_comm(struct task_struct *tsk);
 extern void perf_event_fork(struct task_struct *tsk);
 
-extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs);
+/* Callchains */
+DECLARE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
+
+extern void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs);
+extern void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs);
+
+
+static inline void
+perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
+{
+	if (entry->nr < PERF_MAX_STACK_DEPTH)
+		entry->ip[entry->nr++] = ip;
+}
 
 extern int sysctl_perf_event_paranoid;
 extern int sysctl_perf_event_mlock;
@@ -1019,21 +1085,18 @@
 extern void perf_swevent_put_recursion_context(int rctx);
 extern void perf_event_enable(struct perf_event *event);
 extern void perf_event_disable(struct perf_event *event);
+extern void perf_event_task_tick(void);
 #else
 static inline void
 perf_event_task_sched_in(struct task_struct *task)			{ }
 static inline void
 perf_event_task_sched_out(struct task_struct *task,
 			    struct task_struct *next)			{ }
-static inline void
-perf_event_task_tick(struct task_struct *task)				{ }
 static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
-static inline void perf_event_do_pending(void)				{ }
+static inline void perf_event_delayed_put(struct task_struct *task)	{ }
 static inline void perf_event_print_debug(void)				{ }
-static inline void perf_disable(void)					{ }
-static inline void perf_enable(void)					{ }
 static inline int perf_event_task_disable(void)				{ return -EINVAL; }
 static inline int perf_event_task_enable(void)				{ return -EINVAL; }
 
@@ -1056,6 +1119,7 @@
 static inline void perf_swevent_put_recursion_context(int rctx)		{ }
 static inline void perf_event_enable(struct perf_event *event)		{ }
 static inline void perf_event_disable(struct perf_event *event)		{ }
+static inline void perf_event_task_tick(void)				{ }
 #endif
 
 #define perf_output_put(handle, x) \
diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
index 634b8e6..a39cbed 100644
--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -47,6 +47,8 @@
 {
 	return (void *)((unsigned long)ptr & ~RADIX_TREE_INDIRECT_PTR);
 }
+#define radix_tree_indirect_to_ptr(ptr) \
+	radix_tree_indirect_to_ptr((void __force *)(ptr))
 
 static inline int radix_tree_is_indirect_ptr(void *ptr)
 {
@@ -61,7 +63,7 @@
 struct radix_tree_root {
 	unsigned int		height;
 	gfp_t			gfp_mask;
-	struct radix_tree_node	*rnode;
+	struct radix_tree_node	__rcu *rnode;
 };
 
 #define RADIX_TREE_INIT(mask)	{					\
diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index 4ec3b38..f31ef61 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -10,6 +10,21 @@
 #include <linux/rcupdate.h>
 
 /*
+ * Why is there no list_empty_rcu()?  Because list_empty() serves this
+ * purpose.  The list_empty() function fetches the RCU-protected pointer
+ * and compares it to the address of the list head, but neither dereferences
+ * this pointer itself nor provides this pointer to the caller.  Therefore,
+ * it is not necessary to use rcu_dereference(), so that list_empty() can
+ * be used anywhere you would want to use a list_empty_rcu().
+ */
+
+/*
+ * return the ->next pointer of a list_head in an rcu safe
+ * way, we must not access it directly
+ */
+#define list_next_rcu(list)	(*((struct list_head __rcu **)(&(list)->next)))
+
+/*
  * Insert a new entry between two known consecutive entries.
  *
  * This is only for internal list manipulation where we know
@@ -20,7 +35,7 @@
 {
 	new->next = next;
 	new->prev = prev;
-	rcu_assign_pointer(prev->next, new);
+	rcu_assign_pointer(list_next_rcu(prev), new);
 	next->prev = new;
 }
 
@@ -138,7 +153,7 @@
 {
 	new->next = old->next;
 	new->prev = old->prev;
-	rcu_assign_pointer(new->prev->next, new);
+	rcu_assign_pointer(list_next_rcu(new->prev), new);
 	new->next->prev = new;
 	old->prev = LIST_POISON2;
 }
@@ -193,7 +208,7 @@
 	 */
 
 	last->next = at;
-	rcu_assign_pointer(head->next, first);
+	rcu_assign_pointer(list_next_rcu(head), first);
 	first->prev = head;
 	at->prev = last;
 }
@@ -208,7 +223,9 @@
  * primitives such as list_add_rcu() as long as it's guarded by rcu_read_lock().
  */
 #define list_entry_rcu(ptr, type, member) \
-	container_of(rcu_dereference_raw(ptr), type, member)
+	({typeof (*ptr) __rcu *__ptr = (typeof (*ptr) __rcu __force *)ptr; \
+	 container_of((typeof(ptr))rcu_dereference_raw(__ptr), type, member); \
+	})
 
 /**
  * list_first_entry_rcu - get the first element from a list
@@ -225,9 +242,9 @@
 	list_entry_rcu((ptr)->next, type, member)
 
 #define __list_for_each_rcu(pos, head) \
-	for (pos = rcu_dereference_raw((head)->next); \
+	for (pos = rcu_dereference_raw(list_next_rcu(head)); \
 		pos != (head); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(list_next_rcu((pos)))
 
 /**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
@@ -257,9 +274,9 @@
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_continue_rcu(pos, head) \
-	for ((pos) = rcu_dereference_raw((pos)->next); \
+	for ((pos) = rcu_dereference_raw(list_next_rcu(pos)); \
 		prefetch((pos)->next), (pos) != (head); \
-		(pos) = rcu_dereference_raw((pos)->next))
+		(pos) = rcu_dereference_raw(list_next_rcu(pos)))
 
 /**
  * list_for_each_entry_continue_rcu - continue iteration over list of given type
@@ -314,12 +331,19 @@
 
 	new->next = next;
 	new->pprev = old->pprev;
-	rcu_assign_pointer(*new->pprev, new);
+	rcu_assign_pointer(*(struct hlist_node __rcu **)new->pprev, new);
 	if (next)
 		new->next->pprev = &new->next;
 	old->pprev = LIST_POISON2;
 }
 
+/*
+ * return the first or the next element in an RCU protected hlist
+ */
+#define hlist_first_rcu(head)	(*((struct hlist_node __rcu **)(&(head)->first)))
+#define hlist_next_rcu(node)	(*((struct hlist_node __rcu **)(&(node)->next)))
+#define hlist_pprev_rcu(node)	(*((struct hlist_node __rcu **)((node)->pprev)))
+
 /**
  * hlist_add_head_rcu
  * @n: the element to add to the hash list.
@@ -346,7 +370,7 @@
 
 	n->next = first;
 	n->pprev = &h->first;
-	rcu_assign_pointer(h->first, n);
+	rcu_assign_pointer(hlist_first_rcu(h), n);
 	if (first)
 		first->pprev = &n->next;
 }
@@ -374,7 +398,7 @@
 {
 	n->pprev = next->pprev;
 	n->next = next;
-	rcu_assign_pointer(*(n->pprev), n);
+	rcu_assign_pointer(hlist_pprev_rcu(n), n);
 	next->pprev = &n->next;
 }
 
@@ -401,15 +425,15 @@
 {
 	n->next = prev->next;
 	n->pprev = &prev->next;
-	rcu_assign_pointer(prev->next, n);
+	rcu_assign_pointer(hlist_next_rcu(prev), n);
 	if (n->next)
 		n->next->pprev = &n->next;
 }
 
-#define __hlist_for_each_rcu(pos, head)			\
-	for (pos = rcu_dereference((head)->first);	\
-	     pos && ({ prefetch(pos->next); 1; });	\
-	     pos = rcu_dereference(pos->next))
+#define __hlist_for_each_rcu(pos, head)				\
+	for (pos = rcu_dereference(hlist_first_rcu(head));	\
+	     pos && ({ prefetch(pos->next); 1; });		\
+	     pos = rcu_dereference(hlist_next_rcu(pos)))
 
 /**
  * hlist_for_each_entry_rcu - iterate over rcu list of given type
@@ -422,11 +446,11 @@
  * the _rcu list-mutation primitives such as hlist_add_head_rcu()
  * as long as the traversal is guarded by rcu_read_lock().
  */
-#define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
-	for (pos = rcu_dereference_raw((head)->first);			 \
+#define hlist_for_each_entry_rcu(tpos, pos, head, member)		\
+	for (pos = rcu_dereference_raw(hlist_first_rcu(head));		\
 		pos && ({ prefetch(pos->next); 1; }) &&			 \
 		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1; }); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(hlist_next_rcu(pos)))
 
 /**
  * hlist_for_each_entry_rcu_bh - iterate over rcu list of given type
diff --git a/include/linux/rculist_nulls.h b/include/linux/rculist_nulls.h
index b70ffe5..2ae1371 100644
--- a/include/linux/rculist_nulls.h
+++ b/include/linux/rculist_nulls.h
@@ -37,6 +37,12 @@
 	}
 }
 
+#define hlist_nulls_first_rcu(head) \
+	(*((struct hlist_nulls_node __rcu __force **)&(head)->first))
+
+#define hlist_nulls_next_rcu(node) \
+	(*((struct hlist_nulls_node __rcu __force **)&(node)->next))
+
 /**
  * hlist_nulls_del_rcu - deletes entry from hash list without re-initialization
  * @n: the element to delete from the hash list.
@@ -88,7 +94,7 @@
 
 	n->next = first;
 	n->pprev = &h->first;
-	rcu_assign_pointer(h->first, n);
+	rcu_assign_pointer(hlist_nulls_first_rcu(h), n);
 	if (!is_a_nulls(first))
 		first->pprev = &n->next;
 }
@@ -100,11 +106,11 @@
  * @member:	the name of the hlist_nulls_node within the struct.
  *
  */
-#define hlist_nulls_for_each_entry_rcu(tpos, pos, head, member) \
-	for (pos = rcu_dereference_raw((head)->first);			 \
-		(!is_a_nulls(pos)) &&			\
+#define hlist_nulls_for_each_entry_rcu(tpos, pos, head, member)			\
+	for (pos = rcu_dereference_raw(hlist_nulls_first_rcu(head));		\
+		(!is_a_nulls(pos)) &&						\
 		({ tpos = hlist_nulls_entry(pos, typeof(*tpos), member); 1; }); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(hlist_nulls_next_rcu(pos)))
 
 #endif
 #endif
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 83af1f8..03cda7b 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -41,11 +41,15 @@
 #include <linux/lockdep.h>
 #include <linux/completion.h>
 #include <linux/debugobjects.h>
+#include <linux/compiler.h>
 
 #ifdef CONFIG_RCU_TORTURE_TEST
 extern int rcutorture_runnable; /* for sysctl */
 #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
 
+#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
+#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
+
 /**
  * struct rcu_head - callback structure for use with RCU
  * @next: next update requests in a list
@@ -57,29 +61,94 @@
 };
 
 /* Exported common interfaces */
-extern void rcu_barrier(void);
+extern void call_rcu_sched(struct rcu_head *head,
+			   void (*func)(struct rcu_head *rcu));
+extern void synchronize_sched(void);
 extern void rcu_barrier_bh(void);
 extern void rcu_barrier_sched(void);
 extern void synchronize_sched_expedited(void);
 extern int sched_expedited_torture_stats(char *page);
 
+static inline void __rcu_read_lock_bh(void)
+{
+	local_bh_disable();
+}
+
+static inline void __rcu_read_unlock_bh(void)
+{
+	local_bh_enable();
+}
+
+#ifdef CONFIG_PREEMPT_RCU
+
+extern void __rcu_read_lock(void);
+extern void __rcu_read_unlock(void);
+void synchronize_rcu(void);
+
+/*
+ * Defined as a macro as it is a very low level header included from
+ * areas that don't even know about current.  This gives the rcu_read_lock()
+ * nesting depth, but makes sense only if CONFIG_PREEMPT_RCU -- in other
+ * types of kernel builds, the rcu_read_lock() nesting depth is unknowable.
+ */
+#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
+
+#else /* #ifdef CONFIG_PREEMPT_RCU */
+
+static inline void __rcu_read_lock(void)
+{
+	preempt_disable();
+}
+
+static inline void __rcu_read_unlock(void)
+{
+	preempt_enable();
+}
+
+static inline void synchronize_rcu(void)
+{
+	synchronize_sched();
+}
+
+static inline int rcu_preempt_depth(void)
+{
+	return 0;
+}
+
+#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
+
 /* Internal to kernel */
 extern void rcu_init(void);
+extern void rcu_sched_qs(int cpu);
+extern void rcu_bh_qs(int cpu);
+extern void rcu_check_callbacks(int cpu, int user);
+struct notifier_block;
+
+#ifdef CONFIG_NO_HZ
+
+extern void rcu_enter_nohz(void);
+extern void rcu_exit_nohz(void);
+
+#else /* #ifdef CONFIG_NO_HZ */
+
+static inline void rcu_enter_nohz(void)
+{
+}
+
+static inline void rcu_exit_nohz(void)
+{
+}
+
+#endif /* #else #ifdef CONFIG_NO_HZ */
 
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 #include <linux/rcutree.h>
-#elif defined(CONFIG_TINY_RCU)
+#elif defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 #include <linux/rcutiny.h>
 #else
 #error "Unknown RCU implementation specified to kernel configuration"
 #endif
 
-#define RCU_HEAD_INIT	{ .next = NULL, .func = NULL }
-#define RCU_HEAD(head) struct rcu_head head = RCU_HEAD_INIT
-#define INIT_RCU_HEAD(ptr) do { \
-       (ptr)->next = NULL; (ptr)->func = NULL; \
-} while (0)
-
 /*
  * init_rcu_head_on_stack()/destroy_rcu_head_on_stack() are needed for dynamic
  * initialization and destruction of rcu_head on the stack. rcu_head structures
@@ -120,14 +189,15 @@
 extern int debug_lockdep_rcu_enabled(void);
 
 /**
- * rcu_read_lock_held - might we be in RCU read-side critical section?
+ * rcu_read_lock_held() - might we be in RCU read-side critical section?
  *
  * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an RCU
  * read-side critical section.  In absence of CONFIG_DEBUG_LOCK_ALLOC,
  * this assumes we are in an RCU read-side critical section unless it can
- * prove otherwise.
+ * prove otherwise.  This is useful for debug checks in functions that
+ * require that they be called within an RCU read-side critical section.
  *
- * Check debug_lockdep_rcu_enabled() to prevent false positives during boot
+ * Checks debug_lockdep_rcu_enabled() to prevent false positives during boot
  * and while lockdep is disabled.
  */
 static inline int rcu_read_lock_held(void)
@@ -144,14 +214,16 @@
 extern int rcu_read_lock_bh_held(void);
 
 /**
- * rcu_read_lock_sched_held - might we be in RCU-sched read-side critical section?
+ * rcu_read_lock_sched_held() - might we be in RCU-sched read-side critical section?
  *
  * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an
  * RCU-sched read-side critical section.  In absence of
  * CONFIG_DEBUG_LOCK_ALLOC, this assumes we are in an RCU-sched read-side
  * critical section unless it can prove otherwise.  Note that disabling
  * of preemption (including disabling irqs) counts as an RCU-sched
- * read-side critical section.
+ * read-side critical section.  This is useful for debug checks in functions
+ * that required that they be called within an RCU-sched read-side
+ * critical section.
  *
  * Check debug_lockdep_rcu_enabled() to prevent false positives during boot
  * and while lockdep is disabled.
@@ -211,7 +283,11 @@
 
 extern int rcu_my_thread_group_empty(void);
 
-#define __do_rcu_dereference_check(c)					\
+/**
+ * rcu_lockdep_assert - emit lockdep splat if specified condition not met
+ * @c: condition to check
+ */
+#define rcu_lockdep_assert(c)						\
 	do {								\
 		static bool __warned;					\
 		if (debug_lockdep_rcu_enabled() && !__warned && !(c)) {	\
@@ -220,41 +296,163 @@
 		}							\
 	} while (0)
 
+#else /* #ifdef CONFIG_PROVE_RCU */
+
+#define rcu_lockdep_assert(c) do { } while (0)
+
+#endif /* #else #ifdef CONFIG_PROVE_RCU */
+
+/*
+ * Helper functions for rcu_dereference_check(), rcu_dereference_protected()
+ * and rcu_assign_pointer().  Some of these could be folded into their
+ * callers, but they are left separate in order to ease introduction of
+ * multiple flavors of pointers to match the multiple flavors of RCU
+ * (e.g., __rcu_bh, * __rcu_sched, and __srcu), should this make sense in
+ * the future.
+ */
+
+#ifdef __CHECKER__
+#define rcu_dereference_sparse(p, space) \
+	((void)(((typeof(*p) space *)p) == p))
+#else /* #ifdef __CHECKER__ */
+#define rcu_dereference_sparse(p, space)
+#endif /* #else #ifdef __CHECKER__ */
+
+#define __rcu_access_pointer(p, space) \
+	({ \
+		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
+		rcu_dereference_sparse(p, space); \
+		((typeof(*p) __force __kernel *)(_________p1)); \
+	})
+#define __rcu_dereference_check(p, c, space) \
+	({ \
+		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
+		rcu_lockdep_assert(c); \
+		rcu_dereference_sparse(p, space); \
+		smp_read_barrier_depends(); \
+		((typeof(*p) __force __kernel *)(_________p1)); \
+	})
+#define __rcu_dereference_protected(p, c, space) \
+	({ \
+		rcu_lockdep_assert(c); \
+		rcu_dereference_sparse(p, space); \
+		((typeof(*p) __force __kernel *)(p)); \
+	})
+
+#define __rcu_dereference_index_check(p, c) \
+	({ \
+		typeof(p) _________p1 = ACCESS_ONCE(p); \
+		rcu_lockdep_assert(c); \
+		smp_read_barrier_depends(); \
+		(_________p1); \
+	})
+#define __rcu_assign_pointer(p, v, space) \
+	({ \
+		if (!__builtin_constant_p(v) || \
+		    ((v) != NULL)) \
+			smp_wmb(); \
+		(p) = (typeof(*v) __force space *)(v); \
+	})
+
+
 /**
- * rcu_dereference_check - rcu_dereference with debug checking
+ * rcu_access_pointer() - fetch RCU pointer with no dereferencing
+ * @p: The pointer to read
+ *
+ * Return the value of the specified RCU-protected pointer, but omit the
+ * smp_read_barrier_depends() and keep the ACCESS_ONCE().  This is useful
+ * when the value of this pointer is accessed, but the pointer is not
+ * dereferenced, for example, when testing an RCU-protected pointer against
+ * NULL.  Although rcu_access_pointer() may also be used in cases where
+ * update-side locks prevent the value of the pointer from changing, you
+ * should instead use rcu_dereference_protected() for this use case.
+ */
+#define rcu_access_pointer(p) __rcu_access_pointer((p), __rcu)
+
+/**
+ * rcu_dereference_check() - rcu_dereference with debug checking
  * @p: The pointer to read, prior to dereferencing
  * @c: The conditions under which the dereference will take place
  *
  * Do an rcu_dereference(), but check that the conditions under which the
- * dereference will take place are correct.  Typically the conditions indicate
- * the various locking conditions that should be held at that point.  The check
- * should return true if the conditions are satisfied.
+ * dereference will take place are correct.  Typically the conditions
+ * indicate the various locking conditions that should be held at that
+ * point.  The check should return true if the conditions are satisfied.
+ * An implicit check for being in an RCU read-side critical section
+ * (rcu_read_lock()) is included.
  *
  * For example:
  *
- *	bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() ||
- *					      lockdep_is_held(&foo->lock));
+ *	bar = rcu_dereference_check(foo->bar, lockdep_is_held(&foo->lock));
  *
  * could be used to indicate to lockdep that foo->bar may only be dereferenced
- * if either the RCU read lock is held, or that the lock required to replace
+ * if either rcu_read_lock() is held, or that the lock required to replace
  * the bar struct at foo->bar is held.
  *
  * Note that the list of conditions may also include indications of when a lock
  * need not be held, for example during initialisation or destruction of the
  * target struct:
  *
- *	bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() ||
- *					      lockdep_is_held(&foo->lock) ||
+ *	bar = rcu_dereference_check(foo->bar, lockdep_is_held(&foo->lock) ||
  *					      atomic_read(&foo->usage) == 0);
+ *
+ * Inserts memory barriers on architectures that require them
+ * (currently only the Alpha), prevents the compiler from refetching
+ * (and from merging fetches), and, more importantly, documents exactly
+ * which pointers are protected by RCU and checks that the pointer is
+ * annotated as __rcu.
  */
 #define rcu_dereference_check(p, c) \
-	({ \
-		__do_rcu_dereference_check(c); \
-		rcu_dereference_raw(p); \
-	})
+	__rcu_dereference_check((p), rcu_read_lock_held() || (c), __rcu)
 
 /**
- * rcu_dereference_protected - fetch RCU pointer when updates prevented
+ * rcu_dereference_bh_check() - rcu_dereference_bh with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-bh counterpart to rcu_dereference_check().
+ */
+#define rcu_dereference_bh_check(p, c) \
+	__rcu_dereference_check((p), rcu_read_lock_bh_held() || (c), __rcu)
+
+/**
+ * rcu_dereference_sched_check() - rcu_dereference_sched with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-sched counterpart to rcu_dereference_check().
+ */
+#define rcu_dereference_sched_check(p, c) \
+	__rcu_dereference_check((p), rcu_read_lock_sched_held() || (c), \
+				__rcu)
+
+#define rcu_dereference_raw(p) rcu_dereference_check(p, 1) /*@@@ needed? @@@*/
+
+/**
+ * rcu_dereference_index_check() - rcu_dereference for indices with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * Similar to rcu_dereference_check(), but omits the sparse checking.
+ * This allows rcu_dereference_index_check() to be used on integers,
+ * which can then be used as array indices.  Attempting to use
+ * rcu_dereference_check() on an integer will give compiler warnings
+ * because the sparse address-space mechanism relies on dereferencing
+ * the RCU-protected pointer.  Dereferencing integers is not something
+ * that even gcc will put up with.
+ *
+ * Note that this function does not implicitly check for RCU read-side
+ * critical sections.  If this function gains lots of uses, it might
+ * make sense to provide versions for each flavor of RCU, but it does
+ * not make sense as of early 2010.
+ */
+#define rcu_dereference_index_check(p, c) \
+	__rcu_dereference_index_check((p), (c))
+
+/**
+ * rcu_dereference_protected() - fetch RCU pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
  *
  * Return the value of the specified RCU-protected pointer, but omit
  * both the smp_read_barrier_depends() and the ACCESS_ONCE().  This
@@ -263,35 +461,61 @@
  * prevent the compiler from repeating this reference or combining it
  * with other references, so it should not be used without protection
  * of appropriate locks.
+ *
+ * This function is only for update-side use.  Using this function
+ * when protected only by rcu_read_lock() will result in infrequent
+ * but very ugly failures.
  */
 #define rcu_dereference_protected(p, c) \
-	({ \
-		__do_rcu_dereference_check(c); \
-		(p); \
-	})
-
-#else /* #ifdef CONFIG_PROVE_RCU */
-
-#define rcu_dereference_check(p, c)	rcu_dereference_raw(p)
-#define rcu_dereference_protected(p, c) (p)
-
-#endif /* #else #ifdef CONFIG_PROVE_RCU */
+	__rcu_dereference_protected((p), (c), __rcu)
 
 /**
- * rcu_access_pointer - fetch RCU pointer with no dereferencing
+ * rcu_dereference_bh_protected() - fetch RCU-bh pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
  *
- * Return the value of the specified RCU-protected pointer, but omit the
- * smp_read_barrier_depends() and keep the ACCESS_ONCE().  This is useful
- * when the value of this pointer is accessed, but the pointer is not
- * dereferenced, for example, when testing an RCU-protected pointer against
- * NULL.  This may also be used in cases where update-side locks prevent
- * the value of the pointer from changing, but rcu_dereference_protected()
- * is a lighter-weight primitive for this use case.
+ * This is the RCU-bh counterpart to rcu_dereference_protected().
  */
-#define rcu_access_pointer(p)	ACCESS_ONCE(p)
+#define rcu_dereference_bh_protected(p, c) \
+	__rcu_dereference_protected((p), (c), __rcu)
 
 /**
- * rcu_read_lock - mark the beginning of an RCU read-side critical section.
+ * rcu_dereference_sched_protected() - fetch RCU-sched pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-sched counterpart to rcu_dereference_protected().
+ */
+#define rcu_dereference_sched_protected(p, c) \
+	__rcu_dereference_protected((p), (c), __rcu)
+
+
+/**
+ * rcu_dereference() - fetch RCU-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * This is a simple wrapper around rcu_dereference_check().
+ */
+#define rcu_dereference(p) rcu_dereference_check(p, 0)
+
+/**
+ * rcu_dereference_bh() - fetch an RCU-bh-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Makes rcu_dereference_check() do the dirty work.
+ */
+#define rcu_dereference_bh(p) rcu_dereference_bh_check(p, 0)
+
+/**
+ * rcu_dereference_sched() - fetch RCU-sched-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Makes rcu_dereference_check() do the dirty work.
+ */
+#define rcu_dereference_sched(p) rcu_dereference_sched_check(p, 0)
+
+/**
+ * rcu_read_lock() - mark the beginning of an RCU read-side critical section
  *
  * When synchronize_rcu() is invoked on one CPU while other CPUs
  * are within RCU read-side critical sections, then the
@@ -302,7 +526,7 @@
  * until after the all the other CPUs exit their critical sections.
  *
  * Note, however, that RCU callbacks are permitted to run concurrently
- * with RCU read-side critical sections.  One way that this can happen
+ * with new RCU read-side critical sections.  One way that this can happen
  * is via the following sequence of events: (1) CPU 0 enters an RCU
  * read-side critical section, (2) CPU 1 invokes call_rcu() to register
  * an RCU callback, (3) CPU 0 exits the RCU read-side critical section,
@@ -317,7 +541,20 @@
  * will be deferred until the outermost RCU read-side critical section
  * completes.
  *
- * It is illegal to block while in an RCU read-side critical section.
+ * You can avoid reading and understanding the next paragraph by
+ * following this rule: don't put anything in an rcu_read_lock() RCU
+ * read-side critical section that would block in a !PREEMPT kernel.
+ * But if you want the full story, read on!
+ *
+ * In non-preemptible RCU implementations (TREE_RCU and TINY_RCU), it
+ * is illegal to block while in an RCU read-side critical section.  In
+ * preemptible RCU implementations (TREE_PREEMPT_RCU and TINY_PREEMPT_RCU)
+ * in CONFIG_PREEMPT kernel builds, RCU read-side critical sections may
+ * be preempted, but explicit blocking is illegal.  Finally, in preemptible
+ * RCU implementations in real-time (CONFIG_PREEMPT_RT) kernel builds,
+ * RCU read-side critical sections may be preempted and they may also
+ * block, but only when acquiring spinlocks that are subject to priority
+ * inheritance.
  */
 static inline void rcu_read_lock(void)
 {
@@ -337,7 +574,7 @@
  */
 
 /**
- * rcu_read_unlock - marks the end of an RCU read-side critical section.
+ * rcu_read_unlock() - marks the end of an RCU read-side critical section.
  *
  * See rcu_read_lock() for more information.
  */
@@ -349,15 +586,16 @@
 }
 
 /**
- * rcu_read_lock_bh - mark the beginning of a softirq-only RCU critical section
+ * rcu_read_lock_bh() - mark the beginning of an RCU-bh critical section
  *
  * This is equivalent of rcu_read_lock(), but to be used when updates
- * are being done using call_rcu_bh(). Since call_rcu_bh() callbacks
- * consider completion of a softirq handler to be a quiescent state,
- * a process in RCU read-side critical section must be protected by
- * disabling softirqs. Read-side critical sections in interrupt context
- * can use just rcu_read_lock().
- *
+ * are being done using call_rcu_bh() or synchronize_rcu_bh(). Since
+ * both call_rcu_bh() and synchronize_rcu_bh() consider completion of a
+ * softirq handler to be a quiescent state, a process in RCU read-side
+ * critical section must be protected by disabling softirqs. Read-side
+ * critical sections in interrupt context can use just rcu_read_lock(),
+ * though this should at least be commented to avoid confusing people
+ * reading the code.
  */
 static inline void rcu_read_lock_bh(void)
 {
@@ -379,13 +617,12 @@
 }
 
 /**
- * rcu_read_lock_sched - mark the beginning of a RCU-classic critical section
+ * rcu_read_lock_sched() - mark the beginning of a RCU-sched critical section
  *
- * Should be used with either
- * - synchronize_sched()
- * or
- * - call_rcu_sched() and rcu_barrier_sched()
- * on the write-side to insure proper synchronization.
+ * This is equivalent of rcu_read_lock(), but to be used when updates
+ * are being done using call_rcu_sched() or synchronize_rcu_sched().
+ * Read-side critical sections can also be introduced by anything that
+ * disables preemption, including local_irq_disable() and friends.
  */
 static inline void rcu_read_lock_sched(void)
 {
@@ -420,54 +657,14 @@
 	preempt_enable_notrace();
 }
 
-
 /**
- * rcu_dereference_raw - fetch an RCU-protected pointer
+ * rcu_assign_pointer() - assign to RCU-protected pointer
+ * @p: pointer to assign to
+ * @v: value to assign (publish)
  *
- * The caller must be within some flavor of RCU read-side critical
- * section, or must be otherwise preventing the pointer from changing,
- * for example, by holding an appropriate lock.  This pointer may later
- * be safely dereferenced.  It is the caller's responsibility to have
- * done the right thing, as this primitive does no checking of any kind.
- *
- * Inserts memory barriers on architectures that require them
- * (currently only the Alpha), and, more importantly, documents
- * exactly which pointers are protected by RCU.
- */
-#define rcu_dereference_raw(p)	({ \
-				typeof(p) _________p1 = ACCESS_ONCE(p); \
-				smp_read_barrier_depends(); \
-				(_________p1); \
-				})
-
-/**
- * rcu_dereference - fetch an RCU-protected pointer, checking for RCU
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference(p) \
-	rcu_dereference_check(p, rcu_read_lock_held())
-
-/**
- * rcu_dereference_bh - fetch an RCU-protected pointer, checking for RCU-bh
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference_bh(p) \
-		rcu_dereference_check(p, rcu_read_lock_bh_held() || irqs_disabled())
-
-/**
- * rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference_sched(p) \
-		rcu_dereference_check(p, rcu_read_lock_sched_held())
-
-/**
- * rcu_assign_pointer - assign (publicize) a pointer to a newly
- * initialized structure that will be dereferenced by RCU read-side
- * critical sections.  Returns the value assigned.
+ * Assigns the specified value to the specified RCU-protected
+ * pointer, ensuring that any concurrent RCU readers will see
+ * any prior initialization.  Returns the value assigned.
  *
  * Inserts memory barriers on architectures that require them
  * (pretty much all of them other than x86), and also prevents
@@ -476,14 +673,17 @@
  * call documents which pointers will be dereferenced by RCU read-side
  * code.
  */
-
 #define rcu_assign_pointer(p, v) \
-	({ \
-		if (!__builtin_constant_p(v) || \
-		    ((v) != NULL)) \
-			smp_wmb(); \
-		(p) = (v); \
-	})
+	__rcu_assign_pointer((p), (v), __rcu)
+
+/**
+ * RCU_INIT_POINTER() - initialize an RCU protected pointer
+ *
+ * Initialize an RCU-protected pointer in such a way to avoid RCU-lockdep
+ * splats.
+ */
+#define RCU_INIT_POINTER(p, v) \
+		p = (typeof(*v) __force __rcu *)(v)
 
 /* Infrastructure to implement the synchronize_() primitives. */
 
@@ -494,26 +694,37 @@
 
 extern void wakeme_after_rcu(struct rcu_head  *head);
 
+#ifdef CONFIG_PREEMPT_RCU
+
 /**
- * call_rcu - Queue an RCU callback for invocation after a grace period.
+ * call_rcu() - Queue an RCU callback for invocation after a grace period.
  * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * @func: actual callback function to be invoked after the grace period
  *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
+ * The callback function will be invoked some time after a full grace
+ * period elapses, in other words after all pre-existing RCU read-side
+ * critical sections have completed.  However, the callback function
+ * might well execute concurrently with RCU read-side critical sections
+ * that started after call_rcu() was invoked.  RCU read-side critical
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
 extern void call_rcu(struct rcu_head *head,
 			      void (*func)(struct rcu_head *head));
 
+#else /* #ifdef CONFIG_PREEMPT_RCU */
+
+/* In classic RCU, call_rcu() is just call_rcu_sched(). */
+#define	call_rcu	call_rcu_sched
+
+#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
+
 /**
- * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
+ * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period.
  * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * @func: actual callback function to be invoked after the grace period
  *
- * The update function will be invoked some time after a full grace
+ * The callback function will be invoked some time after a full grace
  * period elapses, in other words after all currently executing RCU
  * read-side critical sections have completed. call_rcu_bh() assumes
  * that the read-side critical sections end on completion of a softirq
@@ -566,37 +777,4 @@
 }
 #endif	/* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
 
-#ifndef CONFIG_PROVE_RCU
-#define __do_rcu_dereference_check(c) do { } while (0)
-#endif /* #ifdef CONFIG_PROVE_RCU */
-
-#define __rcu_dereference_index_check(p, c) \
-	({ \
-		typeof(p) _________p1 = ACCESS_ONCE(p); \
-		__do_rcu_dereference_check(c); \
-		smp_read_barrier_depends(); \
-		(_________p1); \
-	})
-
-/**
- * rcu_dereference_index_check() - rcu_dereference for indices with debug checking
- * @p: The pointer to read, prior to dereferencing
- * @c: The conditions under which the dereference will take place
- *
- * Similar to rcu_dereference_check(), but omits the sparse checking.
- * This allows rcu_dereference_index_check() to be used on integers,
- * which can then be used as array indices.  Attempting to use
- * rcu_dereference_check() on an integer will give compiler warnings
- * because the sparse address-space mechanism relies on dereferencing
- * the RCU-protected pointer.  Dereferencing integers is not something
- * that even gcc will put up with.
- *
- * Note that this function does not implicitly check for RCU read-side
- * critical sections.  If this function gains lots of uses, it might
- * make sense to provide versions for each flavor of RCU, but it does
- * not make sense as of early 2010.
- */
-#define rcu_dereference_index_check(p, c) \
-	__rcu_dereference_index_check((p), (c))
-
 #endif /* __LINUX_RCUPDATE_H */
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index e2e8931..13877cb 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -27,27 +27,71 @@
 
 #include <linux/cache.h>
 
-void rcu_sched_qs(int cpu);
-void rcu_bh_qs(int cpu);
-static inline void rcu_note_context_switch(int cpu)
+#define rcu_init_sched()	do { } while (0)
+
+#ifdef CONFIG_TINY_RCU
+
+static inline void synchronize_rcu_expedited(void)
 {
-	rcu_sched_qs(cpu);
+	synchronize_sched();	/* Only one CPU, so pretty fast anyway!!! */
 }
 
-#define __rcu_read_lock()	preempt_disable()
-#define __rcu_read_unlock()	preempt_enable()
-#define __rcu_read_lock_bh()	local_bh_disable()
-#define __rcu_read_unlock_bh()	local_bh_enable()
-#define call_rcu_sched		call_rcu
+static inline void rcu_barrier(void)
+{
+	rcu_barrier_sched();  /* Only one CPU, so only one list of callbacks! */
+}
 
-#define rcu_init_sched()	do { } while (0)
-extern void rcu_check_callbacks(int cpu, int user);
+#else /* #ifdef CONFIG_TINY_RCU */
+
+void rcu_barrier(void);
+void synchronize_rcu_expedited(void);
+
+#endif /* #else #ifdef CONFIG_TINY_RCU */
+
+static inline void synchronize_rcu_bh(void)
+{
+	synchronize_sched();
+}
+
+static inline void synchronize_rcu_bh_expedited(void)
+{
+	synchronize_sched();
+}
+
+#ifdef CONFIG_TINY_RCU
+
+static inline void rcu_preempt_note_context_switch(void)
+{
+}
+
+static inline void exit_rcu(void)
+{
+}
 
 static inline int rcu_needs_cpu(int cpu)
 {
 	return 0;
 }
 
+#else /* #ifdef CONFIG_TINY_RCU */
+
+void rcu_preempt_note_context_switch(void);
+extern void exit_rcu(void);
+int rcu_preempt_needs_cpu(void);
+
+static inline int rcu_needs_cpu(int cpu)
+{
+	return rcu_preempt_needs_cpu();
+}
+
+#endif /* #else #ifdef CONFIG_TINY_RCU */
+
+static inline void rcu_note_context_switch(int cpu)
+{
+	rcu_sched_qs(cpu);
+	rcu_preempt_note_context_switch();
+}
+
 /*
  * Return the number of grace periods.
  */
@@ -76,54 +120,8 @@
 {
 }
 
-extern void synchronize_sched(void);
-
-static inline void synchronize_rcu(void)
+static inline void rcu_cpu_stall_reset(void)
 {
-	synchronize_sched();
-}
-
-static inline void synchronize_rcu_bh(void)
-{
-	synchronize_sched();
-}
-
-static inline void synchronize_rcu_expedited(void)
-{
-	synchronize_sched();
-}
-
-static inline void synchronize_rcu_bh_expedited(void)
-{
-	synchronize_sched();
-}
-
-struct notifier_block;
-
-#ifdef CONFIG_NO_HZ
-
-extern void rcu_enter_nohz(void);
-extern void rcu_exit_nohz(void);
-
-#else /* #ifdef CONFIG_NO_HZ */
-
-static inline void rcu_enter_nohz(void)
-{
-}
-
-static inline void rcu_exit_nohz(void)
-{
-}
-
-#endif /* #else #ifdef CONFIG_NO_HZ */
-
-static inline void exit_rcu(void)
-{
-}
-
-static inline int rcu_preempt_depth(void)
-{
-	return 0;
 }
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index c0ed1c0..95518e6 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -30,64 +30,23 @@
 #ifndef __LINUX_RCUTREE_H
 #define __LINUX_RCUTREE_H
 
-struct notifier_block;
-
-extern void rcu_sched_qs(int cpu);
-extern void rcu_bh_qs(int cpu);
 extern void rcu_note_context_switch(int cpu);
 extern int rcu_needs_cpu(int cpu);
+extern void rcu_cpu_stall_reset(void);
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
 
-extern void __rcu_read_lock(void);
-extern void __rcu_read_unlock(void);
-extern void synchronize_rcu(void);
 extern void exit_rcu(void);
 
-/*
- * Defined as macro as it is a very low level header
- * included from areas that don't even know about current
- */
-#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
-
 #else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
-static inline void __rcu_read_lock(void)
-{
-	preempt_disable();
-}
-
-static inline void __rcu_read_unlock(void)
-{
-	preempt_enable();
-}
-
-#define synchronize_rcu synchronize_sched
-
 static inline void exit_rcu(void)
 {
 }
 
-static inline int rcu_preempt_depth(void)
-{
-	return 0;
-}
-
 #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */
 
-static inline void __rcu_read_lock_bh(void)
-{
-	local_bh_disable();
-}
-static inline void __rcu_read_unlock_bh(void)
-{
-	local_bh_enable();
-}
-
-extern void call_rcu_sched(struct rcu_head *head,
-			   void (*func)(struct rcu_head *rcu));
 extern void synchronize_rcu_bh(void);
-extern void synchronize_sched(void);
 extern void synchronize_rcu_expedited(void);
 
 static inline void synchronize_rcu_bh_expedited(void)
@@ -95,7 +54,7 @@
 	synchronize_sched_expedited();
 }
 
-extern void rcu_check_callbacks(int cpu, int user);
+extern void rcu_barrier(void);
 
 extern long rcu_batches_completed(void);
 extern long rcu_batches_completed_bh(void);
@@ -104,18 +63,6 @@
 extern void rcu_bh_force_quiescent_state(void);
 extern void rcu_sched_force_quiescent_state(void);
 
-#ifdef CONFIG_NO_HZ
-void rcu_enter_nohz(void);
-void rcu_exit_nohz(void);
-#else /* CONFIG_NO_HZ */
-static inline void rcu_enter_nohz(void)
-{
-}
-static inline void rcu_exit_nohz(void)
-{
-}
-#endif /* CONFIG_NO_HZ */
-
 /* A context switch is a grace period for RCU-sched and RCU-bh. */
 static inline int rcu_blocking_is_gp(void)
 {
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db..0383601 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -875,6 +875,7 @@
 	SD_LV_NONE = 0,
 	SD_LV_SIBLING,
 	SD_LV_MC,
+	SD_LV_BOOK,
 	SD_LV_CPU,
 	SD_LV_NODE,
 	SD_LV_ALLNODES,
@@ -1160,6 +1161,13 @@
 
 struct rcu_node;
 
+enum perf_event_task_context {
+	perf_invalid_context = -1,
+	perf_hw_context = 0,
+	perf_sw_context,
+	perf_nr_task_contexts,
+};
+
 struct task_struct {
 	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
 	void *stack;
@@ -1202,11 +1210,13 @@
 	unsigned int policy;
 	cpumask_t cpus_allowed;
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#ifdef CONFIG_PREEMPT_RCU
 	int rcu_read_lock_nesting;
 	char rcu_read_unlock_special;
-	struct rcu_node *rcu_blocked_node;
 	struct list_head rcu_node_entry;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
+#ifdef CONFIG_TREE_PREEMPT_RCU
+	struct rcu_node *rcu_blocked_node;
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
@@ -1288,9 +1298,9 @@
 	struct list_head cpu_timers[3];
 
 /* process credentials */
-	const struct cred *real_cred;	/* objective and real subjective task
+	const struct cred __rcu *real_cred; /* objective and real subjective task
 					 * credentials (COW) */
-	const struct cred *cred;	/* effective (overridable) subjective task
+	const struct cred __rcu *cred;	/* effective (overridable) subjective task
 					 * credentials (COW) */
 	struct mutex cred_guard_mutex;	/* guard against foreign influences on
 					 * credential calculations
@@ -1418,7 +1428,7 @@
 #endif
 #ifdef CONFIG_CGROUPS
 	/* Control Group info protected by css_set_lock */
-	struct css_set *cgroups;
+	struct css_set __rcu *cgroups;
 	/* cg_list protected by css_set_lock and tsk->alloc_lock */
 	struct list_head cg_list;
 #endif
@@ -1431,7 +1441,7 @@
 	struct futex_pi_state *pi_state_cache;
 #endif
 #ifdef CONFIG_PERF_EVENTS
-	struct perf_event_context *perf_event_ctxp;
+	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
 	struct mutex perf_event_mutex;
 	struct list_head perf_event_list;
 #endif
@@ -1681,8 +1691,7 @@
 /*
  * Per process flags
  */
-#define PF_ALIGNWARN	0x00000001	/* Print alignment warning msgs */
-					/* Not implemented yet, only for 486*/
+#define PF_KSOFTIRQD	0x00000001	/* I am ksoftirqd */
 #define PF_STARTING	0x00000002	/* being created */
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
@@ -1740,7 +1749,7 @@
 #define tsk_used_math(p) ((p)->flags & PF_USED_MATH)
 #define used_math() tsk_used_math(current)
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#ifdef CONFIG_PREEMPT_RCU
 
 #define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */
 #define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */
@@ -1749,7 +1758,9 @@
 {
 	p->rcu_read_lock_nesting = 0;
 	p->rcu_read_unlock_special = 0;
+#ifdef CONFIG_TREE_PREEMPT_RCU
 	p->rcu_blocked_node = NULL;
+#endif
 	INIT_LIST_HEAD(&p->rcu_node_entry);
 }
 
@@ -1826,6 +1837,19 @@
 extern void sched_clock_idle_wakeup_event(u64 delta_ns);
 #endif
 
+#ifdef CONFIG_IRQ_TIME_ACCOUNTING
+/*
+ * An i/f to runtime opt-in for irq time accounting based off of sched_clock.
+ * The reason for this explicit opt-in is not to have perf penalty with
+ * slow sched_clocks.
+ */
+extern void enable_sched_clock_irqtime(void);
+extern void disable_sched_clock_irqtime(void);
+#else
+static inline void enable_sched_clock_irqtime(void) {}
+static inline void disable_sched_clock_irqtime(void) {}
+#endif
+
 extern unsigned long long
 task_sched_runtime(struct task_struct *task);
 extern unsigned long long thread_group_sched_runtime(struct task_struct *task);
@@ -2367,9 +2391,9 @@
 
 extern int __cond_resched_softirq(void);
 
-#define cond_resched_softirq() ({				\
-	__might_sleep(__FILE__, __LINE__, SOFTIRQ_OFFSET);	\
-	__cond_resched_softirq();				\
+#define cond_resched_softirq() ({					\
+	__might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
+	__cond_resched_softirq();					\
 })
 
 /*
diff --git a/include/linux/security.h b/include/linux/security.h
index a22219a..b8246a8 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -74,7 +74,7 @@
 extern int cap_task_fix_setuid(struct cred *new, const struct cred *old, int flags);
 extern int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3,
 			  unsigned long arg4, unsigned long arg5);
-extern int cap_task_setscheduler(struct task_struct *p, int policy, struct sched_param *lp);
+extern int cap_task_setscheduler(struct task_struct *p);
 extern int cap_task_setioprio(struct task_struct *p, int ioprio);
 extern int cap_task_setnice(struct task_struct *p, int nice);
 extern int cap_syslog(int type, bool from_file);
@@ -959,6 +959,12 @@
  *	Sets the new child socket's sid to the openreq sid.
  * @inet_conn_established:
  *	Sets the connection's peersid to the secmark on skb.
+ * @secmark_relabel_packet:
+ *	check if the process should be allowed to relabel packets to the given secid
+ * @security_secmark_refcount_inc
+ *	tells the LSM to increment the number of secmark labeling rules loaded
+ * @security_secmark_refcount_dec
+ *	tells the LSM to decrement the number of secmark labeling rules loaded
  * @req_classify_flow:
  *	Sets the flow's sid to the openreq sid.
  * @tun_dev_create:
@@ -1279,9 +1285,13 @@
  *	Return 0 if permission is granted.
  *
  * @secid_to_secctx:
- *	Convert secid to security context.
+ *	Convert secid to security context.  If secdata is NULL the length of
+ *	the result will be returned in seclen, but no secdata will be returned.
+ *	This does mean that the length could change between calls to check the
+ *	length and the next call which actually allocates and returns the secdata.
  *	@secid contains the security ID.
  *	@secdata contains the pointer that stores the converted security context.
+ *	@seclen pointer which contains the length of the data
  * @secctx_to_secid:
  *	Convert security context to secid.
  *	@secid contains the pointer to the generated security ID.
@@ -1501,8 +1511,7 @@
 	int (*task_getioprio) (struct task_struct *p);
 	int (*task_setrlimit) (struct task_struct *p, unsigned int resource,
 			struct rlimit *new_rlim);
-	int (*task_setscheduler) (struct task_struct *p, int policy,
-				  struct sched_param *lp);
+	int (*task_setscheduler) (struct task_struct *p);
 	int (*task_getscheduler) (struct task_struct *p);
 	int (*task_movememory) (struct task_struct *p);
 	int (*task_kill) (struct task_struct *p,
@@ -1594,6 +1603,9 @@
 				  struct request_sock *req);
 	void (*inet_csk_clone) (struct sock *newsk, const struct request_sock *req);
 	void (*inet_conn_established) (struct sock *sk, struct sk_buff *skb);
+	int (*secmark_relabel_packet) (u32 secid);
+	void (*secmark_refcount_inc) (void);
+	void (*secmark_refcount_dec) (void);
 	void (*req_classify_flow) (const struct request_sock *req, struct flowi *fl);
 	int (*tun_dev_create)(void);
 	void (*tun_dev_post_create)(struct sock *sk);
@@ -1752,8 +1764,7 @@
 int security_task_getioprio(struct task_struct *p);
 int security_task_setrlimit(struct task_struct *p, unsigned int resource,
 		struct rlimit *new_rlim);
-int security_task_setscheduler(struct task_struct *p,
-				int policy, struct sched_param *lp);
+int security_task_setscheduler(struct task_struct *p);
 int security_task_getscheduler(struct task_struct *p);
 int security_task_movememory(struct task_struct *p);
 int security_task_kill(struct task_struct *p, struct siginfo *info,
@@ -2320,11 +2331,9 @@
 	return 0;
 }
 
-static inline int security_task_setscheduler(struct task_struct *p,
-					     int policy,
-					     struct sched_param *lp)
+static inline int security_task_setscheduler(struct task_struct *p)
 {
-	return cap_task_setscheduler(p, policy, lp);
+	return cap_task_setscheduler(p);
 }
 
 static inline int security_task_getscheduler(struct task_struct *p)
@@ -2551,6 +2560,9 @@
 			const struct request_sock *req);
 void security_inet_conn_established(struct sock *sk,
 			struct sk_buff *skb);
+int security_secmark_relabel_packet(u32 secid);
+void security_secmark_refcount_inc(void);
+void security_secmark_refcount_dec(void);
 int security_tun_dev_create(void);
 void security_tun_dev_post_create(struct sock *sk);
 int security_tun_dev_attach(struct sock *sk);
@@ -2705,6 +2717,19 @@
 {
 }
 
+static inline int security_secmark_relabel_packet(u32 secid)
+{
+	return 0;
+}
+
+static inline void security_secmark_refcount_inc(void)
+{
+}
+
+static inline void security_secmark_refcount_dec(void)
+{
+}
+
 static inline int security_tun_dev_create(void)
 {
 	return 0;
diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 82e0f26..44f4596 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -21,74 +21,11 @@
 #ifdef CONFIG_SECURITY_SELINUX
 
 /**
- *     selinux_string_to_sid - map a security context string to a security ID
- *     @str: the security context string to be mapped
- *     @sid: ID value returned via this.
- *
- *     Returns 0 if successful, with the SID stored in sid.  A value
- *     of zero for sid indicates no SID could be determined (but no error
- *     occurred).
- */
-int selinux_string_to_sid(char *str, u32 *sid);
-
-/**
- *     selinux_secmark_relabel_packet_permission - secmark permission check
- *     @sid: SECMARK ID value to be applied to network packet
- *
- *     Returns 0 if the current task is allowed to set the SECMARK label of
- *     packets with the supplied security ID.  Note that it is implicit that
- *     the packet is always being relabeled from the default unlabeled value,
- *     and that the access control decision is made in the AVC.
- */
-int selinux_secmark_relabel_packet_permission(u32 sid);
-
-/**
- *     selinux_secmark_refcount_inc - increments the secmark use counter
- *
- *     SELinux keeps track of the current SECMARK targets in use so it knows
- *     when to apply SECMARK label access checks to network packets.  This
- *     function incements this reference count to indicate that a new SECMARK
- *     target has been configured.
- */
-void selinux_secmark_refcount_inc(void);
-
-/**
- *     selinux_secmark_refcount_dec - decrements the secmark use counter
- *
- *     SELinux keeps track of the current SECMARK targets in use so it knows
- *     when to apply SECMARK label access checks to network packets.  This
- *     function decements this reference count to indicate that one of the
- *     existing SECMARK targets has been removed/flushed.
- */
-void selinux_secmark_refcount_dec(void);
-
-/**
  * selinux_is_enabled - is SELinux enabled?
  */
 bool selinux_is_enabled(void);
 #else
 
-static inline int selinux_string_to_sid(const char *str, u32 *sid)
-{
-       *sid = 0;
-       return 0;
-}
-
-static inline int selinux_secmark_relabel_packet_permission(u32 sid)
-{
-	return 0;
-}
-
-static inline void selinux_secmark_refcount_inc(void)
-{
-	return;
-}
-
-static inline void selinux_secmark_refcount_dec(void)
-{
-	return;
-}
-
 static inline bool selinux_is_enabled(void)
 {
 	return false;
diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index 4d5d2f5..58971e8 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -108,19 +108,43 @@
 #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
 /**
- * srcu_dereference - fetch SRCU-protected pointer with checking
+ * srcu_dereference_check - fetch SRCU-protected pointer for later dereferencing
+ * @p: the pointer to fetch and protect for later dereferencing
+ * @sp: pointer to the srcu_struct, which is used to check that we
+ *	really are in an SRCU read-side critical section.
+ * @c: condition to check for update-side use
  *
- * Makes rcu_dereference_check() do the dirty work.
+ * If PROVE_RCU is enabled, invoking this outside of an RCU read-side
+ * critical section will result in an RCU-lockdep splat, unless @c evaluates
+ * to 1.  The @c argument will normally be a logical expression containing
+ * lockdep_is_held() calls.
  */
-#define srcu_dereference(p, sp) \
-		rcu_dereference_check(p, srcu_read_lock_held(sp))
+#define srcu_dereference_check(p, sp, c) \
+	__rcu_dereference_check((p), srcu_read_lock_held(sp) || (c), __rcu)
+
+/**
+ * srcu_dereference - fetch SRCU-protected pointer for later dereferencing
+ * @p: the pointer to fetch and protect for later dereferencing
+ * @sp: pointer to the srcu_struct, which is used to check that we
+ *	really are in an SRCU read-side critical section.
+ *
+ * Makes rcu_dereference_check() do the dirty work.  If PROVE_RCU
+ * is enabled, invoking this outside of an RCU read-side critical
+ * section will result in an RCU-lockdep splat.
+ */
+#define srcu_dereference(p, sp) srcu_dereference_check((p), (sp), 0)
 
 /**
  * srcu_read_lock - register a new reader for an SRCU-protected structure.
  * @sp: srcu_struct in which to register the new reader.
  *
  * Enter an SRCU read-side critical section.  Note that SRCU read-side
- * critical sections may be nested.
+ * critical sections may be nested.  However, it is illegal to
+ * call anything that waits on an SRCU grace period for the same
+ * srcu_struct, whether directly or indirectly.  Please note that
+ * one way to indirectly wait on an SRCU grace period is to acquire
+ * a mutex that is held elsewhere while calling synchronize_srcu() or
+ * synchronize_srcu_expedited().
  */
 static inline int srcu_read_lock(struct srcu_struct *sp) __acquires(sp)
 {
diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h
index 6b524a0..1808960 100644
--- a/include/linux/stop_machine.h
+++ b/include/linux/stop_machine.h
@@ -126,8 +126,8 @@
 
 #else	 /* CONFIG_STOP_MACHINE && CONFIG_SMP */
 
-static inline int stop_machine(int (*fn)(void *), void *data,
-			       const struct cpumask *cpus)
+static inline int __stop_machine(int (*fn)(void *), void *data,
+				 const struct cpumask *cpus)
 {
 	int ret;
 	local_irq_disable();
@@ -136,5 +136,11 @@
 	return ret;
 }
 
+static inline int stop_machine(int (*fn)(void *), void *data,
+			       const struct cpumask *cpus)
+{
+	return __stop_machine(fn, data, cpus);
+}
+
 #endif	/* CONFIG_STOP_MACHINE && CONFIG_SMP */
 #endif	/* _LINUX_STOP_MACHINE */
diff --git a/include/linux/sunrpc/auth_gss.h b/include/linux/sunrpc/auth_gss.h
index 671538d..8eee9db 100644
--- a/include/linux/sunrpc/auth_gss.h
+++ b/include/linux/sunrpc/auth_gss.h
@@ -69,7 +69,7 @@
 	enum rpc_gss_proc	gc_proc;
 	u32			gc_seq;
 	spinlock_t		gc_seq_lock;
-	struct gss_ctx		*gc_gss_ctx;
+	struct gss_ctx __rcu	*gc_gss_ctx;
 	struct xdr_netobj	gc_wire_ctx;
 	u32			gc_win;
 	unsigned long		gc_expiry;
@@ -80,7 +80,7 @@
 struct gss_cred {
 	struct rpc_cred		gc_base;
 	enum rpc_gss_svc	gc_service;
-	struct gss_cl_ctx	*gc_ctx;
+	struct gss_cl_ctx __rcu	*gc_ctx;
 	struct gss_upcall_msg	*gc_upcall;
 	unsigned long		gc_upcall_timestamp;
 	unsigned char		gc_machine_cred : 1;
diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index a8cc4e1..c906965 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -23,12 +23,12 @@
 		};
 		/* For futex_wait and futex_wait_requeue_pi */
 		struct {
-			u32 *uaddr;
+			u32 __user *uaddr;
 			u32 val;
 			u32 flags;
 			u32 bitset;
 			u64 time;
-			u32 *uaddr2;
+			u32 __user *uaddr2;
 		} futex;
 		/* For nanosleep */
 		struct {
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 64e084f..b91a40e 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -201,6 +201,12 @@
 	.balance_interval	= 64,					\
 }
 
+#ifdef CONFIG_SCHED_BOOK
+#ifndef SD_BOOK_INIT
+#error Please define an appropriate SD_BOOK_INIT in include/asm/topology.h!!!
+#endif
+#endif /* CONFIG_SCHED_BOOK */
+
 #ifdef CONFIG_NUMA
 #ifndef SD_NODE_INIT
 #error Please define an appropriate SD_NODE_INIT in include/asm/topology.h!!!
diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index 103d1b6..a4a90b6 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -17,6 +17,7 @@
 #include <linux/errno.h>
 #include <linux/types.h>
 #include <linux/rcupdate.h>
+#include <linux/jump_label.h>
 
 struct module;
 struct tracepoint;
@@ -145,7 +146,9 @@
 	extern struct tracepoint __tracepoint_##name;			\
 	static inline void trace_##name(proto)				\
 	{								\
-		if (unlikely(__tracepoint_##name.state))		\
+		JUMP_LABEL(&__tracepoint_##name.state, do_trace);	\
+		return;							\
+do_trace:								\
 			__DO_TRACE(&__tracepoint_##name,		\
 				TP_PROTO(data_proto),			\
 				TP_ARGS(data_args));			\
diff --git a/include/linux/types.h b/include/linux/types.h
index 01a082f..357dbc1 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -121,7 +121,15 @@
 typedef		__s64		int64_t;
 #endif
 
-/* this is a special 64bit data type that is 8-byte aligned */
+/*
+ * aligned_u64 should be used in defining kernel<->userspace ABIs to avoid
+ * common 32/64-bit compat problems.
+ * 64-bit values align to 4-byte boundaries on x86_32 (and possibly other
+ * architectures) and to 8-byte boundaries on 64-bit architetures.  The new
+ * aligned_64 type enforces 8-byte alignment so that structs containing
+ * aligned_64 values have the same alignment on 32-bit and 64-bit architectures.
+ * No conversions are necessary between 32-bit user-space and a 64-bit kernel.
+ */
 #define aligned_u64 __u64 __attribute__((aligned(8)))
 #define aligned_be64 __be64 __attribute__((aligned(8)))
 #define aligned_le64 __le64 __attribute__((aligned(8)))
@@ -178,6 +186,11 @@
 typedef __u16 __bitwise __sum16;
 typedef __u32 __bitwise __wsum;
 
+/* this is a special 64bit data type that is 8-byte aligned */
+#define __aligned_u64 __u64 __attribute__((aligned(8)))
+#define __aligned_be64 __be64 __attribute__((aligned(8)))
+#define __aligned_le64 __le64 __attribute__((aligned(8)))
+
 #ifdef __KERNEL__
 typedef unsigned __bitwise__ gfp_t;
 typedef unsigned __bitwise__ fmode_t;
diff --git a/include/media/videobuf-dma-sg.h b/include/media/videobuf-dma-sg.h
index 97e07f4..aa4ebb4 100644
--- a/include/media/videobuf-dma-sg.h
+++ b/include/media/videobuf-dma-sg.h
@@ -48,6 +48,7 @@
 
 	/* for userland buffer */
 	int                 offset;
+	size_t		    size;
 	struct page         **pages;
 
 	/* for kernel buffers */
diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index 27a902d..30fce01 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -161,12 +161,30 @@
 {
 	struct sk_buff *skb;
 
+	release_sock(sk);
 	if ((skb = sock_alloc_send_skb(sk, len + BT_SKB_RESERVE, nb, err))) {
 		skb_reserve(skb, BT_SKB_RESERVE);
 		bt_cb(skb)->incoming  = 0;
 	}
+	lock_sock(sk);
+
+	if (!skb && *err)
+		return NULL;
+
+	*err = sock_error(sk);
+	if (*err)
+		goto out;
+
+	if (sk->sk_shutdown) {
+		*err = -ECONNRESET;
+		goto out;
+	}
 
 	return skb;
+
+out:
+	kfree_skb(skb);
+	return NULL;
 }
 
 int bt_err(__u16 code);
diff --git a/include/net/cls_cgroup.h b/include/net/cls_cgroup.h
index ef6c24a..a4dc5b0 100644
--- a/include/net/cls_cgroup.h
+++ b/include/net/cls_cgroup.h
@@ -51,7 +51,8 @@
 		return 0;
 
 	rcu_read_lock();
-	id = rcu_dereference(net_cls_subsys_id);
+	id = rcu_dereference_index_check(net_cls_subsys_id,
+					 rcu_read_lock_held());
 	if (id >= 0)
 		classid = container_of(task_subsys_state(p, id),
 				       struct cgroup_cls_state, css)->classid;
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index e624dae..caf17db 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -75,7 +75,7 @@
 /* nf_conn feature for connections that have a helper */
 struct nf_conn_help {
 	/* Helper. if any */
-	struct nf_conntrack_helper *helper;
+	struct nf_conntrack_helper __rcu *helper;
 
 	union nf_conntrack_help help;
 
diff --git a/include/pcmcia/cs.h b/include/pcmcia/cs.h
deleted file mode 100644
index 68d8bde..0000000
--- a/include/pcmcia/cs.h
+++ /dev/null
@@ -1,95 +0,0 @@
-/*
- * cs.h
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * The initial developer of the original code is David A. Hinds
- * <dahinds@users.sourceforge.net>.  Portions created by David A. Hinds
- * are Copyright (C) 1999 David A. Hinds.  All Rights Reserved.
- *
- * (C) 1999             David A. Hinds
- */
-
-#ifndef _LINUX_CS_H
-#define _LINUX_CS_H
-
-#ifdef __KERNEL__
-#include <linux/interrupt.h>
-#endif
-
-/* ModifyConfiguration */
-typedef struct modconf_t {
-    u_int	Attributes;
-    u_int	Vcc, Vpp1, Vpp2;
-} modconf_t;
-
-/* Attributes for ModifyConfiguration */
-#define CONF_IRQ_CHANGE_VALID	0x0100
-#define CONF_VCC_CHANGE_VALID	0x0200
-#define CONF_VPP1_CHANGE_VALID	0x0400
-#define CONF_VPP2_CHANGE_VALID	0x0800
-#define CONF_IO_CHANGE_WIDTH	0x1000
-
-/* For RequestConfiguration */
-typedef struct config_req_t {
-    u_int	Attributes;
-    u_int	Vpp; /* both Vpp1 and Vpp2 */
-    u_int	IntType;
-    u_int	ConfigBase;
-    u_char	Status, Pin, Copy, ExtStatus;
-    u_char	ConfigIndex;
-    u_int	Present;
-} config_req_t;
-
-/* Attributes for RequestConfiguration */
-#define CONF_ENABLE_IRQ		0x01
-#define CONF_ENABLE_DMA		0x02
-#define CONF_ENABLE_SPKR	0x04
-#define CONF_ENABLE_PULSE_IRQ	0x08
-#define CONF_VALID_CLIENT	0x100
-
-/* IntType field */
-#define INT_MEMORY		0x01
-#define INT_MEMORY_AND_IO	0x02
-#define INT_CARDBUS		0x04
-#define INT_ZOOMED_VIDEO	0x08
-
-/* Configuration registers present */
-#define PRESENT_OPTION		0x001
-#define PRESENT_STATUS		0x002
-#define PRESENT_PIN_REPLACE	0x004
-#define PRESENT_COPY		0x008
-#define PRESENT_EXT_STATUS	0x010
-#define PRESENT_IOBASE_0	0x020
-#define PRESENT_IOBASE_1	0x040
-#define PRESENT_IOBASE_2	0x080
-#define PRESENT_IOBASE_3	0x100
-#define PRESENT_IOSIZE		0x200
-
-/* For RequestWindow */
-typedef struct win_req_t {
-    u_int	Attributes;
-    u_long	Base;
-    u_int	Size;
-    u_int	AccessSpeed;
-} win_req_t;
-
-/* Attributes for RequestWindow */
-#define WIN_MEMORY_TYPE_CM	0x00 /* default */
-#define WIN_MEMORY_TYPE_AM	0x20 /* MAP_ATTRIB */
-#define WIN_DATA_WIDTH_8	0x00 /* default */
-#define WIN_DATA_WIDTH_16	0x02 /* MAP_16BIT */
-#define WIN_ENABLE		0x01 /* MAP_ACTIVE */
-#define WIN_USE_WAIT		0x40 /* MAP_USE_WAIT */
-
-#define WIN_FLAGS_MAP		0x63 /* MAP_ATTRIB | MAP_16BIT | MAP_ACTIVE |
-					MAP_USE_WAIT */
-#define WIN_FLAGS_REQ		0x1c /* mapping to socket->win[i]:
-					0x04 -> 0
-					0x08 -> 1
-					0x0c -> 2
-					0x10 -> 3 */
-
-#endif /* _LINUX_CS_H */
diff --git a/include/pcmcia/ds.h b/include/pcmcia/ds.h
index 70c58ed..d830c87 100644
--- a/include/pcmcia/ds.h
+++ b/include/pcmcia/ds.h
@@ -24,9 +24,11 @@
 
 #ifdef __KERNEL__
 #include <linux/device.h>
+#include <linux/interrupt.h>
 #include <pcmcia/ss.h>
 #include <asm/atomic.h>
 
+
 /*
  * PCMCIA device drivers (16-bit cards only; 32-bit cards require CardBus
  * a.k.a. PCI drivers
@@ -36,8 +38,6 @@
 struct config_t;
 struct net_device;
 
-typedef struct resource *window_handle_t;
-
 /* dynamic device IDs for PCMCIA device drivers. See
  * Documentation/pcmcia/driver.txt for details.
 */
@@ -47,6 +47,8 @@
 };
 
 struct pcmcia_driver {
+	const char		*name;
+
 	int (*probe)		(struct pcmcia_device *dev);
 	void (*remove)		(struct pcmcia_device *dev);
 
@@ -90,15 +92,17 @@
 
 	struct list_head	socket_device_list;
 
-	/* deprecated, will be cleaned up soon */
-	config_req_t		conf;
-	window_handle_t		win;
-
 	/* device setup */
 	unsigned int		irq;
 	struct resource		*resource[PCMCIA_NUM_RESOURCES];
+	resource_size_t		card_addr;	/* for the 1st IOMEM resource */
+	unsigned int		vpp;
 
-	unsigned int		io_lines; /* number of I/O lines */
+	unsigned int		config_flags;	/* CONF_ENABLE_ flags below */
+	unsigned int		config_base;
+	unsigned int		config_index;
+	unsigned int		config_regs;	/* PRESENT_ flags below */
+	unsigned int		io_lines;	/* number of I/O lines */
 
 	/* Is the device suspended? */
 	u16			suspended:1;
@@ -174,9 +178,6 @@
 /* loop CIS entries for valid configuration */
 int pcmcia_loop_config(struct pcmcia_device *p_dev,
 		       int	(*conf_check)	(struct pcmcia_device *p_dev,
-						 cistpl_cftable_entry_t *cf,
-						 cistpl_cftable_entry_t *dflt,
-						 unsigned int vcc,
 						 void *priv_data),
 		       void *priv_data);
 
@@ -206,16 +207,17 @@
 int __must_check pcmcia_request_irq(struct pcmcia_device *p_dev,
 				irq_handler_t handler);
 
-int pcmcia_request_configuration(struct pcmcia_device *p_dev,
-				 config_req_t *req);
+int pcmcia_enable_device(struct pcmcia_device *p_dev);
 
-int pcmcia_request_window(struct pcmcia_device *p_dev, win_req_t *req,
-			  window_handle_t *wh);
-int pcmcia_release_window(struct pcmcia_device *p_dev, window_handle_t win);
-int pcmcia_map_mem_page(struct pcmcia_device *p_dev, window_handle_t win,
+int pcmcia_request_window(struct pcmcia_device *p_dev, struct resource *res,
+			unsigned int speed);
+int pcmcia_release_window(struct pcmcia_device *p_dev, struct resource *res);
+int pcmcia_map_mem_page(struct pcmcia_device *p_dev, struct resource *res,
 			unsigned int offset);
 
-int pcmcia_modify_configuration(struct pcmcia_device *p_dev, modconf_t *mod);
+int pcmcia_fixup_vpp(struct pcmcia_device *p_dev, unsigned char new_vpp);
+int pcmcia_fixup_iowidth(struct pcmcia_device *p_dev);
+
 void pcmcia_disable_device(struct pcmcia_device *p_dev);
 
 /* IO ports */
@@ -224,15 +226,46 @@
 #define IO_DATA_PATH_WIDTH_16	0x08
 #define IO_DATA_PATH_WIDTH_AUTO	0x10
 
-/* convert flag found in cfgtable to data path width parameter */
-static inline int pcmcia_io_cfg_data_width(unsigned int flags)
-{
-	if (!(flags & CISTPL_IO_8BIT))
-		return IO_DATA_PATH_WIDTH_16;
-	if (!(flags & CISTPL_IO_16BIT))
-		return IO_DATA_PATH_WIDTH_8;
-	return IO_DATA_PATH_WIDTH_AUTO;
-}
+/* IO memory */
+#define WIN_MEMORY_TYPE_CM	0x00 /* default */
+#define WIN_MEMORY_TYPE_AM	0x20 /* MAP_ATTRIB */
+#define WIN_DATA_WIDTH_8	0x00 /* default */
+#define WIN_DATA_WIDTH_16	0x02 /* MAP_16BIT */
+#define WIN_ENABLE		0x01 /* MAP_ACTIVE */
+#define WIN_USE_WAIT		0x40 /* MAP_USE_WAIT */
+
+#define WIN_FLAGS_MAP		0x63 /* MAP_ATTRIB | MAP_16BIT | MAP_ACTIVE |
+					MAP_USE_WAIT */
+#define WIN_FLAGS_REQ		0x1c /* mapping to socket->win[i]:
+					0x04 -> 0
+					0x08 -> 1
+					0x0c -> 2
+					0x10 -> 3 */
+
+/* config_reg{ister}s present for this PCMCIA device */
+#define PRESENT_OPTION		0x001
+#define PRESENT_STATUS		0x002
+#define PRESENT_PIN_REPLACE	0x004
+#define PRESENT_COPY		0x008
+#define PRESENT_EXT_STATUS	0x010
+#define PRESENT_IOBASE_0	0x020
+#define PRESENT_IOBASE_1	0x040
+#define PRESENT_IOBASE_2	0x080
+#define PRESENT_IOBASE_3	0x100
+#define PRESENT_IOSIZE		0x200
+
+/* flags to be passed to pcmcia_enable_device() */
+#define CONF_ENABLE_IRQ         0x0001
+#define CONF_ENABLE_SPKR        0x0002
+#define CONF_ENABLE_PULSE_IRQ   0x0004
+#define CONF_ENABLE_ESR         0x0008
+
+/* flags used by pcmcia_loop_config() autoconfiguration */
+#define CONF_AUTO_CHECK_VCC	0x0100 /* check for matching Vcc? */
+#define CONF_AUTO_SET_VPP	0x0200 /* set Vpp? */
+#define CONF_AUTO_AUDIO		0x0400 /* enable audio line? */
+#define CONF_AUTO_SET_IO	0x0800 /* set ->resource[0,1] */
+#define CONF_AUTO_SET_IOMEM	0x1000 /* set ->resource[2] */
 
 #endif /* __KERNEL__ */
 
diff --git a/include/pcmcia/ss.h b/include/pcmcia/ss.h
index 626b63c..731cde0 100644
--- a/include/pcmcia/ss.h
+++ b/include/pcmcia/ss.h
@@ -19,7 +19,6 @@
 #include <linux/sched.h>	/* task_struct, completion */
 #include <linux/mutex.h>
 
-#include <pcmcia/cs.h>
 #ifdef CONFIG_CARDBUS
 #include <linux/pci.h>
 #endif
diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h
index 0e4cfb6..6fa7cba 100644
--- a/include/trace/events/irq.h
+++ b/include/trace/events/irq.h
@@ -5,7 +5,9 @@
 #define _TRACE_IRQ_H
 
 #include <linux/tracepoint.h>
-#include <linux/interrupt.h>
+
+struct irqaction;
+struct softirq_action;
 
 #define softirq_name(sirq) { sirq##_SOFTIRQ, #sirq }
 #define show_softirq_name(val)				\
@@ -93,7 +95,10 @@
 	),
 
 	TP_fast_assign(
-		__entry->vec = (int)(h - vec);
+		if (vec)
+			__entry->vec = (int)(h - vec);
+		else
+			__entry->vec = (int)(long)h;
 	),
 
 	TP_printk("vec=%d [action=%s]", __entry->vec,
@@ -136,6 +141,23 @@
 	TP_ARGS(h, vec)
 );
 
+/**
+ * softirq_raise - called immediately when a softirq is raised
+ * @h: pointer to struct softirq_action
+ * @vec: pointer to first struct softirq_action in softirq_vec array
+ *
+ * The @h parameter contains a pointer to the softirq vector number which is
+ * raised. @vec is NULL and it means @h includes vector number not
+ * softirq_action. When used in combination with the softirq_entry tracepoint
+ * we can determine the softirq raise latency.
+ */
+DEFINE_EVENT(softirq, softirq_raise,
+
+	TP_PROTO(struct softirq_action *h, struct softirq_action *vec),
+
+	TP_ARGS(h, vec)
+);
+
 #endif /*  _TRACE_IRQ_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h
index 188deca..8fe1e93 100644
--- a/include/trace/events/napi.h
+++ b/include/trace/events/napi.h
@@ -6,10 +6,31 @@
 
 #include <linux/netdevice.h>
 #include <linux/tracepoint.h>
+#include <linux/ftrace.h>
 
-DECLARE_TRACE(napi_poll,
+#define NO_DEV "(no_device)"
+
+TRACE_EVENT(napi_poll,
+
 	TP_PROTO(struct napi_struct *napi),
-	TP_ARGS(napi));
+
+	TP_ARGS(napi),
+
+	TP_STRUCT__entry(
+		__field(	struct napi_struct *,	napi)
+		__string(	dev_name, napi->dev ? napi->dev->name : NO_DEV)
+	),
+
+	TP_fast_assign(
+		__entry->napi = napi;
+		__assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV);
+	),
+
+	TP_printk("napi poll on napi struct %p for device %s",
+		__entry->napi, __get_str(dev_name))
+);
+
+#undef NO_DEV
 
 #endif /* _TRACE_NAPI_H_ */
 
diff --git a/include/trace/events/net.h b/include/trace/events/net.h
new file mode 100644
index 0000000..5f247f5
--- /dev/null
+++ b/include/trace/events/net.h
@@ -0,0 +1,82 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM net
+
+#if !defined(_TRACE_NET_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NET_H
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/ip.h>
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(net_dev_xmit,
+
+	TP_PROTO(struct sk_buff *skb,
+		 int rc),
+
+	TP_ARGS(skb, rc),
+
+	TP_STRUCT__entry(
+		__field(	void *,		skbaddr		)
+		__field(	unsigned int,	len		)
+		__field(	int,		rc		)
+		__string(	name,		skb->dev->name	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+		__entry->len = skb->len;
+		__entry->rc = rc;
+		__assign_str(name, skb->dev->name);
+	),
+
+	TP_printk("dev=%s skbaddr=%p len=%u rc=%d",
+		__get_str(name), __entry->skbaddr, __entry->len, __entry->rc)
+);
+
+DECLARE_EVENT_CLASS(net_dev_template,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(	void *,		skbaddr		)
+		__field(	unsigned int,	len		)
+		__string(	name,		skb->dev->name	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+		__entry->len = skb->len;
+		__assign_str(name, skb->dev->name);
+	),
+
+	TP_printk("dev=%s skbaddr=%p len=%u",
+		__get_str(name), __entry->skbaddr, __entry->len)
+)
+
+DEFINE_EVENT(net_dev_template, net_dev_queue,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+
+DEFINE_EVENT(net_dev_template, netif_receive_skb,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+
+DEFINE_EVENT(net_dev_template, netif_rx,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+#endif /* _TRACE_NET_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index 35a2a6e..286784d 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -10,12 +10,17 @@
 #ifndef _TRACE_POWER_ENUM_
 #define _TRACE_POWER_ENUM_
 enum {
-	POWER_NONE = 0,
-	POWER_CSTATE = 1,
-	POWER_PSTATE = 2,
+	POWER_NONE	= 0,
+	POWER_CSTATE	= 1,	/* C-State */
+	POWER_PSTATE	= 2,	/* Fequency change or DVFS */
+	POWER_SSTATE	= 3,	/* Suspend */
 };
 #endif
 
+/*
+ * The power events are used for cpuidle & suspend (power_start, power_end)
+ *  and for cpufreq (power_frequency)
+ */
 DECLARE_EVENT_CLASS(power,
 
 	TP_PROTO(unsigned int type, unsigned int state, unsigned int cpu_id),
@@ -70,6 +75,85 @@
 
 );
 
+/*
+ * The clock events are used for clock enable/disable and for
+ *  clock rate change
+ */
+DECLARE_EVENT_CLASS(clock,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id),
+
+	TP_STRUCT__entry(
+		__string(       name,           name            )
+		__field(        u64,            state           )
+		__field(        u64,            cpu_id          )
+	),
+
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->state = state;
+		__entry->cpu_id = cpu_id;
+	),
+
+	TP_printk("%s state=%lu cpu_id=%lu", __get_str(name),
+		(unsigned long)__entry->state, (unsigned long)__entry->cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_enable,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_disable,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_set_rate,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+/*
+ * The power domain events are used for power domains transitions
+ */
+DECLARE_EVENT_CLASS(power_domain,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id),
+
+	TP_STRUCT__entry(
+		__string(       name,           name            )
+		__field(        u64,            state           )
+		__field(        u64,            cpu_id          )
+	),
+
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->state = state;
+		__entry->cpu_id = cpu_id;
+),
+
+	TP_printk("%s state=%lu cpu_id=%lu", __get_str(name),
+		(unsigned long)__entry->state, (unsigned long)__entry->cpu_id)
+);
+
+DEFINE_EVENT(power_domain, power_domain_target,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
 #endif /* _TRACE_POWER_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h
index 9208c92..f633478 100644
--- a/include/trace/events/sched.h
+++ b/include/trace/events/sched.h
@@ -362,6 +362,35 @@
 			(unsigned long long)__entry->vruntime)
 );
 
+/*
+ * Tracepoint for showing priority inheritance modifying a tasks
+ * priority.
+ */
+TRACE_EVENT(sched_pi_setprio,
+
+	TP_PROTO(struct task_struct *tsk, int newprio),
+
+	TP_ARGS(tsk, newprio),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN	)
+		__field( pid_t,	pid			)
+		__field( int,	oldprio			)
+		__field( int,	newprio			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid		= tsk->pid;
+		__entry->oldprio	= tsk->prio;
+		__entry->newprio	= newprio;
+	),
+
+	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
+			__entry->comm, __entry->pid,
+			__entry->oldprio, __entry->newprio)
+);
+
 #endif /* _TRACE_SCHED_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/skb.h b/include/trace/events/skb.h
index 4b2be6d..75ce9d5 100644
--- a/include/trace/events/skb.h
+++ b/include/trace/events/skb.h
@@ -35,6 +35,23 @@
 		__entry->skbaddr, __entry->protocol, __entry->location)
 );
 
+TRACE_EVENT(consume_skb,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(	void *,	skbaddr	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+	),
+
+	TP_printk("skbaddr=%p", __entry->skbaddr)
+);
+
 TRACE_EVENT(skb_copy_datagram_iovec,
 
 	TP_PROTO(const struct sk_buff *skb, int len),
diff --git a/init/Kconfig b/init/Kconfig
index 2de5b1c..36890f0 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -21,6 +21,13 @@
 	depends on !UML
 	default y
 
+config HAVE_IRQ_WORK
+	bool
+
+config IRQ_WORK
+	bool
+	depends on HAVE_IRQ_WORK
+
 menu "General setup"
 
 config EXPERIMENTAL
@@ -332,6 +339,8 @@
 	depends on AUDITSYSCALL
 	select FSNOTIFY
 
+source "kernel/irq/Kconfig"
+
 menu "RCU Subsystem"
 
 choice
@@ -340,6 +349,7 @@
 
 config TREE_RCU
 	bool "Tree-based hierarchical RCU"
+	depends on !PREEMPT && SMP
 	help
 	  This option selects the RCU implementation that is
 	  designed for very large SMP system with hundreds or
@@ -347,7 +357,7 @@
 	  smaller systems.
 
 config TREE_PREEMPT_RCU
-	bool "Preemptable tree-based hierarchical RCU"
+	bool "Preemptible tree-based hierarchical RCU"
 	depends on PREEMPT
 	help
 	  This option selects the RCU implementation that is
@@ -365,8 +375,22 @@
 	  is not required.  This option greatly reduces the
 	  memory footprint of RCU.
 
+config TINY_PREEMPT_RCU
+	bool "Preemptible UP-only small-memory-footprint RCU"
+	depends on !SMP && PREEMPT
+	help
+	  This option selects the RCU implementation that is designed
+	  for real-time UP systems.  This option greatly reduces the
+	  memory footprint of RCU.
+
 endchoice
 
+config PREEMPT_RCU
+	def_bool ( TREE_PREEMPT_RCU || TINY_PREEMPT_RCU )
+	help
+	  This option enables preemptible-RCU code that is common between
+	  the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations.
+
 config RCU_TRACE
 	bool "Enable tracing for RCU"
 	depends on TREE_RCU || TREE_PREEMPT_RCU
@@ -387,9 +411,12 @@
 	help
 	  This option controls the fanout of hierarchical implementations
 	  of RCU, allowing RCU to work efficiently on machines with
-	  large numbers of CPUs.  This value must be at least the cube
-	  root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
-	  systems and up to 262,144 for 64-bit systems.
+	  large numbers of CPUs.  This value must be at least the fourth
+	  root of NR_CPUS, which allows NR_CPUS to be insanely large.
+	  The default value of RCU_FANOUT should be used for production
+	  systems, but if you are stress-testing the RCU implementation
+	  itself, small RCU_FANOUT values allow you to test large-system
+	  code paths on small(er) systems.
 
 	  Select a specific number if testing RCU itself.
 	  Take the default if unsure.
@@ -987,6 +1014,7 @@
 	default y if (PROFILING || PERF_COUNTERS)
 	depends on HAVE_PERF_EVENTS
 	select ANON_INODES
+	select IRQ_WORK
 	help
 	  Enable kernel support for various performance events provided
 	  by software and hardware.
diff --git a/init/main.c b/init/main.c
index 94ab488..9684c96 100644
--- a/init/main.c
+++ b/init/main.c
@@ -556,7 +556,6 @@
 
 	local_irq_disable();
 	early_boot_irqs_off();
-	early_init_irq_lock_class();
 
 /*
  * Interrupts are still disabled. Do necessary setups, then
diff --git a/kernel/Makefile b/kernel/Makefile
index 0b72d1a..e2c9d52 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -10,7 +10,7 @@
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
 	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
 	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \
-	    async.o range.o
+	    async.o range.o jump_label.o
 obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o
 obj-y += groups.o
 
@@ -23,6 +23,7 @@
 CFLAGS_REMOVE_cgroup-debug.o = -pg
 CFLAGS_REMOVE_sched_clock.o = -pg
 CFLAGS_REMOVE_perf_event.o = -pg
+CFLAGS_REMOVE_irq_work.o = -pg
 endif
 
 obj-$(CONFIG_FREEZER) += freezer.o
@@ -86,6 +87,7 @@
 obj-$(CONFIG_TREE_PREEMPT_RCU) += rcutree.o
 obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o
 obj-$(CONFIG_TINY_RCU) += rcutiny.o
+obj-$(CONFIG_TINY_PREEMPT_RCU) += rcutiny.o
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
@@ -100,6 +102,7 @@
 obj-$(CONFIG_X86_DS) += trace/
 obj-$(CONFIG_RING_BUFFER) += trace/
 obj-$(CONFIG_SMP) += sched_cpupri.o
+obj-$(CONFIG_IRQ_WORK) += irq_work.o
 obj-$(CONFIG_PERF_EVENTS) += perf_event.o
 obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
 obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index c9483d8..291ba3d 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -138,7 +138,7 @@
 	 * is called after synchronize_rcu(). But for safe use, css_is_removed()
 	 * css_tryget() should be used for avoiding race.
 	 */
-	struct cgroup_subsys_state *css;
+	struct cgroup_subsys_state __rcu *css;
 	/*
 	 * ID of this css.
 	 */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index b23c097..51b143e 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1397,7 +1397,7 @@
 	if (tsk->flags & PF_THREAD_BOUND)
 		return -EINVAL;
 
-	ret = security_task_setscheduler(tsk, 0, NULL);
+	ret = security_task_setscheduler(tsk);
 	if (ret)
 		return ret;
 	if (threadgroup) {
@@ -1405,7 +1405,7 @@
 
 		rcu_read_lock();
 		list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) {
-			ret = security_task_setscheduler(c, 0, NULL);
+			ret = security_task_setscheduler(c);
 			if (ret) {
 				rcu_read_unlock();
 				return ret;
diff --git a/kernel/exit.c b/kernel/exit.c
index 0312022..e2bdf37 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -149,9 +149,7 @@
 {
 	struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);
 
-#ifdef CONFIG_PERF_EVENTS
-	WARN_ON_ONCE(tsk->perf_event_ctxp);
-#endif
+	perf_event_delayed_put(tsk);
 	trace_sched_process_free(tsk);
 	put_task_struct(tsk);
 }
diff --git a/kernel/futex.c b/kernel/futex.c
index 6a3a5fa..a118bf1 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -91,6 +91,7 @@
 
 /**
  * struct futex_q - The hashed futex queue entry, one per waiting task
+ * @list:		priority-sorted list of tasks waiting on this futex
  * @task:		the task waiting on the futex
  * @lock_ptr:		the hash bucket lock
  * @key:		the key the futex is hashed on
@@ -104,7 +105,7 @@
  *
  * A futex_q has a woken state, just like tasks have TASK_RUNNING.
  * It is considered woken when plist_node_empty(&q->list) || q->lock_ptr == 0.
- * The order of wakup is always to make the first condition true, then
+ * The order of wakeup is always to make the first condition true, then
  * the second.
  *
  * PI futexes are typically woken before they are removed from the hash list via
@@ -295,7 +296,7 @@
  * Slow path to fixup the fault we just took in the atomic write
  * access to @uaddr.
  *
- * We have no generic implementation of a non destructive write to the
+ * We have no generic implementation of a non-destructive write to the
  * user address. We know that we faulted in the atomic pagefault
  * disabled section so we can as well avoid the #PF overhead by
  * calling get_user_pages() right away.
@@ -515,7 +516,7 @@
 			 */
 			pi_state = this->pi_state;
 			/*
-			 * Userspace might have messed up non PI and PI futexes
+			 * Userspace might have messed up non-PI and PI futexes
 			 */
 			if (unlikely(!pi_state))
 				return -EINVAL;
@@ -736,8 +737,8 @@
 
 	/*
 	 * We set q->lock_ptr = NULL _before_ we wake up the task. If
-	 * a non futex wake up happens on another CPU then the task
-	 * might exit and p would dereference a non existing task
+	 * a non-futex wake up happens on another CPU then the task
+	 * might exit and p would dereference a non-existing task
 	 * struct. Prevent this by holding a reference on p across the
 	 * wake up.
 	 */
@@ -1131,11 +1132,13 @@
 
 /**
  * futex_requeue() - Requeue waiters from uaddr1 to uaddr2
- * uaddr1:	source futex user address
- * uaddr2:	target futex user address
- * nr_wake:	number of waiters to wake (must be 1 for requeue_pi)
- * nr_requeue:	number of waiters to requeue (0-INT_MAX)
- * requeue_pi:	if we are attempting to requeue from a non-pi futex to a
+ * @uaddr1:	source futex user address
+ * @fshared:	0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED
+ * @uaddr2:	target futex user address
+ * @nr_wake:	number of waiters to wake (must be 1 for requeue_pi)
+ * @nr_requeue:	number of waiters to requeue (0-INT_MAX)
+ * @cmpval:	@uaddr1 expected value (or %NULL)
+ * @requeue_pi:	if we are attempting to requeue from a non-pi futex to a
  * 		pi futex (pi to pi requeue is not supported)
  *
  * Requeue waiters on uaddr1 to uaddr2. In the requeue_pi case, try to acquire
@@ -1360,10 +1363,10 @@
 
 /* The key must be already stored in q->key. */
 static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)
+	__acquires(&hb->lock)
 {
 	struct futex_hash_bucket *hb;
 
-	get_futex_key_refs(&q->key);
 	hb = hash_futex(&q->key);
 	q->lock_ptr = &hb->lock;
 
@@ -1373,9 +1376,9 @@
 
 static inline void
 queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb)
+	__releases(&hb->lock)
 {
 	spin_unlock(&hb->lock);
-	drop_futex_key_refs(&q->key);
 }
 
 /**
@@ -1391,6 +1394,7 @@
  * an example).
  */
 static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb)
+	__releases(&hb->lock)
 {
 	int prio;
 
@@ -1471,6 +1475,7 @@
  * and dropped here.
  */
 static void unqueue_me_pi(struct futex_q *q)
+	__releases(q->lock_ptr)
 {
 	WARN_ON(plist_node_empty(&q->list));
 	plist_del(&q->list, &q->list.plist);
@@ -1480,8 +1485,6 @@
 	q->pi_state = NULL;
 
 	spin_unlock(q->lock_ptr);
-
-	drop_futex_key_refs(&q->key);
 }
 
 /*
@@ -1812,7 +1815,10 @@
 	}
 
 retry:
-	/* Prepare to wait on uaddr. */
+	/*
+	 * Prepare to wait on uaddr. On success, holds hb lock and increments
+	 * q.key refs.
+	 */
 	ret = futex_wait_setup(uaddr, val, fshared, &q, &hb);
 	if (ret)
 		goto out;
@@ -1822,28 +1828,27 @@
 
 	/* If we were woken (and unqueued), we succeeded, whatever. */
 	ret = 0;
+	/* unqueue_me() drops q.key ref */
 	if (!unqueue_me(&q))
-		goto out_put_key;
+		goto out;
 	ret = -ETIMEDOUT;
 	if (to && !to->task)
-		goto out_put_key;
+		goto out;
 
 	/*
 	 * We expect signal_pending(current), but we might be the
 	 * victim of a spurious wakeup as well.
 	 */
-	if (!signal_pending(current)) {
-		put_futex_key(fshared, &q.key);
+	if (!signal_pending(current))
 		goto retry;
-	}
 
 	ret = -ERESTARTSYS;
 	if (!abs_time)
-		goto out_put_key;
+		goto out;
 
 	restart = &current_thread_info()->restart_block;
 	restart->fn = futex_wait_restart;
-	restart->futex.uaddr = (u32 *)uaddr;
+	restart->futex.uaddr = uaddr;
 	restart->futex.val = val;
 	restart->futex.time = abs_time->tv64;
 	restart->futex.bitset = bitset;
@@ -1856,8 +1861,6 @@
 
 	ret = -ERESTART_RESTARTBLOCK;
 
-out_put_key:
-	put_futex_key(fshared, &q.key);
 out:
 	if (to) {
 		hrtimer_cancel(&to->timer);
@@ -1869,7 +1872,7 @@
 
 static long futex_wait_restart(struct restart_block *restart)
 {
-	u32 __user *uaddr = (u32 __user *)restart->futex.uaddr;
+	u32 __user *uaddr = restart->futex.uaddr;
 	int fshared = 0;
 	ktime_t t, *tp = NULL;
 
@@ -2236,7 +2239,10 @@
 	q.rt_waiter = &rt_waiter;
 	q.requeue_pi_key = &key2;
 
-	/* Prepare to wait on uaddr. */
+	/*
+	 * Prepare to wait on uaddr. On success, increments q.key (key1) ref
+	 * count.
+	 */
 	ret = futex_wait_setup(uaddr, val, fshared, &q, &hb);
 	if (ret)
 		goto out_key2;
@@ -2254,7 +2260,9 @@
 	 * In order for us to be here, we know our q.key == key2, and since
 	 * we took the hb->lock above, we also know that futex_requeue() has
 	 * completed and we no longer have to concern ourselves with a wakeup
-	 * race with the atomic proxy lock acquition by the requeue code.
+	 * race with the atomic proxy lock acquisition by the requeue code. The
+	 * futex_requeue dropped our key1 reference and incremented our key2
+	 * reference count.
 	 */
 
 	/* Check if the requeue code acquired the second futex for us. */
@@ -2458,7 +2466,7 @@
  */
 static inline int fetch_robust_entry(struct robust_list __user **entry,
 				     struct robust_list __user * __user *head,
-				     int *pi)
+				     unsigned int *pi)
 {
 	unsigned long uentry;
 
@@ -2647,7 +2655,7 @@
 	 * of the complex code paths. Also we want to prevent
 	 * registration of robust lists in that case. NULL is
 	 * guaranteed to fault and we get -EFAULT on functional
-	 * implementation, the non functional ones will return
+	 * implementation, the non-functional ones will return
 	 * -ENOSYS.
 	 */
 	curval = cmpxchg_futex_value_locked(NULL, 0, 0);
diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c
index d49afb2..06da4df 100644
--- a/kernel/futex_compat.c
+++ b/kernel/futex_compat.c
@@ -19,7 +19,7 @@
  */
 static inline int
 fetch_robust_entry(compat_uptr_t *uentry, struct robust_list __user **entry,
-		   compat_uptr_t __user *head, int *pi)
+		   compat_uptr_t __user *head, unsigned int *pi)
 {
 	if (get_user(*uentry, head))
 		return -EFAULT;
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 1decafb..72206cf 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -931,6 +931,7 @@
 remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base)
 {
 	if (hrtimer_is_queued(timer)) {
+		unsigned long state;
 		int reprogram;
 
 		/*
@@ -944,8 +945,13 @@
 		debug_deactivate(timer);
 		timer_stats_hrtimer_clear_start_info(timer);
 		reprogram = base->cpu_base == &__get_cpu_var(hrtimer_bases);
-		__remove_hrtimer(timer, base, HRTIMER_STATE_INACTIVE,
-				 reprogram);
+		/*
+		 * We must preserve the CALLBACK state flag here,
+		 * otherwise we could move the timer base in
+		 * switch_hrtimer_base.
+		 */
+		state = timer->state & HRTIMER_STATE_CALLBACK;
+		__remove_hrtimer(timer, base, state, reprogram);
 		return 1;
 	}
 	return 0;
@@ -1231,6 +1237,9 @@
 		BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
 		enqueue_hrtimer(timer, base);
 	}
+
+	WARN_ON_ONCE(!(timer->state & HRTIMER_STATE_CALLBACK));
+
 	timer->state &= ~HRTIMER_STATE_CALLBACK;
 }
 
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
index 0c642d5..53ead17 100644
--- a/kernel/hung_task.c
+++ b/kernel/hung_task.c
@@ -98,7 +98,7 @@
 	printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
 			" disables this message.\n");
 	sched_show_task(t);
-	__debug_show_held_locks(t);
+	debug_show_held_locks(t);
 
 	touch_nmi_watchdog();
 
@@ -111,7 +111,7 @@
  * periodically exit the critical section and enter a new one.
  *
  * For preemptible RCU it is sufficient to call rcu_read_unlock in order
- * exit the grace period. For classic RCU, a reschedule is required.
+ * to exit the grace period. For classic RCU, a reschedule is required.
  */
 static void rcu_lock_break(struct task_struct *g, struct task_struct *t)
 {
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index c7c2aed..2c9120f 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -113,12 +113,12 @@
  */
 static int task_bp_pinned(struct perf_event *bp, enum bp_type_idx type)
 {
-	struct perf_event_context *ctx = bp->ctx;
+	struct task_struct *tsk = bp->hw.bp_target;
 	struct perf_event *iter;
 	int count = 0;
 
 	list_for_each_entry(iter, &bp_task_head, hw.bp_list) {
-		if (iter->ctx == ctx && find_slot_idx(iter) == type)
+		if (iter->hw.bp_target == tsk && find_slot_idx(iter) == type)
 			count += hw_breakpoint_weight(iter);
 	}
 
@@ -134,7 +134,7 @@
 		    enum bp_type_idx type)
 {
 	int cpu = bp->cpu;
-	struct task_struct *tsk = bp->ctx->task;
+	struct task_struct *tsk = bp->hw.bp_target;
 
 	if (cpu >= 0) {
 		slots->pinned = per_cpu(nr_cpu_bp_pinned[type], cpu);
@@ -213,7 +213,7 @@
 	       int weight)
 {
 	int cpu = bp->cpu;
-	struct task_struct *tsk = bp->ctx->task;
+	struct task_struct *tsk = bp->hw.bp_target;
 
 	/* Pinned counter cpu profiling */
 	if (!tsk) {
@@ -433,8 +433,7 @@
 			    perf_overflow_handler_t triggered,
 			    struct task_struct *tsk)
 {
-	return perf_event_create_kernel_counter(attr, -1, task_pid_vnr(tsk),
-						triggered);
+	return perf_event_create_kernel_counter(attr, -1, tsk, triggered);
 }
 EXPORT_SYMBOL_GPL(register_user_hw_breakpoint);
 
@@ -516,7 +515,7 @@
 	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		pevent = per_cpu_ptr(cpu_events, cpu);
-		bp = perf_event_create_kernel_counter(attr, cpu, -1, triggered);
+		bp = perf_event_create_kernel_counter(attr, cpu, NULL, triggered);
 
 		*pevent = bp;
 
@@ -566,6 +565,61 @@
 	.priority = 0x7fffffff
 };
 
+static void bp_perf_event_destroy(struct perf_event *event)
+{
+	release_bp_slot(event);
+}
+
+static int hw_breakpoint_event_init(struct perf_event *bp)
+{
+	int err;
+
+	if (bp->attr.type != PERF_TYPE_BREAKPOINT)
+		return -ENOENT;
+
+	err = register_perf_hw_breakpoint(bp);
+	if (err)
+		return err;
+
+	bp->destroy = bp_perf_event_destroy;
+
+	return 0;
+}
+
+static int hw_breakpoint_add(struct perf_event *bp, int flags)
+{
+	if (!(flags & PERF_EF_START))
+		bp->hw.state = PERF_HES_STOPPED;
+
+	return arch_install_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_del(struct perf_event *bp, int flags)
+{
+	arch_uninstall_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_start(struct perf_event *bp, int flags)
+{
+	bp->hw.state = 0;
+}
+
+static void hw_breakpoint_stop(struct perf_event *bp, int flags)
+{
+	bp->hw.state = PERF_HES_STOPPED;
+}
+
+static struct pmu perf_breakpoint = {
+	.task_ctx_nr	= perf_sw_context, /* could eventually get its own */
+
+	.event_init	= hw_breakpoint_event_init,
+	.add		= hw_breakpoint_add,
+	.del		= hw_breakpoint_del,
+	.start		= hw_breakpoint_start,
+	.stop		= hw_breakpoint_stop,
+	.read		= hw_breakpoint_pmu_read,
+};
+
 static int __init init_hw_breakpoint(void)
 {
 	unsigned int **task_bp_pinned;
@@ -587,6 +641,8 @@
 
 	constraints_initialized = 1;
 
+	perf_pmu_register(&perf_breakpoint);
+
 	return register_die_notifier(&hw_breakpoint_exceptions_nb);
 
  err_alloc:
@@ -602,8 +658,3 @@
 core_initcall(init_hw_breakpoint);
 
 
-struct pmu perf_ops_bp = {
-	.enable		= arch_install_hw_breakpoint,
-	.disable	= arch_uninstall_hw_breakpoint,
-	.read		= hw_breakpoint_pmu_read,
-};
diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
new file mode 100644
index 0000000..31d766b
--- /dev/null
+++ b/kernel/irq/Kconfig
@@ -0,0 +1,53 @@
+config HAVE_GENERIC_HARDIRQS
+	def_bool n
+
+if HAVE_GENERIC_HARDIRQS
+menu "IRQ subsystem"
+#
+# Interrupt subsystem related configuration options
+#
+config GENERIC_HARDIRQS
+       def_bool y
+
+config GENERIC_HARDIRQS_NO__DO_IRQ
+       def_bool y
+
+# Select this to disable the deprecated stuff
+config GENERIC_HARDIRQS_NO_DEPRECATED
+       def_bool n
+
+# Options selectable by the architecture code
+config HAVE_SPARSE_IRQ
+       def_bool n
+
+config GENERIC_IRQ_PROBE
+	def_bool n
+
+config GENERIC_PENDING_IRQ
+	def_bool n
+
+config AUTO_IRQ_AFFINITY
+       def_bool n
+
+config IRQ_PER_CPU
+       def_bool n
+
+config HARDIRQS_SW_RESEND
+       def_bool n
+
+config SPARSE_IRQ
+	bool "Support sparse irq numbering"
+	depends on HAVE_SPARSE_IRQ
+	---help---
+
+	  Sparse irq numbering is useful for distro kernels that want
+	  to define a high CONFIG_NR_CPUS value but still want to have
+	  low kernel memory footprint on smaller machines.
+
+	  ( Sparse irqs can also be beneficial on NUMA boxes, as they spread
+	    out the interrupt descriptors in a more NUMA-friendly way. )
+
+	  If you don't know what to do here, say N.
+
+endmenu
+endif
diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile
index 7d04780..54329cd 100644
--- a/kernel/irq/Makefile
+++ b/kernel/irq/Makefile
@@ -1,7 +1,6 @@
 
-obj-y := handle.o manage.o spurious.o resend.o chip.o devres.o
+obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o
 obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o
 obj-$(CONFIG_PROC_FS) += proc.o
 obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o
-obj-$(CONFIG_NUMA_IRQ_DESC) += numa_migrate.o
 obj-$(CONFIG_PM_SLEEP) += pm.o
diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c
index 2295a31..505798f 100644
--- a/kernel/irq/autoprobe.c
+++ b/kernel/irq/autoprobe.c
@@ -57,9 +57,10 @@
 			 * Some chips need to know about probing in
 			 * progress:
 			 */
-			if (desc->chip->set_type)
-				desc->chip->set_type(i, IRQ_TYPE_PROBE);
-			desc->chip->startup(i);
+			if (desc->irq_data.chip->irq_set_type)
+				desc->irq_data.chip->irq_set_type(&desc->irq_data,
+							 IRQ_TYPE_PROBE);
+			desc->irq_data.chip->irq_startup(&desc->irq_data);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
@@ -76,7 +77,7 @@
 		raw_spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			desc->status |= IRQ_AUTODETECT | IRQ_WAITING;
-			if (desc->chip->startup(i))
+			if (desc->irq_data.chip->irq_startup(&desc->irq_data))
 				desc->status |= IRQ_PENDING;
 		}
 		raw_spin_unlock_irq(&desc->lock);
@@ -98,7 +99,7 @@
 			/* It triggered already - consider it spurious. */
 			if (!(status & IRQ_WAITING)) {
 				desc->status = status & ~IRQ_AUTODETECT;
-				desc->chip->shutdown(i);
+				desc->irq_data.chip->irq_shutdown(&desc->irq_data);
 			} else
 				if (i < 32)
 					mask |= 1 << i;
@@ -137,7 +138,7 @@
 				mask |= 1 << i;
 
 			desc->status = status & ~IRQ_AUTODETECT;
-			desc->chip->shutdown(i);
+			desc->irq_data.chip->irq_shutdown(&desc->irq_data);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
@@ -181,7 +182,7 @@
 				nr_of_irqs++;
 			}
 			desc->status = status & ~IRQ_AUTODETECT;
-			desc->chip->shutdown(i);
+			desc->irq_data.chip->irq_shutdown(&desc->irq_data);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index b7091d5..baa5c4a 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -18,108 +18,6 @@
 
 #include "internals.h"
 
-static void dynamic_irq_init_x(unsigned int irq, bool keep_chip_data)
-{
-	struct irq_desc *desc;
-	unsigned long flags;
-
-	desc = irq_to_desc(irq);
-	if (!desc) {
-		WARN(1, KERN_ERR "Trying to initialize invalid IRQ%d\n", irq);
-		return;
-	}
-
-	/* Ensure we don't have left over values from a previous use of this irq */
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status = IRQ_DISABLED;
-	desc->chip = &no_irq_chip;
-	desc->handle_irq = handle_bad_irq;
-	desc->depth = 1;
-	desc->msi_desc = NULL;
-	desc->handler_data = NULL;
-	if (!keep_chip_data)
-		desc->chip_data = NULL;
-	desc->action = NULL;
-	desc->irq_count = 0;
-	desc->irqs_unhandled = 0;
-#ifdef CONFIG_SMP
-	cpumask_setall(desc->affinity);
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_clear(desc->pending_mask);
-#endif
-#endif
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
-
-/**
- *	dynamic_irq_init - initialize a dynamically allocated irq
- *	@irq:	irq number to initialize
- */
-void dynamic_irq_init(unsigned int irq)
-{
-	dynamic_irq_init_x(irq, false);
-}
-
-/**
- *	dynamic_irq_init_keep_chip_data - initialize a dynamically allocated irq
- *	@irq:	irq number to initialize
- *
- *	does not set irq_to_desc(irq)->chip_data to NULL
- */
-void dynamic_irq_init_keep_chip_data(unsigned int irq)
-{
-	dynamic_irq_init_x(irq, true);
-}
-
-static void dynamic_irq_cleanup_x(unsigned int irq, bool keep_chip_data)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	unsigned long flags;
-
-	if (!desc) {
-		WARN(1, KERN_ERR "Trying to cleanup invalid IRQ%d\n", irq);
-		return;
-	}
-
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	if (desc->action) {
-		raw_spin_unlock_irqrestore(&desc->lock, flags);
-		WARN(1, KERN_ERR "Destroying IRQ%d without calling free_irq\n",
-			irq);
-		return;
-	}
-	desc->msi_desc = NULL;
-	desc->handler_data = NULL;
-	if (!keep_chip_data)
-		desc->chip_data = NULL;
-	desc->handle_irq = handle_bad_irq;
-	desc->chip = &no_irq_chip;
-	desc->name = NULL;
-	clear_kstat_irqs(desc);
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
-
-/**
- *	dynamic_irq_cleanup - cleanup a dynamically allocated irq
- *	@irq:	irq number to initialize
- */
-void dynamic_irq_cleanup(unsigned int irq)
-{
-	dynamic_irq_cleanup_x(irq, false);
-}
-
-/**
- *	dynamic_irq_cleanup_keep_chip_data - cleanup a dynamically allocated irq
- *	@irq:	irq number to initialize
- *
- *	does not set irq_to_desc(irq)->chip_data to NULL
- */
-void dynamic_irq_cleanup_keep_chip_data(unsigned int irq)
-{
-	dynamic_irq_cleanup_x(irq, true);
-}
-
-
 /**
  *	set_irq_chip - set the irq chip for an irq
  *	@irq:	irq number
@@ -140,7 +38,7 @@
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
 	irq_chip_set_defaults(chip);
-	desc->chip = chip;
+	desc->irq_data.chip = chip;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 
 	return 0;
@@ -193,7 +91,7 @@
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->handler_data = data;
+	desc->irq_data.handler_data = data;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 	return 0;
 }
@@ -218,7 +116,7 @@
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->msi_desc = entry;
+	desc->irq_data.msi_desc = entry;
 	if (entry)
 		entry->irq = irq;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
@@ -243,19 +141,27 @@
 		return -EINVAL;
 	}
 
-	if (!desc->chip) {
+	if (!desc->irq_data.chip) {
 		printk(KERN_ERR "BUG: bad set_irq_chip_data(IRQ#%d)\n", irq);
 		return -EINVAL;
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->chip_data = data;
+	desc->irq_data.chip_data = data;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 
 	return 0;
 }
 EXPORT_SYMBOL(set_irq_chip_data);
 
+struct irq_data *irq_get_irq_data(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	return desc ? &desc->irq_data : NULL;
+}
+EXPORT_SYMBOL_GPL(irq_get_irq_data);
+
 /**
  *	set_irq_nested_thread - Set/Reset the IRQ_NESTED_THREAD flag of an irq
  *
@@ -287,93 +193,216 @@
 /*
  * default enable function
  */
-static void default_enable(unsigned int irq)
+static void default_enable(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_desc *desc = irq_data_to_desc(data);
 
-	desc->chip->unmask(irq);
+	desc->irq_data.chip->irq_unmask(&desc->irq_data);
 	desc->status &= ~IRQ_MASKED;
 }
 
 /*
  * default disable function
  */
-static void default_disable(unsigned int irq)
+static void default_disable(struct irq_data *data)
 {
 }
 
 /*
  * default startup function
  */
-static unsigned int default_startup(unsigned int irq)
+static unsigned int default_startup(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_desc *desc = irq_data_to_desc(data);
 
-	desc->chip->enable(irq);
+	desc->irq_data.chip->irq_enable(data);
 	return 0;
 }
 
 /*
  * default shutdown function
  */
-static void default_shutdown(unsigned int irq)
+static void default_shutdown(struct irq_data *data)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_desc *desc = irq_data_to_desc(data);
 
-	desc->chip->mask(irq);
+	desc->irq_data.chip->irq_mask(&desc->irq_data);
 	desc->status |= IRQ_MASKED;
 }
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+/* Temporary migration helpers */
+static void compat_irq_mask(struct irq_data *data)
+{
+	data->chip->mask(data->irq);
+}
+
+static void compat_irq_unmask(struct irq_data *data)
+{
+	data->chip->unmask(data->irq);
+}
+
+static void compat_irq_ack(struct irq_data *data)
+{
+	data->chip->ack(data->irq);
+}
+
+static void compat_irq_mask_ack(struct irq_data *data)
+{
+	data->chip->mask_ack(data->irq);
+}
+
+static void compat_irq_eoi(struct irq_data *data)
+{
+	data->chip->eoi(data->irq);
+}
+
+static void compat_irq_enable(struct irq_data *data)
+{
+	data->chip->enable(data->irq);
+}
+
+static void compat_irq_disable(struct irq_data *data)
+{
+	data->chip->disable(data->irq);
+}
+
+static void compat_irq_shutdown(struct irq_data *data)
+{
+	data->chip->shutdown(data->irq);
+}
+
+static unsigned int compat_irq_startup(struct irq_data *data)
+{
+	return data->chip->startup(data->irq);
+}
+
+static int compat_irq_set_affinity(struct irq_data *data,
+				   const struct cpumask *dest, bool force)
+{
+	return data->chip->set_affinity(data->irq, dest);
+}
+
+static int compat_irq_set_type(struct irq_data *data, unsigned int type)
+{
+	return data->chip->set_type(data->irq, type);
+}
+
+static int compat_irq_set_wake(struct irq_data *data, unsigned int on)
+{
+	return data->chip->set_wake(data->irq, on);
+}
+
+static int compat_irq_retrigger(struct irq_data *data)
+{
+	return data->chip->retrigger(data->irq);
+}
+
+static void compat_bus_lock(struct irq_data *data)
+{
+	data->chip->bus_lock(data->irq);
+}
+
+static void compat_bus_sync_unlock(struct irq_data *data)
+{
+	data->chip->bus_sync_unlock(data->irq);
+}
+#endif
+
 /*
  * Fixup enable/disable function pointers
  */
 void irq_chip_set_defaults(struct irq_chip *chip)
 {
-	if (!chip->enable)
-		chip->enable = default_enable;
-	if (!chip->disable)
-		chip->disable = default_disable;
-	if (!chip->startup)
-		chip->startup = default_startup;
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	/*
-	 * We use chip->disable, when the user provided its own. When
-	 * we have default_disable set for chip->disable, then we need
+	 * Compat fixup functions need to be before we set the
+	 * defaults for enable/disable/startup/shutdown
+	 */
+	if (chip->enable)
+		chip->irq_enable = compat_irq_enable;
+	if (chip->disable)
+		chip->irq_disable = compat_irq_disable;
+	if (chip->shutdown)
+		chip->irq_shutdown = compat_irq_shutdown;
+	if (chip->startup)
+		chip->irq_startup = compat_irq_startup;
+#endif
+	/*
+	 * The real defaults
+	 */
+	if (!chip->irq_enable)
+		chip->irq_enable = default_enable;
+	if (!chip->irq_disable)
+		chip->irq_disable = default_disable;
+	if (!chip->irq_startup)
+		chip->irq_startup = default_startup;
+	/*
+	 * We use chip->irq_disable, when the user provided its own. When
+	 * we have default_disable set for chip->irq_disable, then we need
 	 * to use default_shutdown, otherwise the irq line is not
 	 * disabled on free_irq():
 	 */
-	if (!chip->shutdown)
-		chip->shutdown = chip->disable != default_disable ?
-			chip->disable : default_shutdown;
-	if (!chip->name)
-		chip->name = chip->typename;
+	if (!chip->irq_shutdown)
+		chip->irq_shutdown = chip->irq_disable != default_disable ?
+			chip->irq_disable : default_shutdown;
+
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	if (!chip->end)
 		chip->end = dummy_irq_chip.end;
+
+	/*
+	 * Now fix up the remaining compat handlers
+	 */
+	if (chip->bus_lock)
+		chip->irq_bus_lock = compat_bus_lock;
+	if (chip->bus_sync_unlock)
+		chip->irq_bus_sync_unlock = compat_bus_sync_unlock;
+	if (chip->mask)
+		chip->irq_mask = compat_irq_mask;
+	if (chip->unmask)
+		chip->irq_unmask = compat_irq_unmask;
+	if (chip->ack)
+		chip->irq_ack = compat_irq_ack;
+	if (chip->mask_ack)
+		chip->irq_mask_ack = compat_irq_mask_ack;
+	if (chip->eoi)
+		chip->irq_eoi = compat_irq_eoi;
+	if (chip->set_affinity)
+		chip->irq_set_affinity = compat_irq_set_affinity;
+	if (chip->set_type)
+		chip->irq_set_type = compat_irq_set_type;
+	if (chip->set_wake)
+		chip->irq_set_wake = compat_irq_set_wake;
+	if (chip->retrigger)
+		chip->irq_retrigger = compat_irq_retrigger;
+#endif
 }
 
-static inline void mask_ack_irq(struct irq_desc *desc, int irq)
+static inline void mask_ack_irq(struct irq_desc *desc)
 {
-	if (desc->chip->mask_ack)
-		desc->chip->mask_ack(irq);
+	if (desc->irq_data.chip->irq_mask_ack)
+		desc->irq_data.chip->irq_mask_ack(&desc->irq_data);
 	else {
-		desc->chip->mask(irq);
-		if (desc->chip->ack)
-			desc->chip->ack(irq);
+		desc->irq_data.chip->irq_mask(&desc->irq_data);
+		if (desc->irq_data.chip->irq_ack)
+			desc->irq_data.chip->irq_ack(&desc->irq_data);
 	}
 	desc->status |= IRQ_MASKED;
 }
 
-static inline void mask_irq(struct irq_desc *desc, int irq)
+static inline void mask_irq(struct irq_desc *desc)
 {
-	if (desc->chip->mask) {
-		desc->chip->mask(irq);
+	if (desc->irq_data.chip->irq_mask) {
+		desc->irq_data.chip->irq_mask(&desc->irq_data);
 		desc->status |= IRQ_MASKED;
 	}
 }
 
-static inline void unmask_irq(struct irq_desc *desc, int irq)
+static inline void unmask_irq(struct irq_desc *desc)
 {
-	if (desc->chip->unmask) {
-		desc->chip->unmask(irq);
+	if (desc->irq_data.chip->irq_unmask) {
+		desc->irq_data.chip->irq_unmask(&desc->irq_data);
 		desc->status &= ~IRQ_MASKED;
 	}
 }
@@ -476,7 +505,7 @@
 	irqreturn_t action_ret;
 
 	raw_spin_lock(&desc->lock);
-	mask_ack_irq(desc, irq);
+	mask_ack_irq(desc);
 
 	if (unlikely(desc->status & IRQ_INPROGRESS))
 		goto out_unlock;
@@ -502,7 +531,7 @@
 	desc->status &= ~IRQ_INPROGRESS;
 
 	if (!(desc->status & (IRQ_DISABLED | IRQ_ONESHOT)))
-		unmask_irq(desc, irq);
+		unmask_irq(desc);
 out_unlock:
 	raw_spin_unlock(&desc->lock);
 }
@@ -539,7 +568,7 @@
 	action = desc->action;
 	if (unlikely(!action || (desc->status & IRQ_DISABLED))) {
 		desc->status |= IRQ_PENDING;
-		mask_irq(desc, irq);
+		mask_irq(desc);
 		goto out;
 	}
 
@@ -554,7 +583,7 @@
 	raw_spin_lock(&desc->lock);
 	desc->status &= ~IRQ_INPROGRESS;
 out:
-	desc->chip->eoi(irq);
+	desc->irq_data.chip->irq_eoi(&desc->irq_data);
 
 	raw_spin_unlock(&desc->lock);
 }
@@ -590,14 +619,13 @@
 	if (unlikely((desc->status & (IRQ_INPROGRESS | IRQ_DISABLED)) ||
 		    !desc->action)) {
 		desc->status |= (IRQ_PENDING | IRQ_MASKED);
-		mask_ack_irq(desc, irq);
+		mask_ack_irq(desc);
 		goto out_unlock;
 	}
 	kstat_incr_irqs_this_cpu(irq, desc);
 
 	/* Start handling the irq */
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	desc->irq_data.chip->irq_ack(&desc->irq_data);
 
 	/* Mark the IRQ currently in progress.*/
 	desc->status |= IRQ_INPROGRESS;
@@ -607,7 +635,7 @@
 		irqreturn_t action_ret;
 
 		if (unlikely(!action)) {
-			mask_irq(desc, irq);
+			mask_irq(desc);
 			goto out_unlock;
 		}
 
@@ -619,7 +647,7 @@
 		if (unlikely((desc->status &
 			       (IRQ_PENDING | IRQ_MASKED | IRQ_DISABLED)) ==
 			      (IRQ_PENDING | IRQ_MASKED))) {
-			unmask_irq(desc, irq);
+			unmask_irq(desc);
 		}
 
 		desc->status &= ~IRQ_PENDING;
@@ -650,15 +678,15 @@
 
 	kstat_incr_irqs_this_cpu(irq, desc);
 
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	if (desc->irq_data.chip->irq_ack)
+		desc->irq_data.chip->irq_ack(&desc->irq_data);
 
 	action_ret = handle_IRQ_event(irq, desc->action);
 	if (!noirqdebug)
 		note_interrupt(irq, desc, action_ret);
 
-	if (desc->chip->eoi)
-		desc->chip->eoi(irq);
+	if (desc->irq_data.chip->irq_eoi)
+		desc->irq_data.chip->irq_eoi(&desc->irq_data);
 }
 
 void
@@ -676,7 +704,7 @@
 
 	if (!handle)
 		handle = handle_bad_irq;
-	else if (desc->chip == &no_irq_chip) {
+	else if (desc->irq_data.chip == &no_irq_chip) {
 		printk(KERN_WARNING "Trying to install %sinterrupt handler "
 		       "for IRQ%d\n", is_chained ? "chained " : "", irq);
 		/*
@@ -686,16 +714,16 @@
 		 * prevent us to setup the interrupt at all. Switch it to
 		 * dummy_irq_chip for easy transition.
 		 */
-		desc->chip = &dummy_irq_chip;
+		desc->irq_data.chip = &dummy_irq_chip;
 	}
 
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	raw_spin_lock_irqsave(&desc->lock, flags);
 
 	/* Uninstall? */
 	if (handle == handle_bad_irq) {
-		if (desc->chip != &no_irq_chip)
-			mask_ack_irq(desc, irq);
+		if (desc->irq_data.chip != &no_irq_chip)
+			mask_ack_irq(desc);
 		desc->status |= IRQ_DISABLED;
 		desc->depth = 1;
 	}
@@ -706,10 +734,10 @@
 		desc->status &= ~IRQ_DISABLED;
 		desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE;
 		desc->depth = 0;
-		desc->chip->startup(irq);
+		desc->irq_data.chip->irq_startup(&desc->irq_data);
 	}
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 }
 EXPORT_SYMBOL_GPL(__set_irq_handler);
 
@@ -729,32 +757,20 @@
 	__set_irq_handler(irq, handle, 0, name);
 }
 
-void set_irq_noprobe(unsigned int irq)
+void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long flags;
 
-	if (!desc) {
-		printk(KERN_ERR "Trying to mark IRQ%d non-probeable\n", irq);
+	if (!desc)
 		return;
-	}
+
+	/* Sanitize flags */
+	set &= IRQF_MODIFY_MASK;
+	clr &= IRQF_MODIFY_MASK;
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status |= IRQ_NOPROBE;
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
-
-void set_irq_probe(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	unsigned long flags;
-
-	if (!desc) {
-		printk(KERN_ERR "Trying to mark IRQ%d probeable\n", irq);
-		return;
-	}
-
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status &= ~IRQ_NOPROBE;
+	desc->status &= ~clr;
+	desc->status |= set;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 }
diff --git a/kernel/irq/dummychip.c b/kernel/irq/dummychip.c
new file mode 100644
index 0000000..20dc547
--- /dev/null
+++ b/kernel/irq/dummychip.c
@@ -0,0 +1,68 @@
+/*
+ * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar
+ * Copyright (C) 2005-2006, Thomas Gleixner, Russell King
+ *
+ * This file contains the dummy interrupt chip implementation
+ */
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+
+#include "internals.h"
+
+/*
+ * What should we do if we get a hw irq event on an illegal vector?
+ * Each architecture has to answer this themself.
+ */
+static void ack_bad(struct irq_data *data)
+{
+	struct irq_desc *desc = irq_data_to_desc(data);
+
+	print_irq_desc(data->irq, desc);
+	ack_bad_irq(data->irq);
+}
+
+/*
+ * NOP functions
+ */
+static void noop(struct irq_data *data) { }
+
+static unsigned int noop_ret(struct irq_data *data)
+{
+	return 0;
+}
+
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+static void compat_noop(unsigned int irq) { }
+#define END_INIT .end = compat_noop
+#else
+#define END_INIT
+#endif
+
+/*
+ * Generic no controller implementation
+ */
+struct irq_chip no_irq_chip = {
+	.name		= "none",
+	.irq_startup	= noop_ret,
+	.irq_shutdown	= noop,
+	.irq_enable	= noop,
+	.irq_disable	= noop,
+	.irq_ack	= ack_bad,
+	END_INIT
+};
+
+/*
+ * Generic dummy implementation which can be used for
+ * real dumb interrupt sources
+ */
+struct irq_chip dummy_irq_chip = {
+	.name		= "dummy",
+	.irq_startup	= noop_ret,
+	.irq_shutdown	= noop,
+	.irq_enable	= noop,
+	.irq_disable	= noop,
+	.irq_ack	= noop,
+	.irq_mask	= noop,
+	.irq_unmask	= noop,
+	END_INIT
+};
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 27e5c69..e2347eb 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -11,24 +11,15 @@
  */
 
 #include <linux/irq.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/module.h>
 #include <linux/random.h>
+#include <linux/sched.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
-#include <linux/rculist.h>
-#include <linux/hash.h>
-#include <linux/radix-tree.h>
+
 #include <trace/events/irq.h>
 
 #include "internals.h"
 
-/*
- * lockdep: we want to handle all irq_desc locks as a single lock-class:
- */
-struct lock_class_key irq_desc_lock_class;
-
 /**
  * handle_bad_irq - handle spurious and unhandled irqs
  * @irq:       the interrupt number
@@ -43,304 +34,6 @@
 	ack_bad_irq(irq);
 }
 
-#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
-static void __init init_irq_default_affinity(void)
-{
-	alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
-	cpumask_setall(irq_default_affinity);
-}
-#else
-static void __init init_irq_default_affinity(void)
-{
-}
-#endif
-
-/*
- * Linux has a controller-independent interrupt architecture.
- * Every controller has a 'controller-template', that is used
- * by the main code to do the right thing. Each driver-visible
- * interrupt source is transparently wired to the appropriate
- * controller. Thus drivers need not be aware of the
- * interrupt-controller.
- *
- * The code is designed to be easily extended with new/different
- * interrupt controllers, without having to do assembly magic or
- * having to touch the generic code.
- *
- * Controller mappings for all interrupt sources:
- */
-int nr_irqs = NR_IRQS;
-EXPORT_SYMBOL_GPL(nr_irqs);
-
-#ifdef CONFIG_SPARSE_IRQ
-
-static struct irq_desc irq_desc_init = {
-	.irq	    = -1,
-	.status	    = IRQ_DISABLED,
-	.chip	    = &no_irq_chip,
-	.handle_irq = handle_bad_irq,
-	.depth      = 1,
-	.lock       = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
-};
-
-void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr)
-{
-	void *ptr;
-
-	ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs),
-			   GFP_ATOMIC, node);
-
-	/*
-	 * don't overwite if can not get new one
-	 * init_copy_kstat_irqs() could still use old one
-	 */
-	if (ptr) {
-		printk(KERN_DEBUG "  alloc kstat_irqs on node %d\n", node);
-		desc->kstat_irqs = ptr;
-	}
-}
-
-static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
-{
-	memcpy(desc, &irq_desc_init, sizeof(struct irq_desc));
-
-	raw_spin_lock_init(&desc->lock);
-	desc->irq = irq;
-#ifdef CONFIG_SMP
-	desc->node = node;
-#endif
-	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-	init_kstat_irqs(desc, node, nr_cpu_ids);
-	if (!desc->kstat_irqs) {
-		printk(KERN_ERR "can not alloc kstat_irqs\n");
-		BUG_ON(1);
-	}
-	if (!alloc_desc_masks(desc, node, false)) {
-		printk(KERN_ERR "can not alloc irq_desc cpumasks\n");
-		BUG_ON(1);
-	}
-	init_desc_masks(desc);
-	arch_init_chip_data(desc, node);
-}
-
-/*
- * Protect the sparse_irqs:
- */
-DEFINE_RAW_SPINLOCK(sparse_irq_lock);
-
-static RADIX_TREE(irq_desc_tree, GFP_ATOMIC);
-
-static void set_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-	radix_tree_insert(&irq_desc_tree, irq, desc);
-}
-
-struct irq_desc *irq_to_desc(unsigned int irq)
-{
-	return radix_tree_lookup(&irq_desc_tree, irq);
-}
-
-void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-	void **ptr;
-
-	ptr = radix_tree_lookup_slot(&irq_desc_tree, irq);
-	if (ptr)
-		radix_tree_replace_slot(ptr, desc);
-}
-
-static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
-	[0 ... NR_IRQS_LEGACY-1] = {
-		.irq	    = -1,
-		.status	    = IRQ_DISABLED,
-		.chip	    = &no_irq_chip,
-		.handle_irq = handle_bad_irq,
-		.depth	    = 1,
-		.lock	    = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
-	}
-};
-
-static unsigned int *kstat_irqs_legacy;
-
-int __init early_irq_init(void)
-{
-	struct irq_desc *desc;
-	int legacy_count;
-	int node;
-	int i;
-
-	init_irq_default_affinity();
-
-	 /* initialize nr_irqs based on nr_cpu_ids */
-	arch_probe_nr_irqs();
-	printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs);
-
-	desc = irq_desc_legacy;
-	legacy_count = ARRAY_SIZE(irq_desc_legacy);
-	node = first_online_node;
-
-	/* allocate based on nr_cpu_ids */
-	kstat_irqs_legacy = kzalloc_node(NR_IRQS_LEGACY * nr_cpu_ids *
-					  sizeof(int), GFP_NOWAIT, node);
-
-	for (i = 0; i < legacy_count; i++) {
-		desc[i].irq = i;
-#ifdef CONFIG_SMP
-		desc[i].node = node;
-#endif
-		desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids;
-		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
-		alloc_desc_masks(&desc[i], node, true);
-		init_desc_masks(&desc[i]);
-		set_irq_desc(i, &desc[i]);
-	}
-
-	return arch_early_irq_init();
-}
-
-struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
-{
-	struct irq_desc *desc;
-	unsigned long flags;
-
-	if (irq >= nr_irqs) {
-		WARN(1, "irq (%d) >= nr_irqs (%d) in irq_to_desc_alloc\n",
-			irq, nr_irqs);
-		return NULL;
-	}
-
-	desc = irq_to_desc(irq);
-	if (desc)
-		return desc;
-
-	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
-
-	/* We have to check it to avoid races with another CPU */
-	desc = irq_to_desc(irq);
-	if (desc)
-		goto out_unlock;
-
-	desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
-
-	printk(KERN_DEBUG "  alloc irq_desc for %d on node %d\n", irq, node);
-	if (!desc) {
-		printk(KERN_ERR "can not alloc irq_desc\n");
-		BUG_ON(1);
-	}
-	init_one_irq_desc(irq, desc, node);
-
-	set_irq_desc(irq, desc);
-
-out_unlock:
-	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-	return desc;
-}
-
-#else /* !CONFIG_SPARSE_IRQ */
-
-struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
-	[0 ... NR_IRQS-1] = {
-		.status = IRQ_DISABLED,
-		.chip = &no_irq_chip,
-		.handle_irq = handle_bad_irq,
-		.depth = 1,
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
-	}
-};
-
-static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
-int __init early_irq_init(void)
-{
-	struct irq_desc *desc;
-	int count;
-	int i;
-
-	init_irq_default_affinity();
-
-	printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS);
-
-	desc = irq_desc;
-	count = ARRAY_SIZE(irq_desc);
-
-	for (i = 0; i < count; i++) {
-		desc[i].irq = i;
-		alloc_desc_masks(&desc[i], 0, true);
-		init_desc_masks(&desc[i]);
-		desc[i].kstat_irqs = kstat_irqs_all[i];
-	}
-	return arch_early_irq_init();
-}
-
-struct irq_desc *irq_to_desc(unsigned int irq)
-{
-	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
-}
-
-struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
-{
-	return irq_to_desc(irq);
-}
-#endif /* !CONFIG_SPARSE_IRQ */
-
-void clear_kstat_irqs(struct irq_desc *desc)
-{
-	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
-}
-
-/*
- * What should we do if we get a hw irq event on an illegal vector?
- * Each architecture has to answer this themself.
- */
-static void ack_bad(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	print_irq_desc(irq, desc);
-	ack_bad_irq(irq);
-}
-
-/*
- * NOP functions
- */
-static void noop(unsigned int irq)
-{
-}
-
-static unsigned int noop_ret(unsigned int irq)
-{
-	return 0;
-}
-
-/*
- * Generic no controller implementation
- */
-struct irq_chip no_irq_chip = {
-	.name		= "none",
-	.startup	= noop_ret,
-	.shutdown	= noop,
-	.enable		= noop,
-	.disable	= noop,
-	.ack		= ack_bad,
-	.end		= noop,
-};
-
-/*
- * Generic dummy implementation which can be used for
- * real dumb interrupt sources
- */
-struct irq_chip dummy_irq_chip = {
-	.name		= "dummy",
-	.startup	= noop_ret,
-	.shutdown	= noop,
-	.enable		= noop,
-	.disable	= noop,
-	.ack		= noop,
-	.mask		= noop,
-	.unmask		= noop,
-	.end		= noop,
-};
-
 /*
  * Special, empty irq handler:
  */
@@ -457,20 +150,20 @@
 		/*
 		 * No locking required for CPU-local interrupts:
 		 */
-		if (desc->chip->ack)
-			desc->chip->ack(irq);
+		if (desc->irq_data.chip->ack)
+			desc->irq_data.chip->ack(irq);
 		if (likely(!(desc->status & IRQ_DISABLED))) {
 			action_ret = handle_IRQ_event(irq, desc->action);
 			if (!noirqdebug)
 				note_interrupt(irq, desc, action_ret);
 		}
-		desc->chip->end(irq);
+		desc->irq_data.chip->end(irq);
 		return 1;
 	}
 
 	raw_spin_lock(&desc->lock);
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	if (desc->irq_data.chip->ack)
+		desc->irq_data.chip->ack(irq);
 	/*
 	 * REPLAY is when Linux resends an IRQ that was dropped earlier
 	 * WAITING is used by probe to mark irqs that are being tested
@@ -530,27 +223,9 @@
 	 * The ->end() handler has to deal with interrupts which got
 	 * disabled while the handler was running.
 	 */
-	desc->chip->end(irq);
+	desc->irq_data.chip->end(irq);
 	raw_spin_unlock(&desc->lock);
 
 	return 1;
 }
 #endif
-
-void early_init_irq_lock_class(void)
-{
-	struct irq_desc *desc;
-	int i;
-
-	for_each_irq_desc(i, desc) {
-		lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-	}
-}
-
-unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	return desc ? desc->kstat_irqs[cpu] : 0;
-}
-EXPORT_SYMBOL(kstat_irqs_cpu);
-
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index c63f3bc..4571ae7 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -1,9 +1,12 @@
 /*
  * IRQ subsystem internal functions and variables:
  */
+#include <linux/irqdesc.h>
 
 extern int noirqdebug;
 
+#define irq_data_to_desc(data)	container_of(data, struct irq_desc, irq_data)
+
 /* Set default functions for irq_chip structures: */
 extern void irq_chip_set_defaults(struct irq_chip *chip);
 
@@ -15,21 +18,19 @@
 extern void __disable_irq(struct irq_desc *desc, unsigned int irq, bool susp);
 extern void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume);
 
-extern struct lock_class_key irq_desc_lock_class;
 extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr);
-extern void clear_kstat_irqs(struct irq_desc *desc);
-extern raw_spinlock_t sparse_irq_lock;
 
-#ifdef CONFIG_SPARSE_IRQ
-void replace_irq_desc(unsigned int irq, struct irq_desc *desc);
-#endif
+/* Resending of interrupts :*/
+void check_irq_resend(struct irq_desc *desc, unsigned int irq);
 
 #ifdef CONFIG_PROC_FS
 extern void register_irq_proc(unsigned int irq, struct irq_desc *desc);
+extern void unregister_irq_proc(unsigned int irq, struct irq_desc *desc);
 extern void register_handler_proc(unsigned int irq, struct irqaction *action);
 extern void unregister_handler_proc(unsigned int irq, struct irqaction *action);
 #else
 static inline void register_irq_proc(unsigned int irq, struct irq_desc *desc) { }
+static inline void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) { }
 static inline void register_handler_proc(unsigned int irq,
 					 struct irqaction *action) { }
 static inline void unregister_handler_proc(unsigned int irq,
@@ -40,17 +41,27 @@
 
 extern void irq_set_thread_affinity(struct irq_desc *desc);
 
-/* Inline functions for support of irq chips on slow busses */
-static inline void chip_bus_lock(unsigned int irq, struct irq_desc *desc)
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+static inline void irq_end(unsigned int irq, struct irq_desc *desc)
 {
-	if (unlikely(desc->chip->bus_lock))
-		desc->chip->bus_lock(irq);
+	if (desc->irq_data.chip && desc->irq_data.chip->end)
+		desc->irq_data.chip->end(irq);
+}
+#else
+static inline void irq_end(unsigned int irq, struct irq_desc *desc) { }
+#endif
+
+/* Inline functions for support of irq chips on slow busses */
+static inline void chip_bus_lock(struct irq_desc *desc)
+{
+	if (unlikely(desc->irq_data.chip->irq_bus_lock))
+		desc->irq_data.chip->irq_bus_lock(&desc->irq_data);
 }
 
-static inline void chip_bus_sync_unlock(unsigned int irq, struct irq_desc *desc)
+static inline void chip_bus_sync_unlock(struct irq_desc *desc)
 {
-	if (unlikely(desc->chip->bus_sync_unlock))
-		desc->chip->bus_sync_unlock(irq);
+	if (unlikely(desc->irq_data.chip->irq_bus_sync_unlock))
+		desc->irq_data.chip->irq_bus_sync_unlock(&desc->irq_data);
 }
 
 /*
@@ -67,8 +78,8 @@
 		irq, desc, desc->depth, desc->irq_count, desc->irqs_unhandled);
 	printk("->handle_irq():  %p, ", desc->handle_irq);
 	print_symbol("%s\n", (unsigned long)desc->handle_irq);
-	printk("->chip(): %p, ", desc->chip);
-	print_symbol("%s\n", (unsigned long)desc->chip);
+	printk("->irq_data.chip(): %p, ", desc->irq_data.chip);
+	print_symbol("%s\n", (unsigned long)desc->irq_data.chip);
 	printk("->action(): %p\n", desc->action);
 	if (desc->action) {
 		printk("->action->handler(): %p, ", desc->action->handler);
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
new file mode 100644
index 0000000..9d917ff
--- /dev/null
+++ b/kernel/irq/irqdesc.c
@@ -0,0 +1,395 @@
+/*
+ * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar
+ * Copyright (C) 2005-2006, Thomas Gleixner, Russell King
+ *
+ * This file contains the interrupt descriptor management code
+ *
+ * Detailed information is available in Documentation/DocBook/genericirq
+ *
+ */
+#include <linux/irq.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/kernel_stat.h>
+#include <linux/radix-tree.h>
+#include <linux/bitmap.h>
+
+#include "internals.h"
+
+/*
+ * lockdep: we want to handle all irq_desc locks as a single lock-class:
+ */
+static struct lock_class_key irq_desc_lock_class;
+
+#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
+static void __init init_irq_default_affinity(void)
+{
+	alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
+	cpumask_setall(irq_default_affinity);
+}
+#else
+static void __init init_irq_default_affinity(void)
+{
+}
+#endif
+
+#ifdef CONFIG_SMP
+static int alloc_masks(struct irq_desc *desc, gfp_t gfp, int node)
+{
+	if (!zalloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
+		return -ENOMEM;
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	if (!zalloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
+		free_cpumask_var(desc->irq_data.affinity);
+		return -ENOMEM;
+	}
+#endif
+	return 0;
+}
+
+static void desc_smp_init(struct irq_desc *desc, int node)
+{
+	desc->irq_data.node = node;
+	cpumask_copy(desc->irq_data.affinity, irq_default_affinity);
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_clear(desc->pending_mask);
+#endif
+}
+
+static inline int desc_node(struct irq_desc *desc)
+{
+	return desc->irq_data.node;
+}
+
+#else
+static inline int
+alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; }
+static inline void desc_smp_init(struct irq_desc *desc, int node) { }
+static inline int desc_node(struct irq_desc *desc) { return 0; }
+#endif
+
+static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
+{
+	desc->irq_data.irq = irq;
+	desc->irq_data.chip = &no_irq_chip;
+	desc->irq_data.chip_data = NULL;
+	desc->irq_data.handler_data = NULL;
+	desc->irq_data.msi_desc = NULL;
+	desc->status = IRQ_DEFAULT_INIT_FLAGS;
+	desc->handle_irq = handle_bad_irq;
+	desc->depth = 1;
+	desc->irq_count = 0;
+	desc->irqs_unhandled = 0;
+	desc->name = NULL;
+	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+	desc_smp_init(desc, node);
+}
+
+int nr_irqs = NR_IRQS;
+EXPORT_SYMBOL_GPL(nr_irqs);
+
+static DEFINE_MUTEX(sparse_irq_lock);
+static DECLARE_BITMAP(allocated_irqs, NR_IRQS);
+
+#ifdef CONFIG_SPARSE_IRQ
+
+static RADIX_TREE(irq_desc_tree, GFP_KERNEL);
+
+static void irq_insert_desc(unsigned int irq, struct irq_desc *desc)
+{
+	radix_tree_insert(&irq_desc_tree, irq, desc);
+}
+
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+	return radix_tree_lookup(&irq_desc_tree, irq);
+}
+
+static void delete_irq_desc(unsigned int irq)
+{
+	radix_tree_delete(&irq_desc_tree, irq);
+}
+
+#ifdef CONFIG_SMP
+static void free_masks(struct irq_desc *desc)
+{
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	free_cpumask_var(desc->pending_mask);
+#endif
+	free_cpumask_var(desc->irq_data.affinity);
+}
+#else
+static inline void free_masks(struct irq_desc *desc) { }
+#endif
+
+static struct irq_desc *alloc_desc(int irq, int node)
+{
+	struct irq_desc *desc;
+	gfp_t gfp = GFP_KERNEL;
+
+	desc = kzalloc_node(sizeof(*desc), gfp, node);
+	if (!desc)
+		return NULL;
+	/* allocate based on nr_cpu_ids */
+	desc->kstat_irqs = kzalloc_node(nr_cpu_ids * sizeof(*desc->kstat_irqs),
+					 gfp, node);
+	if (!desc->kstat_irqs)
+		goto err_desc;
+
+	if (alloc_masks(desc, gfp, node))
+		goto err_kstat;
+
+	raw_spin_lock_init(&desc->lock);
+	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
+
+	desc_set_defaults(irq, desc, node);
+
+	return desc;
+
+err_kstat:
+	kfree(desc->kstat_irqs);
+err_desc:
+	kfree(desc);
+	return NULL;
+}
+
+static void free_desc(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	unregister_irq_proc(irq, desc);
+
+	mutex_lock(&sparse_irq_lock);
+	delete_irq_desc(irq);
+	mutex_unlock(&sparse_irq_lock);
+
+	free_masks(desc);
+	kfree(desc->kstat_irqs);
+	kfree(desc);
+}
+
+static int alloc_descs(unsigned int start, unsigned int cnt, int node)
+{
+	struct irq_desc *desc;
+	int i;
+
+	for (i = 0; i < cnt; i++) {
+		desc = alloc_desc(start + i, node);
+		if (!desc)
+			goto err;
+		mutex_lock(&sparse_irq_lock);
+		irq_insert_desc(start + i, desc);
+		mutex_unlock(&sparse_irq_lock);
+	}
+	return start;
+
+err:
+	for (i--; i >= 0; i--)
+		free_desc(start + i);
+
+	mutex_lock(&sparse_irq_lock);
+	bitmap_clear(allocated_irqs, start, cnt);
+	mutex_unlock(&sparse_irq_lock);
+	return -ENOMEM;
+}
+
+struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
+{
+	int res = irq_alloc_descs(irq, irq, 1, node);
+
+	if (res == -EEXIST || res == irq)
+		return irq_to_desc(irq);
+	return NULL;
+}
+
+int __init early_irq_init(void)
+{
+	int i, initcnt, node = first_online_node;
+	struct irq_desc *desc;
+
+	init_irq_default_affinity();
+
+	/* Let arch update nr_irqs and return the nr of preallocated irqs */
+	initcnt = arch_probe_nr_irqs();
+	printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt);
+
+	for (i = 0; i < initcnt; i++) {
+		desc = alloc_desc(i, node);
+		set_bit(i, allocated_irqs);
+		irq_insert_desc(i, desc);
+	}
+	return arch_early_irq_init();
+}
+
+#else /* !CONFIG_SPARSE_IRQ */
+
+struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
+	[0 ... NR_IRQS-1] = {
+		.status		= IRQ_DEFAULT_INIT_FLAGS,
+		.handle_irq	= handle_bad_irq,
+		.depth		= 1,
+		.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
+	}
+};
+
+static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
+int __init early_irq_init(void)
+{
+	int count, i, node = first_online_node;
+	struct irq_desc *desc;
+
+	init_irq_default_affinity();
+
+	printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS);
+
+	desc = irq_desc;
+	count = ARRAY_SIZE(irq_desc);
+
+	for (i = 0; i < count; i++) {
+		desc[i].irq_data.irq = i;
+		desc[i].irq_data.chip = &no_irq_chip;
+		desc[i].kstat_irqs = kstat_irqs_all[i];
+		alloc_masks(desc + i, GFP_KERNEL, node);
+		desc_smp_init(desc + i, node);
+		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
+	}
+	return arch_early_irq_init();
+}
+
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
+}
+
+struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
+{
+	return irq_to_desc(irq);
+}
+
+static void free_desc(unsigned int irq)
+{
+	dynamic_irq_cleanup(irq);
+}
+
+static inline int alloc_descs(unsigned int start, unsigned int cnt, int node)
+{
+	return start;
+}
+#endif /* !CONFIG_SPARSE_IRQ */
+
+/* Dynamic interrupt handling */
+
+/**
+ * irq_free_descs - free irq descriptors
+ * @from:	Start of descriptor range
+ * @cnt:	Number of consecutive irqs to free
+ */
+void irq_free_descs(unsigned int from, unsigned int cnt)
+{
+	int i;
+
+	if (from >= nr_irqs || (from + cnt) > nr_irqs)
+		return;
+
+	for (i = 0; i < cnt; i++)
+		free_desc(from + i);
+
+	mutex_lock(&sparse_irq_lock);
+	bitmap_clear(allocated_irqs, from, cnt);
+	mutex_unlock(&sparse_irq_lock);
+}
+
+/**
+ * irq_alloc_descs - allocate and initialize a range of irq descriptors
+ * @irq:	Allocate for specific irq number if irq >= 0
+ * @from:	Start the search from this irq number
+ * @cnt:	Number of consecutive irqs to allocate.
+ * @node:	Preferred node on which the irq descriptor should be allocated
+ *
+ * Returns the first irq number or error code
+ */
+int __ref
+irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node)
+{
+	int start, ret;
+
+	if (!cnt)
+		return -EINVAL;
+
+	mutex_lock(&sparse_irq_lock);
+
+	start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0);
+	ret = -EEXIST;
+	if (irq >=0 && start != irq)
+		goto err;
+
+	ret = -ENOMEM;
+	if (start >= nr_irqs)
+		goto err;
+
+	bitmap_set(allocated_irqs, start, cnt);
+	mutex_unlock(&sparse_irq_lock);
+	return alloc_descs(start, cnt, node);
+
+err:
+	mutex_unlock(&sparse_irq_lock);
+	return ret;
+}
+
+/**
+ * irq_reserve_irqs - mark irqs allocated
+ * @from:	mark from irq number
+ * @cnt:	number of irqs to mark
+ *
+ * Returns 0 on success or an appropriate error code
+ */
+int irq_reserve_irqs(unsigned int from, unsigned int cnt)
+{
+	unsigned int start;
+	int ret = 0;
+
+	if (!cnt || (from + cnt) > nr_irqs)
+		return -EINVAL;
+
+	mutex_lock(&sparse_irq_lock);
+	start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0);
+	if (start == from)
+		bitmap_set(allocated_irqs, start, cnt);
+	else
+		ret = -EEXIST;
+	mutex_unlock(&sparse_irq_lock);
+	return ret;
+}
+
+/**
+ * irq_get_next_irq - get next allocated irq number
+ * @offset:	where to start the search
+ *
+ * Returns next irq number after offset or nr_irqs if none is found.
+ */
+unsigned int irq_get_next_irq(unsigned int offset)
+{
+	return find_next_bit(allocated_irqs, nr_irqs, offset);
+}
+
+/**
+ * dynamic_irq_cleanup - cleanup a dynamically allocated irq
+ * @irq:	irq number to initialize
+ */
+void dynamic_irq_cleanup(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&desc->lock, flags);
+	desc_set_defaults(irq, desc, desc_node(desc));
+	raw_spin_unlock_irqrestore(&desc->lock, flags);
+}
+
+unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	return desc ? desc->kstat_irqs[cpu] : 0;
+}
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index c3003e9..644e8d5 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -73,8 +73,8 @@
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
-	if (CHECK_IRQ_PER_CPU(desc->status) || !desc->chip ||
-	    !desc->chip->set_affinity)
+	if (CHECK_IRQ_PER_CPU(desc->status) || !desc->irq_data.chip ||
+	    !desc->irq_data.chip->irq_set_affinity)
 		return 0;
 
 	return 1;
@@ -109,17 +109,18 @@
 int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_chip *chip = desc->irq_data.chip;
 	unsigned long flags;
 
-	if (!desc->chip->set_affinity)
+	if (!chip->irq_set_affinity)
 		return -EINVAL;
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PCNTXT) {
-		if (!desc->chip->set_affinity(irq, cpumask)) {
-			cpumask_copy(desc->affinity, cpumask);
+		if (!chip->irq_set_affinity(&desc->irq_data, cpumask, false)) {
+			cpumask_copy(desc->irq_data.affinity, cpumask);
 			irq_set_thread_affinity(desc);
 		}
 	}
@@ -128,8 +129,8 @@
 		cpumask_copy(desc->pending_mask, cpumask);
 	}
 #else
-	if (!desc->chip->set_affinity(irq, cpumask)) {
-		cpumask_copy(desc->affinity, cpumask);
+	if (!chip->irq_set_affinity(&desc->irq_data, cpumask, false)) {
+		cpumask_copy(desc->irq_data.affinity, cpumask);
 		irq_set_thread_affinity(desc);
 	}
 #endif
@@ -168,16 +169,16 @@
 	 * one of the targets is online.
 	 */
 	if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) {
-		if (cpumask_any_and(desc->affinity, cpu_online_mask)
+		if (cpumask_any_and(desc->irq_data.affinity, cpu_online_mask)
 		    < nr_cpu_ids)
 			goto set_affinity;
 		else
 			desc->status &= ~IRQ_AFFINITY_SET;
 	}
 
-	cpumask_and(desc->affinity, cpu_online_mask, irq_default_affinity);
+	cpumask_and(desc->irq_data.affinity, cpu_online_mask, irq_default_affinity);
 set_affinity:
-	desc->chip->set_affinity(irq, desc->affinity);
+	desc->irq_data.chip->irq_set_affinity(&desc->irq_data, desc->irq_data.affinity, false);
 
 	return 0;
 }
@@ -223,7 +224,7 @@
 
 	if (!desc->depth++) {
 		desc->status |= IRQ_DISABLED;
-		desc->chip->disable(irq);
+		desc->irq_data.chip->irq_disable(&desc->irq_data);
 	}
 }
 
@@ -246,11 +247,11 @@
 	if (!desc)
 		return;
 
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	raw_spin_lock_irqsave(&desc->lock, flags);
 	__disable_irq(desc, irq, false);
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 }
 EXPORT_SYMBOL(disable_irq_nosync);
 
@@ -313,7 +314,7 @@
  *	IRQ line is re-enabled.
  *
  *	This function may be called from IRQ context only when
- *	desc->chip->bus_lock and desc->chip->bus_sync_unlock are NULL !
+ *	desc->irq_data.chip->bus_lock and desc->chip->bus_sync_unlock are NULL !
  */
 void enable_irq(unsigned int irq)
 {
@@ -323,11 +324,11 @@
 	if (!desc)
 		return;
 
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	raw_spin_lock_irqsave(&desc->lock, flags);
 	__enable_irq(desc, irq, false);
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 }
 EXPORT_SYMBOL(enable_irq);
 
@@ -336,8 +337,8 @@
 	struct irq_desc *desc = irq_to_desc(irq);
 	int ret = -ENXIO;
 
-	if (desc->chip->set_wake)
-		ret = desc->chip->set_wake(irq, on);
+	if (desc->irq_data.chip->irq_set_wake)
+		ret = desc->irq_data.chip->irq_set_wake(&desc->irq_data, on);
 
 	return ret;
 }
@@ -429,12 +430,12 @@
 }
 
 int __irq_set_trigger(struct irq_desc *desc, unsigned int irq,
-		unsigned long flags)
+		      unsigned long flags)
 {
 	int ret;
-	struct irq_chip *chip = desc->chip;
+	struct irq_chip *chip = desc->irq_data.chip;
 
-	if (!chip || !chip->set_type) {
+	if (!chip || !chip->irq_set_type) {
 		/*
 		 * IRQF_TRIGGER_* but the PIC does not support multiple
 		 * flow-types?
@@ -445,11 +446,11 @@
 	}
 
 	/* caller masked out all except trigger mode flags */
-	ret = chip->set_type(irq, flags);
+	ret = chip->irq_set_type(&desc->irq_data, flags);
 
 	if (ret)
-		pr_err("setting trigger mode %d for irq %u failed (%pF)\n",
-				(int)flags, irq, chip->set_type);
+		pr_err("setting trigger mode %lu for irq %u failed (%pF)\n",
+		       flags, irq, chip->irq_set_type);
 	else {
 		if (flags & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_LEVEL_HIGH))
 			flags |= IRQ_LEVEL;
@@ -457,8 +458,8 @@
 		desc->status &= ~(IRQ_LEVEL | IRQ_TYPE_SENSE_MASK);
 		desc->status |= flags;
 
-		if (chip != desc->chip)
-			irq_chip_set_defaults(desc->chip);
+		if (chip != desc->irq_data.chip)
+			irq_chip_set_defaults(desc->irq_data.chip);
 	}
 
 	return ret;
@@ -507,7 +508,7 @@
 static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc)
 {
 again:
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	raw_spin_lock_irq(&desc->lock);
 
 	/*
@@ -521,17 +522,17 @@
 	 */
 	if (unlikely(desc->status & IRQ_INPROGRESS)) {
 		raw_spin_unlock_irq(&desc->lock);
-		chip_bus_sync_unlock(irq, desc);
+		chip_bus_sync_unlock(desc);
 		cpu_relax();
 		goto again;
 	}
 
 	if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) {
 		desc->status &= ~IRQ_MASKED;
-		desc->chip->unmask(irq);
+		desc->irq_data.chip->irq_unmask(&desc->irq_data);
 	}
 	raw_spin_unlock_irq(&desc->lock);
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 }
 
 #ifdef CONFIG_SMP
@@ -556,7 +557,7 @@
 	}
 
 	raw_spin_lock_irq(&desc->lock);
-	cpumask_copy(mask, desc->affinity);
+	cpumask_copy(mask, desc->irq_data.affinity);
 	raw_spin_unlock_irq(&desc->lock);
 
 	set_cpus_allowed_ptr(current, mask);
@@ -657,7 +658,7 @@
 	if (!desc)
 		return -EINVAL;
 
-	if (desc->chip == &no_irq_chip)
+	if (desc->irq_data.chip == &no_irq_chip)
 		return -ENOSYS;
 	/*
 	 * Some drivers like serial.c use request_irq() heavily,
@@ -752,7 +753,7 @@
 	}
 
 	if (!shared) {
-		irq_chip_set_defaults(desc->chip);
+		irq_chip_set_defaults(desc->irq_data.chip);
 
 		init_waitqueue_head(&desc->wait_for_threads);
 
@@ -779,7 +780,7 @@
 		if (!(desc->status & IRQ_NOAUTOEN)) {
 			desc->depth = 0;
 			desc->status &= ~IRQ_DISABLED;
-			desc->chip->startup(irq);
+			desc->irq_data.chip->irq_startup(&desc->irq_data);
 		} else
 			/* Undo nested disables: */
 			desc->depth = 1;
@@ -912,17 +913,17 @@
 
 	/* Currently used only by UML, might disappear one day: */
 #ifdef CONFIG_IRQ_RELEASE_METHOD
-	if (desc->chip->release)
-		desc->chip->release(irq, dev_id);
+	if (desc->irq_data.chip->release)
+		desc->irq_data.chip->release(irq, dev_id);
 #endif
 
 	/* If this was the last handler, shut down the IRQ line: */
 	if (!desc->action) {
 		desc->status |= IRQ_DISABLED;
-		if (desc->chip->shutdown)
-			desc->chip->shutdown(irq);
+		if (desc->irq_data.chip->irq_shutdown)
+			desc->irq_data.chip->irq_shutdown(&desc->irq_data);
 		else
-			desc->chip->disable(irq);
+			desc->irq_data.chip->irq_disable(&desc->irq_data);
 	}
 
 #ifdef CONFIG_SMP
@@ -997,9 +998,9 @@
 	if (!desc)
 		return;
 
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	kfree(__free_irq(irq, dev_id));
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 }
 EXPORT_SYMBOL(free_irq);
 
@@ -1086,9 +1087,9 @@
 	action->name = devname;
 	action->dev_id = dev_id;
 
-	chip_bus_lock(irq, desc);
+	chip_bus_lock(desc);
 	retval = __setup_irq(irq, desc, action);
-	chip_bus_sync_unlock(irq, desc);
+	chip_bus_sync_unlock(desc);
 
 	if (retval)
 		kfree(action);
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 2419622..1d25419 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -7,6 +7,7 @@
 void move_masked_irq(int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
+	struct irq_chip *chip = desc->irq_data.chip;
 
 	if (likely(!(desc->status & IRQ_MOVE_PENDING)))
 		return;
@@ -24,7 +25,7 @@
 	if (unlikely(cpumask_empty(desc->pending_mask)))
 		return;
 
-	if (!desc->chip->set_affinity)
+	if (!chip->irq_set_affinity)
 		return;
 
 	assert_raw_spin_locked(&desc->lock);
@@ -43,8 +44,9 @@
 	 */
 	if (likely(cpumask_any_and(desc->pending_mask, cpu_online_mask)
 		   < nr_cpu_ids))
-		if (!desc->chip->set_affinity(irq, desc->pending_mask)) {
-			cpumask_copy(desc->affinity, desc->pending_mask);
+		if (!chip->irq_set_affinity(&desc->irq_data,
+					    desc->pending_mask, false)) {
+			cpumask_copy(desc->irq_data.affinity, desc->pending_mask);
 			irq_set_thread_affinity(desc);
 		}
 
@@ -61,8 +63,8 @@
 	if (unlikely(desc->status & IRQ_DISABLED))
 		return;
 
-	desc->chip->mask(irq);
+	desc->irq_data.chip->irq_mask(&desc->irq_data);
 	move_masked_irq(irq);
-	desc->chip->unmask(irq);
+	desc->irq_data.chip->irq_unmask(&desc->irq_data);
 }
 
diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c
deleted file mode 100644
index 65d3845..0000000
--- a/kernel/irq/numa_migrate.c
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
- * NUMA irq-desc migration code
- *
- * Migrate IRQ data structures (irq_desc, chip_data, etc.) over to
- * the new "home node" of the IRQ.
- */
-
-#include <linux/irq.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/random.h>
-#include <linux/interrupt.h>
-#include <linux/kernel_stat.h>
-
-#include "internals.h"
-
-static void init_copy_kstat_irqs(struct irq_desc *old_desc,
-				 struct irq_desc *desc,
-				 int node, int nr)
-{
-	init_kstat_irqs(desc, node, nr);
-
-	if (desc->kstat_irqs != old_desc->kstat_irqs)
-		memcpy(desc->kstat_irqs, old_desc->kstat_irqs,
-			 nr * sizeof(*desc->kstat_irqs));
-}
-
-static void free_kstat_irqs(struct irq_desc *old_desc, struct irq_desc *desc)
-{
-	if (old_desc->kstat_irqs == desc->kstat_irqs)
-		return;
-
-	kfree(old_desc->kstat_irqs);
-	old_desc->kstat_irqs = NULL;
-}
-
-static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc,
-		 struct irq_desc *desc, int node)
-{
-	memcpy(desc, old_desc, sizeof(struct irq_desc));
-	if (!alloc_desc_masks(desc, node, false)) {
-		printk(KERN_ERR "irq %d: can not get new irq_desc cpumask "
-				"for migration.\n", irq);
-		return false;
-	}
-	raw_spin_lock_init(&desc->lock);
-	desc->node = node;
-	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-	init_copy_kstat_irqs(old_desc, desc, node, nr_cpu_ids);
-	init_copy_desc_masks(old_desc, desc);
-	arch_init_copy_chip_data(old_desc, desc, node);
-	return true;
-}
-
-static void free_one_irq_desc(struct irq_desc *old_desc, struct irq_desc *desc)
-{
-	free_kstat_irqs(old_desc, desc);
-	free_desc_masks(old_desc, desc);
-	arch_free_chip_data(old_desc, desc);
-}
-
-static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc,
-						int node)
-{
-	struct irq_desc *desc;
-	unsigned int irq;
-	unsigned long flags;
-
-	irq = old_desc->irq;
-
-	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
-
-	/* We have to check it to avoid races with another CPU */
-	desc = irq_to_desc(irq);
-
-	if (desc && old_desc != desc)
-		goto out_unlock;
-
-	desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
-	if (!desc) {
-		printk(KERN_ERR "irq %d: can not get new irq_desc "
-				"for migration.\n", irq);
-		/* still use old one */
-		desc = old_desc;
-		goto out_unlock;
-	}
-	if (!init_copy_one_irq_desc(irq, old_desc, desc, node)) {
-		/* still use old one */
-		kfree(desc);
-		desc = old_desc;
-		goto out_unlock;
-	}
-
-	replace_irq_desc(irq, desc);
-	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-	/* free the old one */
-	free_one_irq_desc(old_desc, desc);
-	kfree(old_desc);
-
-	return desc;
-
-out_unlock:
-	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-	return desc;
-}
-
-struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
-{
-	/* those static or target node is -1, do not move them */
-	if (desc->irq < NR_IRQS_LEGACY || node == -1)
-		return desc;
-
-	if (desc->node != node)
-		desc = __real_move_irq_desc(desc, node);
-
-	return desc;
-}
-
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index 09a2ee5..01b1d3a 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -21,7 +21,7 @@
 static int irq_affinity_proc_show(struct seq_file *m, void *v)
 {
 	struct irq_desc *desc = irq_to_desc((long)m->private);
-	const struct cpumask *mask = desc->affinity;
+	const struct cpumask *mask = desc->irq_data.affinity;
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PENDING)
@@ -65,7 +65,7 @@
 	cpumask_var_t new_value;
 	int err;
 
-	if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity ||
+	if (!irq_to_desc(irq)->irq_data.chip->irq_set_affinity || no_irq_affinity ||
 	    irq_balancing_disabled(irq))
 		return -EIO;
 
@@ -185,7 +185,7 @@
 {
 	struct irq_desc *desc = irq_to_desc((long) m->private);
 
-	seq_printf(m, "%d\n", desc->node);
+	seq_printf(m, "%d\n", desc->irq_data.node);
 	return 0;
 }
 
@@ -269,7 +269,7 @@
 {
 	char name [MAX_NAMELEN];
 
-	if (!root_irq_dir || (desc->chip == &no_irq_chip) || desc->dir)
+	if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip) || desc->dir)
 		return;
 
 	memset(name, 0, MAX_NAMELEN);
@@ -297,6 +297,24 @@
 			 &irq_spurious_proc_fops, (void *)(long)irq);
 }
 
+void unregister_irq_proc(unsigned int irq, struct irq_desc *desc)
+{
+	char name [MAX_NAMELEN];
+
+	if (!root_irq_dir || !desc->dir)
+		return;
+#ifdef CONFIG_SMP
+	remove_proc_entry("smp_affinity", desc->dir);
+	remove_proc_entry("affinity_hint", desc->dir);
+	remove_proc_entry("node", desc->dir);
+#endif
+	remove_proc_entry("spurious", desc->dir);
+
+	memset(name, 0, MAX_NAMELEN);
+	sprintf(name, "%u", irq);
+	remove_proc_entry(name, root_irq_dir);
+}
+
 #undef MAX_NAMELEN
 
 void unregister_handler_proc(unsigned int irq, struct irqaction *action)
diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c
index 090c376..891115a 100644
--- a/kernel/irq/resend.c
+++ b/kernel/irq/resend.c
@@ -60,7 +60,7 @@
 	/*
 	 * Make sure the interrupt is enabled, before resending it:
 	 */
-	desc->chip->enable(irq);
+	desc->irq_data.chip->irq_enable(&desc->irq_data);
 
 	/*
 	 * We do not resend level type interrupts. Level type
@@ -70,7 +70,8 @@
 	if ((status & (IRQ_LEVEL | IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) {
 		desc->status = (status & ~IRQ_PENDING) | IRQ_REPLAY;
 
-		if (!desc->chip->retrigger || !desc->chip->retrigger(irq)) {
+		if (!desc->irq_data.chip->irq_retrigger ||
+		    !desc->irq_data.chip->irq_retrigger(&desc->irq_data)) {
 #ifdef CONFIG_HARDIRQS_SW_RESEND
 			/* Set it pending and activate the softirq: */
 			set_bit(irq, irqs_resend);
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 89fb90a..3089d3b 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -14,6 +14,8 @@
 #include <linux/moduleparam.h>
 #include <linux/timer.h>
 
+#include "internals.h"
+
 static int irqfixup __read_mostly;
 
 #define POLL_SPURIOUS_IRQ_INTERVAL (HZ/10)
@@ -78,8 +80,8 @@
 	 * If we did actual work for the real IRQ line we must let the
 	 * IRQ controller clean up too
 	 */
-	if (work && desc->chip && desc->chip->end)
-		desc->chip->end(irq);
+	if (work)
+		irq_end(irq, desc);
 	raw_spin_unlock(&desc->lock);
 
 	return ok;
@@ -254,7 +256,7 @@
 		printk(KERN_EMERG "Disabling IRQ #%d\n", irq);
 		desc->status |= IRQ_DISABLED | IRQ_SPURIOUS_DISABLED;
 		desc->depth++;
-		desc->chip->disable(irq);
+		desc->irq_data.chip->irq_disable(&desc->irq_data);
 
 		mod_timer(&poll_spurious_irq_timer,
 			  jiffies + POLL_SPURIOUS_IRQ_INTERVAL);
diff --git a/kernel/irq_work.c b/kernel/irq_work.c
new file mode 100644
index 0000000..f16763f
--- /dev/null
+++ b/kernel/irq_work.c
@@ -0,0 +1,164 @@
+/*
+ * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *
+ * Provides a framework for enqueueing and running callbacks from hardirq
+ * context. The enqueueing is NMI-safe.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/irq_work.h>
+#include <linux/hardirq.h>
+
+/*
+ * An entry can be in one of four states:
+ *
+ * free	     NULL, 0 -> {claimed}       : free to be used
+ * claimed   NULL, 3 -> {pending}       : claimed to be enqueued
+ * pending   next, 3 -> {busy}          : queued, pending callback
+ * busy      NULL, 2 -> {free, claimed} : callback in progress, can be claimed
+ *
+ * We use the lower two bits of the next pointer to keep PENDING and BUSY
+ * flags.
+ */
+
+#define IRQ_WORK_PENDING	1UL
+#define IRQ_WORK_BUSY		2UL
+#define IRQ_WORK_FLAGS		3UL
+
+static inline bool irq_work_is_set(struct irq_work *entry, int flags)
+{
+	return (unsigned long)entry->next & flags;
+}
+
+static inline struct irq_work *irq_work_next(struct irq_work *entry)
+{
+	unsigned long next = (unsigned long)entry->next;
+	next &= ~IRQ_WORK_FLAGS;
+	return (struct irq_work *)next;
+}
+
+static inline struct irq_work *next_flags(struct irq_work *entry, int flags)
+{
+	unsigned long next = (unsigned long)entry;
+	next |= flags;
+	return (struct irq_work *)next;
+}
+
+static DEFINE_PER_CPU(struct irq_work *, irq_work_list);
+
+/*
+ * Claim the entry so that no one else will poke at it.
+ */
+static bool irq_work_claim(struct irq_work *entry)
+{
+	struct irq_work *next, *nflags;
+
+	do {
+		next = entry->next;
+		if ((unsigned long)next & IRQ_WORK_PENDING)
+			return false;
+		nflags = next_flags(next, IRQ_WORK_FLAGS);
+	} while (cmpxchg(&entry->next, next, nflags) != next);
+
+	return true;
+}
+
+
+void __weak arch_irq_work_raise(void)
+{
+	/*
+	 * Lame architectures will get the timer tick callback
+	 */
+}
+
+/*
+ * Queue the entry and raise the IPI if needed.
+ */
+static void __irq_work_queue(struct irq_work *entry)
+{
+	struct irq_work **head, *next;
+
+	head = &get_cpu_var(irq_work_list);
+
+	do {
+		next = *head;
+		/* Can assign non-atomic because we keep the flags set. */
+		entry->next = next_flags(next, IRQ_WORK_FLAGS);
+	} while (cmpxchg(head, next, entry) != next);
+
+	/* The list was empty, raise self-interrupt to start processing. */
+	if (!irq_work_next(entry))
+		arch_irq_work_raise();
+
+	put_cpu_var(irq_work_list);
+}
+
+/*
+ * Enqueue the irq_work @entry, returns true on success, failure when the
+ * @entry was already enqueued by someone else.
+ *
+ * Can be re-enqueued while the callback is still in progress.
+ */
+bool irq_work_queue(struct irq_work *entry)
+{
+	if (!irq_work_claim(entry)) {
+		/*
+		 * Already enqueued, can't do!
+		 */
+		return false;
+	}
+
+	__irq_work_queue(entry);
+	return true;
+}
+EXPORT_SYMBOL_GPL(irq_work_queue);
+
+/*
+ * Run the irq_work entries on this cpu. Requires to be ran from hardirq
+ * context with local IRQs disabled.
+ */
+void irq_work_run(void)
+{
+	struct irq_work *list, **head;
+
+	head = &__get_cpu_var(irq_work_list);
+	if (*head == NULL)
+		return;
+
+	BUG_ON(!in_irq());
+	BUG_ON(!irqs_disabled());
+
+	list = xchg(head, NULL);
+	while (list != NULL) {
+		struct irq_work *entry = list;
+
+		list = irq_work_next(list);
+
+		/*
+		 * Clear the PENDING bit, after this point the @entry
+		 * can be re-used.
+		 */
+		entry->next = next_flags(NULL, IRQ_WORK_BUSY);
+		entry->func(entry);
+		/*
+		 * Clear the BUSY bit and return to the free state if
+		 * no-one else claimed it meanwhile.
+		 */
+		cmpxchg(&entry->next, next_flags(NULL, IRQ_WORK_BUSY), NULL);
+	}
+}
+EXPORT_SYMBOL_GPL(irq_work_run);
+
+/*
+ * Synchronize against the irq_work @entry, ensures the entry is not
+ * currently in use.
+ */
+void irq_work_sync(struct irq_work *entry)
+{
+	WARN_ON_ONCE(irqs_disabled());
+
+	while (irq_work_is_set(entry, IRQ_WORK_BUSY))
+		cpu_relax();
+}
+EXPORT_SYMBOL_GPL(irq_work_sync);
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
new file mode 100644
index 0000000..7be868b
--- /dev/null
+++ b/kernel/jump_label.c
@@ -0,0 +1,429 @@
+/*
+ * jump label support
+ *
+ * Copyright (C) 2009 Jason Baron <jbaron@redhat.com>
+ *
+ */
+#include <linux/jump_label.h>
+#include <linux/memory.h>
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/jhash.h>
+#include <linux/slab.h>
+#include <linux/sort.h>
+#include <linux/err.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+#define JUMP_LABEL_HASH_BITS 6
+#define JUMP_LABEL_TABLE_SIZE (1 << JUMP_LABEL_HASH_BITS)
+static struct hlist_head jump_label_table[JUMP_LABEL_TABLE_SIZE];
+
+/* mutex to protect coming/going of the the jump_label table */
+static DEFINE_MUTEX(jump_label_mutex);
+
+struct jump_label_entry {
+	struct hlist_node hlist;
+	struct jump_entry *table;
+	int nr_entries;
+	/* hang modules off here */
+	struct hlist_head modules;
+	unsigned long key;
+};
+
+struct jump_label_module_entry {
+	struct hlist_node hlist;
+	struct jump_entry *table;
+	int nr_entries;
+	struct module *mod;
+};
+
+static int jump_label_cmp(const void *a, const void *b)
+{
+	const struct jump_entry *jea = a;
+	const struct jump_entry *jeb = b;
+
+	if (jea->key < jeb->key)
+		return -1;
+
+	if (jea->key > jeb->key)
+		return 1;
+
+	return 0;
+}
+
+static void
+sort_jump_label_entries(struct jump_entry *start, struct jump_entry *stop)
+{
+	unsigned long size;
+
+	size = (((unsigned long)stop - (unsigned long)start)
+					/ sizeof(struct jump_entry));
+	sort(start, size, sizeof(struct jump_entry), jump_label_cmp, NULL);
+}
+
+static struct jump_label_entry *get_jump_label_entry(jump_label_t key)
+{
+	struct hlist_head *head;
+	struct hlist_node *node;
+	struct jump_label_entry *e;
+	u32 hash = jhash((void *)&key, sizeof(jump_label_t), 0);
+
+	head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)];
+	hlist_for_each_entry(e, node, head, hlist) {
+		if (key == e->key)
+			return e;
+	}
+	return NULL;
+}
+
+static struct jump_label_entry *
+add_jump_label_entry(jump_label_t key, int nr_entries, struct jump_entry *table)
+{
+	struct hlist_head *head;
+	struct jump_label_entry *e;
+	u32 hash;
+
+	e = get_jump_label_entry(key);
+	if (e)
+		return ERR_PTR(-EEXIST);
+
+	e = kmalloc(sizeof(struct jump_label_entry), GFP_KERNEL);
+	if (!e)
+		return ERR_PTR(-ENOMEM);
+
+	hash = jhash((void *)&key, sizeof(jump_label_t), 0);
+	head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)];
+	e->key = key;
+	e->table = table;
+	e->nr_entries = nr_entries;
+	INIT_HLIST_HEAD(&(e->modules));
+	hlist_add_head(&e->hlist, head);
+	return e;
+}
+
+static int
+build_jump_label_hashtable(struct jump_entry *start, struct jump_entry *stop)
+{
+	struct jump_entry *iter, *iter_begin;
+	struct jump_label_entry *entry;
+	int count;
+
+	sort_jump_label_entries(start, stop);
+	iter = start;
+	while (iter < stop) {
+		entry = get_jump_label_entry(iter->key);
+		if (!entry) {
+			iter_begin = iter;
+			count = 0;
+			while ((iter < stop) &&
+				(iter->key == iter_begin->key)) {
+				iter++;
+				count++;
+			}
+			entry = add_jump_label_entry(iter_begin->key,
+							count, iter_begin);
+			if (IS_ERR(entry))
+				return PTR_ERR(entry);
+		 } else {
+			WARN_ONCE(1, KERN_ERR "build_jump_hashtable: unexpected entry!\n");
+			return -1;
+		}
+	}
+	return 0;
+}
+
+/***
+ * jump_label_update - update jump label text
+ * @key -  key value associated with a a jump label
+ * @type - enum set to JUMP_LABEL_ENABLE or JUMP_LABEL_DISABLE
+ *
+ * Will enable/disable the jump for jump label @key, depending on the
+ * value of @type.
+ *
+ */
+
+void jump_label_update(unsigned long key, enum jump_label_type type)
+{
+	struct jump_entry *iter;
+	struct jump_label_entry *entry;
+	struct hlist_node *module_node;
+	struct jump_label_module_entry *e_module;
+	int count;
+
+	mutex_lock(&jump_label_mutex);
+	entry = get_jump_label_entry((jump_label_t)key);
+	if (entry) {
+		count = entry->nr_entries;
+		iter = entry->table;
+		while (count--) {
+			if (kernel_text_address(iter->code))
+				arch_jump_label_transform(iter, type);
+			iter++;
+		}
+		/* eanble/disable jump labels in modules */
+		hlist_for_each_entry(e_module, module_node, &(entry->modules),
+							hlist) {
+			count = e_module->nr_entries;
+			iter = e_module->table;
+			while (count--) {
+				if (kernel_text_address(iter->code))
+					arch_jump_label_transform(iter, type);
+				iter++;
+			}
+		}
+	}
+	mutex_unlock(&jump_label_mutex);
+}
+
+static int addr_conflict(struct jump_entry *entry, void *start, void *end)
+{
+	if (entry->code <= (unsigned long)end &&
+		entry->code + JUMP_LABEL_NOP_SIZE > (unsigned long)start)
+		return 1;
+
+	return 0;
+}
+
+#ifdef CONFIG_MODULES
+
+static int module_conflict(void *start, void *end)
+{
+	struct hlist_head *head;
+	struct hlist_node *node, *node_next, *module_node, *module_node_next;
+	struct jump_label_entry *e;
+	struct jump_label_module_entry *e_module;
+	struct jump_entry *iter;
+	int i, count;
+	int conflict = 0;
+
+	for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) {
+		head = &jump_label_table[i];
+		hlist_for_each_entry_safe(e, node, node_next, head, hlist) {
+			hlist_for_each_entry_safe(e_module, module_node,
+							module_node_next,
+							&(e->modules), hlist) {
+				count = e_module->nr_entries;
+				iter = e_module->table;
+				while (count--) {
+					if (addr_conflict(iter, start, end)) {
+						conflict = 1;
+						goto out;
+					}
+					iter++;
+				}
+			}
+		}
+	}
+out:
+	return conflict;
+}
+
+#endif
+
+/***
+ * jump_label_text_reserved - check if addr range is reserved
+ * @start: start text addr
+ * @end: end text addr
+ *
+ * checks if the text addr located between @start and @end
+ * overlaps with any of the jump label patch addresses. Code
+ * that wants to modify kernel text should first verify that
+ * it does not overlap with any of the jump label addresses.
+ *
+ * returns 1 if there is an overlap, 0 otherwise
+ */
+int jump_label_text_reserved(void *start, void *end)
+{
+	struct jump_entry *iter;
+	struct jump_entry *iter_start = __start___jump_table;
+	struct jump_entry *iter_stop = __start___jump_table;
+	int conflict = 0;
+
+	mutex_lock(&jump_label_mutex);
+	iter = iter_start;
+	while (iter < iter_stop) {
+		if (addr_conflict(iter, start, end)) {
+			conflict = 1;
+			goto out;
+		}
+		iter++;
+	}
+
+	/* now check modules */
+#ifdef CONFIG_MODULES
+	conflict = module_conflict(start, end);
+#endif
+out:
+	mutex_unlock(&jump_label_mutex);
+	return conflict;
+}
+
+static __init int init_jump_label(void)
+{
+	int ret;
+	struct jump_entry *iter_start = __start___jump_table;
+	struct jump_entry *iter_stop = __stop___jump_table;
+	struct jump_entry *iter;
+
+	mutex_lock(&jump_label_mutex);
+	ret = build_jump_label_hashtable(__start___jump_table,
+					 __stop___jump_table);
+	iter = iter_start;
+	while (iter < iter_stop) {
+		arch_jump_label_text_poke_early(iter->code);
+		iter++;
+	}
+	mutex_unlock(&jump_label_mutex);
+	return ret;
+}
+early_initcall(init_jump_label);
+
+#ifdef CONFIG_MODULES
+
+static struct jump_label_module_entry *
+add_jump_label_module_entry(struct jump_label_entry *entry,
+			    struct jump_entry *iter_begin,
+			    int count, struct module *mod)
+{
+	struct jump_label_module_entry *e;
+
+	e = kmalloc(sizeof(struct jump_label_module_entry), GFP_KERNEL);
+	if (!e)
+		return ERR_PTR(-ENOMEM);
+	e->mod = mod;
+	e->nr_entries = count;
+	e->table = iter_begin;
+	hlist_add_head(&e->hlist, &entry->modules);
+	return e;
+}
+
+static int add_jump_label_module(struct module *mod)
+{
+	struct jump_entry *iter, *iter_begin;
+	struct jump_label_entry *entry;
+	struct jump_label_module_entry *module_entry;
+	int count;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return 0;
+
+	sort_jump_label_entries(mod->jump_entries,
+				mod->jump_entries + mod->num_jump_entries);
+	iter = mod->jump_entries;
+	while (iter < mod->jump_entries + mod->num_jump_entries) {
+		entry = get_jump_label_entry(iter->key);
+		iter_begin = iter;
+		count = 0;
+		while ((iter < mod->jump_entries + mod->num_jump_entries) &&
+			(iter->key == iter_begin->key)) {
+				iter++;
+				count++;
+		}
+		if (!entry) {
+			entry = add_jump_label_entry(iter_begin->key, 0, NULL);
+			if (IS_ERR(entry))
+				return PTR_ERR(entry);
+		}
+		module_entry = add_jump_label_module_entry(entry, iter_begin,
+							   count, mod);
+		if (IS_ERR(module_entry))
+			return PTR_ERR(module_entry);
+	}
+	return 0;
+}
+
+static void remove_jump_label_module(struct module *mod)
+{
+	struct hlist_head *head;
+	struct hlist_node *node, *node_next, *module_node, *module_node_next;
+	struct jump_label_entry *e;
+	struct jump_label_module_entry *e_module;
+	int i;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return;
+
+	for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) {
+		head = &jump_label_table[i];
+		hlist_for_each_entry_safe(e, node, node_next, head, hlist) {
+			hlist_for_each_entry_safe(e_module, module_node,
+						  module_node_next,
+						  &(e->modules), hlist) {
+				if (e_module->mod == mod) {
+					hlist_del(&e_module->hlist);
+					kfree(e_module);
+				}
+			}
+			if (hlist_empty(&e->modules) && (e->nr_entries == 0)) {
+				hlist_del(&e->hlist);
+				kfree(e);
+			}
+		}
+	}
+}
+
+static int
+jump_label_module_notify(struct notifier_block *self, unsigned long val,
+			 void *data)
+{
+	struct module *mod = data;
+	int ret = 0;
+
+	switch (val) {
+	case MODULE_STATE_COMING:
+		mutex_lock(&jump_label_mutex);
+		ret = add_jump_label_module(mod);
+		if (ret)
+			remove_jump_label_module(mod);
+		mutex_unlock(&jump_label_mutex);
+		break;
+	case MODULE_STATE_GOING:
+		mutex_lock(&jump_label_mutex);
+		remove_jump_label_module(mod);
+		mutex_unlock(&jump_label_mutex);
+		break;
+	}
+	return ret;
+}
+
+/***
+ * apply_jump_label_nops - patch module jump labels with arch_get_jump_label_nop()
+ * @mod: module to patch
+ *
+ * Allow for run-time selection of the optimal nops. Before the module
+ * loads patch these with arch_get_jump_label_nop(), which is specified by
+ * the arch specific jump label code.
+ */
+void jump_label_apply_nops(struct module *mod)
+{
+	struct jump_entry *iter;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return;
+
+	iter = mod->jump_entries;
+	while (iter < mod->jump_entries + mod->num_jump_entries) {
+		arch_jump_label_text_poke_early(iter->code);
+		iter++;
+	}
+}
+
+struct notifier_block jump_label_module_nb = {
+	.notifier_call = jump_label_module_notify,
+	.priority = 0,
+};
+
+static __init int init_jump_label_module(void)
+{
+	return register_module_notifier(&jump_label_module_nb);
+}
+early_initcall(init_jump_label_module);
+
+#endif /* CONFIG_MODULES */
+
+#endif
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 282035f..ec4210c 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -47,6 +47,7 @@
 #include <linux/memory.h>
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
+#include <linux/jump_label.h>
 
 #include <asm-generic/sections.h>
 #include <asm/cacheflush.h>
@@ -399,7 +400,7 @@
  * Return an optimized kprobe whose optimizing code replaces
  * instructions including addr (exclude breakpoint).
  */
-struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr)
+static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr)
 {
 	int i;
 	struct kprobe *p = NULL;
@@ -831,6 +832,7 @@
 
 void __kprobes kretprobe_hash_lock(struct task_struct *tsk,
 			 struct hlist_head **head, unsigned long *flags)
+__acquires(hlist_lock)
 {
 	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
 	spinlock_t *hlist_lock;
@@ -842,6 +844,7 @@
 
 static void __kprobes kretprobe_table_lock(unsigned long hash,
 	unsigned long *flags)
+__acquires(hlist_lock)
 {
 	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
 	spin_lock_irqsave(hlist_lock, *flags);
@@ -849,6 +852,7 @@
 
 void __kprobes kretprobe_hash_unlock(struct task_struct *tsk,
 	unsigned long *flags)
+__releases(hlist_lock)
 {
 	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
 	spinlock_t *hlist_lock;
@@ -857,7 +861,9 @@
 	spin_unlock_irqrestore(hlist_lock, *flags);
 }
 
-void __kprobes kretprobe_table_unlock(unsigned long hash, unsigned long *flags)
+static void __kprobes kretprobe_table_unlock(unsigned long hash,
+       unsigned long *flags)
+__releases(hlist_lock)
 {
 	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
 	spin_unlock_irqrestore(hlist_lock, *flags);
@@ -1141,7 +1147,8 @@
 	preempt_disable();
 	if (!kernel_text_address((unsigned long) p->addr) ||
 	    in_kprobes_functions((unsigned long) p->addr) ||
-	    ftrace_text_reserved(p->addr, p->addr)) {
+	    ftrace_text_reserved(p->addr, p->addr) ||
+	    jump_label_text_reserved(p->addr, p->addr)) {
 		preempt_enable();
 		return -EINVAL;
 	}
@@ -1339,18 +1346,19 @@
 	if (num <= 0)
 		return -EINVAL;
 	for (i = 0; i < num; i++) {
-		unsigned long addr;
+		unsigned long addr, offset;
 		jp = jps[i];
 		addr = arch_deref_entry_point(jp->entry);
 
-		if (!kernel_text_address(addr))
-			ret = -EINVAL;
-		else {
-			/* Todo: Verify probepoint is a function entry point */
+		/* Verify probepoint is a function entry point */
+		if (kallsyms_lookup_size_offset(addr, NULL, &offset) &&
+		    offset == 0) {
 			jp->kp.pre_handler = setjmp_pre_handler;
 			jp->kp.break_handler = longjmp_break_handler;
 			ret = register_kprobe(&jp->kp);
-		}
+		} else
+			ret = -EINVAL;
+
 		if (ret < 0) {
 			if (i > 0)
 				unregister_jprobes(jps, i);
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index f2852a5..42ba65d 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -639,6 +639,16 @@
 	}
 #endif
 
+	if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
+		debug_locks_off();
+		printk(KERN_ERR
+			"BUG: looking up invalid subclass: %u\n", subclass);
+		printk(KERN_ERR
+			"turning off the locking correctness validator.\n");
+		dump_stack();
+		return NULL;
+	}
+
 	/*
 	 * Static locks do not have their class-keys yet - for them the key
 	 * is the lock object itself:
@@ -774,7 +784,9 @@
 	raw_local_irq_restore(flags);
 
 	if (!subclass || force)
-		lock->class_cache = class;
+		lock->class_cache[0] = class;
+	else if (subclass < NR_LOCKDEP_CACHING_CLASSES)
+		lock->class_cache[subclass] = class;
 
 	if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass))
 		return NULL;
@@ -2679,7 +2691,11 @@
 void lockdep_init_map(struct lockdep_map *lock, const char *name,
 		      struct lock_class_key *key, int subclass)
 {
-	lock->class_cache = NULL;
+	int i;
+
+	for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
+		lock->class_cache[i] = NULL;
+
 #ifdef CONFIG_LOCK_STAT
 	lock->cpu = raw_smp_processor_id();
 #endif
@@ -2739,21 +2755,13 @@
 	if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
 		return 0;
 
-	if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
-		debug_locks_off();
-		printk("BUG: MAX_LOCKDEP_SUBCLASSES too low!\n");
-		printk("turning off the locking correctness validator.\n");
-		dump_stack();
-		return 0;
-	}
-
 	if (lock->key == &__lockdep_no_validate__)
 		check = 1;
 
-	if (!subclass)
-		class = lock->class_cache;
+	if (subclass < NR_LOCKDEP_CACHING_CLASSES)
+		class = lock->class_cache[subclass];
 	/*
-	 * Not cached yet or subclass?
+	 * Not cached?
 	 */
 	if (unlikely(!class)) {
 		class = register_lock_class(lock, subclass, 0);
@@ -2918,7 +2926,7 @@
 		return 1;
 
 	if (hlock->references) {
-		struct lock_class *class = lock->class_cache;
+		struct lock_class *class = lock->class_cache[0];
 
 		if (!class)
 			class = look_up_lock_class(lock, 0);
@@ -3559,7 +3567,12 @@
 		if (list_empty(head))
 			continue;
 		list_for_each_entry_safe(class, next, head, hash_entry) {
-			if (unlikely(class == lock->class_cache)) {
+			int match = 0;
+
+			for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++)
+				match |= class == lock->class_cache[j];
+
+			if (unlikely(match)) {
 				if (debug_locks_off_graph_unlock())
 					WARN_ON(1);
 				goto out_restore;
@@ -3775,7 +3788,7 @@
  * Careful: only use this function if you are sure that
  * the task cannot run in parallel!
  */
-void __debug_show_held_locks(struct task_struct *task)
+void debug_show_held_locks(struct task_struct *task)
 {
 	if (unlikely(!debug_locks)) {
 		printk("INFO: lockdep is turned off.\n");
@@ -3783,12 +3796,6 @@
 	}
 	lockdep_print_held_locks(task);
 }
-EXPORT_SYMBOL_GPL(__debug_show_held_locks);
-
-void debug_show_held_locks(struct task_struct *task)
-{
-		__debug_show_held_locks(task);
-}
 EXPORT_SYMBOL_GPL(debug_show_held_locks);
 
 void lockdep_sys_exit(void)
diff --git a/kernel/module.c b/kernel/module.c
index ccd6419..2df4630 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -55,6 +55,7 @@
 #include <linux/async.h>
 #include <linux/percpu.h>
 #include <linux/kmemleak.h>
+#include <linux/jump_label.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/module.h>
@@ -2309,6 +2310,11 @@
 					sizeof(*mod->tracepoints),
 					&mod->num_tracepoints);
 #endif
+#ifdef HAVE_JUMP_LABEL
+	mod->jump_entries = section_objs(info, "__jump_table",
+					sizeof(*mod->jump_entries),
+					&mod->num_jump_entries);
+#endif
 #ifdef CONFIG_EVENT_TRACING
 	mod->trace_events = section_objs(info, "_ftrace_events",
 					 sizeof(*mod->trace_events),
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index db5b560..f309e80 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -31,24 +31,18 @@
 #include <linux/kernel_stat.h>
 #include <linux/perf_event.h>
 #include <linux/ftrace_event.h>
-#include <linux/hw_breakpoint.h>
 
 #include <asm/irq_regs.h>
 
-/*
- * Each CPU has a list of per CPU events:
- */
-static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context);
-
-int perf_max_events __read_mostly = 1;
-static int perf_reserved_percpu __read_mostly;
-static int perf_overcommit __read_mostly = 1;
-
-static atomic_t nr_events __read_mostly;
+atomic_t perf_task_events __read_mostly;
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 
+static LIST_HEAD(pmus);
+static DEFINE_MUTEX(pmus_lock);
+static struct srcu_struct pmus_srcu;
+
 /*
  * perf event paranoia level:
  *  -1 - not paranoid at all
@@ -67,36 +61,43 @@
 
 static atomic64_t perf_event_id;
 
-/*
- * Lock for (sysadmin-configurable) event reservations:
- */
-static DEFINE_SPINLOCK(perf_resource_lock);
-
-/*
- * Architecture provided APIs - weak aliases:
- */
-extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	return NULL;
-}
-
-void __weak hw_perf_disable(void)		{ barrier(); }
-void __weak hw_perf_enable(void)		{ barrier(); }
-
 void __weak perf_event_print_debug(void)	{ }
 
-static DEFINE_PER_CPU(int, perf_disable_count);
-
-void perf_disable(void)
+extern __weak const char *perf_pmu_name(void)
 {
-	if (!__get_cpu_var(perf_disable_count)++)
-		hw_perf_disable();
+	return "pmu";
 }
 
-void perf_enable(void)
+void perf_pmu_disable(struct pmu *pmu)
 {
-	if (!--__get_cpu_var(perf_disable_count))
-		hw_perf_enable();
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!(*count)++)
+		pmu->pmu_disable(pmu);
+}
+
+void perf_pmu_enable(struct pmu *pmu)
+{
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!--(*count))
+		pmu->pmu_enable(pmu);
+}
+
+static DEFINE_PER_CPU(struct list_head, rotation_list);
+
+/*
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
+ */
+static void perf_pmu_rotate_start(struct pmu *pmu)
+{
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+	struct list_head *head = &__get_cpu_var(rotation_list);
+
+	WARN_ON(!irqs_disabled());
+
+	if (list_empty(&cpuctx->rotation_list))
+		list_add(&cpuctx->rotation_list, head);
 }
 
 static void get_ctx(struct perf_event_context *ctx)
@@ -151,13 +152,13 @@
  * the context could get moved to another task.
  */
 static struct perf_event_context *
-perf_lock_task_context(struct task_struct *task, unsigned long *flags)
+perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags)
 {
 	struct perf_event_context *ctx;
 
 	rcu_read_lock();
- retry:
-	ctx = rcu_dereference(task->perf_event_ctxp);
+retry:
+	ctx = rcu_dereference(task->perf_event_ctxp[ctxn]);
 	if (ctx) {
 		/*
 		 * If this context is a clone of another, it might
@@ -170,7 +171,7 @@
 		 * can't get swapped on us any more.
 		 */
 		raw_spin_lock_irqsave(&ctx->lock, *flags);
-		if (ctx != rcu_dereference(task->perf_event_ctxp)) {
+		if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) {
 			raw_spin_unlock_irqrestore(&ctx->lock, *flags);
 			goto retry;
 		}
@@ -189,12 +190,13 @@
  * can't get swapped to another task.  This also increments its
  * reference count so that the context can't get freed.
  */
-static struct perf_event_context *perf_pin_task_context(struct task_struct *task)
+static struct perf_event_context *
+perf_pin_task_context(struct task_struct *task, int ctxn)
 {
 	struct perf_event_context *ctx;
 	unsigned long flags;
 
-	ctx = perf_lock_task_context(task, &flags);
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		++ctx->pin_count;
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -302,6 +304,8 @@
 	}
 
 	list_add_rcu(&event->event_entry, &ctx->event_list);
+	if (!ctx->nr_events)
+		perf_pmu_rotate_start(ctx->pmu);
 	ctx->nr_events++;
 	if (event->attr.inherit_stat)
 		ctx->nr_stat++;
@@ -311,7 +315,12 @@
 {
 	struct perf_event *group_leader = event->group_leader;
 
-	WARN_ON_ONCE(event->attach_state & PERF_ATTACH_GROUP);
+	/*
+	 * We can have double attach due to group movement in perf_event_open.
+	 */
+	if (event->attach_state & PERF_ATTACH_GROUP)
+		return;
+
 	event->attach_state |= PERF_ATTACH_GROUP;
 
 	if (group_leader == event)
@@ -408,8 +417,8 @@
 	return event->cpu == -1 || event->cpu == smp_processor_id();
 }
 
-static void
-event_sched_out(struct perf_event *event,
+static int
+__event_sched_out(struct perf_event *event,
 		  struct perf_cpu_context *cpuctx,
 		  struct perf_event_context *ctx)
 {
@@ -428,15 +437,14 @@
 	}
 
 	if (event->state != PERF_EVENT_STATE_ACTIVE)
-		return;
+		return 0;
 
 	event->state = PERF_EVENT_STATE_INACTIVE;
 	if (event->pending_disable) {
 		event->pending_disable = 0;
 		event->state = PERF_EVENT_STATE_OFF;
 	}
-	event->tstamp_stopped = ctx->time;
-	event->pmu->disable(event);
+	event->pmu->del(event, 0);
 	event->oncpu = -1;
 
 	if (!is_software_event(event))
@@ -444,6 +452,19 @@
 	ctx->nr_active--;
 	if (event->attr.exclusive || !cpuctx->active_oncpu)
 		cpuctx->exclusive = 0;
+	return 1;
+}
+
+static void
+event_sched_out(struct perf_event *event,
+		  struct perf_cpu_context *cpuctx,
+		  struct perf_event_context *ctx)
+{
+	int ret;
+
+	ret = __event_sched_out(event, cpuctx, ctx);
+	if (ret)
+		event->tstamp_stopped = ctx->time;
 }
 
 static void
@@ -466,6 +487,12 @@
 		cpuctx->exclusive = 0;
 }
 
+static inline struct perf_cpu_context *
+__get_cpu_context(struct perf_event_context *ctx)
+{
+	return this_cpu_ptr(ctx->pmu->pmu_cpu_context);
+}
+
 /*
  * Cross CPU call to remove a performance event
  *
@@ -474,9 +501,9 @@
  */
 static void __perf_event_remove_from_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -487,27 +514,11 @@
 		return;
 
 	raw_spin_lock(&ctx->lock);
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level.
-	 */
-	perf_disable();
 
 	event_sched_out(event, cpuctx, ctx);
 
 	list_del_event(event, ctx);
 
-	if (!ctx->task) {
-		/*
-		 * Allow more per task events with respect to the
-		 * reservation:
-		 */
-		cpuctx->max_pertask =
-			min(perf_max_events - ctx->nr_events,
-			    perf_max_events - perf_reserved_percpu);
-	}
-
-	perf_enable();
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -572,8 +583,8 @@
 static void __perf_event_disable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a per-task event, need to check whether this
@@ -628,7 +639,7 @@
 		return;
 	}
 
- retry:
+retry:
 	task_oncpu_function_call(task, __perf_event_disable, event);
 
 	raw_spin_lock_irq(&ctx->lock);
@@ -653,7 +664,7 @@
 }
 
 static int
-event_sched_in(struct perf_event *event,
+__event_sched_in(struct perf_event *event,
 		 struct perf_cpu_context *cpuctx,
 		 struct perf_event_context *ctx)
 {
@@ -667,14 +678,12 @@
 	 */
 	smp_wmb();
 
-	if (event->pmu->enable(event)) {
+	if (event->pmu->add(event, PERF_EF_START)) {
 		event->state = PERF_EVENT_STATE_INACTIVE;
 		event->oncpu = -1;
 		return -EAGAIN;
 	}
 
-	event->tstamp_running += ctx->time - event->tstamp_stopped;
-
 	if (!is_software_event(event))
 		cpuctx->active_oncpu++;
 	ctx->nr_active++;
@@ -685,28 +694,56 @@
 	return 0;
 }
 
+static inline int
+event_sched_in(struct perf_event *event,
+		 struct perf_cpu_context *cpuctx,
+		 struct perf_event_context *ctx)
+{
+	int ret = __event_sched_in(event, cpuctx, ctx);
+	if (ret)
+		return ret;
+	event->tstamp_running += ctx->time - event->tstamp_stopped;
+	return 0;
+}
+
+static void
+group_commit_event_sched_in(struct perf_event *group_event,
+	       struct perf_cpu_context *cpuctx,
+	       struct perf_event_context *ctx)
+{
+	struct perf_event *event;
+	u64 now = ctx->time;
+
+	group_event->tstamp_running += now - group_event->tstamp_stopped;
+	/*
+	 * Schedule in siblings as one group (if any):
+	 */
+	list_for_each_entry(event, &group_event->sibling_list, group_entry) {
+		event->tstamp_running += now - event->tstamp_stopped;
+	}
+}
+
 static int
 group_sched_in(struct perf_event *group_event,
 	       struct perf_cpu_context *cpuctx,
 	       struct perf_event_context *ctx)
 {
 	struct perf_event *event, *partial_group = NULL;
-	const struct pmu *pmu = group_event->pmu;
-	bool txn = false;
+	struct pmu *pmu = group_event->pmu;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
 		return 0;
 
-	/* Check if group transaction availabe */
-	if (pmu->start_txn)
-		txn = true;
+	pmu->start_txn(pmu);
 
-	if (txn)
-		pmu->start_txn(pmu);
-
-	if (event_sched_in(group_event, cpuctx, ctx)) {
-		if (txn)
-			pmu->cancel_txn(pmu);
+	/*
+	 * use __event_sched_in() to delay updating tstamp_running
+	 * until the transaction is committed. In case of failure
+	 * we will keep an unmodified tstamp_running which is a
+	 * requirement to get correct timing information
+	 */
+	if (__event_sched_in(group_event, cpuctx, ctx)) {
+		pmu->cancel_txn(pmu);
 		return -EAGAIN;
 	}
 
@@ -714,29 +751,33 @@
 	 * Schedule in siblings as one group (if any):
 	 */
 	list_for_each_entry(event, &group_event->sibling_list, group_entry) {
-		if (event_sched_in(event, cpuctx, ctx)) {
+		if (__event_sched_in(event, cpuctx, ctx)) {
 			partial_group = event;
 			goto group_error;
 		}
 	}
 
-	if (!txn || !pmu->commit_txn(pmu))
+	if (!pmu->commit_txn(pmu)) {
+		/* commit tstamp_running */
+		group_commit_event_sched_in(group_event, cpuctx, ctx);
 		return 0;
-
+	}
 group_error:
 	/*
 	 * Groups can be scheduled in as one unit only, so undo any
 	 * partial group before returning:
+	 *
+	 * use __event_sched_out() to avoid updating tstamp_stopped
+	 * because the event never actually ran
 	 */
 	list_for_each_entry(event, &group_event->sibling_list, group_entry) {
 		if (event == partial_group)
 			break;
-		event_sched_out(event, cpuctx, ctx);
+		__event_sched_out(event, cpuctx, ctx);
 	}
-	event_sched_out(group_event, cpuctx, ctx);
+	__event_sched_out(group_event, cpuctx, ctx);
 
-	if (txn)
-		pmu->cancel_txn(pmu);
+	pmu->cancel_txn(pmu);
 
 	return -EAGAIN;
 }
@@ -789,10 +830,10 @@
  */
 static void __perf_install_in_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -812,12 +853,6 @@
 	ctx->is_active = 1;
 	update_context_time(ctx);
 
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level. NOP for non NMI based events.
-	 */
-	perf_disable();
-
 	add_event_to_ctx(event, ctx);
 
 	if (event->cpu != -1 && event->cpu != smp_processor_id())
@@ -855,12 +890,7 @@
 		}
 	}
 
-	if (!err && !ctx->task && cpuctx->max_pertask)
-		cpuctx->max_pertask--;
-
- unlock:
-	perf_enable();
-
+unlock:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -883,6 +913,8 @@
 {
 	struct task_struct *task = ctx->task;
 
+	event->ctx = ctx;
+
 	if (!task) {
 		/*
 		 * Per cpu events are installed via an smp call and
@@ -931,10 +963,12 @@
 
 	event->state = PERF_EVENT_STATE_INACTIVE;
 	event->tstamp_enabled = ctx->time - event->total_time_enabled;
-	list_for_each_entry(sub, &event->sibling_list, group_entry)
-		if (sub->state >= PERF_EVENT_STATE_INACTIVE)
+	list_for_each_entry(sub, &event->sibling_list, group_entry) {
+		if (sub->state >= PERF_EVENT_STATE_INACTIVE) {
 			sub->tstamp_enabled =
 				ctx->time - sub->total_time_enabled;
+		}
+	}
 }
 
 /*
@@ -943,9 +977,9 @@
 static void __perf_event_enable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -979,12 +1013,10 @@
 	if (!group_can_go_on(event, cpuctx, 1)) {
 		err = -EEXIST;
 	} else {
-		perf_disable();
 		if (event == leader)
 			err = group_sched_in(event, cpuctx, ctx);
 		else
 			err = event_sched_in(event, cpuctx, ctx);
-		perf_enable();
 	}
 
 	if (err) {
@@ -1000,7 +1032,7 @@
 		}
 	}
 
- unlock:
+unlock:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1041,7 +1073,7 @@
 	if (event->state == PERF_EVENT_STATE_ERROR)
 		event->state = PERF_EVENT_STATE_OFF;
 
- retry:
+retry:
 	raw_spin_unlock_irq(&ctx->lock);
 	task_oncpu_function_call(task, __perf_event_enable, event);
 
@@ -1061,7 +1093,7 @@
 	if (event->state == PERF_EVENT_STATE_OFF)
 		__perf_event_mark_enabled(event, ctx);
 
- out:
+out:
 	raw_spin_unlock_irq(&ctx->lock);
 }
 
@@ -1092,26 +1124,26 @@
 	struct perf_event *event;
 
 	raw_spin_lock(&ctx->lock);
+	perf_pmu_disable(ctx->pmu);
 	ctx->is_active = 0;
 	if (likely(!ctx->nr_events))
 		goto out;
 	update_context_time(ctx);
 
-	perf_disable();
 	if (!ctx->nr_active)
-		goto out_enable;
+		goto out;
 
-	if (event_type & EVENT_PINNED)
+	if (event_type & EVENT_PINNED) {
 		list_for_each_entry(event, &ctx->pinned_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
+	}
 
-	if (event_type & EVENT_FLEXIBLE)
+	if (event_type & EVENT_FLEXIBLE) {
 		list_for_each_entry(event, &ctx->flexible_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
-
- out_enable:
-	perf_enable();
- out:
+	}
+out:
+	perf_pmu_enable(ctx->pmu);
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1209,34 +1241,25 @@
 	}
 }
 
-/*
- * Called from scheduler to remove the events of the current task,
- * with interrupts disabled.
- *
- * We stop each event and update the event value in event->count.
- *
- * This does not protect us against NMI, but disable()
- * sets the disabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * not restart the event.
- */
-void perf_event_task_sched_out(struct task_struct *task,
-				 struct task_struct *next)
+void perf_event_context_sched_out(struct task_struct *task, int ctxn,
+				  struct task_struct *next)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx = task->perf_event_ctxp[ctxn];
 	struct perf_event_context *next_ctx;
 	struct perf_event_context *parent;
+	struct perf_cpu_context *cpuctx;
 	int do_switch = 1;
 
-	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+	if (likely(!ctx))
+		return;
 
-	if (likely(!ctx || !cpuctx->task_ctx))
+	cpuctx = __get_cpu_context(ctx);
+	if (!cpuctx->task_ctx)
 		return;
 
 	rcu_read_lock();
 	parent = rcu_dereference(ctx->parent_ctx);
-	next_ctx = next->perf_event_ctxp;
+	next_ctx = next->perf_event_ctxp[ctxn];
 	if (parent && next_ctx &&
 	    rcu_dereference(next_ctx->parent_ctx) == parent) {
 		/*
@@ -1255,8 +1278,8 @@
 			 * XXX do we need a memory barrier of sorts
 			 * wrt to rcu_dereference() of perf_event_ctxp
 			 */
-			task->perf_event_ctxp = next_ctx;
-			next->perf_event_ctxp = ctx;
+			task->perf_event_ctxp[ctxn] = next_ctx;
+			next->perf_event_ctxp[ctxn] = ctx;
 			ctx->task = next;
 			next_ctx->task = task;
 			do_switch = 0;
@@ -1274,10 +1297,35 @@
 	}
 }
 
+#define for_each_task_context_nr(ctxn)					\
+	for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++)
+
+/*
+ * Called from scheduler to remove the events of the current task,
+ * with interrupts disabled.
+ *
+ * We stop each event and update the event value in event->count.
+ *
+ * This does not protect us against NMI, but disable()
+ * sets the disabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * not restart the event.
+ */
+void __perf_event_task_sched_out(struct task_struct *task,
+				 struct task_struct *next)
+{
+	int ctxn;
+
+	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+
+	for_each_task_context_nr(ctxn)
+		perf_event_context_sched_out(task, ctxn, next);
+}
+
 static void task_ctx_sched_out(struct perf_event_context *ctx,
 			       enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	if (!cpuctx->task_ctx)
 		return;
@@ -1292,14 +1340,6 @@
 /*
  * Called with IRQs disabled
  */
-static void __perf_event_task_sched_out(struct perf_event_context *ctx)
-{
-	task_ctx_sched_out(ctx, EVENT_ALL);
-}
-
-/*
- * Called with IRQs disabled
- */
 static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx,
 			      enum event_type_t event_type)
 {
@@ -1350,9 +1390,10 @@
 		if (event->cpu != -1 && event->cpu != smp_processor_id())
 			continue;
 
-		if (group_can_go_on(event, cpuctx, can_add_hw))
+		if (group_can_go_on(event, cpuctx, can_add_hw)) {
 			if (group_sched_in(event, cpuctx, ctx))
 				can_add_hw = 0;
+		}
 	}
 }
 
@@ -1368,8 +1409,6 @@
 
 	ctx->timestamp = perf_clock();
 
-	perf_disable();
-
 	/*
 	 * First go through the list and put on any pinned groups
 	 * in order to give them the best chance of going on.
@@ -1381,8 +1420,7 @@
 	if (event_type & EVENT_FLEXIBLE)
 		ctx_flexible_sched_in(ctx, cpuctx);
 
-	perf_enable();
- out:
+out:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1394,43 +1432,28 @@
 	ctx_sched_in(ctx, cpuctx, event_type);
 }
 
-static void task_ctx_sched_in(struct task_struct *task,
+static void task_ctx_sched_in(struct perf_event_context *ctx,
 			      enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
+       	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
+
 	ctx_sched_in(ctx, cpuctx, event_type);
 	cpuctx->task_ctx = ctx;
 }
-/*
- * Called from scheduler to add the events of the current task
- * with interrupts disabled.
- *
- * We restore the event value and then enable it.
- *
- * This does not protect us against NMI, but enable()
- * sets the enabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * keep the event running.
- */
-void perf_event_task_sched_in(struct task_struct *task)
+
+void perf_event_context_sched_in(struct perf_event_context *ctx)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
-
+	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
 
-	perf_disable();
-
+	perf_pmu_disable(ctx->pmu);
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1444,7 +1467,37 @@
 
 	cpuctx->task_ctx = ctx;
 
-	perf_enable();
+	/*
+	 * Since these rotations are per-cpu, we need to ensure the
+	 * cpu-context we got scheduled on is actually rotating.
+	 */
+	perf_pmu_rotate_start(ctx->pmu);
+	perf_pmu_enable(ctx->pmu);
+}
+
+/*
+ * Called from scheduler to add the events of the current task
+ * with interrupts disabled.
+ *
+ * We restore the event value and then enable it.
+ *
+ * This does not protect us against NMI, but enable()
+ * sets the enabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * keep the event running.
+ */
+void __perf_event_task_sched_in(struct task_struct *task)
+{
+	struct perf_event_context *ctx;
+	int ctxn;
+
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (likely(!ctx))
+			continue;
+
+		perf_event_context_sched_in(ctx);
+	}
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1524,22 +1577,6 @@
 	return div64_u64(dividend, divisor);
 }
 
-static void perf_event_stop(struct perf_event *event)
-{
-	if (!event->pmu->stop)
-		return event->pmu->disable(event);
-
-	return event->pmu->stop(event);
-}
-
-static int perf_event_start(struct perf_event *event)
-{
-	if (!event->pmu->start)
-		return event->pmu->enable(event);
-
-	return event->pmu->start(event);
-}
-
 static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -1559,15 +1596,13 @@
 	hwc->sample_period = sample_period;
 
 	if (local64_read(&hwc->period_left) > 8*sample_period) {
-		perf_disable();
-		perf_event_stop(event);
+		event->pmu->stop(event, PERF_EF_UPDATE);
 		local64_set(&hwc->period_left, 0);
-		perf_event_start(event);
-		perf_enable();
+		event->pmu->start(event, PERF_EF_RELOAD);
 	}
 }
 
-static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
+static void perf_ctx_adjust_freq(struct perf_event_context *ctx, u64 period)
 {
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
@@ -1592,23 +1627,19 @@
 		 */
 		if (interrupts == MAX_INTERRUPTS) {
 			perf_log_throttle(event, 1);
-			perf_disable();
-			event->pmu->unthrottle(event);
-			perf_enable();
+			event->pmu->start(event, 0);
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
 			continue;
 
-		perf_disable();
 		event->pmu->read(event);
 		now = local64_read(&event->count);
 		delta = now - hwc->freq_count_stamp;
 		hwc->freq_count_stamp = now;
 
 		if (delta > 0)
-			perf_adjust_period(event, TICK_NSEC, delta);
-		perf_enable();
+			perf_adjust_period(event, period, delta);
 	}
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1626,32 +1657,38 @@
 	raw_spin_unlock(&ctx->lock);
 }
 
-void perf_event_task_tick(struct task_struct *curr)
+/*
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
+ */
+static void perf_rotate_context(struct perf_cpu_context *cpuctx)
 {
-	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
-	int rotate = 0;
+	u64 interval = (u64)cpuctx->jiffies_interval * TICK_NSEC;
+	struct perf_event_context *ctx = NULL;
+	int rotate = 0, remove = 1;
 
-	if (!atomic_read(&nr_events))
-		return;
+	if (cpuctx->ctx.nr_events) {
+		remove = 0;
+		if (cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
+			rotate = 1;
+	}
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	if (cpuctx->ctx.nr_events &&
-	    cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
-		rotate = 1;
+	ctx = cpuctx->task_ctx;
+	if (ctx && ctx->nr_events) {
+		remove = 0;
+		if (ctx->nr_events != ctx->nr_active)
+			rotate = 1;
+	}
 
-	ctx = curr->perf_event_ctxp;
-	if (ctx && ctx->nr_events && ctx->nr_events != ctx->nr_active)
-		rotate = 1;
-
-	perf_ctx_adjust_freq(&cpuctx->ctx);
+	perf_pmu_disable(cpuctx->ctx.pmu);
+	perf_ctx_adjust_freq(&cpuctx->ctx, interval);
 	if (ctx)
-		perf_ctx_adjust_freq(ctx);
+		perf_ctx_adjust_freq(ctx, interval);
 
 	if (!rotate)
-		return;
+		goto done;
 
-	perf_disable();
 	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
 		task_ctx_sched_out(ctx, EVENT_FLEXIBLE);
@@ -1662,8 +1699,27 @@
 
 	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		task_ctx_sched_in(curr, EVENT_FLEXIBLE);
-	perf_enable();
+		task_ctx_sched_in(ctx, EVENT_FLEXIBLE);
+
+done:
+	if (remove)
+		list_del_init(&cpuctx->rotation_list);
+
+	perf_pmu_enable(cpuctx->ctx.pmu);
+}
+
+void perf_event_task_tick(void)
+{
+	struct list_head *head = &__get_cpu_var(rotation_list);
+	struct perf_cpu_context *cpuctx, *tmp;
+
+	WARN_ON(!irqs_disabled());
+
+	list_for_each_entry_safe(cpuctx, tmp, head, rotation_list) {
+		if (cpuctx->jiffies_interval == 1 ||
+				!(jiffies % cpuctx->jiffies_interval))
+			perf_rotate_context(cpuctx);
+	}
 }
 
 static int event_enable_on_exec(struct perf_event *event,
@@ -1685,20 +1741,18 @@
  * Enable all of a task's events that have been marked enable-on-exec.
  * This expects task == current.
  */
-static void perf_event_enable_on_exec(struct task_struct *task)
+static void perf_event_enable_on_exec(struct perf_event_context *ctx)
 {
-	struct perf_event_context *ctx;
 	struct perf_event *event;
 	unsigned long flags;
 	int enabled = 0;
 	int ret;
 
 	local_irq_save(flags);
-	ctx = task->perf_event_ctxp;
 	if (!ctx || !ctx->nr_events)
 		goto out;
 
-	__perf_event_task_sched_out(ctx);
+	task_ctx_sched_out(ctx, EVENT_ALL);
 
 	raw_spin_lock(&ctx->lock);
 
@@ -1722,8 +1776,8 @@
 
 	raw_spin_unlock(&ctx->lock);
 
-	perf_event_task_sched_in(task);
- out:
+	perf_event_context_sched_in(ctx);
+out:
 	local_irq_restore(flags);
 }
 
@@ -1732,9 +1786,9 @@
  */
 static void __perf_event_read(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -1773,7 +1827,13 @@
 		unsigned long flags;
 
 		raw_spin_lock_irqsave(&ctx->lock, flags);
-		update_context_time(ctx);
+		/*
+		 * may read while context is not active
+		 * (e.g., thread is blocked), in that case
+		 * we cannot update context time
+		 */
+		if (ctx->is_active)
+			update_context_time(ctx);
 		update_event_times(event);
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
 	}
@@ -1782,11 +1842,219 @@
 }
 
 /*
+ * Callchain support
+ */
+
+struct callchain_cpus_entries {
+	struct rcu_head			rcu_head;
+	struct perf_callchain_entry	*cpu_entries[0];
+};
+
+static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]);
+static atomic_t nr_callchain_events;
+static DEFINE_MUTEX(callchain_mutex);
+struct callchain_cpus_entries *callchain_cpus_entries;
+
+
+__weak void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs)
+{
+}
+
+__weak void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs)
+{
+}
+
+static void release_callchain_buffers_rcu(struct rcu_head *head)
+{
+	struct callchain_cpus_entries *entries;
+	int cpu;
+
+	entries = container_of(head, struct callchain_cpus_entries, rcu_head);
+
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+
+	kfree(entries);
+}
+
+static void release_callchain_buffers(void)
+{
+	struct callchain_cpus_entries *entries;
+
+	entries = callchain_cpus_entries;
+	rcu_assign_pointer(callchain_cpus_entries, NULL);
+	call_rcu(&entries->rcu_head, release_callchain_buffers_rcu);
+}
+
+static int alloc_callchain_buffers(void)
+{
+	int cpu;
+	int size;
+	struct callchain_cpus_entries *entries;
+
+	/*
+	 * We can't use the percpu allocation API for data that can be
+	 * accessed from NMI. Use a temporary manual per cpu allocation
+	 * until that gets sorted out.
+	 */
+	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
+		num_possible_cpus();
+
+	entries = kzalloc(size, GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS;
+
+	for_each_possible_cpu(cpu) {
+		entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL,
+							 cpu_to_node(cpu));
+		if (!entries->cpu_entries[cpu])
+			goto fail;
+	}
+
+	rcu_assign_pointer(callchain_cpus_entries, entries);
+
+	return 0;
+
+fail:
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+	kfree(entries);
+
+	return -ENOMEM;
+}
+
+static int get_callchain_buffers(void)
+{
+	int err = 0;
+	int count;
+
+	mutex_lock(&callchain_mutex);
+
+	count = atomic_inc_return(&nr_callchain_events);
+	if (WARN_ON_ONCE(count < 1)) {
+		err = -EINVAL;
+		goto exit;
+	}
+
+	if (count > 1) {
+		/* If the allocation failed, give up */
+		if (!callchain_cpus_entries)
+			err = -ENOMEM;
+		goto exit;
+	}
+
+	err = alloc_callchain_buffers();
+	if (err)
+		release_callchain_buffers();
+exit:
+	mutex_unlock(&callchain_mutex);
+
+	return err;
+}
+
+static void put_callchain_buffers(void)
+{
+	if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) {
+		release_callchain_buffers();
+		mutex_unlock(&callchain_mutex);
+	}
+}
+
+static int get_recursion_context(int *recursion)
+{
+	int rctx;
+
+	if (in_nmi())
+		rctx = 3;
+	else if (in_irq())
+		rctx = 2;
+	else if (in_softirq())
+		rctx = 1;
+	else
+		rctx = 0;
+
+	if (recursion[rctx])
+		return -1;
+
+	recursion[rctx]++;
+	barrier();
+
+	return rctx;
+}
+
+static inline void put_recursion_context(int *recursion, int rctx)
+{
+	barrier();
+	recursion[rctx]--;
+}
+
+static struct perf_callchain_entry *get_callchain_entry(int *rctx)
+{
+	int cpu;
+	struct callchain_cpus_entries *entries;
+
+	*rctx = get_recursion_context(__get_cpu_var(callchain_recursion));
+	if (*rctx == -1)
+		return NULL;
+
+	entries = rcu_dereference(callchain_cpus_entries);
+	if (!entries)
+		return NULL;
+
+	cpu = smp_processor_id();
+
+	return &entries->cpu_entries[cpu][*rctx];
+}
+
+static void
+put_callchain_entry(int rctx)
+{
+	put_recursion_context(__get_cpu_var(callchain_recursion), rctx);
+}
+
+static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+{
+	int rctx;
+	struct perf_callchain_entry *entry;
+
+
+	entry = get_callchain_entry(&rctx);
+	if (rctx == -1)
+		return NULL;
+
+	if (!entry)
+		goto exit_put;
+
+	entry->nr = 0;
+
+	if (!user_mode(regs)) {
+		perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+		perf_callchain_kernel(entry, regs);
+		if (current->mm)
+			regs = task_pt_regs(current);
+		else
+			regs = NULL;
+	}
+
+	if (regs) {
+		perf_callchain_store(entry, PERF_CONTEXT_USER);
+		perf_callchain_user(entry, regs);
+	}
+
+exit_put:
+	put_callchain_entry(rctx);
+
+	return entry;
+}
+
+/*
  * Initialize the perf_event context in a task_struct:
  */
-static void
-__perf_event_init_context(struct perf_event_context *ctx,
-			    struct task_struct *task)
+static void __perf_event_init_context(struct perf_event_context *ctx)
 {
 	raw_spin_lock_init(&ctx->lock);
 	mutex_init(&ctx->mutex);
@@ -1794,45 +2062,38 @@
 	INIT_LIST_HEAD(&ctx->flexible_groups);
 	INIT_LIST_HEAD(&ctx->event_list);
 	atomic_set(&ctx->refcount, 1);
-	ctx->task = task;
 }
 
-static struct perf_event_context *find_get_context(pid_t pid, int cpu)
+static struct perf_event_context *
+alloc_perf_context(struct pmu *pmu, struct task_struct *task)
 {
 	struct perf_event_context *ctx;
-	struct perf_cpu_context *cpuctx;
+
+	ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL);
+	if (!ctx)
+		return NULL;
+
+	__perf_event_init_context(ctx);
+	if (task) {
+		ctx->task = task;
+		get_task_struct(task);
+	}
+	ctx->pmu = pmu;
+
+	return ctx;
+}
+
+static struct task_struct *
+find_lively_task_by_vpid(pid_t vpid)
+{
 	struct task_struct *task;
-	unsigned long flags;
 	int err;
 
-	if (pid == -1 && cpu != -1) {
-		/* Must be root to operate on a CPU event: */
-		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
-			return ERR_PTR(-EACCES);
-
-		if (cpu < 0 || cpu >= nr_cpumask_bits)
-			return ERR_PTR(-EINVAL);
-
-		/*
-		 * We could be clever and allow to attach a event to an
-		 * offline CPU and activate it when the CPU comes up, but
-		 * that's for later.
-		 */
-		if (!cpu_online(cpu))
-			return ERR_PTR(-ENODEV);
-
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		ctx = &cpuctx->ctx;
-		get_ctx(ctx);
-
-		return ctx;
-	}
-
 	rcu_read_lock();
-	if (!pid)
+	if (!vpid)
 		task = current;
 	else
-		task = find_task_by_vpid(pid);
+		task = find_task_by_vpid(vpid);
 	if (task)
 		get_task_struct(task);
 	rcu_read_unlock();
@@ -1852,36 +2113,78 @@
 	if (!ptrace_may_access(task, PTRACE_MODE_READ))
 		goto errout;
 
- retry:
-	ctx = perf_lock_task_context(task, &flags);
+	return task;
+errout:
+	put_task_struct(task);
+	return ERR_PTR(err);
+
+}
+
+static struct perf_event_context *
+find_get_context(struct pmu *pmu, struct task_struct *task, int cpu)
+{
+	struct perf_event_context *ctx;
+	struct perf_cpu_context *cpuctx;
+	unsigned long flags;
+	int ctxn, err;
+
+	if (!task && cpu != -1) {
+		/* Must be root to operate on a CPU event: */
+		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
+			return ERR_PTR(-EACCES);
+
+		if (cpu < 0 || cpu >= nr_cpumask_bits)
+			return ERR_PTR(-EINVAL);
+
+		/*
+		 * We could be clever and allow to attach a event to an
+		 * offline CPU and activate it when the CPU comes up, but
+		 * that's for later.
+		 */
+		if (!cpu_online(cpu))
+			return ERR_PTR(-ENODEV);
+
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
+		ctx = &cpuctx->ctx;
+		get_ctx(ctx);
+
+		return ctx;
+	}
+
+	err = -EINVAL;
+	ctxn = pmu->task_ctx_nr;
+	if (ctxn < 0)
+		goto errout;
+
+retry:
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		unclone_ctx(ctx);
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
 	}
 
 	if (!ctx) {
-		ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL);
+		ctx = alloc_perf_context(pmu, task);
 		err = -ENOMEM;
 		if (!ctx)
 			goto errout;
-		__perf_event_init_context(ctx, task);
+
 		get_ctx(ctx);
-		if (cmpxchg(&task->perf_event_ctxp, NULL, ctx)) {
+
+		if (cmpxchg(&task->perf_event_ctxp[ctxn], NULL, ctx)) {
 			/*
 			 * We raced with some other task; use
 			 * the context they set.
 			 */
+			put_task_struct(task);
 			kfree(ctx);
 			goto retry;
 		}
-		get_task_struct(task);
 	}
 
-	put_task_struct(task);
 	return ctx;
 
- errout:
-	put_task_struct(task);
+errout:
 	return ERR_PTR(err);
 }
 
@@ -1898,21 +2201,23 @@
 	kfree(event);
 }
 
-static void perf_pending_sync(struct perf_event *event);
 static void perf_buffer_put(struct perf_buffer *buffer);
 
 static void free_event(struct perf_event *event)
 {
-	perf_pending_sync(event);
+	irq_work_sync(&event->pending);
 
 	if (!event->parent) {
-		atomic_dec(&nr_events);
+		if (event->attach_state & PERF_ATTACH_TASK)
+			jump_label_dec(&perf_task_events);
 		if (event->attr.mmap || event->attr.mmap_data)
 			atomic_dec(&nr_mmap_events);
 		if (event->attr.comm)
 			atomic_dec(&nr_comm_events);
 		if (event->attr.task)
 			atomic_dec(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+			put_callchain_buffers();
 	}
 
 	if (event->buffer) {
@@ -1923,7 +2228,9 @@
 	if (event->destroy)
 		event->destroy(event);
 
-	put_ctx(event->ctx);
+	if (event->ctx)
+		put_ctx(event->ctx);
+
 	call_rcu(&event->rcu_head, free_event_rcu);
 }
 
@@ -2202,15 +2509,13 @@
 static int perf_event_period(struct perf_event *event, u64 __user *arg)
 {
 	struct perf_event_context *ctx = event->ctx;
-	unsigned long size;
 	int ret = 0;
 	u64 value;
 
 	if (!event->attr.sample_period)
 		return -EINVAL;
 
-	size = copy_from_user(&value, arg, sizeof(value));
-	if (size != sizeof(value))
+	if (copy_from_user(&value, arg, sizeof(value)))
 		return -EFAULT;
 
 	if (!value)
@@ -2344,6 +2649,9 @@
 
 static int perf_event_index(struct perf_event *event)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (event->state != PERF_EVENT_STATE_ACTIVE)
 		return 0;
 
@@ -2847,16 +3155,7 @@
 	}
 }
 
-/*
- * Pending wakeups
- *
- * Handle the case where we need to wakeup up from NMI (or rq->lock) context.
- *
- * The NMI bit means we cannot possibly take locks. Therefore, maintain a
- * single linked list and use cmpxchg() to add entries lockless.
- */
-
-static void perf_pending_event(struct perf_pending_entry *entry)
+static void perf_pending_event(struct irq_work *entry)
 {
 	struct perf_event *event = container_of(entry,
 			struct perf_event, pending);
@@ -2872,99 +3171,6 @@
 	}
 }
 
-#define PENDING_TAIL ((struct perf_pending_entry *)-1UL)
-
-static DEFINE_PER_CPU(struct perf_pending_entry *, perf_pending_head) = {
-	PENDING_TAIL,
-};
-
-static void perf_pending_queue(struct perf_pending_entry *entry,
-			       void (*func)(struct perf_pending_entry *))
-{
-	struct perf_pending_entry **head;
-
-	if (cmpxchg(&entry->next, NULL, PENDING_TAIL) != NULL)
-		return;
-
-	entry->func = func;
-
-	head = &get_cpu_var(perf_pending_head);
-
-	do {
-		entry->next = *head;
-	} while (cmpxchg(head, entry->next, entry) != entry->next);
-
-	set_perf_event_pending();
-
-	put_cpu_var(perf_pending_head);
-}
-
-static int __perf_pending_run(void)
-{
-	struct perf_pending_entry *list;
-	int nr = 0;
-
-	list = xchg(&__get_cpu_var(perf_pending_head), PENDING_TAIL);
-	while (list != PENDING_TAIL) {
-		void (*func)(struct perf_pending_entry *);
-		struct perf_pending_entry *entry = list;
-
-		list = list->next;
-
-		func = entry->func;
-		entry->next = NULL;
-		/*
-		 * Ensure we observe the unqueue before we issue the wakeup,
-		 * so that we won't be waiting forever.
-		 * -- see perf_not_pending().
-		 */
-		smp_wmb();
-
-		func(entry);
-		nr++;
-	}
-
-	return nr;
-}
-
-static inline int perf_not_pending(struct perf_event *event)
-{
-	/*
-	 * If we flush on whatever cpu we run, there is a chance we don't
-	 * need to wait.
-	 */
-	get_cpu();
-	__perf_pending_run();
-	put_cpu();
-
-	/*
-	 * Ensure we see the proper queue state before going to sleep
-	 * so that we do not miss the wakeup. -- see perf_pending_handle()
-	 */
-	smp_rmb();
-	return event->pending.next == NULL;
-}
-
-static void perf_pending_sync(struct perf_event *event)
-{
-	wait_event(event->waitq, perf_not_pending(event));
-}
-
-void perf_event_do_pending(void)
-{
-	__perf_pending_run();
-}
-
-/*
- * Callchain support -- arch specific
- */
-
-__weak struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	return NULL;
-}
-
-
 /*
  * We assume there is only KVM supporting the callbacks.
  * Later on, we might change it to a list if there is
@@ -3014,8 +3220,7 @@
 
 	if (handle->nmi) {
 		handle->event->pending_wakeup = 1;
-		perf_pending_queue(&handle->event->pending,
-				   perf_pending_event);
+		irq_work_queue(&handle->event->pending);
 	} else
 		perf_event_wakeup(handle->event);
 }
@@ -3071,7 +3276,7 @@
 	if (handle->wakeup != local_read(&buffer->wakeup))
 		perf_output_wakeup(handle);
 
- out:
+out:
 	preempt_enable();
 }
 
@@ -3459,14 +3664,20 @@
 	struct perf_output_handle handle;
 	struct perf_event_header header;
 
+	/* protect the callchain buffers */
+	rcu_read_lock();
+
 	perf_prepare_sample(&header, data, event, regs);
 
 	if (perf_output_begin(&handle, event, header.size, nmi, 1))
-		return;
+		goto exit;
 
 	perf_output_sample(&handle, &header, data, event);
 
 	perf_output_end(&handle);
+
+exit:
+	rcu_read_unlock();
 }
 
 /*
@@ -3580,16 +3791,27 @@
 static void perf_event_task_event(struct perf_task_event *task_event)
 {
 	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx = task_event->task_ctx;
+	struct perf_event_context *ctx;
+	struct pmu *pmu;
+	int ctxn;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_task_ctx(&cpuctx->ctx, task_event);
-	if (!ctx)
-		ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_task_ctx(ctx, task_event);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_task_ctx(&cpuctx->ctx, task_event);
+
+		ctx = task_event->task_ctx;
+		if (!ctx) {
+			ctxn = pmu->task_ctx_nr;
+			if (ctxn < 0)
+				goto next;
+			ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		}
+		if (ctx)
+			perf_event_task_ctx(ctx, task_event);
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 }
 
@@ -3694,8 +3916,10 @@
 {
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
-	unsigned int size;
 	char comm[TASK_COMM_LEN];
+	unsigned int size;
+	struct pmu *pmu;
+	int ctxn;
 
 	memset(comm, 0, sizeof(comm));
 	strlcpy(comm, comm_event->task->comm, sizeof(comm));
@@ -3707,21 +3931,36 @@
 	comm_event->event_id.header.size = sizeof(comm_event->event_id) + size;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_comm_ctx(&cpuctx->ctx, comm_event);
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_comm_ctx(ctx, comm_event);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_comm_ctx(&cpuctx->ctx, comm_event);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			goto next;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx)
+			perf_event_comm_ctx(ctx, comm_event);
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 }
 
 void perf_event_comm(struct task_struct *task)
 {
 	struct perf_comm_event comm_event;
+	struct perf_event_context *ctx;
+	int ctxn;
 
-	if (task->perf_event_ctxp)
-		perf_event_enable_on_exec(task);
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
+
+		perf_event_enable_on_exec(ctx);
+	}
 
 	if (!atomic_read(&nr_comm_events))
 		return;
@@ -3823,6 +4062,8 @@
 	char tmp[16];
 	char *buf = NULL;
 	const char *name;
+	struct pmu *pmu;
+	int ctxn;
 
 	memset(tmp, 0, sizeof(tmp));
 
@@ -3875,12 +4116,23 @@
 	mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_mmap_ctx(&cpuctx->ctx, mmap_event, vma->vm_flags & VM_EXEC);
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_mmap_ctx(ctx, mmap_event, vma->vm_flags & VM_EXEC);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_mmap_ctx(&cpuctx->ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			goto next;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx) {
+			perf_event_mmap_ctx(ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+		}
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 
 	kfree(buf);
@@ -3962,8 +4214,6 @@
 	struct hw_perf_event *hwc = &event->hw;
 	int ret = 0;
 
-	throttle = (throttle && event->pmu->unthrottle != NULL);
-
 	if (!throttle) {
 		hwc->interrupts++;
 	} else {
@@ -4006,8 +4256,7 @@
 		event->pending_kill = POLL_HUP;
 		if (nmi) {
 			event->pending_disable = 1;
-			perf_pending_queue(&event->pending,
-					   perf_pending_event);
+			irq_work_queue(&event->pending);
 		} else
 			perf_event_disable(event);
 	}
@@ -4031,6 +4280,17 @@
  * Generic software event infrastructure
  */
 
+struct swevent_htable {
+	struct swevent_hlist		*swevent_hlist;
+	struct mutex			hlist_mutex;
+	int				hlist_refcount;
+
+	/* Recursion avoidance in each contexts */
+	int				recursion[PERF_NR_CONTEXTS];
+};
+
+static DEFINE_PER_CPU(struct swevent_htable, swevent_htable);
+
 /*
  * We directly increment event->count and keep a second value in
  * event->hw.period_left to count intervals. This period event
@@ -4088,7 +4348,7 @@
 	}
 }
 
-static void perf_swevent_add(struct perf_event *event, u64 nr,
+static void perf_swevent_event(struct perf_event *event, u64 nr,
 			       int nmi, struct perf_sample_data *data,
 			       struct pt_regs *regs)
 {
@@ -4114,6 +4374,9 @@
 static int perf_exclude_event(struct perf_event *event,
 			      struct pt_regs *regs)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (regs) {
 		if (event->attr.exclude_user && user_mode(regs))
 			return 1;
@@ -4160,11 +4423,11 @@
 
 /* For the read side: events when they trigger */
 static inline struct hlist_head *
-find_swevent_head_rcu(struct perf_cpu_context *ctx, u64 type, u32 event_id)
+find_swevent_head_rcu(struct swevent_htable *swhash, u64 type, u32 event_id)
 {
 	struct swevent_hlist *hlist;
 
-	hlist = rcu_dereference(ctx->swevent_hlist);
+	hlist = rcu_dereference(swhash->swevent_hlist);
 	if (!hlist)
 		return NULL;
 
@@ -4173,7 +4436,7 @@
 
 /* For the event head insertion and removal in the hlist */
 static inline struct hlist_head *
-find_swevent_head(struct perf_cpu_context *ctx, struct perf_event *event)
+find_swevent_head(struct swevent_htable *swhash, struct perf_event *event)
 {
 	struct swevent_hlist *hlist;
 	u32 event_id = event->attr.config;
@@ -4184,7 +4447,7 @@
 	 * and release. Which makes the protected version suitable here.
 	 * The context lock guarantees that.
 	 */
-	hlist = rcu_dereference_protected(ctx->swevent_hlist,
+	hlist = rcu_dereference_protected(swhash->swevent_hlist,
 					  lockdep_is_held(&event->ctx->lock));
 	if (!hlist)
 		return NULL;
@@ -4197,23 +4460,19 @@
 				    struct perf_sample_data *data,
 				    struct pt_regs *regs)
 {
-	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct perf_event *event;
 	struct hlist_node *node;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	rcu_read_lock();
-
-	head = find_swevent_head_rcu(cpuctx, type, event_id);
-
+	head = find_swevent_head_rcu(swhash, type, event_id);
 	if (!head)
 		goto end;
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_swevent_match(event, type, event_id, data, regs))
-			perf_swevent_add(event, nr, nmi, data, regs);
+			perf_swevent_event(event, nr, nmi, data, regs);
 	}
 end:
 	rcu_read_unlock();
@@ -4221,33 +4480,17 @@
 
 int perf_swevent_get_recursion_context(void)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	int rctx;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 
-	if (in_nmi())
-		rctx = 3;
-	else if (in_irq())
-		rctx = 2;
-	else if (in_softirq())
-		rctx = 1;
-	else
-		rctx = 0;
-
-	if (cpuctx->recursion[rctx])
-		return -1;
-
-	cpuctx->recursion[rctx]++;
-	barrier();
-
-	return rctx;
+	return get_recursion_context(swhash->recursion);
 }
 EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context);
 
 void inline perf_swevent_put_recursion_context(int rctx)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	barrier();
-	cpuctx->recursion[rctx]--;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
+
+	put_recursion_context(swhash->recursion, rctx);
 }
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
@@ -4273,20 +4516,20 @@
 {
 }
 
-static int perf_swevent_enable(struct perf_event *event)
+static int perf_swevent_add(struct perf_event *event, int flags)
 {
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct hw_perf_event *hwc = &event->hw;
-	struct perf_cpu_context *cpuctx;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	if (hwc->sample_period) {
 		hwc->last_period = hwc->sample_period;
 		perf_swevent_set_period(event);
 	}
 
-	head = find_swevent_head(cpuctx, event);
+	hwc->state = !(flags & PERF_EF_START);
+
+	head = find_swevent_head(swhash, event);
 	if (WARN_ON_ONCE(!head))
 		return -EINVAL;
 
@@ -4295,202 +4538,27 @@
 	return 0;
 }
 
-static void perf_swevent_disable(struct perf_event *event)
+static void perf_swevent_del(struct perf_event *event, int flags)
 {
 	hlist_del_rcu(&event->hlist_entry);
 }
 
-static void perf_swevent_void(struct perf_event *event)
+static void perf_swevent_start(struct perf_event *event, int flags)
 {
+	event->hw.state = 0;
 }
 
-static int perf_swevent_int(struct perf_event *event)
+static void perf_swevent_stop(struct perf_event *event, int flags)
 {
-	return 0;
+	event->hw.state = PERF_HES_STOPPED;
 }
 
-static const struct pmu perf_ops_generic = {
-	.enable		= perf_swevent_enable,
-	.disable	= perf_swevent_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
-	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void, /* hwc->interrupts already reset */
-};
-
-/*
- * hrtimer based swevent callback
- */
-
-static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
-{
-	enum hrtimer_restart ret = HRTIMER_RESTART;
-	struct perf_sample_data data;
-	struct pt_regs *regs;
-	struct perf_event *event;
-	u64 period;
-
-	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
-	event->pmu->read(event);
-
-	perf_sample_data_init(&data, 0);
-	data.period = event->hw.last_period;
-	regs = get_irq_regs();
-
-	if (regs && !perf_exclude_event(event, regs)) {
-		if (!(event->attr.exclude_idle && current->pid == 0))
-			if (perf_event_overflow(event, 0, &data, regs))
-				ret = HRTIMER_NORESTART;
-	}
-
-	period = max_t(u64, 10000, event->hw.sample_period);
-	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
-
-	return ret;
-}
-
-static void perf_swevent_start_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	hwc->hrtimer.function = perf_swevent_hrtimer;
-	if (hwc->sample_period) {
-		u64 period;
-
-		if (hwc->remaining) {
-			if (hwc->remaining < 0)
-				period = 10000;
-			else
-				period = hwc->remaining;
-			hwc->remaining = 0;
-		} else {
-			period = max_t(u64, 10000, hwc->sample_period);
-		}
-		__hrtimer_start_range_ns(&hwc->hrtimer,
-				ns_to_ktime(period), 0,
-				HRTIMER_MODE_REL, 0);
-	}
-}
-
-static void perf_swevent_cancel_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	if (hwc->sample_period) {
-		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
-		hwc->remaining = ktime_to_ns(remaining);
-
-		hrtimer_cancel(&hwc->hrtimer);
-	}
-}
-
-/*
- * Software event: cpu wall time clock
- */
-
-static void cpu_clock_perf_event_update(struct perf_event *event)
-{
-	int cpu = raw_smp_processor_id();
-	s64 prev;
-	u64 now;
-
-	now = cpu_clock(cpu);
-	prev = local64_xchg(&event->hw.prev_count, now);
-	local64_add(now - prev, &event->count);
-}
-
-static int cpu_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	int cpu = raw_smp_processor_id();
-
-	local64_set(&hwc->prev_count, cpu_clock(cpu));
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void cpu_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	cpu_clock_perf_event_update(event);
-}
-
-static void cpu_clock_perf_event_read(struct perf_event *event)
-{
-	cpu_clock_perf_event_update(event);
-}
-
-static const struct pmu perf_ops_cpu_clock = {
-	.enable		= cpu_clock_perf_event_enable,
-	.disable	= cpu_clock_perf_event_disable,
-	.read		= cpu_clock_perf_event_read,
-};
-
-/*
- * Software event: task time clock
- */
-
-static void task_clock_perf_event_update(struct perf_event *event, u64 now)
-{
-	u64 prev;
-	s64 delta;
-
-	prev = local64_xchg(&event->hw.prev_count, now);
-	delta = now - prev;
-	local64_add(delta, &event->count);
-}
-
-static int task_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	u64 now;
-
-	now = event->ctx->time;
-
-	local64_set(&hwc->prev_count, now);
-
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void task_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	task_clock_perf_event_update(event, event->ctx->time);
-
-}
-
-static void task_clock_perf_event_read(struct perf_event *event)
-{
-	u64 time;
-
-	if (!in_nmi()) {
-		update_context_time(event->ctx);
-		time = event->ctx->time;
-	} else {
-		u64 now = perf_clock();
-		u64 delta = now - event->ctx->timestamp;
-		time = event->ctx->time + delta;
-	}
-
-	task_clock_perf_event_update(event, time);
-}
-
-static const struct pmu perf_ops_task_clock = {
-	.enable		= task_clock_perf_event_enable,
-	.disable	= task_clock_perf_event_disable,
-	.read		= task_clock_perf_event_read,
-};
-
 /* Deref the hlist from the update side */
 static inline struct swevent_hlist *
-swevent_hlist_deref(struct perf_cpu_context *cpuctx)
+swevent_hlist_deref(struct swevent_htable *swhash)
 {
-	return rcu_dereference_protected(cpuctx->swevent_hlist,
-					 lockdep_is_held(&cpuctx->hlist_mutex));
+	return rcu_dereference_protected(swhash->swevent_hlist,
+					 lockdep_is_held(&swhash->hlist_mutex));
 }
 
 static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
@@ -4501,27 +4569,27 @@
 	kfree(hlist);
 }
 
-static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+static void swevent_hlist_release(struct swevent_htable *swhash)
 {
-	struct swevent_hlist *hlist = swevent_hlist_deref(cpuctx);
+	struct swevent_hlist *hlist = swevent_hlist_deref(swhash);
 
 	if (!hlist)
 		return;
 
-	rcu_assign_pointer(cpuctx->swevent_hlist, NULL);
+	rcu_assign_pointer(swhash->swevent_hlist, NULL);
 	call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu);
 }
 
 static void swevent_hlist_put_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!--cpuctx->hlist_refcount)
-		swevent_hlist_release(cpuctx);
+	if (!--swhash->hlist_refcount)
+		swevent_hlist_release(swhash);
 
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 static void swevent_hlist_put(struct perf_event *event)
@@ -4539,12 +4607,12 @@
 
 static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 	int err = 0;
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!swevent_hlist_deref(cpuctx) && cpu_online(cpu)) {
+	if (!swevent_hlist_deref(swhash) && cpu_online(cpu)) {
 		struct swevent_hlist *hlist;
 
 		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
@@ -4552,11 +4620,11 @@
 			err = -ENOMEM;
 			goto exit;
 		}
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	cpuctx->hlist_refcount++;
- exit:
-	mutex_unlock(&cpuctx->hlist_mutex);
+	swhash->hlist_refcount++;
+exit:
+	mutex_unlock(&swhash->hlist_mutex);
 
 	return err;
 }
@@ -4580,7 +4648,7 @@
 	put_online_cpus();
 
 	return 0;
- fail:
+fail:
 	for_each_possible_cpu(cpu) {
 		if (cpu == failed_cpu)
 			break;
@@ -4591,17 +4659,64 @@
 	return err;
 }
 
-#ifdef CONFIG_EVENT_TRACING
+atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
 
-static const struct pmu perf_ops_tracepoint = {
-	.enable		= perf_trace_enable,
-	.disable	= perf_trace_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
+static void sw_perf_event_destroy(struct perf_event *event)
+{
+	u64 event_id = event->attr.config;
+
+	WARN_ON(event->parent);
+
+	jump_label_dec(&perf_swevent_enabled[event_id]);
+	swevent_hlist_put(event);
+}
+
+static int perf_swevent_init(struct perf_event *event)
+{
+	int event_id = event->attr.config;
+
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	switch (event_id) {
+	case PERF_COUNT_SW_CPU_CLOCK:
+	case PERF_COUNT_SW_TASK_CLOCK:
+		return -ENOENT;
+
+	default:
+		break;
+	}
+
+	if (event_id > PERF_COUNT_SW_MAX)
+		return -ENOENT;
+
+	if (!event->parent) {
+		int err;
+
+		err = swevent_hlist_get(event);
+		if (err)
+			return err;
+
+		jump_label_inc(&perf_swevent_enabled[event_id]);
+		event->destroy = sw_perf_event_destroy;
+	}
+
+	return 0;
+}
+
+static struct pmu perf_swevent = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= perf_swevent_init,
+	.add		= perf_swevent_add,
+	.del		= perf_swevent_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
 	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void,
 };
 
+#ifdef CONFIG_EVENT_TRACING
+
 static int perf_tp_filter_match(struct perf_event *event,
 				struct perf_sample_data *data)
 {
@@ -4645,7 +4760,7 @@
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_tp_event_match(event, &data, regs))
-			perf_swevent_add(event, count, 1, &data, regs);
+			perf_swevent_event(event, count, 1, &data, regs);
 	}
 
 	perf_swevent_put_recursion_context(rctx);
@@ -4657,10 +4772,13 @@
 	perf_trace_destroy(event);
 }
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static int perf_tp_event_init(struct perf_event *event)
 {
 	int err;
 
+	if (event->attr.type != PERF_TYPE_TRACEPOINT)
+		return -ENOENT;
+
 	/*
 	 * Raw tracepoint data is a severe data leak, only allow root to
 	 * have these.
@@ -4668,15 +4786,31 @@
 	if ((event->attr.sample_type & PERF_SAMPLE_RAW) &&
 			perf_paranoid_tracepoint_raw() &&
 			!capable(CAP_SYS_ADMIN))
-		return ERR_PTR(-EPERM);
+		return -EPERM;
 
 	err = perf_trace_init(event);
 	if (err)
-		return NULL;
+		return err;
 
 	event->destroy = tp_perf_event_destroy;
 
-	return &perf_ops_tracepoint;
+	return 0;
+}
+
+static struct pmu perf_tracepoint = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= perf_tp_event_init,
+	.add		= perf_trace_add,
+	.del		= perf_trace_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
+	.read		= perf_swevent_read,
+};
+
+static inline void perf_tp_register(void)
+{
+	perf_pmu_register(&perf_tracepoint);
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4704,9 +4838,8 @@
 
 #else
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static inline void perf_tp_register(void)
 {
-	return NULL;
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4721,24 +4854,6 @@
 #endif /* CONFIG_EVENT_TRACING */
 
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
-static void bp_perf_event_destroy(struct perf_event *event)
-{
-	release_bp_slot(event);
-}
-
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
-{
-	int err;
-
-	err = register_perf_hw_breakpoint(bp);
-	if (err)
-		return ERR_PTR(err);
-
-	bp->destroy = bp_perf_event_destroy;
-
-	return &perf_ops_bp;
-}
-
 void perf_bp_event(struct perf_event *bp, void *data)
 {
 	struct perf_sample_data sample;
@@ -4746,80 +4861,382 @@
 
 	perf_sample_data_init(&sample, bp->attr.bp_addr);
 
-	if (!perf_exclude_event(bp, regs))
-		perf_swevent_add(bp, 1, 1, &sample, regs);
-}
-#else
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
-{
-	return NULL;
-}
-
-void perf_bp_event(struct perf_event *bp, void *regs)
-{
+	if (!bp->hw.state && !perf_exclude_event(bp, regs))
+		perf_swevent_event(bp, 1, 1, &sample, regs);
 }
 #endif
 
-atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
+/*
+ * hrtimer based swevent callback
+ */
 
-static void sw_perf_event_destroy(struct perf_event *event)
+static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 {
-	u64 event_id = event->attr.config;
+	enum hrtimer_restart ret = HRTIMER_RESTART;
+	struct perf_sample_data data;
+	struct pt_regs *regs;
+	struct perf_event *event;
+	u64 period;
 
-	WARN_ON(event->parent);
+	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
+	event->pmu->read(event);
 
-	atomic_dec(&perf_swevent_enabled[event_id]);
-	swevent_hlist_put(event);
+	perf_sample_data_init(&data, 0);
+	data.period = event->hw.last_period;
+	regs = get_irq_regs();
+
+	if (regs && !perf_exclude_event(event, regs)) {
+		if (!(event->attr.exclude_idle && current->pid == 0))
+			if (perf_event_overflow(event, 0, &data, regs))
+				ret = HRTIMER_NORESTART;
+	}
+
+	period = max_t(u64, 10000, event->hw.sample_period);
+	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
+
+	return ret;
 }
 
-static const struct pmu *sw_perf_event_init(struct perf_event *event)
+static void perf_swevent_start_hrtimer(struct perf_event *event)
 {
-	const struct pmu *pmu = NULL;
-	u64 event_id = event->attr.config;
+	struct hw_perf_event *hwc = &event->hw;
+
+	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hwc->hrtimer.function = perf_swevent_hrtimer;
+	if (hwc->sample_period) {
+		s64 period = local64_read(&hwc->period_left);
+
+		if (period) {
+			if (period < 0)
+				period = 10000;
+
+			local64_set(&hwc->period_left, 0);
+		} else {
+			period = max_t(u64, 10000, hwc->sample_period);
+		}
+		__hrtimer_start_range_ns(&hwc->hrtimer,
+				ns_to_ktime(period), 0,
+				HRTIMER_MODE_REL_PINNED, 0);
+	}
+}
+
+static void perf_swevent_cancel_hrtimer(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->sample_period) {
+		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
+		local64_set(&hwc->period_left, ktime_to_ns(remaining));
+
+		hrtimer_cancel(&hwc->hrtimer);
+	}
+}
+
+/*
+ * Software event: cpu wall time clock
+ */
+
+static void cpu_clock_event_update(struct perf_event *event)
+{
+	s64 prev;
+	u64 now;
+
+	now = local_clock();
+	prev = local64_xchg(&event->hw.prev_count, now);
+	local64_add(now - prev, &event->count);
+}
+
+static void cpu_clock_event_start(struct perf_event *event, int flags)
+{
+	local64_set(&event->hw.prev_count, local_clock());
+	perf_swevent_start_hrtimer(event);
+}
+
+static void cpu_clock_event_stop(struct perf_event *event, int flags)
+{
+	perf_swevent_cancel_hrtimer(event);
+	cpu_clock_event_update(event);
+}
+
+static int cpu_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		cpu_clock_event_start(event, flags);
+
+	return 0;
+}
+
+static void cpu_clock_event_del(struct perf_event *event, int flags)
+{
+	cpu_clock_event_stop(event, flags);
+}
+
+static void cpu_clock_event_read(struct perf_event *event)
+{
+	cpu_clock_event_update(event);
+}
+
+static int cpu_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_CPU_CLOCK)
+		return -ENOENT;
+
+	return 0;
+}
+
+static struct pmu perf_cpu_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= cpu_clock_event_init,
+	.add		= cpu_clock_event_add,
+	.del		= cpu_clock_event_del,
+	.start		= cpu_clock_event_start,
+	.stop		= cpu_clock_event_stop,
+	.read		= cpu_clock_event_read,
+};
+
+/*
+ * Software event: task time clock
+ */
+
+static void task_clock_event_update(struct perf_event *event, u64 now)
+{
+	u64 prev;
+	s64 delta;
+
+	prev = local64_xchg(&event->hw.prev_count, now);
+	delta = now - prev;
+	local64_add(delta, &event->count);
+}
+
+static void task_clock_event_start(struct perf_event *event, int flags)
+{
+	local64_set(&event->hw.prev_count, event->ctx->time);
+	perf_swevent_start_hrtimer(event);
+}
+
+static void task_clock_event_stop(struct perf_event *event, int flags)
+{
+	perf_swevent_cancel_hrtimer(event);
+	task_clock_event_update(event, event->ctx->time);
+}
+
+static int task_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		task_clock_event_start(event, flags);
+
+	return 0;
+}
+
+static void task_clock_event_del(struct perf_event *event, int flags)
+{
+	task_clock_event_stop(event, PERF_EF_UPDATE);
+}
+
+static void task_clock_event_read(struct perf_event *event)
+{
+	u64 time;
+
+	if (!in_nmi()) {
+		update_context_time(event->ctx);
+		time = event->ctx->time;
+	} else {
+		u64 now = perf_clock();
+		u64 delta = now - event->ctx->timestamp;
+		time = event->ctx->time + delta;
+	}
+
+	task_clock_event_update(event, time);
+}
+
+static int task_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_TASK_CLOCK)
+		return -ENOENT;
+
+	return 0;
+}
+
+static struct pmu perf_task_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= task_clock_event_init,
+	.add		= task_clock_event_add,
+	.del		= task_clock_event_del,
+	.start		= task_clock_event_start,
+	.stop		= task_clock_event_stop,
+	.read		= task_clock_event_read,
+};
+
+static void perf_pmu_nop_void(struct pmu *pmu)
+{
+}
+
+static int perf_pmu_nop_int(struct pmu *pmu)
+{
+	return 0;
+}
+
+static void perf_pmu_start_txn(struct pmu *pmu)
+{
+	perf_pmu_disable(pmu);
+}
+
+static int perf_pmu_commit_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+	return 0;
+}
+
+static void perf_pmu_cancel_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+}
+
+/*
+ * Ensures all contexts with the same task_ctx_nr have the same
+ * pmu_cpu_context too.
+ */
+static void *find_pmu_context(int ctxn)
+{
+	struct pmu *pmu;
+
+	if (ctxn < 0)
+		return NULL;
+
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->task_ctx_nr == ctxn)
+			return pmu->pmu_cpu_context;
+	}
+
+	return NULL;
+}
+
+static void free_pmu_context(void * __percpu cpu_context)
+{
+	struct pmu *pmu;
+
+	mutex_lock(&pmus_lock);
+	/*
+	 * Like a real lame refcount.
+	 */
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->pmu_cpu_context == cpu_context)
+			goto out;
+	}
+
+	free_percpu(cpu_context);
+out:
+	mutex_unlock(&pmus_lock);
+}
+
+int perf_pmu_register(struct pmu *pmu)
+{
+	int cpu, ret;
+
+	mutex_lock(&pmus_lock);
+	ret = -ENOMEM;
+	pmu->pmu_disable_count = alloc_percpu(int);
+	if (!pmu->pmu_disable_count)
+		goto unlock;
+
+	pmu->pmu_cpu_context = find_pmu_context(pmu->task_ctx_nr);
+	if (pmu->pmu_cpu_context)
+		goto got_cpu_context;
+
+	pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context);
+	if (!pmu->pmu_cpu_context)
+		goto free_pdc;
+
+	for_each_possible_cpu(cpu) {
+		struct perf_cpu_context *cpuctx;
+
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
+		__perf_event_init_context(&cpuctx->ctx);
+		cpuctx->ctx.type = cpu_context;
+		cpuctx->ctx.pmu = pmu;
+		cpuctx->jiffies_interval = 1;
+		INIT_LIST_HEAD(&cpuctx->rotation_list);
+	}
+
+got_cpu_context:
+	if (!pmu->start_txn) {
+		if (pmu->pmu_enable) {
+			/*
+			 * If we have pmu_enable/pmu_disable calls, install
+			 * transaction stubs that use that to try and batch
+			 * hardware accesses.
+			 */
+			pmu->start_txn  = perf_pmu_start_txn;
+			pmu->commit_txn = perf_pmu_commit_txn;
+			pmu->cancel_txn = perf_pmu_cancel_txn;
+		} else {
+			pmu->start_txn  = perf_pmu_nop_void;
+			pmu->commit_txn = perf_pmu_nop_int;
+			pmu->cancel_txn = perf_pmu_nop_void;
+		}
+	}
+
+	if (!pmu->pmu_enable) {
+		pmu->pmu_enable  = perf_pmu_nop_void;
+		pmu->pmu_disable = perf_pmu_nop_void;
+	}
+
+	list_add_rcu(&pmu->entry, &pmus);
+	ret = 0;
+unlock:
+	mutex_unlock(&pmus_lock);
+
+	return ret;
+
+free_pdc:
+	free_percpu(pmu->pmu_disable_count);
+	goto unlock;
+}
+
+void perf_pmu_unregister(struct pmu *pmu)
+{
+	mutex_lock(&pmus_lock);
+	list_del_rcu(&pmu->entry);
+	mutex_unlock(&pmus_lock);
 
 	/*
-	 * Software events (currently) can't in general distinguish
-	 * between user, kernel and hypervisor events.
-	 * However, context switches and cpu migrations are considered
-	 * to be kernel events, and page faults are never hypervisor
-	 * events.
+	 * We dereference the pmu list under both SRCU and regular RCU, so
+	 * synchronize against both of those.
 	 */
-	switch (event_id) {
-	case PERF_COUNT_SW_CPU_CLOCK:
-		pmu = &perf_ops_cpu_clock;
+	synchronize_srcu(&pmus_srcu);
+	synchronize_rcu();
 
-		break;
-	case PERF_COUNT_SW_TASK_CLOCK:
-		/*
-		 * If the user instantiates this as a per-cpu event,
-		 * use the cpu_clock event instead.
-		 */
-		if (event->ctx->task)
-			pmu = &perf_ops_task_clock;
-		else
-			pmu = &perf_ops_cpu_clock;
+	free_percpu(pmu->pmu_disable_count);
+	free_pmu_context(pmu->pmu_cpu_context);
+}
 
-		break;
-	case PERF_COUNT_SW_PAGE_FAULTS:
-	case PERF_COUNT_SW_PAGE_FAULTS_MIN:
-	case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
-	case PERF_COUNT_SW_CONTEXT_SWITCHES:
-	case PERF_COUNT_SW_CPU_MIGRATIONS:
-	case PERF_COUNT_SW_ALIGNMENT_FAULTS:
-	case PERF_COUNT_SW_EMULATION_FAULTS:
-		if (!event->parent) {
-			int err;
+struct pmu *perf_init_event(struct perf_event *event)
+{
+	struct pmu *pmu = NULL;
+	int idx;
 
-			err = swevent_hlist_get(event);
-			if (err)
-				return ERR_PTR(err);
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		int ret = pmu->event_init(event);
+		if (!ret)
+			goto unlock;
 
-			atomic_inc(&perf_swevent_enabled[event_id]);
-			event->destroy = sw_perf_event_destroy;
+		if (ret != -ENOENT) {
+			pmu = ERR_PTR(ret);
+			goto unlock;
 		}
-		pmu = &perf_ops_generic;
-		break;
 	}
+	pmu = ERR_PTR(-ENOENT);
+unlock:
+	srcu_read_unlock(&pmus_srcu, idx);
 
 	return pmu;
 }
@@ -4828,20 +5245,18 @@
  * Allocate and initialize a event structure
  */
 static struct perf_event *
-perf_event_alloc(struct perf_event_attr *attr,
-		   int cpu,
-		   struct perf_event_context *ctx,
-		   struct perf_event *group_leader,
-		   struct perf_event *parent_event,
-		   perf_overflow_handler_t overflow_handler,
-		   gfp_t gfpflags)
+perf_event_alloc(struct perf_event_attr *attr, int cpu,
+		 struct task_struct *task,
+		 struct perf_event *group_leader,
+		 struct perf_event *parent_event,
+		 perf_overflow_handler_t overflow_handler)
 {
-	const struct pmu *pmu;
+	struct pmu *pmu;
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
 	long err;
 
-	event = kzalloc(sizeof(*event), gfpflags);
+	event = kzalloc(sizeof(*event), GFP_KERNEL);
 	if (!event)
 		return ERR_PTR(-ENOMEM);
 
@@ -4859,6 +5274,7 @@
 	INIT_LIST_HEAD(&event->event_entry);
 	INIT_LIST_HEAD(&event->sibling_list);
 	init_waitqueue_head(&event->waitq);
+	init_irq_work(&event->pending, perf_pending_event);
 
 	mutex_init(&event->mmap_mutex);
 
@@ -4866,7 +5282,6 @@
 	event->attr		= *attr;
 	event->group_leader	= group_leader;
 	event->pmu		= NULL;
-	event->ctx		= ctx;
 	event->oncpu		= -1;
 
 	event->parent		= parent_event;
@@ -4876,6 +5291,17 @@
 
 	event->state		= PERF_EVENT_STATE_INACTIVE;
 
+	if (task) {
+		event->attach_state = PERF_ATTACH_TASK;
+#ifdef CONFIG_HAVE_HW_BREAKPOINT
+		/*
+		 * hw_breakpoint is a bit difficult here..
+		 */
+		if (attr->type == PERF_TYPE_BREAKPOINT)
+			event->hw.bp_target = task;
+#endif
+	}
+
 	if (!overflow_handler && parent_event)
 		overflow_handler = parent_event->overflow_handler;
 	
@@ -4900,29 +5326,8 @@
 	if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP))
 		goto done;
 
-	switch (attr->type) {
-	case PERF_TYPE_RAW:
-	case PERF_TYPE_HARDWARE:
-	case PERF_TYPE_HW_CACHE:
-		pmu = hw_perf_event_init(event);
-		break;
+	pmu = perf_init_event(event);
 
-	case PERF_TYPE_SOFTWARE:
-		pmu = sw_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_TRACEPOINT:
-		pmu = tp_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_BREAKPOINT:
-		pmu = bp_perf_event_init(event);
-		break;
-
-
-	default:
-		break;
-	}
 done:
 	err = 0;
 	if (!pmu)
@@ -4940,13 +5345,21 @@
 	event->pmu = pmu;
 
 	if (!event->parent) {
-		atomic_inc(&nr_events);
+		if (event->attach_state & PERF_ATTACH_TASK)
+			jump_label_inc(&perf_task_events);
 		if (event->attr.mmap || event->attr.mmap_data)
 			atomic_inc(&nr_mmap_events);
 		if (event->attr.comm)
 			atomic_inc(&nr_comm_events);
 		if (event->attr.task)
 			atomic_inc(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
+			err = get_callchain_buffers();
+			if (err) {
+				free_event(event);
+				return ERR_PTR(err);
+			}
+		}
 	}
 
 	return event;
@@ -5094,12 +5507,16 @@
 		struct perf_event_attr __user *, attr_uptr,
 		pid_t, pid, int, cpu, int, group_fd, unsigned long, flags)
 {
-	struct perf_event *event, *group_leader = NULL, *output_event = NULL;
+	struct perf_event *group_leader = NULL, *output_event = NULL;
+	struct perf_event *event, *sibling;
 	struct perf_event_attr attr;
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
 	struct file *group_file = NULL;
+	struct task_struct *task = NULL;
+	struct pmu *pmu;
 	int event_fd;
+	int move_group = 0;
 	int fput_needed = 0;
 	int err;
 
@@ -5125,20 +5542,11 @@
 	if (event_fd < 0)
 		return event_fd;
 
-	/*
-	 * Get the target context (task or percpu):
-	 */
-	ctx = find_get_context(pid, cpu);
-	if (IS_ERR(ctx)) {
-		err = PTR_ERR(ctx);
-		goto err_fd;
-	}
-
 	if (group_fd != -1) {
 		group_leader = perf_fget_light(group_fd, &fput_needed);
 		if (IS_ERR(group_leader)) {
 			err = PTR_ERR(group_leader);
-			goto err_put_context;
+			goto err_fd;
 		}
 		group_file = group_leader->filp;
 		if (flags & PERF_FLAG_FD_OUTPUT)
@@ -5147,6 +5555,58 @@
 			group_leader = NULL;
 	}
 
+	if (pid != -1) {
+		task = find_lively_task_by_vpid(pid);
+		if (IS_ERR(task)) {
+			err = PTR_ERR(task);
+			goto err_group_fd;
+		}
+	}
+
+	event = perf_event_alloc(&attr, cpu, task, group_leader, NULL, NULL);
+	if (IS_ERR(event)) {
+		err = PTR_ERR(event);
+		goto err_task;
+	}
+
+	/*
+	 * Special case software events and allow them to be part of
+	 * any hardware group.
+	 */
+	pmu = event->pmu;
+
+	if (group_leader &&
+	    (is_software_event(event) != is_software_event(group_leader))) {
+		if (is_software_event(event)) {
+			/*
+			 * If event and group_leader are not both a software
+			 * event, and event is, then group leader is not.
+			 *
+			 * Allow the addition of software events to !software
+			 * groups, this is safe because software events never
+			 * fail to schedule.
+			 */
+			pmu = group_leader->pmu;
+		} else if (is_software_event(group_leader) &&
+			   (group_leader->group_flags & PERF_GROUP_SOFTWARE)) {
+			/*
+			 * In case the group is a pure software group, and we
+			 * try to add a hardware event, move the whole group to
+			 * the hardware context.
+			 */
+			move_group = 1;
+		}
+	}
+
+	/*
+	 * Get the target context (task or percpu):
+	 */
+	ctx = find_get_context(pmu, task, cpu);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto err_alloc;
+	}
+
 	/*
 	 * Look up the group leader (we will attach this event to it):
 	 */
@@ -5158,42 +5618,66 @@
 		 * becoming part of another group-sibling):
 		 */
 		if (group_leader->group_leader != group_leader)
-			goto err_put_context;
+			goto err_context;
 		/*
 		 * Do not allow to attach to a group in a different
 		 * task or CPU context:
 		 */
-		if (group_leader->ctx != ctx)
-			goto err_put_context;
+		if (move_group) {
+			if (group_leader->ctx->type != ctx->type)
+				goto err_context;
+		} else {
+			if (group_leader->ctx != ctx)
+				goto err_context;
+		}
+
 		/*
 		 * Only a group leader can be exclusive or pinned
 		 */
 		if (attr.exclusive || attr.pinned)
-			goto err_put_context;
-	}
-
-	event = perf_event_alloc(&attr, cpu, ctx, group_leader,
-				     NULL, NULL, GFP_KERNEL);
-	if (IS_ERR(event)) {
-		err = PTR_ERR(event);
-		goto err_put_context;
+			goto err_context;
 	}
 
 	if (output_event) {
 		err = perf_event_set_output(event, output_event);
 		if (err)
-			goto err_free_put_context;
+			goto err_context;
 	}
 
 	event_file = anon_inode_getfile("[perf_event]", &perf_fops, event, O_RDWR);
 	if (IS_ERR(event_file)) {
 		err = PTR_ERR(event_file);
-		goto err_free_put_context;
+		goto err_context;
+	}
+
+	if (move_group) {
+		struct perf_event_context *gctx = group_leader->ctx;
+
+		mutex_lock(&gctx->mutex);
+		perf_event_remove_from_context(group_leader);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_event_remove_from_context(sibling);
+			put_ctx(gctx);
+		}
+		mutex_unlock(&gctx->mutex);
+		put_ctx(gctx);
 	}
 
 	event->filp = event_file;
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
+
+	if (move_group) {
+		perf_install_in_context(ctx, group_leader, cpu);
+		get_ctx(ctx);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_install_in_context(ctx, sibling, cpu);
+			get_ctx(ctx);
+		}
+	}
+
 	perf_install_in_context(ctx, event, cpu);
 	++ctx->generation;
 	mutex_unlock(&ctx->mutex);
@@ -5214,11 +5698,15 @@
 	fd_install(event_fd, event_file);
 	return event_fd;
 
-err_free_put_context:
-	free_event(event);
-err_put_context:
-	fput_light(group_file, fput_needed);
+err_context:
 	put_ctx(ctx);
+err_alloc:
+	free_event(event);
+err_task:
+	if (task)
+		put_task_struct(task);
+err_group_fd:
+	fput_light(group_file, fput_needed);
 err_fd:
 	put_unused_fd(event_fd);
 	return err;
@@ -5229,32 +5717,31 @@
  *
  * @attr: attributes of the counter to create
  * @cpu: cpu in which the counter is bound
- * @pid: task to profile
+ * @task: task to profile (NULL for percpu)
  */
 struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
-				 pid_t pid,
+				 struct task_struct *task,
 				 perf_overflow_handler_t overflow_handler)
 {
-	struct perf_event *event;
 	struct perf_event_context *ctx;
+	struct perf_event *event;
 	int err;
 
 	/*
 	 * Get the target context (task or percpu):
 	 */
 
-	ctx = find_get_context(pid, cpu);
-	if (IS_ERR(ctx)) {
-		err = PTR_ERR(ctx);
-		goto err_exit;
-	}
-
-	event = perf_event_alloc(attr, cpu, ctx, NULL,
-				 NULL, overflow_handler, GFP_KERNEL);
+	event = perf_event_alloc(attr, cpu, task, NULL, NULL, overflow_handler);
 	if (IS_ERR(event)) {
 		err = PTR_ERR(event);
-		goto err_put_context;
+		goto err;
+	}
+
+	ctx = find_get_context(event->pmu, task, cpu);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto err_free;
 	}
 
 	event->filp = NULL;
@@ -5272,112 +5759,13 @@
 
 	return event;
 
- err_put_context:
-	put_ctx(ctx);
- err_exit:
+err_free:
+	free_event(event);
+err:
 	return ERR_PTR(err);
 }
 EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter);
 
-/*
- * inherit a event from parent task to child task:
- */
-static struct perf_event *
-inherit_event(struct perf_event *parent_event,
-	      struct task_struct *parent,
-	      struct perf_event_context *parent_ctx,
-	      struct task_struct *child,
-	      struct perf_event *group_leader,
-	      struct perf_event_context *child_ctx)
-{
-	struct perf_event *child_event;
-
-	/*
-	 * Instead of creating recursive hierarchies of events,
-	 * we link inherited events back to the original parent,
-	 * which has a filp for sure, which we use as the reference
-	 * count:
-	 */
-	if (parent_event->parent)
-		parent_event = parent_event->parent;
-
-	child_event = perf_event_alloc(&parent_event->attr,
-					   parent_event->cpu, child_ctx,
-					   group_leader, parent_event,
-					   NULL, GFP_KERNEL);
-	if (IS_ERR(child_event))
-		return child_event;
-	get_ctx(child_ctx);
-
-	/*
-	 * Make the child state follow the state of the parent event,
-	 * not its attr.disabled bit.  We hold the parent's mutex,
-	 * so we won't race with perf_event_{en, dis}able_family.
-	 */
-	if (parent_event->state >= PERF_EVENT_STATE_INACTIVE)
-		child_event->state = PERF_EVENT_STATE_INACTIVE;
-	else
-		child_event->state = PERF_EVENT_STATE_OFF;
-
-	if (parent_event->attr.freq) {
-		u64 sample_period = parent_event->hw.sample_period;
-		struct hw_perf_event *hwc = &child_event->hw;
-
-		hwc->sample_period = sample_period;
-		hwc->last_period   = sample_period;
-
-		local64_set(&hwc->period_left, sample_period);
-	}
-
-	child_event->overflow_handler = parent_event->overflow_handler;
-
-	/*
-	 * Link it up in the child's context:
-	 */
-	add_event_to_ctx(child_event, child_ctx);
-
-	/*
-	 * Get a reference to the parent filp - we will fput it
-	 * when the child event exits. This is safe to do because
-	 * we are in the parent and we know that the filp still
-	 * exists and has a nonzero count:
-	 */
-	atomic_long_inc(&parent_event->filp->f_count);
-
-	/*
-	 * Link this into the parent event's child list
-	 */
-	WARN_ON_ONCE(parent_event->ctx->parent_ctx);
-	mutex_lock(&parent_event->child_mutex);
-	list_add_tail(&child_event->child_list, &parent_event->child_list);
-	mutex_unlock(&parent_event->child_mutex);
-
-	return child_event;
-}
-
-static int inherit_group(struct perf_event *parent_event,
-	      struct task_struct *parent,
-	      struct perf_event_context *parent_ctx,
-	      struct task_struct *child,
-	      struct perf_event_context *child_ctx)
-{
-	struct perf_event *leader;
-	struct perf_event *sub;
-	struct perf_event *child_ctr;
-
-	leader = inherit_event(parent_event, parent, parent_ctx,
-				 child, NULL, child_ctx);
-	if (IS_ERR(leader))
-		return PTR_ERR(leader);
-	list_for_each_entry(sub, &parent_event->sibling_list, group_entry) {
-		child_ctr = inherit_event(sub, parent, parent_ctx,
-					    child, leader, child_ctx);
-		if (IS_ERR(child_ctr))
-			return PTR_ERR(child_ctr);
-	}
-	return 0;
-}
-
 static void sync_child_event(struct perf_event *child_event,
 			       struct task_struct *child)
 {
@@ -5434,16 +5822,13 @@
 	}
 }
 
-/*
- * When a child task exits, feed back event values to parent events.
- */
-void perf_event_exit_task(struct task_struct *child)
+static void perf_event_exit_task_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event *child_event, *tmp;
 	struct perf_event_context *child_ctx;
 	unsigned long flags;
 
-	if (likely(!child->perf_event_ctxp)) {
+	if (likely(!child->perf_event_ctxp[ctxn])) {
 		perf_event_task(child, NULL, 0);
 		return;
 	}
@@ -5455,8 +5840,8 @@
 	 * scheduled, so we are now safe from rescheduling changing
 	 * our context.
 	 */
-	child_ctx = child->perf_event_ctxp;
-	__perf_event_task_sched_out(child_ctx);
+	child_ctx = child->perf_event_ctxp[ctxn];
+	task_ctx_sched_out(child_ctx, EVENT_ALL);
 
 	/*
 	 * Take the context lock here so that if find_get_context is
@@ -5464,7 +5849,7 @@
 	 * incremented the context's refcount before we do put_ctx below.
 	 */
 	raw_spin_lock(&child_ctx->lock);
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 	/*
 	 * If this context is a clone; unclone it so it can't get
 	 * swapped to another process while we're removing all
@@ -5517,6 +5902,17 @@
 	put_ctx(child_ctx);
 }
 
+/*
+ * When a child task exits, feed back event values to parent events.
+ */
+void perf_event_exit_task(struct task_struct *child)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		perf_event_exit_task_context(child, ctxn);
+}
+
 static void perf_free_event(struct perf_event *event,
 			    struct perf_event_context *ctx)
 {
@@ -5538,48 +5934,166 @@
 
 /*
  * free an unexposed, unused context as created by inheritance by
- * init_task below, used by fork() in case of fail.
+ * perf_event_init_task below, used by fork() in case of fail.
  */
 void perf_event_free_task(struct task_struct *task)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx;
 	struct perf_event *event, *tmp;
+	int ctxn;
 
-	if (!ctx)
-		return;
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
 
-	mutex_lock(&ctx->mutex);
+		mutex_lock(&ctx->mutex);
 again:
-	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->pinned_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
-				 group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	if (!list_empty(&ctx->pinned_groups) ||
-	    !list_empty(&ctx->flexible_groups))
-		goto again;
+		if (!list_empty(&ctx->pinned_groups) ||
+				!list_empty(&ctx->flexible_groups))
+			goto again;
 
-	mutex_unlock(&ctx->mutex);
+		mutex_unlock(&ctx->mutex);
 
-	put_ctx(ctx);
+		put_ctx(ctx);
+	}
+}
+
+void perf_event_delayed_put(struct task_struct *task)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		WARN_ON_ONCE(task->perf_event_ctxp[ctxn]);
+}
+
+/*
+ * inherit a event from parent task to child task:
+ */
+static struct perf_event *
+inherit_event(struct perf_event *parent_event,
+	      struct task_struct *parent,
+	      struct perf_event_context *parent_ctx,
+	      struct task_struct *child,
+	      struct perf_event *group_leader,
+	      struct perf_event_context *child_ctx)
+{
+	struct perf_event *child_event;
+	unsigned long flags;
+
+	/*
+	 * Instead of creating recursive hierarchies of events,
+	 * we link inherited events back to the original parent,
+	 * which has a filp for sure, which we use as the reference
+	 * count:
+	 */
+	if (parent_event->parent)
+		parent_event = parent_event->parent;
+
+	child_event = perf_event_alloc(&parent_event->attr,
+					   parent_event->cpu,
+					   child,
+					   group_leader, parent_event,
+					   NULL);
+	if (IS_ERR(child_event))
+		return child_event;
+	get_ctx(child_ctx);
+
+	/*
+	 * Make the child state follow the state of the parent event,
+	 * not its attr.disabled bit.  We hold the parent's mutex,
+	 * so we won't race with perf_event_{en, dis}able_family.
+	 */
+	if (parent_event->state >= PERF_EVENT_STATE_INACTIVE)
+		child_event->state = PERF_EVENT_STATE_INACTIVE;
+	else
+		child_event->state = PERF_EVENT_STATE_OFF;
+
+	if (parent_event->attr.freq) {
+		u64 sample_period = parent_event->hw.sample_period;
+		struct hw_perf_event *hwc = &child_event->hw;
+
+		hwc->sample_period = sample_period;
+		hwc->last_period   = sample_period;
+
+		local64_set(&hwc->period_left, sample_period);
+	}
+
+	child_event->ctx = child_ctx;
+	child_event->overflow_handler = parent_event->overflow_handler;
+
+	/*
+	 * Link it up in the child's context:
+	 */
+	raw_spin_lock_irqsave(&child_ctx->lock, flags);
+	add_event_to_ctx(child_event, child_ctx);
+	raw_spin_unlock_irqrestore(&child_ctx->lock, flags);
+
+	/*
+	 * Get a reference to the parent filp - we will fput it
+	 * when the child event exits. This is safe to do because
+	 * we are in the parent and we know that the filp still
+	 * exists and has a nonzero count:
+	 */
+	atomic_long_inc(&parent_event->filp->f_count);
+
+	/*
+	 * Link this into the parent event's child list
+	 */
+	WARN_ON_ONCE(parent_event->ctx->parent_ctx);
+	mutex_lock(&parent_event->child_mutex);
+	list_add_tail(&child_event->child_list, &parent_event->child_list);
+	mutex_unlock(&parent_event->child_mutex);
+
+	return child_event;
+}
+
+static int inherit_group(struct perf_event *parent_event,
+	      struct task_struct *parent,
+	      struct perf_event_context *parent_ctx,
+	      struct task_struct *child,
+	      struct perf_event_context *child_ctx)
+{
+	struct perf_event *leader;
+	struct perf_event *sub;
+	struct perf_event *child_ctr;
+
+	leader = inherit_event(parent_event, parent, parent_ctx,
+				 child, NULL, child_ctx);
+	if (IS_ERR(leader))
+		return PTR_ERR(leader);
+	list_for_each_entry(sub, &parent_event->sibling_list, group_entry) {
+		child_ctr = inherit_event(sub, parent, parent_ctx,
+					    child, leader, child_ctx);
+		if (IS_ERR(child_ctr))
+			return PTR_ERR(child_ctr);
+	}
+	return 0;
 }
 
 static int
 inherit_task_group(struct perf_event *event, struct task_struct *parent,
 		   struct perf_event_context *parent_ctx,
-		   struct task_struct *child,
+		   struct task_struct *child, int ctxn,
 		   int *inherited_all)
 {
 	int ret;
-	struct perf_event_context *child_ctx = child->perf_event_ctxp;
+	struct perf_event_context *child_ctx;
 
 	if (!event->attr.inherit) {
 		*inherited_all = 0;
 		return 0;
 	}
 
+       	child_ctx = child->perf_event_ctxp[ctxn];
 	if (!child_ctx) {
 		/*
 		 * This is executed from the parent task context, so
@@ -5588,14 +6102,11 @@
 		 * child.
 		 */
 
-		child_ctx = kzalloc(sizeof(struct perf_event_context),
-				    GFP_KERNEL);
+		child_ctx = alloc_perf_context(event->pmu, child);
 		if (!child_ctx)
 			return -ENOMEM;
 
-		__perf_event_init_context(child_ctx, child);
-		child->perf_event_ctxp = child_ctx;
-		get_task_struct(child);
+		child->perf_event_ctxp[ctxn] = child_ctx;
 	}
 
 	ret = inherit_group(event, parent, parent_ctx,
@@ -5607,11 +6118,10 @@
 	return ret;
 }
 
-
 /*
  * Initialize the perf_event context in task_struct
  */
-int perf_event_init_task(struct task_struct *child)
+int perf_event_init_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event_context *child_ctx, *parent_ctx;
 	struct perf_event_context *cloned_ctx;
@@ -5620,19 +6130,19 @@
 	int inherited_all = 1;
 	int ret = 0;
 
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 
 	mutex_init(&child->perf_event_mutex);
 	INIT_LIST_HEAD(&child->perf_event_list);
 
-	if (likely(!parent->perf_event_ctxp))
+	if (likely(!parent->perf_event_ctxp[ctxn]))
 		return 0;
 
 	/*
 	 * If the parent's context is a clone, pin it so it won't get
 	 * swapped under us.
 	 */
-	parent_ctx = perf_pin_task_context(parent);
+	parent_ctx = perf_pin_task_context(parent, ctxn);
 
 	/*
 	 * No need to check if parent_ctx != NULL here; since we saw
@@ -5652,20 +6162,20 @@
 	 * the list, not manipulating it:
 	 */
 	list_for_each_entry(event, &parent_ctx->pinned_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
 	list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
-	child_ctx = child->perf_event_ctxp;
+	child_ctx = child->perf_event_ctxp[ctxn];
 
 	if (child_ctx && inherited_all) {
 		/*
@@ -5694,63 +6204,98 @@
 	return ret;
 }
 
+/*
+ * Initialize the perf_event context in task_struct
+ */
+int perf_event_init_task(struct task_struct *child)
+{
+	int ctxn, ret;
+
+	for_each_task_context_nr(ctxn) {
+		ret = perf_event_init_context(child, ctxn);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static void __init perf_event_init_all_cpus(void)
 {
+	struct swevent_htable *swhash;
 	int cpu;
-	struct perf_cpu_context *cpuctx;
 
 	for_each_possible_cpu(cpu) {
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		mutex_init(&cpuctx->hlist_mutex);
-		__perf_event_init_context(&cpuctx->ctx, NULL);
+		swhash = &per_cpu(swevent_htable, cpu);
+		mutex_init(&swhash->hlist_mutex);
+		INIT_LIST_HEAD(&per_cpu(rotation_list, cpu));
 	}
 }
 
 static void __cpuinit perf_event_init_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	cpuctx = &per_cpu(perf_cpu_context, cpu);
-
-	spin_lock(&perf_resource_lock);
-	cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
-	spin_unlock(&perf_resource_lock);
-
-	mutex_lock(&cpuctx->hlist_mutex);
-	if (cpuctx->hlist_refcount > 0) {
+	mutex_lock(&swhash->hlist_mutex);
+	if (swhash->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
 
-		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
-		WARN_ON_ONCE(!hlist);
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu));
+		WARN_ON(!hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static void __perf_event_exit_cpu(void *info)
+static void perf_pmu_rotate_stop(struct pmu *pmu)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+
+	WARN_ON(!irqs_disabled());
+
+	list_del_init(&cpuctx->rotation_list);
+}
+
+static void __perf_event_exit_context(void *__info)
+{
+	struct perf_event_context *ctx = __info;
 	struct perf_event *event, *tmp;
 
+	perf_pmu_rotate_stop(ctx->pmu);
+
 	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
 		__perf_event_remove_from_context(event);
 	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry)
 		__perf_event_remove_from_context(event);
 }
+
+static void perf_event_exit_cpu_context(int cpu)
+{
+	struct perf_event_context *ctx;
+	struct pmu *pmu;
+	int idx;
+
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		ctx = &per_cpu_ptr(pmu->pmu_cpu_context, cpu)->ctx;
+
+		mutex_lock(&ctx->mutex);
+		smp_call_function_single(cpu, __perf_event_exit_context, ctx, 1);
+		mutex_unlock(&ctx->mutex);
+	}
+	srcu_read_unlock(&pmus_srcu, idx);
+}
+
 static void perf_event_exit_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
-	swevent_hlist_release(cpuctx);
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
+	swevent_hlist_release(swhash);
+	mutex_unlock(&swhash->hlist_mutex);
 
-	mutex_lock(&ctx->mutex);
-	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
-	mutex_unlock(&ctx->mutex);
+	perf_event_exit_cpu_context(cpu);
 }
 #else
 static inline void perf_event_exit_cpu(int cpu) { }
@@ -5780,118 +6325,13 @@
 	return NOTIFY_OK;
 }
 
-/*
- * This has to have a higher priority than migration_notifier in sched.c.
- */
-static struct notifier_block __cpuinitdata perf_cpu_nb = {
-	.notifier_call		= perf_cpu_notify,
-	.priority		= 20,
-};
-
 void __init perf_event_init(void)
 {
 	perf_event_init_all_cpus();
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE,
-			(void *)(long)smp_processor_id());
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE,
-			(void *)(long)smp_processor_id());
-	register_cpu_notifier(&perf_cpu_nb);
+	init_srcu_struct(&pmus_srcu);
+	perf_pmu_register(&perf_swevent);
+	perf_pmu_register(&perf_cpu_clock);
+	perf_pmu_register(&perf_task_clock);
+	perf_tp_register();
+	perf_cpu_notifier(perf_cpu_notify);
 }
-
-static ssize_t perf_show_reserve_percpu(struct sysdev_class *class,
-					struct sysdev_class_attribute *attr,
-					char *buf)
-{
-	return sprintf(buf, "%d\n", perf_reserved_percpu);
-}
-
-static ssize_t
-perf_set_reserve_percpu(struct sysdev_class *class,
-			struct sysdev_class_attribute *attr,
-			const char *buf,
-			size_t count)
-{
-	struct perf_cpu_context *cpuctx;
-	unsigned long val;
-	int err, cpu, mpt;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > perf_max_events)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_reserved_percpu = val;
-	for_each_online_cpu(cpu) {
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		raw_spin_lock_irq(&cpuctx->ctx.lock);
-		mpt = min(perf_max_events - cpuctx->ctx.nr_events,
-			  perf_max_events - perf_reserved_percpu);
-		cpuctx->max_pertask = mpt;
-		raw_spin_unlock_irq(&cpuctx->ctx.lock);
-	}
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static ssize_t perf_show_overcommit(struct sysdev_class *class,
-				    struct sysdev_class_attribute *attr,
-				    char *buf)
-{
-	return sprintf(buf, "%d\n", perf_overcommit);
-}
-
-static ssize_t
-perf_set_overcommit(struct sysdev_class *class,
-		    struct sysdev_class_attribute *attr,
-		    const char *buf, size_t count)
-{
-	unsigned long val;
-	int err;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > 1)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_overcommit = val;
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static SYSDEV_CLASS_ATTR(
-				reserve_percpu,
-				0644,
-				perf_show_reserve_percpu,
-				perf_set_reserve_percpu
-			);
-
-static SYSDEV_CLASS_ATTR(
-				overcommit,
-				0644,
-				perf_show_overcommit,
-				perf_set_overcommit
-			);
-
-static struct attribute *perfclass_attrs[] = {
-	&attr_reserve_percpu.attr,
-	&attr_overcommit.attr,
-	NULL
-};
-
-static struct attribute_group perfclass_attr_group = {
-	.attrs			= perfclass_attrs,
-	.name			= "perf_events",
-};
-
-static int __init perf_event_sysfs_init(void)
-{
-	return sysfs_create_group(&cpu_sysdev_class.kset.kobj,
-				  &perfclass_attr_group);
-}
-device_initcall(perf_event_sysfs_init);
diff --git a/kernel/pid.c b/kernel/pid.c
index d55c6fb..39b65b6 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -401,7 +401,7 @@
 	struct task_struct *result = NULL;
 	if (pid) {
 		struct hlist_node *first;
-		first = rcu_dereference_check(pid->tasks[type].first,
+		first = rcu_dereference_check(hlist_first_rcu(&pid->tasks[type]),
 					      rcu_read_lock_held() ||
 					      lockdep_tasklist_lock_is_held());
 		if (first)
@@ -416,6 +416,7 @@
  */
 struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns)
 {
+	rcu_lockdep_assert(rcu_read_lock_held());
 	return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID);
 }
 
diff --git a/kernel/printk.c b/kernel/printk.c
index 8fe465a..2531017 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -85,7 +85,7 @@
  * provides serialisation for access to the entire console
  * driver system.
  */
-static DECLARE_MUTEX(console_sem);
+static DEFINE_SEMAPHORE(console_sem);
 struct console *console_drivers;
 EXPORT_SYMBOL_GPL(console_drivers);
 
@@ -556,7 +556,7 @@
 	/* If a crash is occurring, make sure we can't deadlock */
 	spin_lock_init(&logbuf_lock);
 	/* And make sure that we print immediately */
-	init_MUTEX(&console_sem);
+	sema_init(&console_sem, 1);
 }
 
 #if defined(CONFIG_PRINTK_TIME)
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index 4d16983..a23a57a 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -73,12 +73,14 @@
 EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled);
 
 /**
- * rcu_read_lock_bh_held - might we be in RCU-bh read-side critical section?
+ * rcu_read_lock_bh_held() - might we be in RCU-bh read-side critical section?
  *
  * Check for bottom half being disabled, which covers both the
  * CONFIG_PROVE_RCU and not cases.  Note that if someone uses
  * rcu_read_lock_bh(), but then later enables BH, lockdep (if enabled)
- * will show the situation.
+ * will show the situation.  This is useful for debug checks in functions
+ * that require that they be called within an RCU read-side critical
+ * section.
  *
  * Check debug_lockdep_rcu_enabled() to prevent false positives during boot.
  */
@@ -86,7 +88,7 @@
 {
 	if (!debug_lockdep_rcu_enabled())
 		return 1;
-	return in_softirq();
+	return in_softirq() || irqs_disabled();
 }
 EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held);
 
diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c
index 196ec02..d806735 100644
--- a/kernel/rcutiny.c
+++ b/kernel/rcutiny.c
@@ -59,6 +59,14 @@
 EXPORT_SYMBOL_GPL(rcu_scheduler_active);
 #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
+/* Forward declarations for rcutiny_plugin.h. */
+static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp);
+static void __call_rcu(struct rcu_head *head,
+		       void (*func)(struct rcu_head *rcu),
+		       struct rcu_ctrlblk *rcp);
+
+#include "rcutiny_plugin.h"
+
 #ifdef CONFIG_NO_HZ
 
 static long rcu_dynticks_nesting = 1;
@@ -140,6 +148,7 @@
 		rcu_sched_qs(cpu);
 	else if (!in_softirq())
 		rcu_bh_qs(cpu);
+	rcu_preempt_check_callbacks();
 }
 
 /*
@@ -162,6 +171,7 @@
 	*rcp->donetail = NULL;
 	if (rcp->curtail == rcp->donetail)
 		rcp->curtail = &rcp->rcucblist;
+	rcu_preempt_remove_callbacks(rcp);
 	rcp->donetail = &rcp->rcucblist;
 	local_irq_restore(flags);
 
@@ -182,6 +192,7 @@
 {
 	__rcu_process_callbacks(&rcu_sched_ctrlblk);
 	__rcu_process_callbacks(&rcu_bh_ctrlblk);
+	rcu_preempt_process_callbacks();
 }
 
 /*
@@ -223,15 +234,15 @@
 }
 
 /*
- * Post an RCU callback to be invoked after the end of an RCU grace
+ * Post an RCU callback to be invoked after the end of an RCU-sched grace
  * period.  But since we have but one CPU, that would be after any
  * quiescent state.
  */
-void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
 {
 	__call_rcu(head, func, &rcu_sched_ctrlblk);
 }
-EXPORT_SYMBOL_GPL(call_rcu);
+EXPORT_SYMBOL_GPL(call_rcu_sched);
 
 /*
  * Post an RCU bottom-half callback to be invoked after any subsequent
@@ -243,20 +254,6 @@
 }
 EXPORT_SYMBOL_GPL(call_rcu_bh);
 
-void rcu_barrier(void)
-{
-	struct rcu_synchronize rcu;
-
-	init_rcu_head_on_stack(&rcu.head);
-	init_completion(&rcu.completion);
-	/* Will wake me after RCU finished. */
-	call_rcu(&rcu.head, wakeme_after_rcu);
-	/* Wait for it. */
-	wait_for_completion(&rcu.completion);
-	destroy_rcu_head_on_stack(&rcu.head);
-}
-EXPORT_SYMBOL_GPL(rcu_barrier);
-
 void rcu_barrier_bh(void)
 {
 	struct rcu_synchronize rcu;
@@ -289,5 +286,3 @@
 {
 	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 }
-
-#include "rcutiny_plugin.h"
diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h
index d223a92bc..6ceca4f 100644
--- a/kernel/rcutiny_plugin.h
+++ b/kernel/rcutiny_plugin.h
@@ -1,7 +1,7 @@
 /*
- * Read-Copy Update mechanism for mutual exclusion (tree-based version)
+ * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition
  * Internal non-public definitions that provide either classic
- * or preemptable semantics.
+ * or preemptible semantics.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -17,11 +17,587 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  *
- * Copyright IBM Corporation, 2009
+ * Copyright (c) 2010 Linaro
  *
  * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
  */
 
+#ifdef CONFIG_TINY_PREEMPT_RCU
+
+#include <linux/delay.h>
+
+/* Global control variables for preemptible RCU. */
+struct rcu_preempt_ctrlblk {
+	struct rcu_ctrlblk rcb;	/* curtail: ->next ptr of last CB for GP. */
+	struct rcu_head **nexttail;
+				/* Tasks blocked in a preemptible RCU */
+				/*  read-side critical section while an */
+				/*  preemptible-RCU grace period is in */
+				/*  progress must wait for a later grace */
+				/*  period.  This pointer points to the */
+				/*  ->next pointer of the last task that */
+				/*  must wait for a later grace period, or */
+				/*  to &->rcb.rcucblist if there is no */
+				/*  such task. */
+	struct list_head blkd_tasks;
+				/* Tasks blocked in RCU read-side critical */
+				/*  section.  Tasks are placed at the head */
+				/*  of this list and age towards the tail. */
+	struct list_head *gp_tasks;
+				/* Pointer to the first task blocking the */
+				/*  current grace period, or NULL if there */
+				/*  is not such task. */
+	struct list_head *exp_tasks;
+				/* Pointer to first task blocking the */
+				/*  current expedited grace period, or NULL */
+				/*  if there is no such task.  If there */
+				/*  is no current expedited grace period, */
+				/*  then there cannot be any such task. */
+	u8 gpnum;		/* Current grace period. */
+	u8 gpcpu;		/* Last grace period blocked by the CPU. */
+	u8 completed;		/* Last grace period completed. */
+				/*  If all three are equal, RCU is idle. */
+};
+
+static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = {
+	.rcb.donetail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.rcb.curtail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.nexttail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.blkd_tasks = LIST_HEAD_INIT(rcu_preempt_ctrlblk.blkd_tasks),
+};
+
+static int rcu_preempted_readers_exp(void);
+static void rcu_report_exp_done(void);
+
+/*
+ * Return true if the CPU has not yet responded to the current grace period.
+ */
+static int rcu_cpu_blocking_cur_gp(void)
+{
+	return rcu_preempt_ctrlblk.gpcpu != rcu_preempt_ctrlblk.gpnum;
+}
+
+/*
+ * Check for a running RCU reader.  Because there is only one CPU,
+ * there can be but one running RCU reader at a time.  ;-)
+ */
+static int rcu_preempt_running_reader(void)
+{
+	return current->rcu_read_lock_nesting;
+}
+
+/*
+ * Check for preempted RCU readers blocking any grace period.
+ * If the caller needs a reliable answer, it must disable hard irqs.
+ */
+static int rcu_preempt_blocked_readers_any(void)
+{
+	return !list_empty(&rcu_preempt_ctrlblk.blkd_tasks);
+}
+
+/*
+ * Check for preempted RCU readers blocking the current grace period.
+ * If the caller needs a reliable answer, it must disable hard irqs.
+ */
+static int rcu_preempt_blocked_readers_cgp(void)
+{
+	return rcu_preempt_ctrlblk.gp_tasks != NULL;
+}
+
+/*
+ * Return true if another preemptible-RCU grace period is needed.
+ */
+static int rcu_preempt_needs_another_gp(void)
+{
+	return *rcu_preempt_ctrlblk.rcb.curtail != NULL;
+}
+
+/*
+ * Return true if a preemptible-RCU grace period is in progress.
+ * The caller must disable hardirqs.
+ */
+static int rcu_preempt_gp_in_progress(void)
+{
+	return rcu_preempt_ctrlblk.completed != rcu_preempt_ctrlblk.gpnum;
+}
+
+/*
+ * Record a preemptible-RCU quiescent state for the specified CPU.  Note
+ * that this just means that the task currently running on the CPU is
+ * in a quiescent state.  There might be any number of tasks blocked
+ * while in an RCU read-side critical section.
+ *
+ * Unlike the other rcu_*_qs() functions, callers to this function
+ * must disable irqs in order to protect the assignment to
+ * ->rcu_read_unlock_special.
+ *
+ * Because this is a single-CPU implementation, the only way a grace
+ * period can end is if the CPU is in a quiescent state.  The reason is
+ * that a blocked preemptible-RCU reader can exit its critical section
+ * only if the CPU is running it at the time.  Therefore, when the
+ * last task blocking the current grace period exits its RCU read-side
+ * critical section, neither the CPU nor blocked tasks will be stopping
+ * the current grace period.  (In contrast, SMP implementations
+ * might have CPUs running in RCU read-side critical sections that
+ * block later grace periods -- but this is not possible given only
+ * one CPU.)
+ */
+static void rcu_preempt_cpu_qs(void)
+{
+	/* Record both CPU and task as having responded to current GP. */
+	rcu_preempt_ctrlblk.gpcpu = rcu_preempt_ctrlblk.gpnum;
+	current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
+
+	/*
+	 * If there is no GP, or if blocked readers are still blocking GP,
+	 * then there is nothing more to do.
+	 */
+	if (!rcu_preempt_gp_in_progress() || rcu_preempt_blocked_readers_cgp())
+		return;
+
+	/* Advance callbacks. */
+	rcu_preempt_ctrlblk.completed = rcu_preempt_ctrlblk.gpnum;
+	rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.rcb.curtail;
+	rcu_preempt_ctrlblk.rcb.curtail = rcu_preempt_ctrlblk.nexttail;
+
+	/* If there are no blocked readers, next GP is done instantly. */
+	if (!rcu_preempt_blocked_readers_any())
+		rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.nexttail;
+
+	/* If there are done callbacks, make RCU_SOFTIRQ process them. */
+	if (*rcu_preempt_ctrlblk.rcb.donetail != NULL)
+		raise_softirq(RCU_SOFTIRQ);
+}
+
+/*
+ * Start a new RCU grace period if warranted.  Hard irqs must be disabled.
+ */
+static void rcu_preempt_start_gp(void)
+{
+	if (!rcu_preempt_gp_in_progress() && rcu_preempt_needs_another_gp()) {
+
+		/* Official start of GP. */
+		rcu_preempt_ctrlblk.gpnum++;
+
+		/* Any blocked RCU readers block new GP. */
+		if (rcu_preempt_blocked_readers_any())
+			rcu_preempt_ctrlblk.gp_tasks =
+				rcu_preempt_ctrlblk.blkd_tasks.next;
+
+		/* If there is no running reader, CPU is done with GP. */
+		if (!rcu_preempt_running_reader())
+			rcu_preempt_cpu_qs();
+	}
+}
+
+/*
+ * We have entered the scheduler, and the current task might soon be
+ * context-switched away from.  If this task is in an RCU read-side
+ * critical section, we will no longer be able to rely on the CPU to
+ * record that fact, so we enqueue the task on the blkd_tasks list.
+ * If the task started after the current grace period began, as recorded
+ * by ->gpcpu, we enqueue at the beginning of the list.  Otherwise
+ * before the element referenced by ->gp_tasks (or at the tail if
+ * ->gp_tasks is NULL) and point ->gp_tasks at the newly added element.
+ * The task will dequeue itself when it exits the outermost enclosing
+ * RCU read-side critical section.  Therefore, the current grace period
+ * cannot be permitted to complete until the ->gp_tasks pointer becomes
+ * NULL.
+ *
+ * Caller must disable preemption.
+ */
+void rcu_preempt_note_context_switch(void)
+{
+	struct task_struct *t = current;
+	unsigned long flags;
+
+	local_irq_save(flags); /* must exclude scheduler_tick(). */
+	if (rcu_preempt_running_reader() &&
+	    (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) {
+
+		/* Possibly blocking in an RCU read-side critical section. */
+		t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
+
+		/*
+		 * If this CPU has already checked in, then this task
+		 * will hold up the next grace period rather than the
+		 * current grace period.  Queue the task accordingly.
+		 * If the task is queued for the current grace period
+		 * (i.e., this CPU has not yet passed through a quiescent
+		 * state for the current grace period), then as long
+		 * as that task remains queued, the current grace period
+		 * cannot end.
+		 */
+		list_add(&t->rcu_node_entry, &rcu_preempt_ctrlblk.blkd_tasks);
+		if (rcu_cpu_blocking_cur_gp())
+			rcu_preempt_ctrlblk.gp_tasks = &t->rcu_node_entry;
+	}
+
+	/*
+	 * Either we were not in an RCU read-side critical section to
+	 * begin with, or we have now recorded that critical section
+	 * globally.  Either way, we can now note a quiescent state
+	 * for this CPU.  Again, if we were in an RCU read-side critical
+	 * section, and if that critical section was blocking the current
+	 * grace period, then the fact that the task has been enqueued
+	 * means that current grace period continues to be blocked.
+	 */
+	rcu_preempt_cpu_qs();
+	local_irq_restore(flags);
+}
+
+/*
+ * Tiny-preemptible RCU implementation for rcu_read_lock().
+ * Just increment ->rcu_read_lock_nesting, shared state will be updated
+ * if we block.
+ */
+void __rcu_read_lock(void)
+{
+	current->rcu_read_lock_nesting++;
+	barrier();  /* needed if we ever invoke rcu_read_lock in rcutiny.c */
+}
+EXPORT_SYMBOL_GPL(__rcu_read_lock);
+
+/*
+ * Handle special cases during rcu_read_unlock(), such as needing to
+ * notify RCU core processing or task having blocked during the RCU
+ * read-side critical section.
+ */
+static void rcu_read_unlock_special(struct task_struct *t)
+{
+	int empty;
+	int empty_exp;
+	unsigned long flags;
+	struct list_head *np;
+	int special;
+
+	/*
+	 * NMI handlers cannot block and cannot safely manipulate state.
+	 * They therefore cannot possibly be special, so just leave.
+	 */
+	if (in_nmi())
+		return;
+
+	local_irq_save(flags);
+
+	/*
+	 * If RCU core is waiting for this CPU to exit critical section,
+	 * let it know that we have done so.
+	 */
+	special = t->rcu_read_unlock_special;
+	if (special & RCU_READ_UNLOCK_NEED_QS)
+		rcu_preempt_cpu_qs();
+
+	/* Hardware IRQ handlers cannot block. */
+	if (in_irq()) {
+		local_irq_restore(flags);
+		return;
+	}
+
+	/* Clean up if blocked during RCU read-side critical section. */
+	if (special & RCU_READ_UNLOCK_BLOCKED) {
+		t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED;
+
+		/*
+		 * Remove this task from the ->blkd_tasks list and adjust
+		 * any pointers that might have been referencing it.
+		 */
+		empty = !rcu_preempt_blocked_readers_cgp();
+		empty_exp = rcu_preempt_ctrlblk.exp_tasks == NULL;
+		np = t->rcu_node_entry.next;
+		if (np == &rcu_preempt_ctrlblk.blkd_tasks)
+			np = NULL;
+		list_del(&t->rcu_node_entry);
+		if (&t->rcu_node_entry == rcu_preempt_ctrlblk.gp_tasks)
+			rcu_preempt_ctrlblk.gp_tasks = np;
+		if (&t->rcu_node_entry == rcu_preempt_ctrlblk.exp_tasks)
+			rcu_preempt_ctrlblk.exp_tasks = np;
+		INIT_LIST_HEAD(&t->rcu_node_entry);
+
+		/*
+		 * If this was the last task on the current list, and if
+		 * we aren't waiting on the CPU, report the quiescent state
+		 * and start a new grace period if needed.
+		 */
+		if (!empty && !rcu_preempt_blocked_readers_cgp()) {
+			rcu_preempt_cpu_qs();
+			rcu_preempt_start_gp();
+		}
+
+		/*
+		 * If this was the last task on the expedited lists,
+		 * then we need wake up the waiting task.
+		 */
+		if (!empty_exp && rcu_preempt_ctrlblk.exp_tasks == NULL)
+			rcu_report_exp_done();
+	}
+	local_irq_restore(flags);
+}
+
+/*
+ * Tiny-preemptible RCU implementation for rcu_read_unlock().
+ * Decrement ->rcu_read_lock_nesting.  If the result is zero (outermost
+ * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then
+ * invoke rcu_read_unlock_special() to clean up after a context switch
+ * in an RCU read-side critical section and other special cases.
+ */
+void __rcu_read_unlock(void)
+{
+	struct task_struct *t = current;
+
+	barrier();  /* needed if we ever invoke rcu_read_unlock in rcutiny.c */
+	--t->rcu_read_lock_nesting;
+	barrier();  /* decrement before load of ->rcu_read_unlock_special */
+	if (t->rcu_read_lock_nesting == 0 &&
+	    unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
+		rcu_read_unlock_special(t);
+#ifdef CONFIG_PROVE_LOCKING
+	WARN_ON_ONCE(t->rcu_read_lock_nesting < 0);
+#endif /* #ifdef CONFIG_PROVE_LOCKING */
+}
+EXPORT_SYMBOL_GPL(__rcu_read_unlock);
+
+/*
+ * Check for a quiescent state from the current CPU.  When a task blocks,
+ * the task is recorded in the rcu_preempt_ctrlblk structure, which is
+ * checked elsewhere.  This is called from the scheduling-clock interrupt.
+ *
+ * Caller must disable hard irqs.
+ */
+static void rcu_preempt_check_callbacks(void)
+{
+	struct task_struct *t = current;
+
+	if (rcu_preempt_gp_in_progress() &&
+	    (!rcu_preempt_running_reader() ||
+	     !rcu_cpu_blocking_cur_gp()))
+		rcu_preempt_cpu_qs();
+	if (&rcu_preempt_ctrlblk.rcb.rcucblist !=
+	    rcu_preempt_ctrlblk.rcb.donetail)
+		raise_softirq(RCU_SOFTIRQ);
+	if (rcu_preempt_gp_in_progress() &&
+	    rcu_cpu_blocking_cur_gp() &&
+	    rcu_preempt_running_reader())
+		t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS;
+}
+
+/*
+ * TINY_PREEMPT_RCU has an extra callback-list tail pointer to
+ * update, so this is invoked from __rcu_process_callbacks() to
+ * handle that case.  Of course, it is invoked for all flavors of
+ * RCU, but RCU callbacks can appear only on one of the lists, and
+ * neither ->nexttail nor ->donetail can possibly be NULL, so there
+ * is no need for an explicit check.
+ */
+static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp)
+{
+	if (rcu_preempt_ctrlblk.nexttail == rcp->donetail)
+		rcu_preempt_ctrlblk.nexttail = &rcp->rcucblist;
+}
+
+/*
+ * Process callbacks for preemptible RCU.
+ */
+static void rcu_preempt_process_callbacks(void)
+{
+	__rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb);
+}
+
+/*
+ * Queue a preemptible -RCU callback for invocation after a grace period.
+ */
+void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+
+	debug_rcu_head_queue(head);
+	head->func = func;
+	head->next = NULL;
+
+	local_irq_save(flags);
+	*rcu_preempt_ctrlblk.nexttail = head;
+	rcu_preempt_ctrlblk.nexttail = &head->next;
+	rcu_preempt_start_gp();  /* checks to see if GP needed. */
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+void rcu_barrier(void)
+{
+	struct rcu_synchronize rcu;
+
+	init_rcu_head_on_stack(&rcu.head);
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished. */
+	call_rcu(&rcu.head, wakeme_after_rcu);
+	/* Wait for it. */
+	wait_for_completion(&rcu.completion);
+	destroy_rcu_head_on_stack(&rcu.head);
+}
+EXPORT_SYMBOL_GPL(rcu_barrier);
+
+/*
+ * synchronize_rcu - wait until a grace period has elapsed.
+ *
+ * Control will return to the caller some time after a full grace
+ * period has elapsed, in other words after all currently executing RCU
+ * read-side critical sections have completed.  RCU read-side critical
+ * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
+ * and may be nested.
+ */
+void synchronize_rcu(void)
+{
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (!rcu_scheduler_active)
+		return;
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+	WARN_ON_ONCE(rcu_preempt_running_reader());
+	if (!rcu_preempt_blocked_readers_any())
+		return;
+
+	/* Once we get past the fastpath checks, same code as rcu_barrier(). */
+	rcu_barrier();
+}
+EXPORT_SYMBOL_GPL(synchronize_rcu);
+
+static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq);
+static unsigned long sync_rcu_preempt_exp_count;
+static DEFINE_MUTEX(sync_rcu_preempt_exp_mutex);
+
+/*
+ * Return non-zero if there are any tasks in RCU read-side critical
+ * sections blocking the current preemptible-RCU expedited grace period.
+ * If there is no preemptible-RCU expedited grace period currently in
+ * progress, returns zero unconditionally.
+ */
+static int rcu_preempted_readers_exp(void)
+{
+	return rcu_preempt_ctrlblk.exp_tasks != NULL;
+}
+
+/*
+ * Report the exit from RCU read-side critical section for the last task
+ * that queued itself during or before the current expedited preemptible-RCU
+ * grace period.
+ */
+static void rcu_report_exp_done(void)
+{
+	wake_up(&sync_rcu_preempt_exp_wq);
+}
+
+/*
+ * Wait for an rcu-preempt grace period, but expedite it.  The basic idea
+ * is to rely in the fact that there is but one CPU, and that it is
+ * illegal for a task to invoke synchronize_rcu_expedited() while in a
+ * preemptible-RCU read-side critical section.  Therefore, any such
+ * critical sections must correspond to blocked tasks, which must therefore
+ * be on the ->blkd_tasks list.  So just record the current head of the
+ * list in the ->exp_tasks pointer, and wait for all tasks including and
+ * after the task pointed to by ->exp_tasks to drain.
+ */
+void synchronize_rcu_expedited(void)
+{
+	unsigned long flags;
+	struct rcu_preempt_ctrlblk *rpcp = &rcu_preempt_ctrlblk;
+	unsigned long snap;
+
+	barrier(); /* ensure prior action seen before grace period. */
+
+	WARN_ON_ONCE(rcu_preempt_running_reader());
+
+	/*
+	 * Acquire lock so that there is only one preemptible RCU grace
+	 * period in flight.  Of course, if someone does the expedited
+	 * grace period for us while we are acquiring the lock, just leave.
+	 */
+	snap = sync_rcu_preempt_exp_count + 1;
+	mutex_lock(&sync_rcu_preempt_exp_mutex);
+	if (ULONG_CMP_LT(snap, sync_rcu_preempt_exp_count))
+		goto unlock_mb_ret; /* Others did our work for us. */
+
+	local_irq_save(flags);
+
+	/*
+	 * All RCU readers have to already be on blkd_tasks because
+	 * we cannot legally be executing in an RCU read-side critical
+	 * section.
+	 */
+
+	/* Snapshot current head of ->blkd_tasks list. */
+	rpcp->exp_tasks = rpcp->blkd_tasks.next;
+	if (rpcp->exp_tasks == &rpcp->blkd_tasks)
+		rpcp->exp_tasks = NULL;
+	local_irq_restore(flags);
+
+	/* Wait for tail of ->blkd_tasks list to drain. */
+	if (rcu_preempted_readers_exp())
+		wait_event(sync_rcu_preempt_exp_wq,
+			   !rcu_preempted_readers_exp());
+
+	/* Clean up and exit. */
+	barrier(); /* ensure expedited GP seen before counter increment. */
+	sync_rcu_preempt_exp_count++;
+unlock_mb_ret:
+	mutex_unlock(&sync_rcu_preempt_exp_mutex);
+	barrier(); /* ensure subsequent action seen after grace period. */
+}
+EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
+
+/*
+ * Does preemptible RCU need the CPU to stay out of dynticks mode?
+ */
+int rcu_preempt_needs_cpu(void)
+{
+	if (!rcu_preempt_running_reader())
+		rcu_preempt_cpu_qs();
+	return rcu_preempt_ctrlblk.rcb.rcucblist != NULL;
+}
+
+/*
+ * Check for a task exiting while in a preemptible -RCU read-side
+ * critical section, clean up if so.  No need to issue warnings,
+ * as debug_check_no_locks_held() already does this if lockdep
+ * is enabled.
+ */
+void exit_rcu(void)
+{
+	struct task_struct *t = current;
+
+	if (t->rcu_read_lock_nesting == 0)
+		return;
+	t->rcu_read_lock_nesting = 1;
+	rcu_read_unlock();
+}
+
+#else /* #ifdef CONFIG_TINY_PREEMPT_RCU */
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to check.
+ */
+static void rcu_preempt_check_callbacks(void)
+{
+}
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to remove.
+ */
+static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp)
+{
+}
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to process.
+ */
+static void rcu_preempt_process_callbacks(void)
+{
+}
+
+#endif /* #else #ifdef CONFIG_TINY_PREEMPT_RCU */
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
 #include <linux/kernel_stat.h>
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index 2e2726d..9d8e8fb 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -120,7 +120,7 @@
 };
 
 static LIST_HEAD(rcu_torture_freelist);
-static struct rcu_torture *rcu_torture_current;
+static struct rcu_torture __rcu *rcu_torture_current;
 static long rcu_torture_current_version;
 static struct rcu_torture rcu_tortures[10 * RCU_TORTURE_PIPE_LEN];
 static DEFINE_SPINLOCK(rcu_torture_lock);
@@ -153,8 +153,10 @@
 #define FULLSTOP_SHUTDOWN 1	/* System shutdown with rcutorture running. */
 #define FULLSTOP_RMMOD    2	/* Normal rmmod of rcutorture. */
 static int fullstop = FULLSTOP_RMMOD;
-DEFINE_MUTEX(fullstop_mutex);	/* Protect fullstop transitions and spawning */
-				/*  of kthreads. */
+/*
+ * Protect fullstop transitions and spawning of kthreads.
+ */
+static DEFINE_MUTEX(fullstop_mutex);
 
 /*
  * Detect and respond to a system shutdown.
@@ -303,6 +305,10 @@
 		mdelay(longdelay_ms);
 	if (!(rcu_random(rrsp) % (nrealreaders * 2 * shortdelay_us)))
 		udelay(shortdelay_us);
+#ifdef CONFIG_PREEMPT
+	if (!preempt_count() && !(rcu_random(rrsp) % (nrealreaders * 20000)))
+		preempt_schedule();  /* No QS if preempt_disable() in effect */
+#endif
 }
 
 static void rcu_torture_read_unlock(int idx) __releases(RCU)
@@ -536,6 +542,8 @@
 	delay = rcu_random(rrsp) % (nrealreaders * 2 * longdelay * uspertick);
 	if (!delay)
 		schedule_timeout_interruptible(longdelay);
+	else
+		rcu_read_delay(rrsp);
 }
 
 static void srcu_torture_read_unlock(int idx) __releases(&srcu_ctl)
@@ -731,7 +739,8 @@
 			continue;
 		rp->rtort_pipe_count = 0;
 		udelay(rcu_random(&rand) & 0x3ff);
-		old_rp = rcu_torture_current;
+		old_rp = rcu_dereference_check(rcu_torture_current,
+					       current == writer_task);
 		rp->rtort_mbtest = 1;
 		rcu_assign_pointer(rcu_torture_current, rp);
 		smp_wmb(); /* Mods to old_rp must follow rcu_assign_pointer() */
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index d5bc439..ccdc04c 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -143,6 +143,11 @@
 module_param(qhimark, int, 0);
 module_param(qlowmark, int, 0);
 
+#ifdef CONFIG_RCU_CPU_STALL_DETECTOR
+int rcu_cpu_stall_suppress __read_mostly = RCU_CPU_STALL_SUPPRESS_INIT;
+module_param(rcu_cpu_stall_suppress, int, 0644);
+#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
+
 static void force_quiescent_state(struct rcu_state *rsp, int relaxed);
 static int rcu_pending(int cpu);
 
@@ -450,7 +455,7 @@
 
 #ifdef CONFIG_RCU_CPU_STALL_DETECTOR
 
-int rcu_cpu_stall_panicking __read_mostly;
+int rcu_cpu_stall_suppress __read_mostly;
 
 static void record_gp_stall_check_time(struct rcu_state *rsp)
 {
@@ -482,8 +487,11 @@
 	rcu_print_task_stall(rnp);
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
 
-	/* OK, time to rat on our buddy... */
-
+	/*
+	 * OK, time to rat on our buddy...
+	 * See Documentation/RCU/stallwarn.txt for info on how to debug
+	 * RCU CPU stall warnings.
+	 */
 	printk(KERN_ERR "INFO: %s detected stalls on CPUs/tasks: {",
 	       rsp->name);
 	rcu_for_each_leaf_node(rsp, rnp) {
@@ -512,6 +520,11 @@
 	unsigned long flags;
 	struct rcu_node *rnp = rcu_get_root(rsp);
 
+	/*
+	 * OK, time to rat on ourselves...
+	 * See Documentation/RCU/stallwarn.txt for info on how to debug
+	 * RCU CPU stall warnings.
+	 */
 	printk(KERN_ERR "INFO: %s detected stall on CPU %d (t=%lu jiffies)\n",
 	       rsp->name, smp_processor_id(), jiffies - rsp->gp_start);
 	trigger_all_cpu_backtrace();
@@ -530,11 +543,11 @@
 	long delta;
 	struct rcu_node *rnp;
 
-	if (rcu_cpu_stall_panicking)
+	if (rcu_cpu_stall_suppress)
 		return;
-	delta = jiffies - rsp->jiffies_stall;
+	delta = jiffies - ACCESS_ONCE(rsp->jiffies_stall);
 	rnp = rdp->mynode;
-	if ((rnp->qsmask & rdp->grpmask) && delta >= 0) {
+	if ((ACCESS_ONCE(rnp->qsmask) & rdp->grpmask) && delta >= 0) {
 
 		/* We haven't checked in, so go dump stack. */
 		print_cpu_stall(rsp);
@@ -548,10 +561,26 @@
 
 static int rcu_panic(struct notifier_block *this, unsigned long ev, void *ptr)
 {
-	rcu_cpu_stall_panicking = 1;
+	rcu_cpu_stall_suppress = 1;
 	return NOTIFY_DONE;
 }
 
+/**
+ * rcu_cpu_stall_reset - prevent further stall warnings in current grace period
+ *
+ * Set the stall-warning timeout way off into the future, thus preventing
+ * any RCU CPU stall-warning messages from appearing in the current set of
+ * RCU grace periods.
+ *
+ * The caller must disable hard irqs.
+ */
+void rcu_cpu_stall_reset(void)
+{
+	rcu_sched_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+	rcu_bh_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+	rcu_preempt_stall_reset();
+}
+
 static struct notifier_block rcu_panic_block = {
 	.notifier_call = rcu_panic,
 };
@@ -571,6 +600,10 @@
 {
 }
 
+void rcu_cpu_stall_reset(void)
+{
+}
+
 static void __init check_cpu_stall_init(void)
 {
 }
@@ -712,7 +745,7 @@
 rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
 	__releases(rcu_get_root(rsp)->lock)
 {
-	struct rcu_data *rdp = rsp->rda[smp_processor_id()];
+	struct rcu_data *rdp = this_cpu_ptr(rsp->rda);
 	struct rcu_node *rnp = rcu_get_root(rsp);
 
 	if (!cpu_needs_another_gp(rsp, rdp) || rsp->fqs_active) {
@@ -960,7 +993,7 @@
 static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp)
 {
 	int i;
-	struct rcu_data *rdp = rsp->rda[smp_processor_id()];
+	struct rcu_data *rdp = this_cpu_ptr(rsp->rda);
 
 	if (rdp->nxtlist == NULL)
 		return;  /* irqs disabled, so comparison is stable. */
@@ -971,6 +1004,7 @@
 	for (i = 0; i < RCU_NEXT_SIZE; i++)
 		rdp->nxttail[i] = &rdp->nxtlist;
 	rsp->orphan_qlen += rdp->qlen;
+	rdp->n_cbs_orphaned += rdp->qlen;
 	rdp->qlen = 0;
 	raw_spin_unlock(&rsp->onofflock);  /* irqs remain disabled. */
 }
@@ -984,7 +1018,7 @@
 	struct rcu_data *rdp;
 
 	raw_spin_lock_irqsave(&rsp->onofflock, flags);
-	rdp = rsp->rda[smp_processor_id()];
+	rdp = this_cpu_ptr(rsp->rda);
 	if (rsp->orphan_cbs_list == NULL) {
 		raw_spin_unlock_irqrestore(&rsp->onofflock, flags);
 		return;
@@ -992,6 +1026,7 @@
 	*rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_list;
 	rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_tail;
 	rdp->qlen += rsp->orphan_qlen;
+	rdp->n_cbs_adopted += rsp->orphan_qlen;
 	rsp->orphan_cbs_list = NULL;
 	rsp->orphan_cbs_tail = &rsp->orphan_cbs_list;
 	rsp->orphan_qlen = 0;
@@ -1007,7 +1042,7 @@
 	unsigned long flags;
 	unsigned long mask;
 	int need_report = 0;
-	struct rcu_data *rdp = rsp->rda[cpu];
+	struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
 	struct rcu_node *rnp;
 
 	/* Exclude any attempts to start a new grace period. */
@@ -1123,6 +1158,7 @@
 
 	/* Update count, and requeue any remaining callbacks. */
 	rdp->qlen -= count;
+	rdp->n_cbs_invoked += count;
 	if (list != NULL) {
 		*tail = rdp->nxtlist;
 		rdp->nxtlist = list;
@@ -1226,7 +1262,8 @@
 		cpu = rnp->grplo;
 		bit = 1;
 		for (; cpu <= rnp->grphi; cpu++, bit <<= 1) {
-			if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu]))
+			if ((rnp->qsmask & bit) != 0 &&
+			    f(per_cpu_ptr(rsp->rda, cpu)))
 				mask |= bit;
 		}
 		if (mask != 0) {
@@ -1402,7 +1439,7 @@
 	 * a quiescent state betweentimes.
 	 */
 	local_irq_save(flags);
-	rdp = rsp->rda[smp_processor_id()];
+	rdp = this_cpu_ptr(rsp->rda);
 	rcu_process_gp_end(rsp, rdp);
 	check_for_new_grace_period(rsp, rdp);
 
@@ -1701,7 +1738,7 @@
 {
 	unsigned long flags;
 	int i;
-	struct rcu_data *rdp = rsp->rda[cpu];
+	struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
 	struct rcu_node *rnp = rcu_get_root(rsp);
 
 	/* Set up local state, ensuring consistent view of global state. */
@@ -1729,7 +1766,7 @@
 {
 	unsigned long flags;
 	unsigned long mask;
-	struct rcu_data *rdp = rsp->rda[cpu];
+	struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
 	struct rcu_node *rnp = rcu_get_root(rsp);
 
 	/* Set up local state, ensuring consistent view of global state. */
@@ -1865,7 +1902,8 @@
 /*
  * Helper function for rcu_init() that initializes one rcu_state structure.
  */
-static void __init rcu_init_one(struct rcu_state *rsp)
+static void __init rcu_init_one(struct rcu_state *rsp,
+		struct rcu_data __percpu *rda)
 {
 	static char *buf[] = { "rcu_node_level_0",
 			       "rcu_node_level_1",
@@ -1918,37 +1956,23 @@
 		}
 	}
 
+	rsp->rda = rda;
 	rnp = rsp->level[NUM_RCU_LVLS - 1];
 	for_each_possible_cpu(i) {
 		while (i > rnp->grphi)
 			rnp++;
-		rsp->rda[i]->mynode = rnp;
+		per_cpu_ptr(rsp->rda, i)->mynode = rnp;
 		rcu_boot_init_percpu_data(i, rsp);
 	}
 }
 
-/*
- * Helper macro for __rcu_init() and __rcu_init_preempt().  To be used
- * nowhere else!  Assigns leaf node pointers into each CPU's rcu_data
- * structure.
- */
-#define RCU_INIT_FLAVOR(rsp, rcu_data) \
-do { \
-	int i; \
-	\
-	for_each_possible_cpu(i) { \
-		(rsp)->rda[i] = &per_cpu(rcu_data, i); \
-	} \
-	rcu_init_one(rsp); \
-} while (0)
-
 void __init rcu_init(void)
 {
 	int cpu;
 
 	rcu_bootup_announce();
-	RCU_INIT_FLAVOR(&rcu_sched_state, rcu_sched_data);
-	RCU_INIT_FLAVOR(&rcu_bh_state, rcu_bh_data);
+	rcu_init_one(&rcu_sched_state, &rcu_sched_data);
+	rcu_init_one(&rcu_bh_state, &rcu_bh_data);
 	__rcu_init_preempt();
 	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index 14c040b..91d4170 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -202,6 +202,9 @@
 	long		qlen;		/* # of queued callbacks */
 	long		qlen_last_fqs_check;
 					/* qlen at last check for QS forcing */
+	unsigned long	n_cbs_invoked;	/* count of RCU cbs invoked. */
+	unsigned long	n_cbs_orphaned;	/* RCU cbs sent to orphanage. */
+	unsigned long	n_cbs_adopted;	/* RCU cbs adopted from orphanage. */
 	unsigned long	n_force_qs_snap;
 					/* did other CPU force QS recently? */
 	long		blimit;		/* Upper limit on a processed batch */
@@ -254,19 +257,23 @@
 #define RCU_STALL_DELAY_DELTA	       0
 #endif
 
-#define RCU_SECONDS_TILL_STALL_CHECK   (10 * HZ + RCU_STALL_DELAY_DELTA)
+#define RCU_SECONDS_TILL_STALL_CHECK   (CONFIG_RCU_CPU_STALL_TIMEOUT * HZ + \
+					RCU_STALL_DELAY_DELTA)
 						/* for rsp->jiffies_stall */
-#define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ + RCU_STALL_DELAY_DELTA)
+#define RCU_SECONDS_TILL_STALL_RECHECK (3 * RCU_SECONDS_TILL_STALL_CHECK + 30)
 						/* for rsp->jiffies_stall */
 #define RCU_STALL_RAT_DELAY		2	/* Allow other CPUs time */
 						/*  to take at least one */
 						/*  scheduling clock irq */
 						/*  before ratting on them. */
 
-#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
+#ifdef CONFIG_RCU_CPU_STALL_DETECTOR_RUNNABLE
+#define RCU_CPU_STALL_SUPPRESS_INIT 0
+#else
+#define RCU_CPU_STALL_SUPPRESS_INIT 1
+#endif
 
-#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
-#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
+#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
 /*
  * RCU global state, including node hierarchy.  This hierarchy is
@@ -283,7 +290,7 @@
 	struct rcu_node *level[NUM_RCU_LVLS];	/* Hierarchy levels. */
 	u32 levelcnt[MAX_RCU_LVLS + 1];		/* # nodes in each level. */
 	u8 levelspread[NUM_RCU_LVLS];		/* kids/node in each level. */
-	struct rcu_data *rda[NR_CPUS];		/* array of rdp pointers. */
+	struct rcu_data __percpu *rda;		/* pointer of percu rcu_data. */
 
 	/* The following fields are guarded by the root rcu_node's lock. */
 
@@ -365,6 +372,7 @@
 #ifdef CONFIG_RCU_CPU_STALL_DETECTOR
 static void rcu_print_detail_task_stall(struct rcu_state *rsp);
 static void rcu_print_task_stall(struct rcu_node *rnp);
+static void rcu_preempt_stall_reset(void);
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp);
 #ifdef CONFIG_HOTPLUG_CPU
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 0e4f420..71a4147 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -57,7 +57,7 @@
 	printk(KERN_INFO
 	       "\tRCU-based detection of stalled CPUs is disabled.\n");
 #endif
-#ifndef CONFIG_RCU_CPU_STALL_VERBOSE
+#if defined(CONFIG_TREE_PREEMPT_RCU) && !defined(CONFIG_RCU_CPU_STALL_VERBOSE)
 	printk(KERN_INFO "\tVerbose stalled-CPUs detection is disabled.\n");
 #endif
 #if NUM_RCU_LVL_4 != 0
@@ -154,7 +154,7 @@
 	    (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) {
 
 		/* Possibly blocking in an RCU read-side critical section. */
-		rdp = rcu_preempt_state.rda[cpu];
+		rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu);
 		rnp = rdp->mynode;
 		raw_spin_lock_irqsave(&rnp->lock, flags);
 		t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
@@ -201,7 +201,7 @@
  */
 void __rcu_read_lock(void)
 {
-	ACCESS_ONCE(current->rcu_read_lock_nesting)++;
+	current->rcu_read_lock_nesting++;
 	barrier();  /* needed if we ever invoke rcu_read_lock in rcutree.c */
 }
 EXPORT_SYMBOL_GPL(__rcu_read_lock);
@@ -344,7 +344,9 @@
 	struct task_struct *t = current;
 
 	barrier();  /* needed if we ever invoke rcu_read_unlock in rcutree.c */
-	if (--ACCESS_ONCE(t->rcu_read_lock_nesting) == 0 &&
+	--t->rcu_read_lock_nesting;
+	barrier();  /* decrement before load of ->rcu_read_unlock_special */
+	if (t->rcu_read_lock_nesting == 0 &&
 	    unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
 		rcu_read_unlock_special(t);
 #ifdef CONFIG_PROVE_LOCKING
@@ -417,6 +419,16 @@
 	}
 }
 
+/*
+ * Suppress preemptible RCU's CPU stall warnings by pushing the
+ * time of the next stall-warning message comfortably far into the
+ * future.
+ */
+static void rcu_preempt_stall_reset(void)
+{
+	rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+}
+
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
 /*
@@ -546,9 +558,11 @@
  *
  * Control will return to the caller some time after a full grace
  * period has elapsed, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
- * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
- * and may be nested.
+ * read-side critical sections have completed.  Note, however, that
+ * upon return from synchronize_rcu(), the caller might well be executing
+ * concurrently with new RCU read-side critical sections that began while
+ * synchronize_rcu() was waiting.  RCU read-side critical sections are
+ * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested.
  */
 void synchronize_rcu(void)
 {
@@ -771,7 +785,7 @@
  */
 static void __init __rcu_init_preempt(void)
 {
-	RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_data);
+	rcu_init_one(&rcu_preempt_state, &rcu_preempt_data);
 }
 
 /*
@@ -865,6 +879,14 @@
 {
 }
 
+/*
+ * Because preemptible RCU does not exist, there is no need to suppress
+ * its CPU stall warnings.
+ */
+static void rcu_preempt_stall_reset(void)
+{
+}
+
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
 /*
@@ -919,15 +941,6 @@
 }
 
 /*
- * In classic RCU, call_rcu() is just call_rcu_sched().
- */
-void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
-{
-	call_rcu_sched(head, func);
-}
-EXPORT_SYMBOL_GPL(call_rcu);
-
-/*
  * Wait for an rcu-preempt grace period, but make it happen quickly.
  * But because preemptable RCU does not exist, map to rcu-sched.
  */
diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c
index 36c95b4..d15430b 100644
--- a/kernel/rcutree_trace.c
+++ b/kernel/rcutree_trace.c
@@ -64,7 +64,9 @@
 		   rdp->dynticks_fqs);
 #endif /* #ifdef CONFIG_NO_HZ */
 	seq_printf(m, " of=%lu ri=%lu", rdp->offline_fqs, rdp->resched_ipi);
-	seq_printf(m, " ql=%ld b=%ld\n", rdp->qlen, rdp->blimit);
+	seq_printf(m, " ql=%ld b=%ld", rdp->qlen, rdp->blimit);
+	seq_printf(m, " ci=%lu co=%lu ca=%lu\n",
+		   rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted);
 }
 
 #define PRINT_RCU_DATA(name, func, m) \
@@ -119,7 +121,9 @@
 		   rdp->dynticks_fqs);
 #endif /* #ifdef CONFIG_NO_HZ */
 	seq_printf(m, ",%lu,%lu", rdp->offline_fqs, rdp->resched_ipi);
-	seq_printf(m, ",%ld,%ld\n", rdp->qlen, rdp->blimit);
+	seq_printf(m, ",%ld,%ld", rdp->qlen, rdp->blimit);
+	seq_printf(m, ",%lu,%lu,%lu\n",
+		   rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted);
 }
 
 static int show_rcudata_csv(struct seq_file *m, void *unused)
@@ -128,7 +132,7 @@
 #ifdef CONFIG_NO_HZ
 	seq_puts(m, "\"dt\",\"dt nesting\",\"dn\",\"df\",");
 #endif /* #ifdef CONFIG_NO_HZ */
-	seq_puts(m, "\"of\",\"ri\",\"ql\",\"b\"\n");
+	seq_puts(m, "\"of\",\"ri\",\"ql\",\"b\",\"ci\",\"co\",\"ca\"\n");
 #ifdef CONFIG_TREE_PREEMPT_RCU
 	seq_puts(m, "\"rcu_preempt:\"\n");
 	PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data_csv, m);
@@ -262,7 +266,7 @@
 	struct rcu_data *rdp;
 
 	for_each_possible_cpu(cpu) {
-		rdp = rsp->rda[cpu];
+		rdp = per_cpu_ptr(rsp->rda, cpu);
 		if (rdp->beenonline)
 			print_one_rcu_pending(m, rdp);
 	}
diff --git a/kernel/sched.c b/kernel/sched.c
index dc85ceb..d42992b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -426,9 +426,7 @@
 	 */
 	cpumask_var_t rto_mask;
 	atomic_t rto_count;
-#ifdef CONFIG_SMP
 	struct cpupri cpupri;
-#endif
 };
 
 /*
@@ -437,7 +435,7 @@
  */
 static struct root_domain def_root_domain;
 
-#endif
+#endif /* CONFIG_SMP */
 
 /*
  * This is the main, per-CPU runqueue data structure.
@@ -488,11 +486,12 @@
 	 */
 	unsigned long nr_uninterruptible;
 
-	struct task_struct *curr, *idle;
+	struct task_struct *curr, *idle, *stop;
 	unsigned long next_balance;
 	struct mm_struct *prev_mm;
 
 	u64 clock;
+	u64 clock_task;
 
 	atomic_t nr_iowait;
 
@@ -520,6 +519,10 @@
 	u64 avg_idle;
 #endif
 
+#ifdef CONFIG_IRQ_TIME_ACCOUNTING
+	u64 prev_irq_time;
+#endif
+
 	/* calc_load related fields */
 	unsigned long calc_load_update;
 	long calc_load_active;
@@ -643,10 +646,22 @@
 
 #endif /* CONFIG_CGROUP_SCHED */
 
+static u64 irq_time_cpu(int cpu);
+static void sched_irq_time_avg_update(struct rq *rq, u64 irq_time);
+
 inline void update_rq_clock(struct rq *rq)
 {
-	if (!rq->skip_clock_update)
-		rq->clock = sched_clock_cpu(cpu_of(rq));
+	if (!rq->skip_clock_update) {
+		int cpu = cpu_of(rq);
+		u64 irq_time;
+
+		rq->clock = sched_clock_cpu(cpu);
+		irq_time = irq_time_cpu(cpu);
+		if (rq->clock - irq_time > rq->clock_task)
+			rq->clock_task = rq->clock - irq_time;
+
+		sched_irq_time_avg_update(rq, irq_time);
+	}
 }
 
 /*
@@ -723,7 +738,7 @@
 		size_t cnt, loff_t *ppos)
 {
 	char buf[64];
-	char *cmp = buf;
+	char *cmp;
 	int neg = 0;
 	int i;
 
@@ -734,6 +749,7 @@
 		return -EFAULT;
 
 	buf[cnt] = 0;
+	cmp = strstrip(buf);
 
 	if (strncmp(buf, "NO_", 3) == 0) {
 		neg = 1;
@@ -741,9 +757,7 @@
 	}
 
 	for (i = 0; sched_feat_names[i]; i++) {
-		int len = strlen(sched_feat_names[i]);
-
-		if (strncmp(cmp, sched_feat_names[i], len) == 0) {
+		if (strcmp(cmp, sched_feat_names[i]) == 0) {
 			if (neg)
 				sysctl_sched_features &= ~(1UL << i);
 			else
@@ -1840,7 +1854,7 @@
 
 static const struct sched_class rt_sched_class;
 
-#define sched_class_highest (&rt_sched_class)
+#define sched_class_highest (&stop_sched_class)
 #define for_each_class(class) \
    for (class = sched_class_highest; class; class = class->next)
 
@@ -1858,12 +1872,6 @@
 
 static void set_load_weight(struct task_struct *p)
 {
-	if (task_has_rt_policy(p)) {
-		p->se.load.weight = 0;
-		p->se.load.inv_weight = WMULT_CONST;
-		return;
-	}
-
 	/*
 	 * SCHED_IDLE tasks get minimal weight:
 	 */
@@ -1917,13 +1925,132 @@
 	dec_nr_running(rq);
 }
 
+#ifdef CONFIG_IRQ_TIME_ACCOUNTING
+
+/*
+ * There are no locks covering percpu hardirq/softirq time.
+ * They are only modified in account_system_vtime, on corresponding CPU
+ * with interrupts disabled. So, writes are safe.
+ * They are read and saved off onto struct rq in update_rq_clock().
+ * This may result in other CPU reading this CPU's irq time and can
+ * race with irq/account_system_vtime on this CPU. We would either get old
+ * or new value (or semi updated value on 32 bit) with a side effect of
+ * accounting a slice of irq time to wrong task when irq is in progress
+ * while we read rq->clock. That is a worthy compromise in place of having
+ * locks on each irq in account_system_time.
+ */
+static DEFINE_PER_CPU(u64, cpu_hardirq_time);
+static DEFINE_PER_CPU(u64, cpu_softirq_time);
+
+static DEFINE_PER_CPU(u64, irq_start_time);
+static int sched_clock_irqtime;
+
+void enable_sched_clock_irqtime(void)
+{
+	sched_clock_irqtime = 1;
+}
+
+void disable_sched_clock_irqtime(void)
+{
+	sched_clock_irqtime = 0;
+}
+
+static u64 irq_time_cpu(int cpu)
+{
+	if (!sched_clock_irqtime)
+		return 0;
+
+	return per_cpu(cpu_softirq_time, cpu) + per_cpu(cpu_hardirq_time, cpu);
+}
+
+void account_system_vtime(struct task_struct *curr)
+{
+	unsigned long flags;
+	int cpu;
+	u64 now, delta;
+
+	if (!sched_clock_irqtime)
+		return;
+
+	local_irq_save(flags);
+
+	cpu = smp_processor_id();
+	now = sched_clock_cpu(cpu);
+	delta = now - per_cpu(irq_start_time, cpu);
+	per_cpu(irq_start_time, cpu) = now;
+	/*
+	 * We do not account for softirq time from ksoftirqd here.
+	 * We want to continue accounting softirq time to ksoftirqd thread
+	 * in that case, so as not to confuse scheduler with a special task
+	 * that do not consume any time, but still wants to run.
+	 */
+	if (hardirq_count())
+		per_cpu(cpu_hardirq_time, cpu) += delta;
+	else if (in_serving_softirq() && !(curr->flags & PF_KSOFTIRQD))
+		per_cpu(cpu_softirq_time, cpu) += delta;
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(account_system_vtime);
+
+static void sched_irq_time_avg_update(struct rq *rq, u64 curr_irq_time)
+{
+	if (sched_clock_irqtime && sched_feat(NONIRQ_POWER)) {
+		u64 delta_irq = curr_irq_time - rq->prev_irq_time;
+		rq->prev_irq_time = curr_irq_time;
+		sched_rt_avg_update(rq, delta_irq);
+	}
+}
+
+#else
+
+static u64 irq_time_cpu(int cpu)
+{
+	return 0;
+}
+
+static void sched_irq_time_avg_update(struct rq *rq, u64 curr_irq_time) { }
+
+#endif
+
 #include "sched_idletask.c"
 #include "sched_fair.c"
 #include "sched_rt.c"
+#include "sched_stoptask.c"
 #ifdef CONFIG_SCHED_DEBUG
 # include "sched_debug.c"
 #endif
 
+void sched_set_stop_task(int cpu, struct task_struct *stop)
+{
+	struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 };
+	struct task_struct *old_stop = cpu_rq(cpu)->stop;
+
+	if (stop) {
+		/*
+		 * Make it appear like a SCHED_FIFO task, its something
+		 * userspace knows about and won't get confused about.
+		 *
+		 * Also, it will make PI more or less work without too
+		 * much confusion -- but then, stop work should not
+		 * rely on PI working anyway.
+		 */
+		sched_setscheduler_nocheck(stop, SCHED_FIFO, &param);
+
+		stop->sched_class = &stop_sched_class;
+	}
+
+	cpu_rq(cpu)->stop = stop;
+
+	if (old_stop) {
+		/*
+		 * Reset it back to a normal scheduling class so that
+		 * it can die in pieces.
+		 */
+		old_stop->sched_class = &rt_sched_class;
+	}
+}
+
 /*
  * __normal_prio - return the priority that is based on the static prio
  */
@@ -2003,6 +2130,9 @@
 	if (p->sched_class != &fair_sched_class)
 		return 0;
 
+	if (unlikely(p->policy == SCHED_IDLE))
+		return 0;
+
 	/*
 	 * Buddy candidates are cache hot:
 	 */
@@ -2852,14 +2982,14 @@
 	 */
 	arch_start_context_switch(prev);
 
-	if (likely(!mm)) {
+	if (!mm) {
 		next->active_mm = oldmm;
 		atomic_inc(&oldmm->mm_count);
 		enter_lazy_tlb(oldmm, next);
 	} else
 		switch_mm(oldmm, mm, next);
 
-	if (likely(!prev->mm)) {
+	if (!prev->mm) {
 		prev->active_mm = NULL;
 		rq->prev_mm = oldmm;
 	}
@@ -3248,7 +3378,7 @@
 
 	if (task_current(rq, p)) {
 		update_rq_clock(rq);
-		ns = rq->clock - p->se.exec_start;
+		ns = rq->clock_task - p->se.exec_start;
 		if ((s64)ns < 0)
 			ns = 0;
 	}
@@ -3397,7 +3527,7 @@
 	tmp = cputime_to_cputime64(cputime);
 	if (hardirq_count() - hardirq_offset)
 		cpustat->irq = cputime64_add(cpustat->irq, tmp);
-	else if (softirq_count())
+	else if (in_serving_softirq())
 		cpustat->softirq = cputime64_add(cpustat->softirq, tmp);
 	else
 		cpustat->system = cputime64_add(cpustat->system, tmp);
@@ -3584,7 +3714,7 @@
 	curr->sched_class->task_tick(rq, curr, 0);
 	raw_spin_unlock(&rq->lock);
 
-	perf_event_task_tick(curr);
+	perf_event_task_tick();
 
 #ifdef CONFIG_SMP
 	rq->idle_at_tick = idle_cpu(cpu);
@@ -3723,17 +3853,13 @@
 			return p;
 	}
 
-	class = sched_class_highest;
-	for ( ; ; ) {
+	for_each_class(class) {
 		p = class->pick_next_task(rq);
 		if (p)
 			return p;
-		/*
-		 * Will never be NULL as the idle class always
-		 * returns a non-NULL p:
-		 */
-		class = class->next;
 	}
+
+	BUG(); /* the idle class will always have a runnable task */
 }
 
 /*
@@ -4358,6 +4484,7 @@
 
 	rq = task_rq_lock(p, &flags);
 
+	trace_sched_pi_setprio(p, prio);
 	oldprio = p->prio;
 	prev_class = p->sched_class;
 	on_rq = p->se.on_rq;
@@ -4645,7 +4772,7 @@
 	}
 
 	if (user) {
-		retval = security_task_setscheduler(p, policy, param);
+		retval = security_task_setscheduler(p);
 		if (retval)
 			return retval;
 	}
@@ -4661,6 +4788,15 @@
 	 */
 	rq = __task_rq_lock(p);
 
+	/*
+	 * Changing the policy of the stop threads its a very bad idea
+	 */
+	if (p == rq->stop) {
+		__task_rq_unlock(rq);
+		raw_spin_unlock_irqrestore(&p->pi_lock, flags);
+		return -EINVAL;
+	}
+
 #ifdef CONFIG_RT_GROUP_SCHED
 	if (user) {
 		/*
@@ -4887,13 +5023,13 @@
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
 
-	retval = security_task_setscheduler(p, 0, NULL);
+	retval = security_task_setscheduler(p);
 	if (retval)
 		goto out_unlock;
 
 	cpuset_cpus_allowed(p, cpus_allowed);
 	cpumask_and(new_mask, in_mask, cpus_allowed);
- again:
+again:
 	retval = set_cpus_allowed_ptr(p, new_mask);
 
 	if (!retval) {
@@ -5337,7 +5473,19 @@
 	idle->se.exec_start = sched_clock();
 
 	cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu));
+	/*
+	 * We're having a chicken and egg problem, even though we are
+	 * holding rq->lock, the cpu isn't yet set to this cpu so the
+	 * lockdep check in task_group() will fail.
+	 *
+	 * Similar case to sched_fork(). / Alternatively we could
+	 * use task_rq_lock() here and obtain the other rq->lock.
+	 *
+	 * Silence PROVE_RCU
+	 */
+	rcu_read_lock();
 	__set_task_cpu(idle, cpu);
+	rcu_read_unlock();
 
 	rq->curr = rq->idle = idle;
 #if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
@@ -6514,6 +6662,7 @@
 	cpumask_var_t		nodemask;
 	cpumask_var_t		this_sibling_map;
 	cpumask_var_t		this_core_map;
+	cpumask_var_t		this_book_map;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
 	struct sched_group	**sched_group_nodes;
@@ -6525,6 +6674,7 @@
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
+	sa_this_book_map,
 	sa_this_core_map,
 	sa_this_sibling_map,
 	sa_nodemask,
@@ -6560,31 +6710,48 @@
 #ifdef CONFIG_SCHED_MC
 static DEFINE_PER_CPU(struct static_sched_domain, core_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
-#endif /* CONFIG_SCHED_MC */
 
-#if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT)
 static int
 cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
 		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
-
+#ifdef CONFIG_SCHED_SMT
 	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
 	group = cpumask_first(mask);
+#else
+	group = cpu;
+#endif
 	if (sg)
 		*sg = &per_cpu(sched_group_core, group).sg;
 	return group;
 }
-#elif defined(CONFIG_SCHED_MC)
+#endif /* CONFIG_SCHED_MC */
+
+/*
+ * book sched-domains:
+ */
+#ifdef CONFIG_SCHED_BOOK
+static DEFINE_PER_CPU(struct static_sched_domain, book_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_book);
+
 static int
-cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *unused)
+cpu_to_book_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
 {
-	if (sg)
-		*sg = &per_cpu(sched_group_core, cpu).sg;
-	return cpu;
-}
+	int group = cpu;
+#ifdef CONFIG_SCHED_MC
+	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
+	group = cpumask_first(mask);
+#elif defined(CONFIG_SCHED_SMT)
+	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
+	group = cpumask_first(mask);
 #endif
+	if (sg)
+		*sg = &per_cpu(sched_group_book, group).sg;
+	return group;
+}
+#endif /* CONFIG_SCHED_BOOK */
 
 static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
@@ -6594,7 +6761,10 @@
 		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
-#ifdef CONFIG_SCHED_MC
+#ifdef CONFIG_SCHED_BOOK
+	cpumask_and(mask, cpu_book_mask(cpu), cpu_map);
+	group = cpumask_first(mask);
+#elif defined(CONFIG_SCHED_MC)
 	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
 	group = cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
@@ -6855,6 +7025,9 @@
 #ifdef CONFIG_SCHED_MC
  SD_INIT_FUNC(MC)
 #endif
+#ifdef CONFIG_SCHED_BOOK
+ SD_INIT_FUNC(BOOK)
+#endif
 
 static int default_relax_domain_level = -1;
 
@@ -6904,6 +7077,8 @@
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
+	case sa_this_book_map:
+		free_cpumask_var(d->this_book_map); /* fall through */
 	case sa_this_core_map:
 		free_cpumask_var(d->this_core_map); /* fall through */
 	case sa_this_sibling_map:
@@ -6950,8 +7125,10 @@
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->this_core_map, GFP_KERNEL))
 		return sa_this_sibling_map;
-	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
+	if (!alloc_cpumask_var(&d->this_book_map, GFP_KERNEL))
 		return sa_this_core_map;
+	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
+		return sa_this_book_map;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
 	d->rd = alloc_rootdomain();
@@ -7009,6 +7186,23 @@
 	return sd;
 }
 
+static struct sched_domain *__build_book_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
+{
+	struct sched_domain *sd = parent;
+#ifdef CONFIG_SCHED_BOOK
+	sd = &per_cpu(book_domains, i).sd;
+	SD_INIT(sd, BOOK);
+	set_domain_attribute(sd, attr);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
+	sd->parent = parent;
+	parent->child = sd;
+	cpu_to_book_group(i, cpu_map, &sd->groups, d->tmpmask);
+#endif
+	return sd;
+}
+
 static struct sched_domain *__build_mc_sched_domain(struct s_data *d,
 	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
 	struct sched_domain *parent, int i)
@@ -7066,6 +7260,15 @@
 						d->send_covered, d->tmpmask);
 		break;
 #endif
+#ifdef CONFIG_SCHED_BOOK
+	case SD_LV_BOOK: /* set up book groups */
+		cpumask_and(d->this_book_map, cpu_map, cpu_book_mask(cpu));
+		if (cpu == cpumask_first(d->this_book_map))
+			init_sched_build_groups(d->this_book_map, cpu_map,
+						&cpu_to_book_group,
+						d->send_covered, d->tmpmask);
+		break;
+#endif
 	case SD_LV_CPU: /* set up physical groups */
 		cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map);
 		if (!cpumask_empty(d->nodemask))
@@ -7113,12 +7316,14 @@
 
 		sd = __build_numa_sched_domains(&d, cpu_map, attr, i);
 		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
+		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 	}
 
 	for_each_cpu(i, cpu_map) {
 		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
+		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 	}
 
@@ -7149,6 +7354,12 @@
 		init_sched_groups_power(i, sd);
 	}
 #endif
+#ifdef CONFIG_SCHED_BOOK
+	for_each_cpu(i, cpu_map) {
+		sd = &per_cpu(book_domains, i).sd;
+		init_sched_groups_power(i, sd);
+	}
+#endif
 
 	for_each_cpu(i, cpu_map) {
 		sd = &per_cpu(phys_domains, i).sd;
@@ -7174,6 +7385,8 @@
 		sd = &per_cpu(cpu_domains, i).sd;
 #elif defined(CONFIG_SCHED_MC)
 		sd = &per_cpu(core_domains, i).sd;
+#elif defined(CONFIG_SCHED_BOOK)
+		sd = &per_cpu(book_domains, i).sd;
 #else
 		sd = &per_cpu(phys_domains, i).sd;
 #endif
@@ -8078,9 +8291,9 @@
 
 	return 1;
 
- err_free_rq:
+err_free_rq:
 	kfree(cfs_rq);
- err:
+err:
 	return 0;
 }
 
@@ -8168,9 +8381,9 @@
 
 	return 1;
 
- err_free_rq:
+err_free_rq:
 	kfree(rt_rq);
- err:
+err:
 	return 0;
 }
 
@@ -8528,7 +8741,7 @@
 		raw_spin_unlock(&rt_rq->rt_runtime_lock);
 	}
 	raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock);
- unlock:
+unlock:
 	read_unlock(&tasklist_lock);
 	mutex_unlock(&rt_constraints_mutex);
 
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index db3f674..933f3d1 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -25,7 +25,7 @@
 
 /*
  * Targeted preemption latency for CPU-bound tasks:
- * (default: 5ms * (1 + ilog(ncpus)), units: nanoseconds)
+ * (default: 6ms * (1 + ilog(ncpus)), units: nanoseconds)
  *
  * NOTE: this latency value is not the same as the concept of
  * 'timeslice length' - timeslices in CFS are of variable length
@@ -52,7 +52,7 @@
 
 /*
  * Minimal preemption granularity for CPU-bound tasks:
- * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds)
+ * (default: 0.75 msec * (1 + ilog(ncpus)), units: nanoseconds)
  */
 unsigned int sysctl_sched_min_granularity = 750000ULL;
 unsigned int normalized_sysctl_sched_min_granularity = 750000ULL;
@@ -519,7 +519,7 @@
 static void update_curr(struct cfs_rq *cfs_rq)
 {
 	struct sched_entity *curr = cfs_rq->curr;
-	u64 now = rq_of(cfs_rq)->clock;
+	u64 now = rq_of(cfs_rq)->clock_task;
 	unsigned long delta_exec;
 
 	if (unlikely(!curr))
@@ -602,7 +602,7 @@
 	/*
 	 * We are starting a new run period:
 	 */
-	se->exec_start = rq_of(cfs_rq)->clock;
+	se->exec_start = rq_of(cfs_rq)->clock_task;
 }
 
 /**************************************************
@@ -1764,6 +1764,10 @@
 	set_task_cpu(p, this_cpu);
 	activate_task(this_rq, p, 0);
 	check_preempt_curr(this_rq, p, 0);
+
+	/* re-arm NEWIDLE balancing when moving tasks */
+	src_rq->avg_idle = this_rq->avg_idle = 2*sysctl_sched_migration_cost;
+	this_rq->idle_stamp = 0;
 }
 
 /*
@@ -1798,7 +1802,7 @@
 	 * 2) too many balance attempts have failed.
 	 */
 
-	tsk_cache_hot = task_hot(p, rq->clock, sd);
+	tsk_cache_hot = task_hot(p, rq->clock_task, sd);
 	if (!tsk_cache_hot ||
 		sd->nr_balance_failed > sd->cache_nice_tries) {
 #ifdef CONFIG_SCHEDSTATS
@@ -2030,12 +2034,14 @@
 	unsigned long this_load;
 	unsigned long this_load_per_task;
 	unsigned long this_nr_running;
+	unsigned long this_has_capacity;
 
 	/* Statistics of the busiest group */
 	unsigned long max_load;
 	unsigned long busiest_load_per_task;
 	unsigned long busiest_nr_running;
 	unsigned long busiest_group_capacity;
+	unsigned long busiest_has_capacity;
 
 	int group_imb; /* Is there imbalance in this sd */
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
@@ -2058,6 +2064,7 @@
 	unsigned long sum_weighted_load; /* Weighted load of group's tasks */
 	unsigned long group_capacity;
 	int group_imb; /* Is there an imbalance in the group ? */
+	int group_has_capacity; /* Is there extra capacity in the group? */
 };
 
 /**
@@ -2268,7 +2275,13 @@
 	u64 total, available;
 
 	total = sched_avg_period() + (rq->clock - rq->age_stamp);
-	available = total - rq->rt_avg;
+
+	if (unlikely(total < rq->rt_avg)) {
+		/* Ensures that power won't end up being negative */
+		available = 0;
+	} else {
+		available = total - rq->rt_avg;
+	}
 
 	if (unlikely((s64)total < SCHED_LOAD_SCALE))
 		total = SCHED_LOAD_SCALE;
@@ -2378,7 +2391,7 @@
 			int local_group, const struct cpumask *cpus,
 			int *balance, struct sg_lb_stats *sgs)
 {
-	unsigned long load, max_cpu_load, min_cpu_load;
+	unsigned long load, max_cpu_load, min_cpu_load, max_nr_running;
 	int i;
 	unsigned int balance_cpu = -1, first_idle_cpu = 0;
 	unsigned long avg_load_per_task = 0;
@@ -2389,6 +2402,7 @@
 	/* Tally up the load of all CPUs in the group */
 	max_cpu_load = 0;
 	min_cpu_load = ~0UL;
+	max_nr_running = 0;
 
 	for_each_cpu_and(i, sched_group_cpus(group), cpus) {
 		struct rq *rq = cpu_rq(i);
@@ -2406,8 +2420,10 @@
 			load = target_load(i, load_idx);
 		} else {
 			load = source_load(i, load_idx);
-			if (load > max_cpu_load)
+			if (load > max_cpu_load) {
 				max_cpu_load = load;
+				max_nr_running = rq->nr_running;
+			}
 			if (min_cpu_load > load)
 				min_cpu_load = load;
 		}
@@ -2447,13 +2463,15 @@
 	if (sgs->sum_nr_running)
 		avg_load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running;
 
-	if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task)
+	if ((max_cpu_load - min_cpu_load) > 2*avg_load_per_task && max_nr_running > 1)
 		sgs->group_imb = 1;
 
-	sgs->group_capacity =
-		DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE);
+	sgs->group_capacity = DIV_ROUND_CLOSEST(group->cpu_power, SCHED_LOAD_SCALE);
 	if (!sgs->group_capacity)
 		sgs->group_capacity = fix_small_capacity(sd, group);
+
+	if (sgs->group_capacity > sgs->sum_nr_running)
+		sgs->group_has_capacity = 1;
 }
 
 /**
@@ -2542,9 +2560,14 @@
 		/*
 		 * In case the child domain prefers tasks go to siblings
 		 * first, lower the sg capacity to one so that we'll try
-		 * and move all the excess tasks away.
+		 * and move all the excess tasks away. We lower the capacity
+		 * of a group only if the local group has the capacity to fit
+		 * these excess tasks, i.e. nr_running < group_capacity. The
+		 * extra check prevents the case where you always pull from the
+		 * heaviest group when it is already under-utilized (possible
+		 * with a large weight task outweighs the tasks on the system).
 		 */
-		if (prefer_sibling)
+		if (prefer_sibling && !local_group && sds->this_has_capacity)
 			sgs.group_capacity = min(sgs.group_capacity, 1UL);
 
 		if (local_group) {
@@ -2552,12 +2575,14 @@
 			sds->this = sg;
 			sds->this_nr_running = sgs.sum_nr_running;
 			sds->this_load_per_task = sgs.sum_weighted_load;
+			sds->this_has_capacity = sgs.group_has_capacity;
 		} else if (update_sd_pick_busiest(sd, sds, sg, &sgs, this_cpu)) {
 			sds->max_load = sgs.avg_load;
 			sds->busiest = sg;
 			sds->busiest_nr_running = sgs.sum_nr_running;
 			sds->busiest_group_capacity = sgs.group_capacity;
 			sds->busiest_load_per_task = sgs.sum_weighted_load;
+			sds->busiest_has_capacity = sgs.group_has_capacity;
 			sds->group_imb = sgs.group_imb;
 		}
 
@@ -2754,6 +2779,7 @@
 		return fix_small_imbalance(sds, this_cpu, imbalance);
 
 }
+
 /******* find_busiest_group() helpers end here *********************/
 
 /**
@@ -2805,6 +2831,11 @@
 	 * 4) This group is more busy than the avg busieness at this
 	 *    sched_domain.
 	 * 5) The imbalance is within the specified limit.
+	 *
+	 * Note: when doing newidle balance, if the local group has excess
+	 * capacity (i.e. nr_running < group_capacity) and the busiest group
+	 * does not have any capacity, we force a load balance to pull tasks
+	 * to the local group. In this case, we skip past checks 3, 4 and 5.
 	 */
 	if (!(*balance))
 		goto ret;
@@ -2816,6 +2847,11 @@
 	if (!sds.busiest || sds.busiest_nr_running == 0)
 		goto out_balanced;
 
+	/*  SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */
+	if (idle == CPU_NEWLY_IDLE && sds.this_has_capacity &&
+			!sds.busiest_has_capacity)
+		goto force_balance;
+
 	if (sds.this_load >= sds.max_load)
 		goto out_balanced;
 
@@ -2827,6 +2863,7 @@
 	if (100 * sds.max_load <= sd->imbalance_pct * sds.this_load)
 		goto out_balanced;
 
+force_balance:
 	/* Looks like there is an imbalance. Compute it */
 	calculate_imbalance(&sds, this_cpu, imbalance);
 	return sds.busiest;
@@ -3031,7 +3068,14 @@
 
 	if (!ld_moved) {
 		schedstat_inc(sd, lb_failed[idle]);
-		sd->nr_balance_failed++;
+		/*
+		 * Increment the failure counter only on periodic balance.
+		 * We do not want newidle balance, which can be very
+		 * frequent, pollute the failure counter causing
+		 * excessive cache_hot migrations and active balances.
+		 */
+		if (idle != CPU_NEWLY_IDLE)
+			sd->nr_balance_failed++;
 
 		if (need_active_balance(sd, sd_idle, idle, cpu_of(busiest),
 					this_cpu)) {
@@ -3153,10 +3197,8 @@
 		interval = msecs_to_jiffies(sd->balance_interval);
 		if (time_after(next_balance, sd->last_balance + interval))
 			next_balance = sd->last_balance + interval;
-		if (pulled_task) {
-			this_rq->idle_stamp = 0;
+		if (pulled_task)
 			break;
-		}
 	}
 
 	raw_spin_lock(&this_rq->lock);
@@ -3751,8 +3793,11 @@
 
 	update_rq_clock(rq);
 
-	if (unlikely(task_cpu(p) != this_cpu))
+	if (unlikely(task_cpu(p) != this_cpu)) {
+		rcu_read_lock();
 		__set_task_cpu(p, this_cpu);
+		rcu_read_unlock();
+	}
 
 	update_curr(cfs_rq);
 
diff --git a/kernel/sched_features.h b/kernel/sched_features.h
index 83c66e8..185f920 100644
--- a/kernel/sched_features.h
+++ b/kernel/sched_features.h
@@ -61,3 +61,8 @@
  * release the lock. Decreases scheduling overhead.
  */
 SCHED_FEAT(OWNER_SPIN, 1)
+
+/*
+ * Decrement CPU power based on irq activity
+ */
+SCHED_FEAT(NONIRQ_POWER, 1)
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index d10c80e..bea7d79 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -609,7 +609,7 @@
 	if (!task_has_rt_policy(curr))
 		return;
 
-	delta_exec = rq->clock - curr->se.exec_start;
+	delta_exec = rq->clock_task - curr->se.exec_start;
 	if (unlikely((s64)delta_exec < 0))
 		delta_exec = 0;
 
@@ -618,7 +618,7 @@
 	curr->se.sum_exec_runtime += delta_exec;
 	account_group_exec_runtime(curr, delta_exec);
 
-	curr->se.exec_start = rq->clock;
+	curr->se.exec_start = rq->clock_task;
 	cpuacct_charge(curr, delta_exec);
 
 	sched_rt_avg_update(rq, delta_exec);
@@ -960,18 +960,19 @@
 	 * runqueue. Otherwise simply start this RT task
 	 * on its current runqueue.
 	 *
-	 * We want to avoid overloading runqueues. Even if
-	 * the RT task is of higher priority than the current RT task.
-	 * RT tasks behave differently than other tasks. If
-	 * one gets preempted, we try to push it off to another queue.
-	 * So trying to keep a preempting RT task on the same
-	 * cache hot CPU will force the running RT task to
-	 * a cold CPU. So we waste all the cache for the lower
-	 * RT task in hopes of saving some of a RT task
-	 * that is just being woken and probably will have
-	 * cold cache anyway.
+	 * We want to avoid overloading runqueues. If the woken
+	 * task is a higher priority, then it will stay on this CPU
+	 * and the lower prio task should be moved to another CPU.
+	 * Even though this will probably make the lower prio task
+	 * lose its cache, we do not want to bounce a higher task
+	 * around just because it gave up its CPU, perhaps for a
+	 * lock?
+	 *
+	 * For equal prio tasks, we just let the scheduler sort it out.
 	 */
 	if (unlikely(rt_task(rq->curr)) &&
+	    (rq->curr->rt.nr_cpus_allowed < 2 ||
+	     rq->curr->prio < p->prio) &&
 	    (p->rt.nr_cpus_allowed > 1)) {
 		int cpu = find_lowest_rq(p);
 
@@ -1074,7 +1075,7 @@
 	} while (rt_rq);
 
 	p = rt_task_of(rt_se);
-	p->se.exec_start = rq->clock;
+	p->se.exec_start = rq->clock_task;
 
 	return p;
 }
@@ -1139,7 +1140,7 @@
 	for_each_leaf_rt_rq(rt_rq, rq) {
 		array = &rt_rq->active;
 		idx = sched_find_first_bit(array->bitmap);
- next_idx:
+next_idx:
 		if (idx >= MAX_RT_PRIO)
 			continue;
 		if (next && next->prio < idx)
@@ -1315,7 +1316,7 @@
 	if (!next_task)
 		return 0;
 
- retry:
+retry:
 	if (unlikely(next_task == rq->curr)) {
 		WARN_ON(1);
 		return 0;
@@ -1463,7 +1464,7 @@
 			 * but possible)
 			 */
 		}
- skip:
+skip:
 		double_unlock_balance(this_rq, src_rq);
 	}
 
@@ -1491,7 +1492,10 @@
 	if (!task_running(rq, p) &&
 	    !test_tsk_need_resched(rq->curr) &&
 	    has_pushable_tasks(rq) &&
-	    p->rt.nr_cpus_allowed > 1)
+	    p->rt.nr_cpus_allowed > 1 &&
+	    rt_task(rq->curr) &&
+	    (rq->curr->rt.nr_cpus_allowed < 2 ||
+	     rq->curr->prio < p->prio))
 		push_rt_tasks(rq);
 }
 
@@ -1709,7 +1713,7 @@
 {
 	struct task_struct *p = rq->curr;
 
-	p->se.exec_start = rq->clock;
+	p->se.exec_start = rq->clock_task;
 
 	/* The running task is never eligible for pushing */
 	dequeue_pushable_task(rq, p);
diff --git a/kernel/sched_stoptask.c b/kernel/sched_stoptask.c
new file mode 100644
index 0000000..45bddc0
--- /dev/null
+++ b/kernel/sched_stoptask.c
@@ -0,0 +1,108 @@
+/*
+ * stop-task scheduling class.
+ *
+ * The stop task is the highest priority task in the system, it preempts
+ * everything and will be preempted by nothing.
+ *
+ * See kernel/stop_machine.c
+ */
+
+#ifdef CONFIG_SMP
+static int
+select_task_rq_stop(struct rq *rq, struct task_struct *p,
+		    int sd_flag, int flags)
+{
+	return task_cpu(p); /* stop tasks as never migrate */
+}
+#endif /* CONFIG_SMP */
+
+static void
+check_preempt_curr_stop(struct rq *rq, struct task_struct *p, int flags)
+{
+	resched_task(rq->curr); /* we preempt everything */
+}
+
+static struct task_struct *pick_next_task_stop(struct rq *rq)
+{
+	struct task_struct *stop = rq->stop;
+
+	if (stop && stop->state == TASK_RUNNING)
+		return stop;
+
+	return NULL;
+}
+
+static void
+enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags)
+{
+}
+
+static void
+dequeue_task_stop(struct rq *rq, struct task_struct *p, int flags)
+{
+}
+
+static void yield_task_stop(struct rq *rq)
+{
+	BUG(); /* the stop task should never yield, its pointless. */
+}
+
+static void put_prev_task_stop(struct rq *rq, struct task_struct *prev)
+{
+}
+
+static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued)
+{
+}
+
+static void set_curr_task_stop(struct rq *rq)
+{
+}
+
+static void switched_to_stop(struct rq *rq, struct task_struct *p,
+			     int running)
+{
+	BUG(); /* its impossible to change to this class */
+}
+
+static void prio_changed_stop(struct rq *rq, struct task_struct *p,
+			      int oldprio, int running)
+{
+	BUG(); /* how!?, what priority? */
+}
+
+static unsigned int
+get_rr_interval_stop(struct rq *rq, struct task_struct *task)
+{
+	return 0;
+}
+
+/*
+ * Simple, special scheduling class for the per-CPU stop tasks:
+ */
+static const struct sched_class stop_sched_class = {
+	.next			= &rt_sched_class,
+
+	.enqueue_task		= enqueue_task_stop,
+	.dequeue_task		= dequeue_task_stop,
+	.yield_task		= yield_task_stop,
+
+	.check_preempt_curr	= check_preempt_curr_stop,
+
+	.pick_next_task		= pick_next_task_stop,
+	.put_prev_task		= put_prev_task_stop,
+
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_stop,
+#endif
+
+	.set_curr_task          = set_curr_task_stop,
+	.task_tick		= task_tick_stop,
+
+	.get_rr_interval	= get_rr_interval_stop,
+
+	.prio_changed		= prio_changed_stop,
+	.switched_to		= switched_to_stop,
+
+	/* no .task_new for stop tasks */
+};
diff --git a/kernel/signal.c b/kernel/signal.c
index bded651..919562c 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2215,6 +2215,14 @@
 #ifdef __ARCH_SI_TRAPNO
 		err |= __put_user(from->si_trapno, &to->si_trapno);
 #endif
+#ifdef BUS_MCEERR_AO
+		/* 
+		 * Other callers might not initialize the si_lsb field,
+	 	 * so check explicitely for the right codes here.
+		 */
+		if (from->si_code == BUS_MCEERR_AR || from->si_code == BUS_MCEERR_AO)
+			err |= __put_user(from->si_addr_lsb, &to->si_addr_lsb);
+#endif
 		break;
 	case __SI_CHLD:
 		err |= __put_user(from->si_pid, &to->si_pid);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 07b4f1b..fc97888 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -77,11 +77,21 @@
 }
 
 /*
+ * preempt_count and SOFTIRQ_OFFSET usage:
+ * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving
+ *   softirq processing.
+ * - preempt_count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET)
+ *   on local_bh_disable or local_bh_enable.
+ * This lets us distinguish between whether we are currently processing
+ * softirq and whether we just have bh disabled.
+ */
+
+/*
  * This one is for softirq.c-internal use,
  * where hardirqs are disabled legitimately:
  */
 #ifdef CONFIG_TRACE_IRQFLAGS
-static void __local_bh_disable(unsigned long ip)
+static void __local_bh_disable(unsigned long ip, unsigned int cnt)
 {
 	unsigned long flags;
 
@@ -95,32 +105,43 @@
 	 * We must manually increment preempt_count here and manually
 	 * call the trace_preempt_off later.
 	 */
-	preempt_count() += SOFTIRQ_OFFSET;
+	preempt_count() += cnt;
 	/*
 	 * Were softirqs turned off above:
 	 */
-	if (softirq_count() == SOFTIRQ_OFFSET)
+	if (softirq_count() == cnt)
 		trace_softirqs_off(ip);
 	raw_local_irq_restore(flags);
 
-	if (preempt_count() == SOFTIRQ_OFFSET)
+	if (preempt_count() == cnt)
 		trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 }
 #else /* !CONFIG_TRACE_IRQFLAGS */
-static inline void __local_bh_disable(unsigned long ip)
+static inline void __local_bh_disable(unsigned long ip, unsigned int cnt)
 {
-	add_preempt_count(SOFTIRQ_OFFSET);
+	add_preempt_count(cnt);
 	barrier();
 }
 #endif /* CONFIG_TRACE_IRQFLAGS */
 
 void local_bh_disable(void)
 {
-	__local_bh_disable((unsigned long)__builtin_return_address(0));
+	__local_bh_disable((unsigned long)__builtin_return_address(0),
+				SOFTIRQ_DISABLE_OFFSET);
 }
 
 EXPORT_SYMBOL(local_bh_disable);
 
+static void __local_bh_enable(unsigned int cnt)
+{
+	WARN_ON_ONCE(in_irq());
+	WARN_ON_ONCE(!irqs_disabled());
+
+	if (softirq_count() == cnt)
+		trace_softirqs_on((unsigned long)__builtin_return_address(0));
+	sub_preempt_count(cnt);
+}
+
 /*
  * Special-case - softirqs can safely be enabled in
  * cond_resched_softirq(), or by __do_softirq(),
@@ -128,12 +149,7 @@
  */
 void _local_bh_enable(void)
 {
-	WARN_ON_ONCE(in_irq());
-	WARN_ON_ONCE(!irqs_disabled());
-
-	if (softirq_count() == SOFTIRQ_OFFSET)
-		trace_softirqs_on((unsigned long)__builtin_return_address(0));
-	sub_preempt_count(SOFTIRQ_OFFSET);
+	__local_bh_enable(SOFTIRQ_DISABLE_OFFSET);
 }
 
 EXPORT_SYMBOL(_local_bh_enable);
@@ -147,13 +163,13 @@
 	/*
 	 * Are softirqs going to be turned on now:
 	 */
-	if (softirq_count() == SOFTIRQ_OFFSET)
+	if (softirq_count() == SOFTIRQ_DISABLE_OFFSET)
 		trace_softirqs_on(ip);
 	/*
 	 * Keep preemption disabled until we are done with
 	 * softirq processing:
  	 */
- 	sub_preempt_count(SOFTIRQ_OFFSET - 1);
+	sub_preempt_count(SOFTIRQ_DISABLE_OFFSET - 1);
 
 	if (unlikely(!in_interrupt() && local_softirq_pending()))
 		do_softirq();
@@ -198,7 +214,8 @@
 	pending = local_softirq_pending();
 	account_system_vtime(current);
 
-	__local_bh_disable((unsigned long)__builtin_return_address(0));
+	__local_bh_disable((unsigned long)__builtin_return_address(0),
+				SOFTIRQ_OFFSET);
 	lockdep_softirq_enter();
 
 	cpu = smp_processor_id();
@@ -245,7 +262,7 @@
 	lockdep_softirq_exit();
 
 	account_system_vtime(current);
-	_local_bh_enable();
+	__local_bh_enable(SOFTIRQ_OFFSET);
 }
 
 #ifndef __ARCH_HAS_DO_SOFTIRQ
@@ -279,10 +296,16 @@
 
 	rcu_irq_enter();
 	if (idle_cpu(cpu) && !in_interrupt()) {
-		__irq_enter();
+		/*
+		 * Prevent raise_softirq from needlessly waking up ksoftirqd
+		 * here, as softirq will be serviced on return from interrupt.
+		 */
+		local_bh_disable();
 		tick_check_idle(cpu);
-	} else
-		__irq_enter();
+		_local_bh_enable();
+	}
+
+	__irq_enter();
 }
 
 #ifdef __ARCH_IRQ_EXIT_IRQS_DISABLED
@@ -696,6 +719,7 @@
 {
 	set_current_state(TASK_INTERRUPTIBLE);
 
+	current->flags |= PF_KSOFTIRQD;
 	while (!kthread_should_stop()) {
 		preempt_disable();
 		if (!local_softirq_pending()) {
@@ -886,17 +910,14 @@
 	return 0;
 }
 
+#ifdef CONFIG_GENERIC_HARDIRQS
 int __init __weak arch_probe_nr_irqs(void)
 {
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 
 int __init __weak arch_early_irq_init(void)
 {
 	return 0;
 }
-
-int __weak arch_init_chip_data(struct irq_desc *desc, int node)
-{
-	return 0;
-}
+#endif
diff --git a/kernel/srcu.c b/kernel/srcu.c
index 2980da3..c71e075 100644
--- a/kernel/srcu.c
+++ b/kernel/srcu.c
@@ -46,11 +46,9 @@
 int __init_srcu_struct(struct srcu_struct *sp, const char *name,
 		       struct lock_class_key *key)
 {
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
 	/* Don't re-initialize a lock while it is held. */
 	debug_check_no_locks_freed((void *)sp, sizeof(*sp));
 	lockdep_init_map(&sp->dep_map, name, key, 0);
-#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 	return init_srcu_struct_fields(sp);
 }
 EXPORT_SYMBOL_GPL(__init_srcu_struct);
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 4372ccb..090c288 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -287,11 +287,12 @@
 	goto repeat;
 }
 
+extern void sched_set_stop_task(int cpu, struct task_struct *stop);
+
 /* manage stopper for a cpu, mostly lifted from sched migration thread mgmt */
 static int __cpuinit cpu_stop_cpu_callback(struct notifier_block *nfb,
 					   unsigned long action, void *hcpu)
 {
-	struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 };
 	unsigned int cpu = (unsigned long)hcpu;
 	struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
 	struct task_struct *p;
@@ -304,13 +305,13 @@
 				   cpu);
 		if (IS_ERR(p))
 			return NOTIFY_BAD;
-		sched_setscheduler_nocheck(p, SCHED_FIFO, &param);
 		get_task_struct(p);
+		kthread_bind(p, cpu);
+		sched_set_stop_task(cpu, p);
 		stopper->thread = p;
 		break;
 
 	case CPU_ONLINE:
-		kthread_bind(stopper->thread, cpu);
 		/* strictly unnecessary, as first user will wake it */
 		wake_up_process(stopper->thread);
 		/* mark enabled */
@@ -325,6 +326,7 @@
 	{
 		struct cpu_stop_work *work;
 
+		sched_set_stop_task(cpu, NULL);
 		/* kill the stopper */
 		kthread_stop(stopper->thread);
 		/* drain remaining works */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index f88552c..3a45c22 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -2485,7 +2485,7 @@
 		kbuf[left] = 0;
 	}
 
-	for (; left && vleft--; i++, min++, max++, first=0) {
+	for (; left && vleft--; i++, first = 0) {
 		unsigned long val;
 
 		if (write) {
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index 04cdcf7..10b90d8 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -143,15 +143,6 @@
 				if (!table->maxlen)
 					set_fail(&fail, table, "No maxlen");
 			}
-			if ((table->proc_handler == proc_doulongvec_minmax) ||
-			    (table->proc_handler == proc_doulongvec_ms_jiffies_minmax)) {
-				if (table->maxlen > sizeof (unsigned long)) {
-					if (!table->extra1)
-						set_fail(&fail, table, "No min");
-					if (!table->extra2)
-						set_fail(&fail, table, "No max");
-				}
-			}
 #ifdef CONFIG_PROC_SYSCTL
 			if (table->procname && !table->proc_handler)
 				set_fail(&fail, table, "No proc_handler");
diff --git a/kernel/test_kprobes.c b/kernel/test_kprobes.c
index 4f10451..f8b11a2 100644
--- a/kernel/test_kprobes.c
+++ b/kernel/test_kprobes.c
@@ -115,7 +115,9 @@
 	int ret;
 	struct kprobe *kps[2] = {&kp, &kp2};
 
-	kp.addr = 0; /* addr should be cleard for reusing kprobe. */
+	/* addr and flags should be cleard for reusing kprobe. */
+	kp.addr = NULL;
+	kp.flags = 0;
 	ret = register_kprobes(kps, 2);
 	if (ret < 0) {
 		printk(KERN_ERR "Kprobe smoke test failed: "
@@ -210,7 +212,9 @@
 	int ret;
 	struct jprobe *jps[2] = {&jp, &jp2};
 
-	jp.kp.addr = 0; /* addr should be cleard for reusing kprobe. */
+	/* addr and flags should be cleard for reusing kprobe. */
+	jp.kp.addr = NULL;
+	jp.kp.flags = 0;
 	ret = register_jprobes(jps, 2);
 	if (ret < 0) {
 		printk(KERN_ERR "Kprobe smoke test failed: "
@@ -323,7 +327,9 @@
 	int ret;
 	struct kretprobe *rps[2] = {&rp, &rp2};
 
-	rp.kp.addr = 0; /* addr should be cleard for reusing kprobe. */
+	/* addr and flags should be cleard for reusing kprobe. */
+	rp.kp.addr = NULL;
+	rp.kp.flags = 0;
 	ret = register_kretprobes(rps, 2);
 	if (ret < 0) {
 		printk(KERN_ERR "Kprobe smoke test failed: "
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index c631168..d232189 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -149,10 +149,18 @@
 	time_reftime = get_seconds();
 
 	offset64    = offset;
-	freq_adj    = (offset64 * secs) <<
-			(NTP_SCALE_SHIFT - 2 * (SHIFT_PLL + 2 + time_constant));
+	freq_adj    = ntp_update_offset_fll(offset64, secs);
 
-	freq_adj    += ntp_update_offset_fll(offset64, secs);
+	/*
+	 * Clamp update interval to reduce PLL gain with low
+	 * sampling rate (e.g. intermittent network connection)
+	 * to avoid instability.
+	 */
+	if (unlikely(secs > 1 << (SHIFT_PLL + 1 + time_constant)))
+		secs = 1 << (SHIFT_PLL + 1 + time_constant);
+
+	freq_adj    += (offset64 * secs) <<
+			(NTP_SCALE_SHIFT - 2 * (SHIFT_PLL + 2 + time_constant));
 
 	freq_adj    = min(freq_adj + time_freq, MAXFREQ_SCALED);
 
diff --git a/kernel/timer.c b/kernel/timer.c
index 97bf05b..68a9ae7 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -37,7 +37,7 @@
 #include <linux/delay.h>
 #include <linux/tick.h>
 #include <linux/kallsyms.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 
@@ -1279,7 +1279,10 @@
 	run_local_timers();
 	rcu_check_callbacks(cpu, user_tick);
 	printk_tick();
-	perf_event_do_pending();
+#ifdef CONFIG_IRQ_WORK
+	if (in_irq())
+		irq_work_run();
+#endif
 	scheduler_tick();
 	run_posix_cpu_timers(p);
 }
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 538501c..e550d2ed 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -49,6 +49,11 @@
 	help
 	  See Documentation/trace/ftrace-design.txt
 
+config HAVE_C_RECORDMCOUNT
+	bool
+	help
+	  C version of recordmcount available?
+
 config TRACER_MAX_TRACE
 	bool
 
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index fa7ece6..ebd80d5 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -884,10 +884,8 @@
 	FTRACE_ENABLE_CALLS		= (1 << 0),
 	FTRACE_DISABLE_CALLS		= (1 << 1),
 	FTRACE_UPDATE_TRACE_FUNC	= (1 << 2),
-	FTRACE_ENABLE_MCOUNT		= (1 << 3),
-	FTRACE_DISABLE_MCOUNT		= (1 << 4),
-	FTRACE_START_FUNC_RET		= (1 << 5),
-	FTRACE_STOP_FUNC_RET		= (1 << 6),
+	FTRACE_START_FUNC_RET		= (1 << 3),
+	FTRACE_STOP_FUNC_RET		= (1 << 4),
 };
 
 static int ftrace_filtered;
@@ -1226,8 +1224,6 @@
 
 static void ftrace_startup_sysctl(void)
 {
-	int command = FTRACE_ENABLE_MCOUNT;
-
 	if (unlikely(ftrace_disabled))
 		return;
 
@@ -1235,23 +1231,17 @@
 	saved_ftrace_func = NULL;
 	/* ftrace_start_up is true if we want ftrace running */
 	if (ftrace_start_up)
-		command |= FTRACE_ENABLE_CALLS;
-
-	ftrace_run_update_code(command);
+		ftrace_run_update_code(FTRACE_ENABLE_CALLS);
 }
 
 static void ftrace_shutdown_sysctl(void)
 {
-	int command = FTRACE_DISABLE_MCOUNT;
-
 	if (unlikely(ftrace_disabled))
 		return;
 
 	/* ftrace_start_up is true if ftrace is running */
 	if (ftrace_start_up)
-		command |= FTRACE_DISABLE_CALLS;
-
-	ftrace_run_update_code(command);
+		ftrace_run_update_code(FTRACE_DISABLE_CALLS);
 }
 
 static cycle_t		ftrace_update_time;
@@ -1368,24 +1358,29 @@
 #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */
 
 struct ftrace_iterator {
-	struct ftrace_page	*pg;
-	int			hidx;
-	int			idx;
-	unsigned		flags;
-	struct trace_parser	parser;
+	loff_t				pos;
+	loff_t				func_pos;
+	struct ftrace_page		*pg;
+	struct dyn_ftrace		*func;
+	struct ftrace_func_probe	*probe;
+	struct trace_parser		parser;
+	int				hidx;
+	int				idx;
+	unsigned			flags;
 };
 
 static void *
-t_hash_next(struct seq_file *m, void *v, loff_t *pos)
+t_hash_next(struct seq_file *m, loff_t *pos)
 {
 	struct ftrace_iterator *iter = m->private;
-	struct hlist_node *hnd = v;
+	struct hlist_node *hnd = NULL;
 	struct hlist_head *hhd;
 
-	WARN_ON(!(iter->flags & FTRACE_ITER_HASH));
-
 	(*pos)++;
+	iter->pos = *pos;
 
+	if (iter->probe)
+		hnd = &iter->probe->node;
  retry:
 	if (iter->hidx >= FTRACE_FUNC_HASHSIZE)
 		return NULL;
@@ -1408,7 +1403,12 @@
 		}
 	}
 
-	return hnd;
+	if (WARN_ON_ONCE(!hnd))
+		return NULL;
+
+	iter->probe = hlist_entry(hnd, struct ftrace_func_probe, node);
+
+	return iter;
 }
 
 static void *t_hash_start(struct seq_file *m, loff_t *pos)
@@ -1417,26 +1417,32 @@
 	void *p = NULL;
 	loff_t l;
 
-	if (!(iter->flags & FTRACE_ITER_HASH))
-		*pos = 0;
-
-	iter->flags |= FTRACE_ITER_HASH;
+	if (iter->func_pos > *pos)
+		return NULL;
 
 	iter->hidx = 0;
-	for (l = 0; l <= *pos; ) {
-		p = t_hash_next(m, p, &l);
+	for (l = 0; l <= (*pos - iter->func_pos); ) {
+		p = t_hash_next(m, &l);
 		if (!p)
 			break;
 	}
-	return p;
+	if (!p)
+		return NULL;
+
+	/* Only set this if we have an item */
+	iter->flags |= FTRACE_ITER_HASH;
+
+	return iter;
 }
 
-static int t_hash_show(struct seq_file *m, void *v)
+static int
+t_hash_show(struct seq_file *m, struct ftrace_iterator *iter)
 {
 	struct ftrace_func_probe *rec;
-	struct hlist_node *hnd = v;
 
-	rec = hlist_entry(hnd, struct ftrace_func_probe, node);
+	rec = iter->probe;
+	if (WARN_ON_ONCE(!rec))
+		return -EIO;
 
 	if (rec->ops->print)
 		return rec->ops->print(m, rec->ip, rec->ops, rec->data);
@@ -1457,12 +1463,13 @@
 	struct dyn_ftrace *rec = NULL;
 
 	if (iter->flags & FTRACE_ITER_HASH)
-		return t_hash_next(m, v, pos);
+		return t_hash_next(m, pos);
 
 	(*pos)++;
+	iter->pos = *pos;
 
 	if (iter->flags & FTRACE_ITER_PRINTALL)
-		return NULL;
+		return t_hash_start(m, pos);
 
  retry:
 	if (iter->idx >= iter->pg->index) {
@@ -1491,7 +1498,20 @@
 		}
 	}
 
-	return rec;
+	if (!rec)
+		return t_hash_start(m, pos);
+
+	iter->func_pos = *pos;
+	iter->func = rec;
+
+	return iter;
+}
+
+static void reset_iter_read(struct ftrace_iterator *iter)
+{
+	iter->pos = 0;
+	iter->func_pos = 0;
+	iter->flags &= ~(FTRACE_ITER_PRINTALL & FTRACE_ITER_HASH);
 }
 
 static void *t_start(struct seq_file *m, loff_t *pos)
@@ -1502,6 +1522,12 @@
 
 	mutex_lock(&ftrace_lock);
 	/*
+	 * If an lseek was done, then reset and start from beginning.
+	 */
+	if (*pos < iter->pos)
+		reset_iter_read(iter);
+
+	/*
 	 * For set_ftrace_filter reading, if we have the filter
 	 * off, we can short cut and just print out that all
 	 * functions are enabled.
@@ -1518,6 +1544,11 @@
 	if (iter->flags & FTRACE_ITER_HASH)
 		return t_hash_start(m, pos);
 
+	/*
+	 * Unfortunately, we need to restart at ftrace_pages_start
+	 * every time we let go of the ftrace_mutex. This is because
+	 * those pointers can change without the lock.
+	 */
 	iter->pg = ftrace_pages_start;
 	iter->idx = 0;
 	for (l = 0; l <= *pos; ) {
@@ -1526,10 +1557,14 @@
 			break;
 	}
 
-	if (!p && iter->flags & FTRACE_ITER_FILTER)
-		return t_hash_start(m, pos);
+	if (!p) {
+		if (iter->flags & FTRACE_ITER_FILTER)
+			return t_hash_start(m, pos);
 
-	return p;
+		return NULL;
+	}
+
+	return iter;
 }
 
 static void t_stop(struct seq_file *m, void *p)
@@ -1540,16 +1575,18 @@
 static int t_show(struct seq_file *m, void *v)
 {
 	struct ftrace_iterator *iter = m->private;
-	struct dyn_ftrace *rec = v;
+	struct dyn_ftrace *rec;
 
 	if (iter->flags & FTRACE_ITER_HASH)
-		return t_hash_show(m, v);
+		return t_hash_show(m, iter);
 
 	if (iter->flags & FTRACE_ITER_PRINTALL) {
 		seq_printf(m, "#### all functions enabled ####\n");
 		return 0;
 	}
 
+	rec = iter->func;
+
 	if (!rec)
 		return 0;
 
@@ -1601,8 +1638,8 @@
 
 	ret = ftrace_avail_open(inode, file);
 	if (!ret) {
-		m = (struct seq_file *)file->private_data;
-		iter = (struct ftrace_iterator *)m->private;
+		m = file->private_data;
+		iter = m->private;
 		iter->flags = FTRACE_ITER_FAILURES;
 	}
 
@@ -2418,7 +2455,7 @@
 	.open = ftrace_filter_open,
 	.read = seq_read,
 	.write = ftrace_filter_write,
-	.llseek = no_llseek,
+	.llseek = ftrace_regex_lseek,
 	.release = ftrace_filter_release,
 };
 
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 492197e..c5a632a 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -405,7 +405,7 @@
 #define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2))
 
 /* Max number of timestamps that can fit on a page */
-#define RB_TIMESTAMPS_PER_PAGE	(BUF_PAGE_SIZE / RB_LEN_TIME_STAMP)
+#define RB_TIMESTAMPS_PER_PAGE	(BUF_PAGE_SIZE / RB_LEN_TIME_EXTEND)
 
 int ring_buffer_print_page_header(struct trace_seq *s)
 {
@@ -2606,6 +2606,19 @@
 }
 EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu);
 
+/*
+ * The total entries in the ring buffer is the running counter
+ * of entries entered into the ring buffer, minus the sum of
+ * the entries read from the ring buffer and the number of
+ * entries that were overwritten.
+ */
+static inline unsigned long
+rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer)
+{
+	return local_read(&cpu_buffer->entries) -
+		(local_read(&cpu_buffer->overrun) + cpu_buffer->read);
+}
+
 /**
  * ring_buffer_entries_cpu - get the number of entries in a cpu buffer
  * @buffer: The ring buffer
@@ -2614,16 +2627,13 @@
 unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu)
 {
 	struct ring_buffer_per_cpu *cpu_buffer;
-	unsigned long ret;
 
 	if (!cpumask_test_cpu(cpu, buffer->cpumask))
 		return 0;
 
 	cpu_buffer = buffer->buffers[cpu];
-	ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun))
-		- cpu_buffer->read;
 
-	return ret;
+	return rb_num_of_entries(cpu_buffer);
 }
 EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu);
 
@@ -2684,8 +2694,7 @@
 	/* if you care about this being correct, lock the buffer */
 	for_each_buffer_cpu(buffer, cpu) {
 		cpu_buffer = buffer->buffers[cpu];
-		entries += (local_read(&cpu_buffer->entries) -
-			    local_read(&cpu_buffer->overrun)) - cpu_buffer->read;
+		entries += rb_num_of_entries(cpu_buffer);
 	}
 
 	return entries;
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 9ec59f5..001bcd2 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2196,7 +2196,7 @@
 
 static int tracing_release(struct inode *inode, struct file *file)
 {
-	struct seq_file *m = (struct seq_file *)file->private_data;
+	struct seq_file *m = file->private_data;
 	struct trace_iterator *iter;
 	int cpu;
 
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index d39b3c5..9021f8c 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -343,6 +343,10 @@
 		    unsigned long ip,
 		    unsigned long parent_ip,
 		    unsigned long flags, int pc);
+void trace_graph_function(struct trace_array *tr,
+		    unsigned long ip,
+		    unsigned long parent_ip,
+		    unsigned long flags, int pc);
 void trace_default_header(struct seq_file *m);
 void print_trace_header(struct seq_file *m, struct trace_iterator *iter);
 int trace_empty(struct trace_iterator *iter);
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 31cc4cb..39c059c 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -9,7 +9,7 @@
 #include <linux/kprobes.h>
 #include "trace.h"
 
-static char *perf_trace_buf[4];
+static char __percpu *perf_trace_buf[PERF_NR_CONTEXTS];
 
 /*
  * Force it to be aligned to unsigned long to avoid misaligned accesses
@@ -24,7 +24,7 @@
 static int perf_trace_event_init(struct ftrace_event_call *tp_event,
 				 struct perf_event *p_event)
 {
-	struct hlist_head *list;
+	struct hlist_head __percpu *list;
 	int ret = -ENOMEM;
 	int cpu;
 
@@ -42,11 +42,11 @@
 	tp_event->perf_events = list;
 
 	if (!total_ref_count) {
-		char *buf;
+		char __percpu *buf;
 		int i;
 
-		for (i = 0; i < 4; i++) {
-			buf = (char *)alloc_percpu(perf_trace_t);
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
+			buf = (char __percpu *)alloc_percpu(perf_trace_t);
 			if (!buf)
 				goto fail;
 
@@ -65,7 +65,7 @@
 	if (!total_ref_count) {
 		int i;
 
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
@@ -101,22 +101,26 @@
 	return ret;
 }
 
-int perf_trace_enable(struct perf_event *p_event)
+int perf_trace_add(struct perf_event *p_event, int flags)
 {
 	struct ftrace_event_call *tp_event = p_event->tp_event;
+	struct hlist_head __percpu *pcpu_list;
 	struct hlist_head *list;
 
-	list = tp_event->perf_events;
-	if (WARN_ON_ONCE(!list))
+	pcpu_list = tp_event->perf_events;
+	if (WARN_ON_ONCE(!pcpu_list))
 		return -EINVAL;
 
-	list = this_cpu_ptr(list);
+	if (!(flags & PERF_EF_START))
+		p_event->hw.state = PERF_HES_STOPPED;
+
+	list = this_cpu_ptr(pcpu_list);
 	hlist_add_head_rcu(&p_event->hlist_entry, list);
 
 	return 0;
 }
 
-void perf_trace_disable(struct perf_event *p_event)
+void perf_trace_del(struct perf_event *p_event, int flags)
 {
 	hlist_del_rcu(&p_event->hlist_entry);
 }
@@ -142,7 +146,7 @@
 	tp_event->perf_events = NULL;
 
 	if (!--total_ref_count) {
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 4c758f1..398c0e8 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -600,21 +600,29 @@
 
 enum {
 	FORMAT_HEADER		= 1,
-	FORMAT_PRINTFMT		= 2,
+	FORMAT_FIELD_SEPERATOR	= 2,
+	FORMAT_PRINTFMT		= 3,
 };
 
 static void *f_next(struct seq_file *m, void *v, loff_t *pos)
 {
 	struct ftrace_event_call *call = m->private;
 	struct ftrace_event_field *field;
-	struct list_head *head;
+	struct list_head *common_head = &ftrace_common_fields;
+	struct list_head *head = trace_get_fields(call);
 
 	(*pos)++;
 
 	switch ((unsigned long)v) {
 	case FORMAT_HEADER:
-		head = &ftrace_common_fields;
+		if (unlikely(list_empty(common_head)))
+			return NULL;
 
+		field = list_entry(common_head->prev,
+				   struct ftrace_event_field, link);
+		return field;
+
+	case FORMAT_FIELD_SEPERATOR:
 		if (unlikely(list_empty(head)))
 			return NULL;
 
@@ -626,31 +634,10 @@
 		return NULL;
 	}
 
-	head = trace_get_fields(call);
-
-	/*
-	 * To separate common fields from event fields, the
-	 * LSB is set on the first event field. Clear it in case.
-	 */
-	v = (void *)((unsigned long)v & ~1L);
-
 	field = v;
-	/*
-	 * If this is a common field, and at the end of the list, then
-	 * continue with main list.
-	 */
-	if (field->link.prev == &ftrace_common_fields) {
-		if (unlikely(list_empty(head)))
-			return NULL;
-		field = list_entry(head->prev, struct ftrace_event_field, link);
-		/* Set the LSB to notify f_show to print an extra newline */
-		field = (struct ftrace_event_field *)
-			((unsigned long)field | 1);
-		return field;
-	}
-
-	/* If we are done tell f_show to print the format */
-	if (field->link.prev == head)
+	if (field->link.prev == common_head)
+		return (void *)FORMAT_FIELD_SEPERATOR;
+	else if (field->link.prev == head)
 		return (void *)FORMAT_PRINTFMT;
 
 	field = list_entry(field->link.prev, struct ftrace_event_field, link);
@@ -688,22 +675,16 @@
 		seq_printf(m, "format:\n");
 		return 0;
 
+	case FORMAT_FIELD_SEPERATOR:
+		seq_putc(m, '\n');
+		return 0;
+
 	case FORMAT_PRINTFMT:
 		seq_printf(m, "\nprint fmt: %s\n",
 			   call->print_fmt);
 		return 0;
 	}
 
-	/*
-	 * To separate common fields from event fields, the
-	 * LSB is set on the first event field. Clear it and
-	 * print a newline if it is set.
-	 */
-	if ((unsigned long)v & 1) {
-		seq_putc(m, '\n');
-		v = (void *)((unsigned long)v & ~1L);
-	}
-
 	field = v;
 
 	/*
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index 6f23369..76b0598 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -15,15 +15,19 @@
 #include "trace.h"
 #include "trace_output.h"
 
+/* When set, irq functions will be ignored */
+static int ftrace_graph_skip_irqs;
+
 struct fgraph_cpu_data {
 	pid_t		last_pid;
 	int		depth;
+	int		depth_irq;
 	int		ignore;
 	unsigned long	enter_funcs[FTRACE_RETFUNC_DEPTH];
 };
 
 struct fgraph_data {
-	struct fgraph_cpu_data		*cpu_data;
+	struct fgraph_cpu_data __percpu *cpu_data;
 
 	/* Place to preserve last processed entry. */
 	struct ftrace_graph_ent_entry	ent;
@@ -41,6 +45,7 @@
 #define TRACE_GRAPH_PRINT_PROC		0x8
 #define TRACE_GRAPH_PRINT_DURATION	0x10
 #define TRACE_GRAPH_PRINT_ABS_TIME	0x20
+#define TRACE_GRAPH_PRINT_IRQS		0x40
 
 static struct tracer_opt trace_opts[] = {
 	/* Display overruns? (for self-debug purpose) */
@@ -55,13 +60,15 @@
 	{ TRACER_OPT(funcgraph-duration, TRACE_GRAPH_PRINT_DURATION) },
 	/* Display absolute time of an entry */
 	{ TRACER_OPT(funcgraph-abstime, TRACE_GRAPH_PRINT_ABS_TIME) },
+	/* Display interrupts */
+	{ TRACER_OPT(funcgraph-irqs, TRACE_GRAPH_PRINT_IRQS) },
 	{ } /* Empty entry */
 };
 
 static struct tracer_flags tracer_flags = {
 	/* Don't display overruns and proc by default */
 	.val = TRACE_GRAPH_PRINT_CPU | TRACE_GRAPH_PRINT_OVERHEAD |
-	       TRACE_GRAPH_PRINT_DURATION,
+	       TRACE_GRAPH_PRINT_DURATION | TRACE_GRAPH_PRINT_IRQS,
 	.opts = trace_opts
 };
 
@@ -204,6 +211,14 @@
 	return 1;
 }
 
+static inline int ftrace_graph_ignore_irqs(void)
+{
+	if (!ftrace_graph_skip_irqs)
+		return 0;
+
+	return in_irq();
+}
+
 int trace_graph_entry(struct ftrace_graph_ent *trace)
 {
 	struct trace_array *tr = graph_array;
@@ -218,7 +233,8 @@
 		return 0;
 
 	/* trace it when it is-nested-in or is a function enabled. */
-	if (!(trace->depth || ftrace_graph_addr(trace->func)))
+	if (!(trace->depth || ftrace_graph_addr(trace->func)) ||
+	      ftrace_graph_ignore_irqs())
 		return 0;
 
 	local_irq_save(flags);
@@ -246,6 +262,34 @@
 		return trace_graph_entry(trace);
 }
 
+static void
+__trace_graph_function(struct trace_array *tr,
+		unsigned long ip, unsigned long flags, int pc)
+{
+	u64 time = trace_clock_local();
+	struct ftrace_graph_ent ent = {
+		.func  = ip,
+		.depth = 0,
+	};
+	struct ftrace_graph_ret ret = {
+		.func     = ip,
+		.depth    = 0,
+		.calltime = time,
+		.rettime  = time,
+	};
+
+	__trace_graph_entry(tr, &ent, flags, pc);
+	__trace_graph_return(tr, &ret, flags, pc);
+}
+
+void
+trace_graph_function(struct trace_array *tr,
+		unsigned long ip, unsigned long parent_ip,
+		unsigned long flags, int pc)
+{
+	__trace_graph_function(tr, ip, flags, pc);
+}
+
 void __trace_graph_return(struct trace_array *tr,
 				struct ftrace_graph_ret *trace,
 				unsigned long flags,
@@ -649,8 +693,9 @@
 
 	/* Print nsecs (we don't want to exceed 7 numbers) */
 	if (len < 7) {
-		snprintf(nsecs_str, min(sizeof(nsecs_str), 8UL - len), "%03lu",
-			 nsecs_rem);
+		size_t slen = min_t(size_t, sizeof(nsecs_str), 8UL - len);
+
+		snprintf(nsecs_str, slen, "%03lu", nsecs_rem);
 		ret = trace_seq_printf(s, ".%s", nsecs_str);
 		if (!ret)
 			return TRACE_TYPE_PARTIAL_LINE;
@@ -855,6 +900,108 @@
 	return 0;
 }
 
+/*
+ * Entry check for irq code
+ *
+ * returns 1 if
+ *  - we are inside irq code
+ *  - we just extered irq code
+ *
+ * retunns 0 if
+ *  - funcgraph-interrupts option is set
+ *  - we are not inside irq code
+ */
+static int
+check_irq_entry(struct trace_iterator *iter, u32 flags,
+		unsigned long addr, int depth)
+{
+	int cpu = iter->cpu;
+	int *depth_irq;
+	struct fgraph_data *data = iter->private;
+
+	/*
+	 * If we are either displaying irqs, or we got called as
+	 * a graph event and private data does not exist,
+	 * then we bypass the irq check.
+	 */
+	if ((flags & TRACE_GRAPH_PRINT_IRQS) ||
+	    (!data))
+		return 0;
+
+	depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
+	/*
+	 * We are inside the irq code
+	 */
+	if (*depth_irq >= 0)
+		return 1;
+
+	if ((addr < (unsigned long)__irqentry_text_start) ||
+	    (addr >= (unsigned long)__irqentry_text_end))
+		return 0;
+
+	/*
+	 * We are entering irq code.
+	 */
+	*depth_irq = depth;
+	return 1;
+}
+
+/*
+ * Return check for irq code
+ *
+ * returns 1 if
+ *  - we are inside irq code
+ *  - we just left irq code
+ *
+ * returns 0 if
+ *  - funcgraph-interrupts option is set
+ *  - we are not inside irq code
+ */
+static int
+check_irq_return(struct trace_iterator *iter, u32 flags, int depth)
+{
+	int cpu = iter->cpu;
+	int *depth_irq;
+	struct fgraph_data *data = iter->private;
+
+	/*
+	 * If we are either displaying irqs, or we got called as
+	 * a graph event and private data does not exist,
+	 * then we bypass the irq check.
+	 */
+	if ((flags & TRACE_GRAPH_PRINT_IRQS) ||
+	    (!data))
+		return 0;
+
+	depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
+	/*
+	 * We are not inside the irq code.
+	 */
+	if (*depth_irq == -1)
+		return 0;
+
+	/*
+	 * We are inside the irq code, and this is returning entry.
+	 * Let's not trace it and clear the entry depth, since
+	 * we are out of irq code.
+	 *
+	 * This condition ensures that we 'leave the irq code' once
+	 * we are out of the entry depth. Thus protecting us from
+	 * the RETURN entry loss.
+	 */
+	if (*depth_irq >= depth) {
+		*depth_irq = -1;
+		return 1;
+	}
+
+	/*
+	 * We are inside the irq code, and this is not the entry.
+	 */
+	return 1;
+}
+
 static enum print_line_t
 print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s,
 			struct trace_iterator *iter, u32 flags)
@@ -865,6 +1012,9 @@
 	static enum print_line_t ret;
 	int cpu = iter->cpu;
 
+	if (check_irq_entry(iter, flags, call->func, call->depth))
+		return TRACE_TYPE_HANDLED;
+
 	if (print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func, flags))
 		return TRACE_TYPE_PARTIAL_LINE;
 
@@ -902,6 +1052,9 @@
 	int ret;
 	int i;
 
+	if (check_irq_return(iter, flags, trace->depth))
+		return TRACE_TYPE_HANDLED;
+
 	if (data) {
 		struct fgraph_cpu_data *cpu_data;
 		int cpu = iter->cpu;
@@ -1054,7 +1207,7 @@
 
 
 enum print_line_t
-print_graph_function_flags(struct trace_iterator *iter, u32 flags)
+__print_graph_function_flags(struct trace_iterator *iter, u32 flags)
 {
 	struct ftrace_graph_ent_entry *field;
 	struct fgraph_data *data = iter->private;
@@ -1117,7 +1270,18 @@
 static enum print_line_t
 print_graph_function(struct trace_iterator *iter)
 {
-	return print_graph_function_flags(iter, tracer_flags.val);
+	return __print_graph_function_flags(iter, tracer_flags.val);
+}
+
+enum print_line_t print_graph_function_flags(struct trace_iterator *iter,
+					     u32 flags)
+{
+	if (trace_flags & TRACE_ITER_LATENCY_FMT)
+		flags |= TRACE_GRAPH_PRINT_DURATION;
+	else
+		flags |= TRACE_GRAPH_PRINT_ABS_TIME;
+
+	return __print_graph_function_flags(iter, flags);
 }
 
 static enum print_line_t
@@ -1149,7 +1313,7 @@
 	seq_printf(s, "#%.*s|||| /                     \n", size, spaces);
 }
 
-void print_graph_headers_flags(struct seq_file *s, u32 flags)
+static void __print_graph_headers_flags(struct seq_file *s, u32 flags)
 {
 	int lat = trace_flags & TRACE_ITER_LATENCY_FMT;
 
@@ -1190,6 +1354,23 @@
 	print_graph_headers_flags(s, tracer_flags.val);
 }
 
+void print_graph_headers_flags(struct seq_file *s, u32 flags)
+{
+	struct trace_iterator *iter = s->private;
+
+	if (trace_flags & TRACE_ITER_LATENCY_FMT) {
+		/* print nothing if the buffers are empty */
+		if (trace_empty(iter))
+			return;
+
+		print_trace_header(s, iter);
+		flags |= TRACE_GRAPH_PRINT_DURATION;
+	} else
+		flags |= TRACE_GRAPH_PRINT_ABS_TIME;
+
+	__print_graph_headers_flags(s, flags);
+}
+
 void graph_trace_open(struct trace_iterator *iter)
 {
 	/* pid and depth on the last trace processed */
@@ -1210,9 +1391,12 @@
 		pid_t *pid = &(per_cpu_ptr(data->cpu_data, cpu)->last_pid);
 		int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth);
 		int *ignore = &(per_cpu_ptr(data->cpu_data, cpu)->ignore);
+		int *depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
 		*pid = -1;
 		*depth = 0;
 		*ignore = 0;
+		*depth_irq = -1;
 	}
 
 	iter->private = data;
@@ -1235,6 +1419,14 @@
 	}
 }
 
+static int func_graph_set_flag(u32 old_flags, u32 bit, int set)
+{
+	if (bit == TRACE_GRAPH_PRINT_IRQS)
+		ftrace_graph_skip_irqs = !set;
+
+	return 0;
+}
+
 static struct trace_event_functions graph_functions = {
 	.trace		= print_graph_function_event,
 };
@@ -1261,6 +1453,7 @@
 	.print_line	= print_graph_function,
 	.print_header	= print_graph_headers,
 	.flags		= &tracer_flags,
+	.set_flag	= func_graph_set_flag,
 #ifdef CONFIG_FTRACE_SELFTEST
 	.selftest	= trace_selftest_startup_function_graph,
 #endif
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index 73a6b06..5cf8c60 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -87,14 +87,22 @@
 
 #ifdef CONFIG_FUNCTION_TRACER
 /*
- * irqsoff uses its own tracer function to keep the overhead down:
+ * Prologue for the preempt and irqs off function tracers.
+ *
+ * Returns 1 if it is OK to continue, and data->disabled is
+ *            incremented.
+ *         0 if the trace is to be ignored, and data->disabled
+ *            is kept the same.
+ *
+ * Note, this function is also used outside this ifdef but
+ *  inside the #ifdef of the function graph tracer below.
+ *  This is OK, since the function graph tracer is
+ *  dependent on the function tracer.
  */
-static void
-irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip)
+static int func_prolog_dec(struct trace_array *tr,
+			   struct trace_array_cpu **data,
+			   unsigned long *flags)
 {
-	struct trace_array *tr = irqsoff_trace;
-	struct trace_array_cpu *data;
-	unsigned long flags;
 	long disabled;
 	int cpu;
 
@@ -106,18 +114,38 @@
 	 */
 	cpu = raw_smp_processor_id();
 	if (likely(!per_cpu(tracing_cpu, cpu)))
-		return;
+		return 0;
 
-	local_save_flags(flags);
+	local_save_flags(*flags);
 	/* slight chance to get a false positive on tracing_cpu */
-	if (!irqs_disabled_flags(flags))
-		return;
+	if (!irqs_disabled_flags(*flags))
+		return 0;
 
-	data = tr->data[cpu];
-	disabled = atomic_inc_return(&data->disabled);
+	*data = tr->data[cpu];
+	disabled = atomic_inc_return(&(*data)->disabled);
 
 	if (likely(disabled == 1))
-		trace_function(tr, ip, parent_ip, flags, preempt_count());
+		return 1;
+
+	atomic_dec(&(*data)->disabled);
+
+	return 0;
+}
+
+/*
+ * irqsoff uses its own tracer function to keep the overhead down:
+ */
+static void
+irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip)
+{
+	struct trace_array *tr = irqsoff_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+
+	if (!func_prolog_dec(tr, &data, &flags))
+		return;
+
+	trace_function(tr, ip, parent_ip, flags, preempt_count());
 
 	atomic_dec(&data->disabled);
 }
@@ -155,30 +183,16 @@
 	struct trace_array *tr = irqsoff_trace;
 	struct trace_array_cpu *data;
 	unsigned long flags;
-	long disabled;
 	int ret;
-	int cpu;
 	int pc;
 
-	cpu = raw_smp_processor_id();
-	if (likely(!per_cpu(tracing_cpu, cpu)))
+	if (!func_prolog_dec(tr, &data, &flags))
 		return 0;
 
-	local_save_flags(flags);
-	/* slight chance to get a false positive on tracing_cpu */
-	if (!irqs_disabled_flags(flags))
-		return 0;
-
-	data = tr->data[cpu];
-	disabled = atomic_inc_return(&data->disabled);
-
-	if (likely(disabled == 1)) {
-		pc = preempt_count();
-		ret = __trace_graph_entry(tr, trace, flags, pc);
-	} else
-		ret = 0;
-
+	pc = preempt_count();
+	ret = __trace_graph_entry(tr, trace, flags, pc);
 	atomic_dec(&data->disabled);
+
 	return ret;
 }
 
@@ -187,27 +201,13 @@
 	struct trace_array *tr = irqsoff_trace;
 	struct trace_array_cpu *data;
 	unsigned long flags;
-	long disabled;
-	int cpu;
 	int pc;
 
-	cpu = raw_smp_processor_id();
-	if (likely(!per_cpu(tracing_cpu, cpu)))
+	if (!func_prolog_dec(tr, &data, &flags))
 		return;
 
-	local_save_flags(flags);
-	/* slight chance to get a false positive on tracing_cpu */
-	if (!irqs_disabled_flags(flags))
-		return;
-
-	data = tr->data[cpu];
-	disabled = atomic_inc_return(&data->disabled);
-
-	if (likely(disabled == 1)) {
-		pc = preempt_count();
-		__trace_graph_return(tr, trace, flags, pc);
-	}
-
+	pc = preempt_count();
+	__trace_graph_return(tr, trace, flags, pc);
 	atomic_dec(&data->disabled);
 }
 
@@ -229,75 +229,33 @@
 
 static enum print_line_t irqsoff_print_line(struct trace_iterator *iter)
 {
-	u32 flags = GRAPH_TRACER_FLAGS;
-
-	if (trace_flags & TRACE_ITER_LATENCY_FMT)
-		flags |= TRACE_GRAPH_PRINT_DURATION;
-	else
-		flags |= TRACE_GRAPH_PRINT_ABS_TIME;
-
 	/*
 	 * In graph mode call the graph tracer output function,
 	 * otherwise go with the TRACE_FN event handler
 	 */
 	if (is_graph())
-		return print_graph_function_flags(iter, flags);
+		return print_graph_function_flags(iter, GRAPH_TRACER_FLAGS);
 
 	return TRACE_TYPE_UNHANDLED;
 }
 
 static void irqsoff_print_header(struct seq_file *s)
 {
-	if (is_graph()) {
-		struct trace_iterator *iter = s->private;
-		u32 flags = GRAPH_TRACER_FLAGS;
-
-		if (trace_flags & TRACE_ITER_LATENCY_FMT) {
-			/* print nothing if the buffers are empty */
-			if (trace_empty(iter))
-				return;
-
-			print_trace_header(s, iter);
-			flags |= TRACE_GRAPH_PRINT_DURATION;
-		} else
-			flags |= TRACE_GRAPH_PRINT_ABS_TIME;
-
-		print_graph_headers_flags(s, flags);
-	} else
+	if (is_graph())
+		print_graph_headers_flags(s, GRAPH_TRACER_FLAGS);
+	else
 		trace_default_header(s);
 }
 
 static void
-trace_graph_function(struct trace_array *tr,
-		 unsigned long ip, unsigned long flags, int pc)
-{
-	u64 time = trace_clock_local();
-	struct ftrace_graph_ent ent = {
-		.func  = ip,
-		.depth = 0,
-	};
-	struct ftrace_graph_ret ret = {
-		.func     = ip,
-		.depth    = 0,
-		.calltime = time,
-		.rettime  = time,
-	};
-
-	__trace_graph_entry(tr, &ent, flags, pc);
-	__trace_graph_return(tr, &ret, flags, pc);
-}
-
-static void
 __trace_function(struct trace_array *tr,
 		 unsigned long ip, unsigned long parent_ip,
 		 unsigned long flags, int pc)
 {
-	if (!is_graph())
+	if (is_graph())
+		trace_graph_function(tr, ip, parent_ip, flags, pc);
+	else
 		trace_function(tr, ip, parent_ip, flags, pc);
-	else {
-		trace_graph_function(tr, parent_ip, flags, pc);
-		trace_graph_function(tr, ip, flags, pc);
-	}
 }
 
 #else
diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
index 4086eae..7319559 100644
--- a/kernel/trace/trace_sched_wakeup.c
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -31,13 +31,81 @@
 static arch_spinlock_t wakeup_lock =
 	(arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
 
+static void wakeup_reset(struct trace_array *tr);
 static void __wakeup_reset(struct trace_array *tr);
+static int wakeup_graph_entry(struct ftrace_graph_ent *trace);
+static void wakeup_graph_return(struct ftrace_graph_ret *trace);
 
 static int save_lat_flag;
 
+#define TRACE_DISPLAY_GRAPH     1
+
+static struct tracer_opt trace_opts[] = {
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	/* display latency trace as call graph */
+	{ TRACER_OPT(display-graph, TRACE_DISPLAY_GRAPH) },
+#endif
+	{ } /* Empty entry */
+};
+
+static struct tracer_flags tracer_flags = {
+	.val  = 0,
+	.opts = trace_opts,
+};
+
+#define is_graph() (tracer_flags.val & TRACE_DISPLAY_GRAPH)
+
 #ifdef CONFIG_FUNCTION_TRACER
+
 /*
- * irqsoff uses its own tracer function to keep the overhead down:
+ * Prologue for the wakeup function tracers.
+ *
+ * Returns 1 if it is OK to continue, and preemption
+ *            is disabled and data->disabled is incremented.
+ *         0 if the trace is to be ignored, and preemption
+ *            is not disabled and data->disabled is
+ *            kept the same.
+ *
+ * Note, this function is also used outside this ifdef but
+ *  inside the #ifdef of the function graph tracer below.
+ *  This is OK, since the function graph tracer is
+ *  dependent on the function tracer.
+ */
+static int
+func_prolog_preempt_disable(struct trace_array *tr,
+			    struct trace_array_cpu **data,
+			    int *pc)
+{
+	long disabled;
+	int cpu;
+
+	if (likely(!wakeup_task))
+		return 0;
+
+	*pc = preempt_count();
+	preempt_disable_notrace();
+
+	cpu = raw_smp_processor_id();
+	if (cpu != wakeup_current_cpu)
+		goto out_enable;
+
+	*data = tr->data[cpu];
+	disabled = atomic_inc_return(&(*data)->disabled);
+	if (unlikely(disabled != 1))
+		goto out;
+
+	return 1;
+
+out:
+	atomic_dec(&(*data)->disabled);
+
+out_enable:
+	preempt_enable_notrace();
+	return 0;
+}
+
+/*
+ * wakeup uses its own tracer function to keep the overhead down:
  */
 static void
 wakeup_tracer_call(unsigned long ip, unsigned long parent_ip)
@@ -45,34 +113,16 @@
 	struct trace_array *tr = wakeup_trace;
 	struct trace_array_cpu *data;
 	unsigned long flags;
-	long disabled;
-	int cpu;
 	int pc;
 
-	if (likely(!wakeup_task))
+	if (!func_prolog_preempt_disable(tr, &data, &pc))
 		return;
 
-	pc = preempt_count();
-	preempt_disable_notrace();
-
-	cpu = raw_smp_processor_id();
-	if (cpu != wakeup_current_cpu)
-		goto out_enable;
-
-	data = tr->data[cpu];
-	disabled = atomic_inc_return(&data->disabled);
-	if (unlikely(disabled != 1))
-		goto out;
-
 	local_irq_save(flags);
-
 	trace_function(tr, ip, parent_ip, flags, pc);
-
 	local_irq_restore(flags);
 
- out:
 	atomic_dec(&data->disabled);
- out_enable:
 	preempt_enable_notrace();
 }
 
@@ -82,6 +132,156 @@
 };
 #endif /* CONFIG_FUNCTION_TRACER */
 
+static int start_func_tracer(int graph)
+{
+	int ret;
+
+	if (!graph)
+		ret = register_ftrace_function(&trace_ops);
+	else
+		ret = register_ftrace_graph(&wakeup_graph_return,
+					    &wakeup_graph_entry);
+
+	if (!ret && tracing_is_enabled())
+		tracer_enabled = 1;
+	else
+		tracer_enabled = 0;
+
+	return ret;
+}
+
+static void stop_func_tracer(int graph)
+{
+	tracer_enabled = 0;
+
+	if (!graph)
+		unregister_ftrace_function(&trace_ops);
+	else
+		unregister_ftrace_graph();
+}
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+static int wakeup_set_flag(u32 old_flags, u32 bit, int set)
+{
+
+	if (!(bit & TRACE_DISPLAY_GRAPH))
+		return -EINVAL;
+
+	if (!(is_graph() ^ set))
+		return 0;
+
+	stop_func_tracer(!set);
+
+	wakeup_reset(wakeup_trace);
+	tracing_max_latency = 0;
+
+	return start_func_tracer(set);
+}
+
+static int wakeup_graph_entry(struct ftrace_graph_ent *trace)
+{
+	struct trace_array *tr = wakeup_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+	int pc, ret = 0;
+
+	if (!func_prolog_preempt_disable(tr, &data, &pc))
+		return 0;
+
+	local_save_flags(flags);
+	ret = __trace_graph_entry(tr, trace, flags, pc);
+	atomic_dec(&data->disabled);
+	preempt_enable_notrace();
+
+	return ret;
+}
+
+static void wakeup_graph_return(struct ftrace_graph_ret *trace)
+{
+	struct trace_array *tr = wakeup_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+	int pc;
+
+	if (!func_prolog_preempt_disable(tr, &data, &pc))
+		return;
+
+	local_save_flags(flags);
+	__trace_graph_return(tr, trace, flags, pc);
+	atomic_dec(&data->disabled);
+
+	preempt_enable_notrace();
+	return;
+}
+
+static void wakeup_trace_open(struct trace_iterator *iter)
+{
+	if (is_graph())
+		graph_trace_open(iter);
+}
+
+static void wakeup_trace_close(struct trace_iterator *iter)
+{
+	if (iter->private)
+		graph_trace_close(iter);
+}
+
+#define GRAPH_TRACER_FLAGS (TRACE_GRAPH_PRINT_PROC)
+
+static enum print_line_t wakeup_print_line(struct trace_iterator *iter)
+{
+	/*
+	 * In graph mode call the graph tracer output function,
+	 * otherwise go with the TRACE_FN event handler
+	 */
+	if (is_graph())
+		return print_graph_function_flags(iter, GRAPH_TRACER_FLAGS);
+
+	return TRACE_TYPE_UNHANDLED;
+}
+
+static void wakeup_print_header(struct seq_file *s)
+{
+	if (is_graph())
+		print_graph_headers_flags(s, GRAPH_TRACER_FLAGS);
+	else
+		trace_default_header(s);
+}
+
+static void
+__trace_function(struct trace_array *tr,
+		 unsigned long ip, unsigned long parent_ip,
+		 unsigned long flags, int pc)
+{
+	if (is_graph())
+		trace_graph_function(tr, ip, parent_ip, flags, pc);
+	else
+		trace_function(tr, ip, parent_ip, flags, pc);
+}
+#else
+#define __trace_function trace_function
+
+static int wakeup_set_flag(u32 old_flags, u32 bit, int set)
+{
+	return -EINVAL;
+}
+
+static int wakeup_graph_entry(struct ftrace_graph_ent *trace)
+{
+	return -1;
+}
+
+static enum print_line_t wakeup_print_line(struct trace_iterator *iter)
+{
+	return TRACE_TYPE_UNHANDLED;
+}
+
+static void wakeup_graph_return(struct ftrace_graph_ret *trace) { }
+static void wakeup_print_header(struct seq_file *s) { }
+static void wakeup_trace_open(struct trace_iterator *iter) { }
+static void wakeup_trace_close(struct trace_iterator *iter) { }
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
 /*
  * Should this new latency be reported/recorded?
  */
@@ -152,7 +352,7 @@
 	/* The task we are waiting for is waking up */
 	data = wakeup_trace->data[wakeup_cpu];
 
-	trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc);
+	__trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc);
 	tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc);
 
 	T0 = data->preempt_timestamp;
@@ -252,7 +452,7 @@
 	 * is not called by an assembly function  (where as schedule is)
 	 * it should be safe to use it here.
 	 */
-	trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc);
+	__trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc);
 
 out_locked:
 	arch_spin_unlock(&wakeup_lock);
@@ -303,12 +503,8 @@
 	 */
 	smp_wmb();
 
-	register_ftrace_function(&trace_ops);
-
-	if (tracing_is_enabled())
-		tracer_enabled = 1;
-	else
-		tracer_enabled = 0;
+	if (start_func_tracer(is_graph()))
+		printk(KERN_ERR "failed to start wakeup tracer\n");
 
 	return;
 fail_deprobe_wake_new:
@@ -320,7 +516,7 @@
 static void stop_wakeup_tracer(struct trace_array *tr)
 {
 	tracer_enabled = 0;
-	unregister_ftrace_function(&trace_ops);
+	stop_func_tracer(is_graph());
 	unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL);
 	unregister_trace_sched_wakeup_new(probe_wakeup, NULL);
 	unregister_trace_sched_wakeup(probe_wakeup, NULL);
@@ -379,9 +575,15 @@
 	.start		= wakeup_tracer_start,
 	.stop		= wakeup_tracer_stop,
 	.print_max	= 1,
+	.print_header	= wakeup_print_header,
+	.print_line	= wakeup_print_line,
+	.flags		= &tracer_flags,
+	.set_flag	= wakeup_set_flag,
 #ifdef CONFIG_FTRACE_SELFTEST
 	.selftest    = trace_selftest_startup_wakeup,
 #endif
+	.open		= wakeup_trace_open,
+	.close		= wakeup_trace_close,
 	.use_max_tr	= 1,
 };
 
@@ -394,9 +596,15 @@
 	.stop		= wakeup_tracer_stop,
 	.wait_pipe	= poll_wait_pipe,
 	.print_max	= 1,
+	.print_header	= wakeup_print_header,
+	.print_line	= wakeup_print_line,
+	.flags		= &tracer_flags,
+	.set_flag	= wakeup_set_flag,
 #ifdef CONFIG_FTRACE_SELFTEST
 	.selftest    = trace_selftest_startup_wakeup,
 #endif
+	.open		= wakeup_trace_open,
+	.close		= wakeup_trace_close,
 	.use_max_tr	= 1,
 };
 
diff --git a/kernel/trace/trace_workqueue.c b/kernel/trace/trace_workqueue.c
index a7cc379..209b379 100644
--- a/kernel/trace/trace_workqueue.c
+++ b/kernel/trace/trace_workqueue.c
@@ -263,6 +263,11 @@
 {
 	int ret, cpu;
 
+	for_each_possible_cpu(cpu) {
+		spin_lock_init(&workqueue_cpu_stat(cpu)->lock);
+		INIT_LIST_HEAD(&workqueue_cpu_stat(cpu)->list);
+	}
+
 	ret = register_trace_workqueue_insertion(probe_workqueue_insertion, NULL);
 	if (ret)
 		goto out;
@@ -279,11 +284,6 @@
 	if (ret)
 		goto no_creation;
 
-	for_each_possible_cpu(cpu) {
-		spin_lock_init(&workqueue_cpu_stat(cpu)->lock);
-		INIT_LIST_HEAD(&workqueue_cpu_stat(cpu)->list);
-	}
-
 	return 0;
 
 no_creation:
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index c77f3ec..e95ee7f 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -25,6 +25,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/jump_label.h>
 
 extern struct tracepoint __start___tracepoints[];
 extern struct tracepoint __stop___tracepoints[];
@@ -263,7 +264,13 @@
 	 * is used.
 	 */
 	rcu_assign_pointer(elem->funcs, (*entry)->funcs);
-	elem->state = active;
+	if (!elem->state && active) {
+		jump_label_enable(&elem->state);
+		elem->state = active;
+	} else if (elem->state && !active) {
+		jump_label_disable(&elem->state);
+		elem->state = active;
+	}
 }
 
 /*
@@ -277,7 +284,10 @@
 	if (elem->unregfunc && elem->state)
 		elem->unregfunc();
 
-	elem->state = 0;
+	if (elem->state) {
+		jump_label_disable(&elem->state);
+		elem->state = 0;
+	}
 	rcu_assign_pointer(elem->funcs, NULL);
 }
 
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index 7f9c3c5..bafba68 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -43,7 +43,6 @@
 static DEFINE_PER_CPU(struct perf_event *, watchdog_ev);
 #endif
 
-static int __read_mostly did_panic;
 static int __initdata no_watchdog;
 
 
@@ -187,18 +186,6 @@
 	return 0;
 }
 
-static int
-watchdog_panic(struct notifier_block *this, unsigned long event, void *ptr)
-{
-	did_panic = 1;
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block panic_block = {
-	.notifier_call = watchdog_panic,
-};
-
 #ifdef CONFIG_HARDLOCKUP_DETECTOR
 static struct perf_event_attr wd_hw_attr = {
 	.type		= PERF_TYPE_HARDWARE,
@@ -209,7 +196,7 @@
 };
 
 /* Callback function for perf event subsystem */
-void watchdog_overflow_callback(struct perf_event *event, int nmi,
+static void watchdog_overflow_callback(struct perf_event *event, int nmi,
 		 struct perf_sample_data *data,
 		 struct pt_regs *regs)
 {
@@ -371,14 +358,14 @@
 	/* Try to register using hardware perf events */
 	wd_attr = &wd_hw_attr;
 	wd_attr->sample_period = hw_nmi_get_sample_period();
-	event = perf_event_create_kernel_counter(wd_attr, cpu, -1, watchdog_overflow_callback);
+	event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback);
 	if (!IS_ERR(event)) {
 		printk(KERN_INFO "NMI watchdog enabled, takes one hw-pmu counter.\n");
 		goto out_save;
 	}
 
 	printk(KERN_ERR "NMI watchdog failed to create perf event on cpu%i: %p\n", cpu, event);
-	return -1;
+	return PTR_ERR(event);
 
 	/* success path */
 out_save:
@@ -422,17 +409,19 @@
 static int watchdog_enable(int cpu)
 {
 	struct task_struct *p = per_cpu(softlockup_watchdog, cpu);
+	int err;
 
 	/* enable the perf event */
-	if (watchdog_nmi_enable(cpu) != 0)
-		return -1;
+	err = watchdog_nmi_enable(cpu);
+	if (err)
+		return err;
 
 	/* create the watchdog thread */
 	if (!p) {
 		p = kthread_create(watchdog, (void *)(unsigned long)cpu, "watchdog/%d", cpu);
 		if (IS_ERR(p)) {
 			printk(KERN_ERR "softlockup watchdog for %i failed\n", cpu);
-			return -1;
+			return PTR_ERR(p);
 		}
 		kthread_bind(p, cpu);
 		per_cpu(watchdog_touch_ts, cpu) = 0;
@@ -484,6 +473,9 @@
 {
 	int cpu;
 
+	if (no_watchdog)
+		return;
+
 	for_each_online_cpu(cpu)
 		watchdog_disable(cpu);
 
@@ -526,17 +518,16 @@
 cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 {
 	int hotcpu = (unsigned long)hcpu;
+	int err = 0;
 
 	switch (action) {
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
-		if (watchdog_prepare_cpu(hotcpu))
-			return NOTIFY_BAD;
+		err = watchdog_prepare_cpu(hotcpu);
 		break;
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
-		if (watchdog_enable(hotcpu))
-			return NOTIFY_BAD;
+		err = watchdog_enable(hotcpu);
 		break;
 #ifdef CONFIG_HOTPLUG_CPU
 	case CPU_UP_CANCELED:
@@ -549,7 +540,7 @@
 		break;
 #endif /* CONFIG_HOTPLUG_CPU */
 	}
-	return NOTIFY_OK;
+	return notifier_from_errno(err);
 }
 
 static struct notifier_block __cpuinitdata cpu_nfb = {
@@ -565,13 +556,11 @@
 		return 0;
 
 	err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);
-	WARN_ON(err == NOTIFY_BAD);
+	WARN_ON(notifier_to_errno(err));
 
 	cpu_callback(&cpu_nfb, CPU_ONLINE, cpu);
 	register_cpu_notifier(&cpu_nfb);
 
-	atomic_notifier_chain_register(&panic_notifier_list, &panic_block);
-
 	return 0;
 }
 early_initcall(spawn_watchdog_task);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 1b4afd2..21ac830 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -482,6 +482,7 @@
 	select DEBUG_SPINLOCK
 	select DEBUG_MUTEXES
 	select DEBUG_LOCK_ALLOC
+	select TRACE_IRQFLAGS
 	default n
 	help
 	 This feature enables the kernel to prove that all locking
@@ -539,6 +540,23 @@
 	 disabling, allowing multiple RCU-lockdep warnings to be printed
 	 on a single reboot.
 
+	 Say Y to allow multiple RCU-lockdep warnings per boot.
+
+	 Say N if you are unsure.
+
+config SPARSE_RCU_POINTER
+	bool "RCU debugging: sparse-based checks for pointer usage"
+	default n
+	help
+	 This feature enables the __rcu sparse annotation for
+	 RCU-protected pointers.  This annotation will cause sparse
+	 to flag any non-RCU used of annotated pointers.  This can be
+	 helpful when debugging RCU usage.  Please note that this feature
+	 is not intended to enforce code cleanliness; it is instead merely
+	 a debugging aid.
+
+	 Say Y to make sparse flag questionable use of RCU-protected pointers
+
 	 Say N if you are unsure.
 
 config LOCKDEP
@@ -579,11 +597,10 @@
 	  of more runtime overhead.
 
 config TRACE_IRQFLAGS
-	depends on DEBUG_KERNEL
 	bool
-	default y
-	depends on TRACE_IRQFLAGS_SUPPORT
-	depends on PROVE_LOCKING
+	help
+	  Enables hooks to interrupt enabling and disabling for
+	  either tracing or lock debugging.
 
 config DEBUG_SPINLOCK_SLEEP
 	bool "Spinlock debugging: sleep-inside-spinlock checking"
@@ -832,6 +849,30 @@
 
 	  Say Y if you are unsure.
 
+config RCU_CPU_STALL_TIMEOUT
+	int "RCU CPU stall timeout in seconds"
+	depends on RCU_CPU_STALL_DETECTOR
+	range 3 300
+	default 60
+	help
+	  If a given RCU grace period extends more than the specified
+	  number of seconds, a CPU stall warning is printed.  If the
+	  RCU grace period persists, additional CPU stall warnings are
+	  printed at more widely spaced intervals.
+
+config RCU_CPU_STALL_DETECTOR_RUNNABLE
+	bool "RCU CPU stall checking starts automatically at boot"
+	depends on RCU_CPU_STALL_DETECTOR
+	default y
+	help
+	  If set, start checking for RCU CPU stalls immediately on
+	  boot.  Otherwise, RCU CPU stall checking must be manually
+	  enabled.
+
+	  Say Y if you are unsure.
+
+	  Say N if you wish to suppress RCU CPU stall checking during boot.
+
 config RCU_CPU_STALL_VERBOSE
 	bool "Print additional per-task information for RCU_CPU_STALL_DETECTOR"
 	depends on RCU_CPU_STALL_DETECTOR && TREE_PREEMPT_RCU
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index 02afc25..7bd6df7 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
@@ -26,19 +26,11 @@
 #include <linux/dynamic_debug.h>
 #include <linux/debugfs.h>
 #include <linux/slab.h>
+#include <linux/jump_label.h>
 
 extern struct _ddebug __start___verbose[];
 extern struct _ddebug __stop___verbose[];
 
-/* dynamic_debug_enabled, and dynamic_debug_enabled2 are bitmasks in which
- * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They
- * use independent hash functions, to reduce the chance of false positives.
- */
-long long dynamic_debug_enabled;
-EXPORT_SYMBOL_GPL(dynamic_debug_enabled);
-long long dynamic_debug_enabled2;
-EXPORT_SYMBOL_GPL(dynamic_debug_enabled2);
-
 struct ddebug_table {
 	struct list_head link;
 	char *mod_name;
@@ -88,26 +80,6 @@
 }
 
 /*
- * must be called with ddebug_lock held
- */
-
-static int disabled_hash(char hash, bool first_table)
-{
-	struct ddebug_table *dt;
-	char table_hash_value;
-
-	list_for_each_entry(dt, &ddebug_tables, link) {
-		if (first_table)
-			table_hash_value = dt->ddebugs->primary_hash;
-		else
-			table_hash_value = dt->ddebugs->secondary_hash;
-		if (dt->num_enabled && (hash == table_hash_value))
-			return 0;
-	}
-	return 1;
-}
-
-/*
  * Search the tables for _ddebug's which match the given
  * `query' and apply the `flags' and `mask' to them.  Tells
  * the user which ddebug's were changed, or whether none
@@ -170,17 +142,9 @@
 				dt->num_enabled++;
 			dp->flags = newflags;
 			if (newflags) {
-				dynamic_debug_enabled |=
-						(1LL << dp->primary_hash);
-				dynamic_debug_enabled2 |=
-						(1LL << dp->secondary_hash);
+				jump_label_enable(&dp->enabled);
 			} else {
-				if (disabled_hash(dp->primary_hash, true))
-					dynamic_debug_enabled &=
-						~(1LL << dp->primary_hash);
-				if (disabled_hash(dp->secondary_hash, false))
-					dynamic_debug_enabled2 &=
-						~(1LL << dp->secondary_hash);
+				jump_label_disable(&dp->enabled);
 			}
 			if (verbose)
 				printk(KERN_INFO
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index efd16fa..6f412ab4 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -49,7 +49,7 @@
 	unsigned int	height;		/* Height from the bottom */
 	unsigned int	count;
 	struct rcu_head	rcu_head;
-	void		*slots[RADIX_TREE_MAP_SIZE];
+	void __rcu	*slots[RADIX_TREE_MAP_SIZE];
 	unsigned long	tags[RADIX_TREE_MAX_TAGS][RADIX_TREE_TAG_LONGS];
 };
 
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 34e3082..7c06ee5 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -70,7 +70,7 @@
  */
 static unsigned long io_tlb_overflow = 32*1024;
 
-void *io_tlb_overflow_buffer;
+static void *io_tlb_overflow_buffer;
 
 /*
  * This is a free list describing the number of free entries available from
@@ -147,16 +147,16 @@
 	 * to find contiguous free memory regions of size up to IO_TLB_SEGSIZE
 	 * between io_tlb_start and io_tlb_end.
 	 */
-	io_tlb_list = alloc_bootmem(io_tlb_nslabs * sizeof(int));
+	io_tlb_list = alloc_bootmem_pages(PAGE_ALIGN(io_tlb_nslabs * sizeof(int)));
 	for (i = 0; i < io_tlb_nslabs; i++)
  		io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE);
 	io_tlb_index = 0;
-	io_tlb_orig_addr = alloc_bootmem(io_tlb_nslabs * sizeof(phys_addr_t));
+	io_tlb_orig_addr = alloc_bootmem_pages(PAGE_ALIGN(io_tlb_nslabs * sizeof(phys_addr_t)));
 
 	/*
 	 * Get the overflow emergency buffer
 	 */
-	io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow);
+	io_tlb_overflow_buffer = alloc_bootmem_low_pages(PAGE_ALIGN(io_tlb_overflow));
 	if (!io_tlb_overflow_buffer)
 		panic("Cannot allocate SWIOTLB overflow buffer!\n");
 	if (verbose)
@@ -182,7 +182,7 @@
 	/*
 	 * Get IO TLB memory from the low pages
 	 */
-	io_tlb_start = alloc_bootmem_low_pages(bytes);
+	io_tlb_start = alloc_bootmem_low_pages(PAGE_ALIGN(bytes));
 	if (!io_tlb_start)
 		panic("Cannot allocate SWIOTLB buffer");
 
@@ -308,13 +308,13 @@
 			   get_order(io_tlb_nslabs << IO_TLB_SHIFT));
 	} else {
 		free_bootmem_late(__pa(io_tlb_overflow_buffer),
-				  io_tlb_overflow);
+				  PAGE_ALIGN(io_tlb_overflow));
 		free_bootmem_late(__pa(io_tlb_orig_addr),
-				  io_tlb_nslabs * sizeof(phys_addr_t));
+				  PAGE_ALIGN(io_tlb_nslabs * sizeof(phys_addr_t)));
 		free_bootmem_late(__pa(io_tlb_list),
-				  io_tlb_nslabs * sizeof(int));
+				  PAGE_ALIGN(io_tlb_nslabs * sizeof(int)));
 		free_bootmem_late(__pa(io_tlb_start),
-				  io_tlb_nslabs << IO_TLB_SHIFT);
+				  PAGE_ALIGN(io_tlb_nslabs << IO_TLB_SHIFT));
 	}
 }
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 3eed583..9be3cf8 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3587,9 +3587,13 @@
 
 static void mem_cgroup_threshold(struct mem_cgroup *memcg)
 {
-	__mem_cgroup_threshold(memcg, false);
-	if (do_swap_account)
-		__mem_cgroup_threshold(memcg, true);
+	while (memcg) {
+		__mem_cgroup_threshold(memcg, false);
+		if (do_swap_account)
+			__mem_cgroup_threshold(memcg, true);
+
+		memcg = parent_mem_cgroup(memcg);
+	}
 }
 
 static int compare_thresholds(const void *a, const void *b)
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 9c26eec..757f6b0 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -183,7 +183,7 @@
  * signal.
  */
 static int kill_proc_ao(struct task_struct *t, unsigned long addr, int trapno,
-			unsigned long pfn)
+			unsigned long pfn, struct page *page)
 {
 	struct siginfo si;
 	int ret;
@@ -198,7 +198,7 @@
 #ifdef __ARCH_SI_TRAPNO
 	si.si_trapno = trapno;
 #endif
-	si.si_addr_lsb = PAGE_SHIFT;
+	si.si_addr_lsb = compound_order(compound_head(page)) + PAGE_SHIFT;
 	/*
 	 * Don't use force here, it's convenient if the signal
 	 * can be temporarily blocked.
@@ -235,7 +235,7 @@
 		int nr;
 		do {
 			nr = shrink_slab(1000, GFP_KERNEL, 1000);
-			if (page_count(p) == 0)
+			if (page_count(p) == 1)
 				break;
 		} while (nr > 10);
 	}
@@ -327,7 +327,7 @@
  * wrong earlier.
  */
 static void kill_procs_ao(struct list_head *to_kill, int doit, int trapno,
-			  int fail, unsigned long pfn)
+			  int fail, struct page *page, unsigned long pfn)
 {
 	struct to_kill *tk, *next;
 
@@ -352,7 +352,7 @@
 			 * process anyways.
 			 */
 			else if (kill_proc_ao(tk->tsk, tk->addr, trapno,
-					      pfn) < 0)
+					      pfn, page) < 0)
 				printk(KERN_ERR
 		"MCE %#lx: Cannot send advisory machine check signal to %s:%d\n",
 					pfn, tk->tsk->comm, tk->tsk->pid);
@@ -928,7 +928,7 @@
 	 * any accesses to the poisoned memory.
 	 */
 	kill_procs_ao(&tokill, !!PageDirty(hpage), trapno,
-		      ret != SWAP_SUCCESS, pfn);
+		      ret != SWAP_SUCCESS, p, pfn);
 
 	return ret;
 }
diff --git a/mm/memory.c b/mm/memory.c
index 0e18b4d..98b58fe 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3185,7 +3185,7 @@
 		 * with threads.
 		 */
 		if (flags & FAULT_FLAG_WRITE)
-			flush_tlb_page(vma, address);
+			flush_tlb_fix_spurious_fault(vma, address);
 	}
 unlock:
 	pte_unmap_unlock(pte, ptl);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a8cfa9c..f12ad18 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5182,9 +5182,9 @@
 	if (!table)
 		panic("Failed to allocate %s hash table\n", tablename);
 
-	printk(KERN_INFO "%s hash table entries: %d (order: %d, %lu bytes)\n",
+	printk(KERN_INFO "%s hash table entries: %ld (order: %d, %lu bytes)\n",
 	       tablename,
-	       (1U << log2qty),
+	       (1UL << log2qty),
 	       ilog2(size) - PAGE_SHIFT,
 	       size);
 
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 6b8889d..d8087f0 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -517,6 +517,15 @@
 static void purge_fragmented_blocks_allcpus(void);
 
 /*
+ * called before a call to iounmap() if the caller wants vm_area_struct's
+ * immediately freed.
+ */
+void set_iounmap_nonlazy(void)
+{
+	atomic_set(&vmap_lazy_nr, lazy_max_pages()+1);
+}
+
+/*
  * Purges all lazily-freed vmap areas.
  *
  * If sync is 0 then don't purge if there is already a purge in progress.
diff --git a/net/Kconfig b/net/Kconfig
index e926884..55fd82e 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -293,6 +293,7 @@
 source "net/rfkill/Kconfig"
 source "net/9p/Kconfig"
 source "net/caif/Kconfig"
+source "net/ceph/Kconfig"
 
 
 endif   # if NET
diff --git a/net/Makefile b/net/Makefile
index ea60fbc..6b7bfd7 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -68,3 +68,4 @@
 endif
 obj-$(CONFIG_WIMAX)		+= wimax/
 obj-$(CONFIG_DNS_RESOLVER)	+= dns_resolver/
+obj-$(CONFIG_CEPH_LIB)		+= ceph/
diff --git a/net/atm/mpc.c b/net/atm/mpc.c
index 622b471..74bcc66 100644
--- a/net/atm/mpc.c
+++ b/net/atm/mpc.c
@@ -778,7 +778,7 @@
 	eg->packets_rcvd++;
 	mpc->eg_ops->put(eg);
 
-	memset(ATM_SKB(skb), 0, sizeof(struct atm_skb_data));
+	memset(ATM_SKB(new_skb), 0, sizeof(struct atm_skb_data));
 	netif_rx(new_skb);
 }
 
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index fadf26b..0b54b7d 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -1441,33 +1441,23 @@
 
 static void l2cap_streaming_send(struct sock *sk)
 {
-	struct sk_buff *skb, *tx_skb;
+	struct sk_buff *skb;
 	struct l2cap_pinfo *pi = l2cap_pi(sk);
 	u16 control, fcs;
 
-	while ((skb = sk->sk_send_head)) {
-		tx_skb = skb_clone(skb, GFP_ATOMIC);
-
-		control = get_unaligned_le16(tx_skb->data + L2CAP_HDR_SIZE);
+	while ((skb = skb_dequeue(TX_QUEUE(sk)))) {
+		control = get_unaligned_le16(skb->data + L2CAP_HDR_SIZE);
 		control |= pi->next_tx_seq << L2CAP_CTRL_TXSEQ_SHIFT;
-		put_unaligned_le16(control, tx_skb->data + L2CAP_HDR_SIZE);
+		put_unaligned_le16(control, skb->data + L2CAP_HDR_SIZE);
 
 		if (pi->fcs == L2CAP_FCS_CRC16) {
-			fcs = crc16(0, (u8 *)tx_skb->data, tx_skb->len - 2);
-			put_unaligned_le16(fcs, tx_skb->data + tx_skb->len - 2);
+			fcs = crc16(0, (u8 *)skb->data, skb->len - 2);
+			put_unaligned_le16(fcs, skb->data + skb->len - 2);
 		}
 
-		l2cap_do_send(sk, tx_skb);
+		l2cap_do_send(sk, skb);
 
 		pi->next_tx_seq = (pi->next_tx_seq + 1) % 64;
-
-		if (skb_queue_is_last(TX_QUEUE(sk), skb))
-			sk->sk_send_head = NULL;
-		else
-			sk->sk_send_head = skb_queue_next(TX_QUEUE(sk), skb);
-
-		skb = skb_dequeue(TX_QUEUE(sk));
-		kfree_skb(skb);
 	}
 }
 
@@ -1960,6 +1950,11 @@
 
 	switch (optname) {
 	case L2CAP_OPTIONS:
+		if (sk->sk_state == BT_CONNECTED) {
+			err = -EINVAL;
+			break;
+		}
+
 		opts.imtu     = l2cap_pi(sk)->imtu;
 		opts.omtu     = l2cap_pi(sk)->omtu;
 		opts.flush_to = l2cap_pi(sk)->flush_to;
@@ -2771,10 +2766,10 @@
 		case L2CAP_CONF_MTU:
 			if (val < L2CAP_DEFAULT_MIN_MTU) {
 				*result = L2CAP_CONF_UNACCEPT;
-				pi->omtu = L2CAP_DEFAULT_MIN_MTU;
+				pi->imtu = L2CAP_DEFAULT_MIN_MTU;
 			} else
-				pi->omtu = val;
-			l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, pi->omtu);
+				pi->imtu = val;
+			l2cap_add_conf_opt(&ptr, L2CAP_CONF_MTU, 2, pi->imtu);
 			break;
 
 		case L2CAP_CONF_FLUSH_TO:
@@ -3071,6 +3066,17 @@
 	return 0;
 }
 
+static inline void set_default_fcs(struct l2cap_pinfo *pi)
+{
+	/* FCS is enabled only in ERTM or streaming mode, if one or both
+	 * sides request it.
+	 */
+	if (pi->mode != L2CAP_MODE_ERTM && pi->mode != L2CAP_MODE_STREAMING)
+		pi->fcs = L2CAP_FCS_NONE;
+	else if (!(pi->conf_state & L2CAP_CONF_NO_FCS_RECV))
+		pi->fcs = L2CAP_FCS_CRC16;
+}
+
 static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd, u16 cmd_len, u8 *data)
 {
 	struct l2cap_conf_req *req = (struct l2cap_conf_req *) data;
@@ -3088,14 +3094,8 @@
 	if (!sk)
 		return -ENOENT;
 
-	if (sk->sk_state != BT_CONFIG) {
-		struct l2cap_cmd_rej rej;
-
-		rej.reason = cpu_to_le16(0x0002);
-		l2cap_send_cmd(conn, cmd->ident, L2CAP_COMMAND_REJ,
-				sizeof(rej), &rej);
+	if (sk->sk_state == BT_DISCONN)
 		goto unlock;
-	}
 
 	/* Reject if config buffer is too small. */
 	len = cmd_len - sizeof(*req);
@@ -3135,9 +3135,7 @@
 		goto unlock;
 
 	if (l2cap_pi(sk)->conf_state & L2CAP_CONF_INPUT_DONE) {
-		if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_NO_FCS_RECV) ||
-		    l2cap_pi(sk)->fcs != L2CAP_FCS_NONE)
-			l2cap_pi(sk)->fcs = L2CAP_FCS_CRC16;
+		set_default_fcs(l2cap_pi(sk));
 
 		sk->sk_state = BT_CONNECTED;
 
@@ -3225,9 +3223,7 @@
 	l2cap_pi(sk)->conf_state |= L2CAP_CONF_INPUT_DONE;
 
 	if (l2cap_pi(sk)->conf_state & L2CAP_CONF_OUTPUT_DONE) {
-		if (!(l2cap_pi(sk)->conf_state & L2CAP_CONF_NO_FCS_RECV) ||
-		    l2cap_pi(sk)->fcs != L2CAP_FCS_NONE)
-			l2cap_pi(sk)->fcs = L2CAP_FCS_CRC16;
+		set_default_fcs(l2cap_pi(sk));
 
 		sk->sk_state = BT_CONNECTED;
 		l2cap_pi(sk)->next_tx_seq = 0;
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 44a6232..194b3a0 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -82,11 +82,14 @@
 static void rfcomm_sk_state_change(struct rfcomm_dlc *d, int err)
 {
 	struct sock *sk = d->owner, *parent;
+	unsigned long flags;
+
 	if (!sk)
 		return;
 
 	BT_DBG("dlc %p state %ld err %d", d, d->state, err);
 
+	local_irq_save(flags);
 	bh_lock_sock(sk);
 
 	if (err)
@@ -108,6 +111,7 @@
 	}
 
 	bh_unlock_sock(sk);
+	local_irq_restore(flags);
 
 	if (parent && sock_flag(sk, SOCK_ZAPPED)) {
 		/* We have to drop DLC lock here, otherwise
diff --git a/net/caif/caif_socket.c b/net/caif/caif_socket.c
index 8ce9047..4bf28f2 100644
--- a/net/caif/caif_socket.c
+++ b/net/caif/caif_socket.c
@@ -827,6 +827,7 @@
 	long timeo;
 	int err;
 	int ifindex, headroom, tailroom;
+	unsigned int mtu;
 	struct net_device *dev;
 
 	lock_sock(sk);
@@ -896,15 +897,23 @@
 		cf_sk->sk.sk_state = CAIF_DISCONNECTED;
 		goto out;
 	}
-	dev = dev_get_by_index(sock_net(sk), ifindex);
+
+	err = -ENODEV;
+	rcu_read_lock();
+	dev = dev_get_by_index_rcu(sock_net(sk), ifindex);
+	if (!dev) {
+		rcu_read_unlock();
+		goto out;
+	}
 	cf_sk->headroom = LL_RESERVED_SPACE_EXTRA(dev, headroom);
+	mtu = dev->mtu;
+	rcu_read_unlock();
+
 	cf_sk->tailroom = tailroom;
-	cf_sk->maxframe = dev->mtu - (headroom + tailroom);
-	dev_put(dev);
+	cf_sk->maxframe = mtu - (headroom + tailroom);
 	if (cf_sk->maxframe < 1) {
-		pr_warning("CAIF: %s(): CAIF Interface MTU too small (%d)\n",
-			__func__, dev->mtu);
-		err = -ENODEV;
+		pr_warning("CAIF: %s(): CAIF Interface MTU too small (%u)\n",
+			   __func__, mtu);
 		goto out;
 	}
 
diff --git a/net/ceph/Kconfig b/net/ceph/Kconfig
new file mode 100644
index 0000000..ad42404
--- /dev/null
+++ b/net/ceph/Kconfig
@@ -0,0 +1,28 @@
+config CEPH_LIB
+        tristate "Ceph core library (EXPERIMENTAL)"
+	depends on INET && EXPERIMENTAL
+	select LIBCRC32C
+	select CRYPTO_AES
+	select CRYPTO
+	default n
+	help
+	  Choose Y or M here to include cephlib, which provides the
+	  common functionality to both the Ceph filesystem and
+	  to the rados block device (rbd).
+
+	  More information at http://ceph.newdream.net/.
+
+	  If unsure, say N.
+
+config CEPH_LIB_PRETTYDEBUG
+	bool "Include file:line in ceph debug output"
+	depends on CEPH_LIB
+	default n
+	help
+	  If you say Y here, debug output will include a filename and
+	  line to aid debugging.  This increases kernel size and slows
+	  execution slightly when debug call sites are enabled (e.g.,
+	  via CONFIG_DYNAMIC_DEBUG).
+
+	  If unsure, say N.
+
diff --git a/net/ceph/Makefile b/net/ceph/Makefile
new file mode 100644
index 0000000..aab1cab
--- /dev/null
+++ b/net/ceph/Makefile
@@ -0,0 +1,37 @@
+#
+# Makefile for CEPH filesystem.
+#
+
+ifneq ($(KERNELRELEASE),)
+
+obj-$(CONFIG_CEPH_LIB) += libceph.o
+
+libceph-objs := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \
+	mon_client.o \
+	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
+	debugfs.o \
+	auth.o auth_none.o \
+	crypto.o armor.o \
+	auth_x.o \
+	ceph_fs.o ceph_strings.o ceph_hash.o \
+	pagevec.o
+
+else
+#Otherwise we were called directly from the command
+# line; invoke the kernel build system.
+
+KERNELDIR ?= /lib/modules/$(shell uname -r)/build
+PWD := $(shell pwd)
+
+default: all
+
+all:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules
+
+modules_install:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules_install
+
+clean:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) clean
+
+endif
diff --git a/fs/ceph/armor.c b/net/ceph/armor.c
similarity index 100%
rename from fs/ceph/armor.c
rename to net/ceph/armor.c
diff --git a/fs/ceph/auth.c b/net/ceph/auth.c
similarity index 96%
rename from fs/ceph/auth.c
rename to net/ceph/auth.c
index 6d2e306..549c1f4 100644
--- a/fs/ceph/auth.c
+++ b/net/ceph/auth.c
@@ -1,16 +1,16 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/module.h>
 #include <linux/err.h>
 #include <linux/slab.h>
 
-#include "types.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/messenger.h>
 #include "auth_none.h"
 #include "auth_x.h"
-#include "decode.h"
-#include "super.h"
 
-#include "messenger.h"
 
 /*
  * get protocol handler
diff --git a/fs/ceph/auth_none.c b/net/ceph/auth_none.c
similarity index 96%
rename from fs/ceph/auth_none.c
rename to net/ceph/auth_none.c
index ad1dc21..214c2bb 100644
--- a/fs/ceph/auth_none.c
+++ b/net/ceph/auth_none.c
@@ -1,14 +1,15 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/slab.h>
 
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
 #include "auth_none.h"
-#include "auth.h"
-#include "decode.h"
 
 static void reset(struct ceph_auth_client *ac)
 {
diff --git a/fs/ceph/auth_none.h b/net/ceph/auth_none.h
similarity index 94%
rename from fs/ceph/auth_none.h
rename to net/ceph/auth_none.h
index 8164df1..ed7d088 100644
--- a/fs/ceph/auth_none.h
+++ b/net/ceph/auth_none.h
@@ -2,8 +2,7 @@
 #define _FS_CEPH_AUTH_NONE_H
 
 #include <linux/slab.h>
-
-#include "auth.h"
+#include <linux/ceph/auth.h>
 
 /*
  * null security mode.
diff --git a/fs/ceph/auth_x.c b/net/ceph/auth_x.c
similarity index 99%
rename from fs/ceph/auth_x.c
rename to net/ceph/auth_x.c
index a2d002c..7fd5dfc 100644
--- a/fs/ceph/auth_x.c
+++ b/net/ceph/auth_x.c
@@ -1,16 +1,17 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/slab.h>
 
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
+#include "crypto.h"
 #include "auth_x.h"
 #include "auth_x_protocol.h"
-#include "crypto.h"
-#include "auth.h"
-#include "decode.h"
 
 #define TEMP_TICKET_BUF_LEN	256
 
diff --git a/fs/ceph/auth_x.h b/net/ceph/auth_x.h
similarity index 96%
rename from fs/ceph/auth_x.h
rename to net/ceph/auth_x.h
index ff6f818..e02da7a 100644
--- a/fs/ceph/auth_x.h
+++ b/net/ceph/auth_x.h
@@ -3,8 +3,9 @@
 
 #include <linux/rbtree.h>
 
+#include <linux/ceph/auth.h>
+
 #include "crypto.h"
-#include "auth.h"
 #include "auth_x_protocol.h"
 
 /*
diff --git a/fs/ceph/auth_x_protocol.h b/net/ceph/auth_x_protocol.h
similarity index 100%
rename from fs/ceph/auth_x_protocol.h
rename to net/ceph/auth_x_protocol.h
diff --git a/fs/ceph/buffer.c b/net/ceph/buffer.c
similarity index 85%
rename from fs/ceph/buffer.c
rename to net/ceph/buffer.c
index cd39f17..53d8abf 100644
--- a/fs/ceph/buffer.c
+++ b/net/ceph/buffer.c
@@ -1,10 +1,11 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/slab.h>
 
-#include "buffer.h"
-#include "decode.h"
+#include <linux/ceph/buffer.h>
+#include <linux/ceph/decode.h>
 
 struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp)
 {
@@ -32,6 +33,7 @@
 	dout("buffer_new %p\n", b);
 	return b;
 }
+EXPORT_SYMBOL(ceph_buffer_new);
 
 void ceph_buffer_release(struct kref *kref)
 {
@@ -46,6 +48,7 @@
 	}
 	kfree(b);
 }
+EXPORT_SYMBOL(ceph_buffer_release);
 
 int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end)
 {
diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c
new file mode 100644
index 0000000..f3e4a13
--- /dev/null
+++ b/net/ceph/ceph_common.c
@@ -0,0 +1,529 @@
+
+#include <linux/ceph/ceph_debug.h>
+#include <linux/backing-dev.h>
+#include <linux/ctype.h>
+#include <linux/fs.h>
+#include <linux/inet.h>
+#include <linux/in6.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/parser.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/statfs.h>
+#include <linux/string.h>
+
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/debugfs.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+
+
+
+/*
+ * find filename portion of a path (/foo/bar/baz -> baz)
+ */
+const char *ceph_file_part(const char *s, int len)
+{
+	const char *e = s + len;
+
+	while (e != s && *(e-1) != '/')
+		e--;
+	return e;
+}
+EXPORT_SYMBOL(ceph_file_part);
+
+const char *ceph_msg_type_name(int type)
+{
+	switch (type) {
+	case CEPH_MSG_SHUTDOWN: return "shutdown";
+	case CEPH_MSG_PING: return "ping";
+	case CEPH_MSG_AUTH: return "auth";
+	case CEPH_MSG_AUTH_REPLY: return "auth_reply";
+	case CEPH_MSG_MON_MAP: return "mon_map";
+	case CEPH_MSG_MON_GET_MAP: return "mon_get_map";
+	case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe";
+	case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack";
+	case CEPH_MSG_STATFS: return "statfs";
+	case CEPH_MSG_STATFS_REPLY: return "statfs_reply";
+	case CEPH_MSG_MDS_MAP: return "mds_map";
+	case CEPH_MSG_CLIENT_SESSION: return "client_session";
+	case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect";
+	case CEPH_MSG_CLIENT_REQUEST: return "client_request";
+	case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward";
+	case CEPH_MSG_CLIENT_REPLY: return "client_reply";
+	case CEPH_MSG_CLIENT_CAPS: return "client_caps";
+	case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release";
+	case CEPH_MSG_CLIENT_SNAP: return "client_snap";
+	case CEPH_MSG_CLIENT_LEASE: return "client_lease";
+	case CEPH_MSG_OSD_MAP: return "osd_map";
+	case CEPH_MSG_OSD_OP: return "osd_op";
+	case CEPH_MSG_OSD_OPREPLY: return "osd_opreply";
+	default: return "unknown";
+	}
+}
+EXPORT_SYMBOL(ceph_msg_type_name);
+
+/*
+ * Initially learn our fsid, or verify an fsid matches.
+ */
+int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
+{
+	if (client->have_fsid) {
+		if (ceph_fsid_compare(&client->fsid, fsid)) {
+			pr_err("bad fsid, had %pU got %pU",
+			       &client->fsid, fsid);
+			return -1;
+		}
+	} else {
+		pr_info("client%lld fsid %pU\n", ceph_client_id(client), fsid);
+		memcpy(&client->fsid, fsid, sizeof(*fsid));
+		ceph_debugfs_client_init(client);
+		client->have_fsid = true;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_check_fsid);
+
+static int strcmp_null(const char *s1, const char *s2)
+{
+	if (!s1 && !s2)
+		return 0;
+	if (s1 && !s2)
+		return -1;
+	if (!s1 && s2)
+		return 1;
+	return strcmp(s1, s2);
+}
+
+int ceph_compare_options(struct ceph_options *new_opt,
+			 struct ceph_client *client)
+{
+	struct ceph_options *opt1 = new_opt;
+	struct ceph_options *opt2 = client->options;
+	int ofs = offsetof(struct ceph_options, mon_addr);
+	int i;
+	int ret;
+
+	ret = memcmp(opt1, opt2, ofs);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->name, opt2->name);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->secret, opt2->secret);
+	if (ret)
+		return ret;
+
+	/* any matching mon ip implies a match */
+	for (i = 0; i < opt1->num_mon; i++) {
+		if (ceph_monmap_contains(client->monc.monmap,
+				 &opt1->mon_addr[i]))
+			return 0;
+	}
+	return -1;
+}
+EXPORT_SYMBOL(ceph_compare_options);
+
+
+static int parse_fsid(const char *str, struct ceph_fsid *fsid)
+{
+	int i = 0;
+	char tmp[3];
+	int err = -EINVAL;
+	int d;
+
+	dout("parse_fsid '%s'\n", str);
+	tmp[2] = 0;
+	while (*str && i < 16) {
+		if (ispunct(*str)) {
+			str++;
+			continue;
+		}
+		if (!isxdigit(str[0]) || !isxdigit(str[1]))
+			break;
+		tmp[0] = str[0];
+		tmp[1] = str[1];
+		if (sscanf(tmp, "%x", &d) < 1)
+			break;
+		fsid->fsid[i] = d & 0xff;
+		i++;
+		str += 2;
+	}
+
+	if (i == 16)
+		err = 0;
+	dout("parse_fsid ret %d got fsid %pU", err, fsid);
+	return err;
+}
+
+/*
+ * ceph options
+ */
+enum {
+	Opt_osdtimeout,
+	Opt_osdkeepalivetimeout,
+	Opt_mount_timeout,
+	Opt_osd_idle_ttl,
+	Opt_last_int,
+	/* int args above */
+	Opt_fsid,
+	Opt_name,
+	Opt_secret,
+	Opt_ip,
+	Opt_last_string,
+	/* string args above */
+	Opt_noshare,
+	Opt_nocrc,
+};
+
+static match_table_t opt_tokens = {
+	{Opt_osdtimeout, "osdtimeout=%d"},
+	{Opt_osdkeepalivetimeout, "osdkeepalive=%d"},
+	{Opt_mount_timeout, "mount_timeout=%d"},
+	{Opt_osd_idle_ttl, "osd_idle_ttl=%d"},
+	/* int args above */
+	{Opt_fsid, "fsid=%s"},
+	{Opt_name, "name=%s"},
+	{Opt_secret, "secret=%s"},
+	{Opt_ip, "ip=%s"},
+	/* string args above */
+	{Opt_noshare, "noshare"},
+	{Opt_nocrc, "nocrc"},
+	{-1, NULL}
+};
+
+void ceph_destroy_options(struct ceph_options *opt)
+{
+	dout("destroy_options %p\n", opt);
+	kfree(opt->name);
+	kfree(opt->secret);
+	kfree(opt);
+}
+EXPORT_SYMBOL(ceph_destroy_options);
+
+int ceph_parse_options(struct ceph_options **popt, char *options,
+		       const char *dev_name, const char *dev_name_end,
+		       int (*parse_extra_token)(char *c, void *private),
+		       void *private)
+{
+	struct ceph_options *opt;
+	const char *c;
+	int err = -ENOMEM;
+	substring_t argstr[MAX_OPT_ARGS];
+
+	opt = kzalloc(sizeof(*opt), GFP_KERNEL);
+	if (!opt)
+		return err;
+	opt->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*opt->mon_addr),
+				GFP_KERNEL);
+	if (!opt->mon_addr)
+		goto out;
+
+	dout("parse_options %p options '%s' dev_name '%s'\n", opt, options,
+	     dev_name);
+
+	/* start with defaults */
+	opt->flags = CEPH_OPT_DEFAULT;
+	opt->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT;
+	opt->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT;
+	opt->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */
+	opt->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT;   /* seconds */
+
+	/* get mon ip(s) */
+	/* ip1[:port1][,ip2[:port2]...] */
+	err = ceph_parse_ips(dev_name, dev_name_end, opt->mon_addr,
+			     CEPH_MAX_MON, &opt->num_mon);
+	if (err < 0)
+		goto out;
+
+	/* parse mount options */
+	while ((c = strsep(&options, ",")) != NULL) {
+		int token, intval, ret;
+		if (!*c)
+			continue;
+		err = -EINVAL;
+		token = match_token((char *)c, opt_tokens, argstr);
+		if (token < 0 && parse_extra_token) {
+			/* extra? */
+			err = parse_extra_token((char *)c, private);
+			if (err < 0) {
+				pr_err("bad option at '%s'\n", c);
+				goto out;
+			}
+			continue;
+		}
+		if (token < Opt_last_int) {
+			ret = match_int(&argstr[0], &intval);
+			if (ret < 0) {
+				pr_err("bad mount option arg (not int) "
+				       "at '%s'\n", c);
+				continue;
+			}
+			dout("got int token %d val %d\n", token, intval);
+		} else if (token > Opt_last_int && token < Opt_last_string) {
+			dout("got string token %d val %s\n", token,
+			     argstr[0].from);
+		} else {
+			dout("got token %d\n", token);
+		}
+		switch (token) {
+		case Opt_ip:
+			err = ceph_parse_ips(argstr[0].from,
+					     argstr[0].to,
+					     &opt->my_addr,
+					     1, NULL);
+			if (err < 0)
+				goto out;
+			opt->flags |= CEPH_OPT_MYIP;
+			break;
+
+		case Opt_fsid:
+			err = parse_fsid(argstr[0].from, &opt->fsid);
+			if (err == 0)
+				opt->flags |= CEPH_OPT_FSID;
+			break;
+		case Opt_name:
+			opt->name = kstrndup(argstr[0].from,
+					      argstr[0].to-argstr[0].from,
+					      GFP_KERNEL);
+			break;
+		case Opt_secret:
+			opt->secret = kstrndup(argstr[0].from,
+						argstr[0].to-argstr[0].from,
+						GFP_KERNEL);
+			break;
+
+			/* misc */
+		case Opt_osdtimeout:
+			opt->osd_timeout = intval;
+			break;
+		case Opt_osdkeepalivetimeout:
+			opt->osd_keepalive_timeout = intval;
+			break;
+		case Opt_osd_idle_ttl:
+			opt->osd_idle_ttl = intval;
+			break;
+		case Opt_mount_timeout:
+			opt->mount_timeout = intval;
+			break;
+
+		case Opt_noshare:
+			opt->flags |= CEPH_OPT_NOSHARE;
+			break;
+
+		case Opt_nocrc:
+			opt->flags |= CEPH_OPT_NOCRC;
+			break;
+
+		default:
+			BUG_ON(token);
+		}
+	}
+
+	/* success */
+	*popt = opt;
+	return 0;
+
+out:
+	ceph_destroy_options(opt);
+	return err;
+}
+EXPORT_SYMBOL(ceph_parse_options);
+
+u64 ceph_client_id(struct ceph_client *client)
+{
+	return client->monc.auth->global_id;
+}
+EXPORT_SYMBOL(ceph_client_id);
+
+/*
+ * create a fresh client instance
+ */
+struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private)
+{
+	struct ceph_client *client;
+	int err = -ENOMEM;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (client == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	client->private = private;
+	client->options = opt;
+
+	mutex_init(&client->mount_mutex);
+	init_waitqueue_head(&client->auth_wq);
+	client->auth_err = 0;
+
+	client->extra_mon_dispatch = NULL;
+	client->supported_features = CEPH_FEATURE_SUPPORTED_DEFAULT;
+	client->required_features = CEPH_FEATURE_REQUIRED_DEFAULT;
+
+	client->msgr = NULL;
+
+	/* subsystems */
+	err = ceph_monc_init(&client->monc, client);
+	if (err < 0)
+		goto fail;
+	err = ceph_osdc_init(&client->osdc, client);
+	if (err < 0)
+		goto fail_monc;
+
+	return client;
+
+fail_monc:
+	ceph_monc_stop(&client->monc);
+fail:
+	kfree(client);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL(ceph_create_client);
+
+void ceph_destroy_client(struct ceph_client *client)
+{
+	dout("destroy_client %p\n", client);
+
+	/* unmount */
+	ceph_osdc_stop(&client->osdc);
+
+	/*
+	 * make sure mds and osd connections close out before destroying
+	 * the auth module, which is needed to free those connections'
+	 * ceph_authorizers.
+	 */
+	ceph_msgr_flush();
+
+	ceph_monc_stop(&client->monc);
+
+	ceph_debugfs_client_cleanup(client);
+
+	if (client->msgr)
+		ceph_messenger_destroy(client->msgr);
+
+	ceph_destroy_options(client->options);
+
+	kfree(client);
+	dout("destroy_client %p done\n", client);
+}
+EXPORT_SYMBOL(ceph_destroy_client);
+
+/*
+ * true if we have the mon map (and have thus joined the cluster)
+ */
+static int have_mon_and_osd_map(struct ceph_client *client)
+{
+	return client->monc.monmap && client->monc.monmap->epoch &&
+	       client->osdc.osdmap && client->osdc.osdmap->epoch;
+}
+
+/*
+ * mount: join the ceph cluster, and open root directory.
+ */
+int __ceph_open_session(struct ceph_client *client, unsigned long started)
+{
+	struct ceph_entity_addr *myaddr = NULL;
+	int err;
+	unsigned long timeout = client->options->mount_timeout * HZ;
+
+	/* initialize the messenger */
+	if (client->msgr == NULL) {
+		if (ceph_test_opt(client, MYIP))
+			myaddr = &client->options->my_addr;
+		client->msgr = ceph_messenger_create(myaddr,
+					client->supported_features,
+					client->required_features);
+		if (IS_ERR(client->msgr)) {
+			client->msgr = NULL;
+			return PTR_ERR(client->msgr);
+		}
+		client->msgr->nocrc = ceph_test_opt(client, NOCRC);
+	}
+
+	/* open session, and wait for mon and osd maps */
+	err = ceph_monc_open_session(&client->monc);
+	if (err < 0)
+		return err;
+
+	while (!have_mon_and_osd_map(client)) {
+		err = -EIO;
+		if (timeout && time_after_eq(jiffies, started + timeout))
+			return err;
+
+		/* wait */
+		dout("mount waiting for mon_map\n");
+		err = wait_event_interruptible_timeout(client->auth_wq,
+			have_mon_and_osd_map(client) || (client->auth_err < 0),
+			timeout);
+		if (err == -EINTR || err == -ERESTARTSYS)
+			return err;
+		if (client->auth_err < 0)
+			return client->auth_err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(__ceph_open_session);
+
+
+int ceph_open_session(struct ceph_client *client)
+{
+	int ret;
+	unsigned long started = jiffies;  /* note the start time */
+
+	dout("open_session start\n");
+	mutex_lock(&client->mount_mutex);
+
+	ret = __ceph_open_session(client, started);
+
+	mutex_unlock(&client->mount_mutex);
+	return ret;
+}
+EXPORT_SYMBOL(ceph_open_session);
+
+
+static int __init init_ceph_lib(void)
+{
+	int ret = 0;
+
+	ret = ceph_debugfs_init();
+	if (ret < 0)
+		goto out;
+
+	ret = ceph_msgr_init();
+	if (ret < 0)
+		goto out_debugfs;
+
+	pr_info("loaded (mon/osd proto %d/%d, osdmap %d/%d %d/%d)\n",
+		CEPH_MONC_PROTOCOL, CEPH_OSDC_PROTOCOL,
+		CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT,
+		CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT);
+
+	return 0;
+
+out_debugfs:
+	ceph_debugfs_cleanup();
+out:
+	return ret;
+}
+
+static void __exit exit_ceph_lib(void)
+{
+	dout("exit_ceph_lib\n");
+	ceph_msgr_exit();
+	ceph_debugfs_cleanup();
+}
+
+module_init(init_ceph_lib);
+module_exit(exit_ceph_lib);
+
+MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
+MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
+MODULE_AUTHOR("Patience Warnick <patience@newdream.net>");
+MODULE_DESCRIPTION("Ceph filesystem for Linux");
+MODULE_LICENSE("GPL");
diff --git a/fs/ceph/ceph_fs.c b/net/ceph/ceph_fs.c
similarity index 92%
rename from fs/ceph/ceph_fs.c
rename to net/ceph/ceph_fs.c
index 3ac6cc7..a3a3a31 100644
--- a/fs/ceph/ceph_fs.c
+++ b/net/ceph/ceph_fs.c
@@ -1,7 +1,8 @@
 /*
  * Some non-inline ceph helpers
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
 /*
  * return true if @layout appears to be valid
@@ -52,6 +53,7 @@
 
 	return mode;
 }
+EXPORT_SYMBOL(ceph_flags_to_mode);
 
 int ceph_caps_for_mode(int mode)
 {
@@ -70,3 +72,4 @@
 
 	return caps;
 }
+EXPORT_SYMBOL(ceph_caps_for_mode);
diff --git a/fs/ceph/ceph_hash.c b/net/ceph/ceph_hash.c
similarity index 98%
rename from fs/ceph/ceph_hash.c
rename to net/ceph/ceph_hash.c
index bd57001..815ef88 100644
--- a/fs/ceph/ceph_hash.c
+++ b/net/ceph/ceph_hash.c
@@ -1,5 +1,5 @@
 
-#include "types.h"
+#include <linux/ceph/types.h>
 
 /*
  * Robert Jenkin's hash function.
diff --git a/net/ceph/ceph_strings.c b/net/ceph/ceph_strings.c
new file mode 100644
index 0000000..3fbda04
--- /dev/null
+++ b/net/ceph/ceph_strings.c
@@ -0,0 +1,84 @@
+/*
+ * Ceph string constants
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+const char *ceph_entity_type_name(int type)
+{
+	switch (type) {
+	case CEPH_ENTITY_TYPE_MDS: return "mds";
+	case CEPH_ENTITY_TYPE_OSD: return "osd";
+	case CEPH_ENTITY_TYPE_MON: return "mon";
+	case CEPH_ENTITY_TYPE_CLIENT: return "client";
+	case CEPH_ENTITY_TYPE_AUTH: return "auth";
+	default: return "unknown";
+	}
+}
+
+const char *ceph_osd_op_name(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_READ: return "read";
+	case CEPH_OSD_OP_STAT: return "stat";
+
+	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
+
+	case CEPH_OSD_OP_WRITE: return "write";
+	case CEPH_OSD_OP_DELETE: return "delete";
+	case CEPH_OSD_OP_TRUNCATE: return "truncate";
+	case CEPH_OSD_OP_ZERO: return "zero";
+	case CEPH_OSD_OP_WRITEFULL: return "writefull";
+	case CEPH_OSD_OP_ROLLBACK: return "rollback";
+
+	case CEPH_OSD_OP_APPEND: return "append";
+	case CEPH_OSD_OP_STARTSYNC: return "startsync";
+	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
+	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
+
+	case CEPH_OSD_OP_TMAPUP: return "tmapup";
+	case CEPH_OSD_OP_TMAPGET: return "tmapget";
+	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
+
+	case CEPH_OSD_OP_GETXATTR: return "getxattr";
+	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
+	case CEPH_OSD_OP_SETXATTR: return "setxattr";
+	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
+	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
+	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
+	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
+
+	case CEPH_OSD_OP_PULL: return "pull";
+	case CEPH_OSD_OP_PUSH: return "push";
+	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
+	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
+	case CEPH_OSD_OP_SCRUB: return "scrub";
+
+	case CEPH_OSD_OP_WRLOCK: return "wrlock";
+	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
+	case CEPH_OSD_OP_RDLOCK: return "rdlock";
+	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
+	case CEPH_OSD_OP_UPLOCK: return "uplock";
+	case CEPH_OSD_OP_DNLOCK: return "dnlock";
+
+	case CEPH_OSD_OP_CALL: return "call";
+
+	case CEPH_OSD_OP_PGLS: return "pgls";
+	}
+	return "???";
+}
+
+
+const char *ceph_pool_op_name(int op)
+{
+	switch (op) {
+	case POOL_OP_CREATE: return "create";
+	case POOL_OP_DELETE: return "delete";
+	case POOL_OP_AUID_CHANGE: return "auid change";
+	case POOL_OP_CREATE_SNAP: return "create snap";
+	case POOL_OP_DELETE_SNAP: return "delete snap";
+	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
+	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
+	}
+	return "???";
+}
diff --git a/fs/ceph/crush/crush.c b/net/ceph/crush/crush.c
similarity index 98%
rename from fs/ceph/crush/crush.c
rename to net/ceph/crush/crush.c
index fabd302..d6ebb13 100644
--- a/fs/ceph/crush/crush.c
+++ b/net/ceph/crush/crush.c
@@ -8,7 +8,7 @@
 # define BUG_ON(x) assert(!(x))
 #endif
 
-#include "crush.h"
+#include <linux/crush/crush.h>
 
 const char *crush_bucket_alg_name(int alg)
 {
diff --git a/fs/ceph/crush/hash.c b/net/ceph/crush/hash.c
similarity index 98%
rename from fs/ceph/crush/hash.c
rename to net/ceph/crush/hash.c
index 5873aed..5bb63e3 100644
--- a/fs/ceph/crush/hash.c
+++ b/net/ceph/crush/hash.c
@@ -1,6 +1,6 @@
 
 #include <linux/types.h>
-#include "hash.h"
+#include <linux/crush/hash.h>
 
 /*
  * Robert Jenkins' function for mixing 32-bit values
diff --git a/fs/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
similarity index 99%
rename from fs/ceph/crush/mapper.c
rename to net/ceph/crush/mapper.c
index a4eec13..42599e3 100644
--- a/fs/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -18,8 +18,8 @@
 # define kfree(x) free(x)
 #endif
 
-#include "crush.h"
-#include "hash.h"
+#include <linux/crush/crush.h>
+#include <linux/crush/hash.h>
 
 /*
  * Implement the core CRUSH mapping algorithm.
diff --git a/fs/ceph/crypto.c b/net/ceph/crypto.c
similarity index 99%
rename from fs/ceph/crypto.c
rename to net/ceph/crypto.c
index a3e627f..7b505b0 100644
--- a/fs/ceph/crypto.c
+++ b/net/ceph/crypto.c
@@ -1,13 +1,13 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/scatterlist.h>
 #include <linux/slab.h>
 #include <crypto/hash.h>
 
+#include <linux/ceph/decode.h>
 #include "crypto.h"
-#include "decode.h"
 
 int ceph_crypto_key_encode(struct ceph_crypto_key *key, void **p, void *end)
 {
diff --git a/fs/ceph/crypto.h b/net/ceph/crypto.h
similarity index 95%
rename from fs/ceph/crypto.h
rename to net/ceph/crypto.h
index bdf3860..f9eccac 100644
--- a/fs/ceph/crypto.h
+++ b/net/ceph/crypto.h
@@ -1,8 +1,8 @@
 #ifndef _FS_CEPH_CRYPTO_H
 #define _FS_CEPH_CRYPTO_H
 
-#include "types.h"
-#include "buffer.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/buffer.h>
 
 /*
  * cryptographic secret
diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c
new file mode 100644
index 0000000..27d4ea31
--- /dev/null
+++ b/net/ceph/debugfs.c
@@ -0,0 +1,267 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
+
+#ifdef CONFIG_DEBUG_FS
+
+/*
+ * Implement /sys/kernel/debug/ceph fun
+ *
+ * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
+ *      .../osdmap      - current osdmap
+ *      .../monmap      - current monmap
+ *      .../osdc        - active osd requests
+ *      .../monc        - mon client state
+ *      .../dentry_lru  - dump contents of dentry lru
+ *      .../caps        - expose cap (reservation) stats
+ *      .../bdi         - symlink to ../../bdi/something
+ */
+
+static struct dentry *ceph_debugfs_dir;
+
+static int monmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+
+	if (client->monc.monmap == NULL)
+		return 0;
+
+	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
+	for (i = 0; i < client->monc.monmap->num_mon; i++) {
+		struct ceph_entity_inst *inst =
+			&client->monc.monmap->mon_inst[i];
+
+		seq_printf(s, "\t%s%lld\t%s\n",
+			   ENTITY_NAME(inst->name),
+			   ceph_pr_addr(&inst->addr.in_addr));
+	}
+	return 0;
+}
+
+static int osdmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+	struct rb_node *n;
+
+	if (client->osdc.osdmap == NULL)
+		return 0;
+	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
+	seq_printf(s, "flags%s%s\n",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
+		   " NEARFULL" : "",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
+		   " FULL" : "");
+	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
+		struct ceph_pg_pool_info *pool =
+			rb_entry(n, struct ceph_pg_pool_info, node);
+		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
+			   pool->id, pool->v.pg_num, pool->pg_num_mask,
+			   pool->v.lpg_num, pool->lpg_num_mask);
+	}
+	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
+		struct ceph_entity_addr *addr =
+			&client->osdc.osdmap->osd_addr[i];
+		int state = client->osdc.osdmap->osd_state[i];
+		char sb[64];
+
+		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
+			   i, ceph_pr_addr(&addr->in_addr),
+			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
+			   ceph_osdmap_state_str(sb, sizeof(sb), state));
+	}
+	return 0;
+}
+
+static int monc_show(struct seq_file *s, void *p)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_client *monc = &client->monc;
+	struct rb_node *rp;
+
+	mutex_lock(&monc->mutex);
+
+	if (monc->have_mdsmap)
+		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
+	if (monc->have_osdmap)
+		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
+	if (monc->want_next_osdmap)
+		seq_printf(s, "want next osdmap\n");
+
+	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
+		__u16 op;
+		req = rb_entry(rp, struct ceph_mon_generic_request, node);
+		op = le16_to_cpu(req->request->hdr.type);
+		if (op == CEPH_MSG_STATFS)
+			seq_printf(s, "%lld statfs\n", req->tid);
+		else
+			seq_printf(s, "%lld unknown\n", req->tid);
+	}
+
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+
+static int osdc_show(struct seq_file *s, void *pp)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_osd_client *osdc = &client->osdc;
+	struct rb_node *p;
+
+	mutex_lock(&osdc->request_mutex);
+	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
+		struct ceph_osd_request *req;
+		struct ceph_osd_request_head *head;
+		struct ceph_osd_op *op;
+		int num_ops;
+		int opcode, olen;
+		int i;
+
+		req = rb_entry(p, struct ceph_osd_request, r_node);
+
+		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
+			   req->r_osd ? req->r_osd->o_osd : -1,
+			   le32_to_cpu(req->r_pgid.pool),
+			   le16_to_cpu(req->r_pgid.ps));
+
+		head = req->r_request->front.iov_base;
+		op = (void *)(head + 1);
+
+		num_ops = le16_to_cpu(head->num_ops);
+		olen = le32_to_cpu(head->object_len);
+		seq_printf(s, "%.*s", olen,
+			   (const char *)(head->ops + num_ops));
+
+		if (req->r_reassert_version.epoch)
+			seq_printf(s, "\t%u'%llu",
+			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
+			   le64_to_cpu(req->r_reassert_version.version));
+		else
+			seq_printf(s, "\t");
+
+		for (i = 0; i < num_ops; i++) {
+			opcode = le16_to_cpu(op->op);
+			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
+			op++;
+		}
+
+		seq_printf(s, "\n");
+	}
+	mutex_unlock(&osdc->request_mutex);
+	return 0;
+}
+
+CEPH_DEFINE_SHOW_FUNC(monmap_show)
+CEPH_DEFINE_SHOW_FUNC(osdmap_show)
+CEPH_DEFINE_SHOW_FUNC(monc_show)
+CEPH_DEFINE_SHOW_FUNC(osdc_show)
+
+int ceph_debugfs_init(void)
+{
+	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
+	if (!ceph_debugfs_dir)
+		return -ENOMEM;
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+	debugfs_remove(ceph_debugfs_dir);
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client)
+{
+	int ret = -ENOMEM;
+	char name[80];
+
+	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
+		 client->monc.auth->global_id);
+
+	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
+	if (!client->debugfs_dir)
+		goto out;
+
+	client->monc.debugfs_file = debugfs_create_file("monc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &monc_show_fops);
+	if (!client->monc.debugfs_file)
+		goto out;
+
+	client->osdc.debugfs_file = debugfs_create_file("osdc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &osdc_show_fops);
+	if (!client->osdc.debugfs_file)
+		goto out;
+
+	client->debugfs_monmap = debugfs_create_file("monmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&monmap_show_fops);
+	if (!client->debugfs_monmap)
+		goto out;
+
+	client->debugfs_osdmap = debugfs_create_file("osdmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&osdmap_show_fops);
+	if (!client->debugfs_osdmap)
+		goto out;
+
+	return 0;
+
+out:
+	ceph_debugfs_client_cleanup(client);
+	return ret;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+	debugfs_remove(client->debugfs_osdmap);
+	debugfs_remove(client->debugfs_monmap);
+	debugfs_remove(client->osdc.debugfs_file);
+	debugfs_remove(client->monc.debugfs_file);
+	debugfs_remove(client->debugfs_dir);
+}
+
+#else  /* CONFIG_DEBUG_FS */
+
+int ceph_debugfs_init(void)
+{
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client)
+{
+	return 0;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+}
+
+#endif  /* CONFIG_DEBUG_FS */
+
+EXPORT_SYMBOL(ceph_debugfs_init);
+EXPORT_SYMBOL(ceph_debugfs_cleanup);
diff --git a/fs/ceph/messenger.c b/net/ceph/messenger.c
similarity index 89%
rename from fs/ceph/messenger.c
rename to net/ceph/messenger.c
index 2502d76..0e8157e 100644
--- a/fs/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/crc32c.h>
 #include <linux/ctype.h>
@@ -9,12 +9,14 @@
 #include <linux/slab.h>
 #include <linux/socket.h>
 #include <linux/string.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
 #include <net/tcp.h>
 
-#include "super.h"
-#include "messenger.h"
-#include "decode.h"
-#include "pagelist.h"
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/pagelist.h>
 
 /*
  * Ceph uses the messenger to exchange ceph_msg messages with other
@@ -48,7 +50,7 @@
 static DEFINE_SPINLOCK(addr_str_lock);
 static int last_addr_str;
 
-const char *pr_addr(const struct sockaddr_storage *ss)
+const char *ceph_pr_addr(const struct sockaddr_storage *ss)
 {
 	int i;
 	char *s;
@@ -79,6 +81,7 @@
 
 	return s;
 }
+EXPORT_SYMBOL(ceph_pr_addr);
 
 static void encode_my_addr(struct ceph_messenger *msgr)
 {
@@ -91,7 +94,7 @@
  */
 struct workqueue_struct *ceph_msgr_wq;
 
-int __init ceph_msgr_init(void)
+int ceph_msgr_init(void)
 {
 	ceph_msgr_wq = create_workqueue("ceph-msgr");
 	if (IS_ERR(ceph_msgr_wq)) {
@@ -102,16 +105,19 @@
 	}
 	return 0;
 }
+EXPORT_SYMBOL(ceph_msgr_init);
 
 void ceph_msgr_exit(void)
 {
 	destroy_workqueue(ceph_msgr_wq);
 }
+EXPORT_SYMBOL(ceph_msgr_exit);
 
 void ceph_msgr_flush(void)
 {
 	flush_workqueue(ceph_msgr_wq);
 }
+EXPORT_SYMBOL(ceph_msgr_flush);
 
 
 /*
@@ -221,19 +227,19 @@
 
 	set_sock_callbacks(sock, con);
 
-	dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
+	dout("connect %s\n", ceph_pr_addr(&con->peer_addr.in_addr));
 
 	ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
 				 O_NONBLOCK);
 	if (ret == -EINPROGRESS) {
 		dout("connect %s EINPROGRESS sk_state = %u\n",
-		     pr_addr(&con->peer_addr.in_addr),
+		     ceph_pr_addr(&con->peer_addr.in_addr),
 		     sock->sk->sk_state);
 		ret = 0;
 	}
 	if (ret < 0) {
 		pr_err("connect %s error %d\n",
-		       pr_addr(&con->peer_addr.in_addr), ret);
+		       ceph_pr_addr(&con->peer_addr.in_addr), ret);
 		sock_release(sock);
 		con->sock = NULL;
 		con->error_msg = "connect error";
@@ -334,7 +340,8 @@
  */
 void ceph_con_close(struct ceph_connection *con)
 {
-	dout("con_close %p peer %s\n", con, pr_addr(&con->peer_addr.in_addr));
+	dout("con_close %p peer %s\n", con,
+	     ceph_pr_addr(&con->peer_addr.in_addr));
 	set_bit(CLOSED, &con->state);  /* in case there's queued work */
 	clear_bit(STANDBY, &con->state);  /* avoid connect_seq bump */
 	clear_bit(LOSSYTX, &con->state);  /* so we retry next connect */
@@ -347,19 +354,21 @@
 	mutex_unlock(&con->mutex);
 	queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_close);
 
 /*
  * Reopen a closed connection, with a new peer address.
  */
 void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
 {
-	dout("con_open %p %s\n", con, pr_addr(&addr->in_addr));
+	dout("con_open %p %s\n", con, ceph_pr_addr(&addr->in_addr));
 	set_bit(OPENING, &con->state);
 	clear_bit(CLOSED, &con->state);
 	memcpy(&con->peer_addr, addr, sizeof(*addr));
 	con->delay = 0;      /* reset backoff memory */
 	queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_open);
 
 /*
  * return true if this connection ever successfully opened
@@ -406,6 +415,7 @@
 	INIT_LIST_HEAD(&con->out_sent);
 	INIT_DELAYED_WORK(&con->work, con_work);
 }
+EXPORT_SYMBOL(ceph_con_init);
 
 
 /*
@@ -529,8 +539,11 @@
 	if (le32_to_cpu(m->hdr.data_len) > 0) {
 		/* initialize page iterator */
 		con->out_msg_pos.page = 0;
-		con->out_msg_pos.page_pos =
-			le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
+		if (m->pages)
+			con->out_msg_pos.page_pos =
+				le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
+		else
+			con->out_msg_pos.page_pos = 0;
 		con->out_msg_pos.data_pos = 0;
 		con->out_msg_pos.did_page_crc = 0;
 		con->out_more = 1;  /* data + footer will follow */
@@ -647,7 +660,7 @@
 	dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
 	     con->connect_seq, global_seq, proto);
 
-	con->out_connect.features = cpu_to_le64(CEPH_FEATURE_SUPPORTED);
+	con->out_connect.features = cpu_to_le64(msgr->supported_features);
 	con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
 	con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
 	con->out_connect.global_seq = cpu_to_le32(global_seq);
@@ -712,6 +725,31 @@
 	return ret;  /* done! */
 }
 
+#ifdef CONFIG_BLOCK
+static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg)
+{
+	if (!bio) {
+		*iter = NULL;
+		*seg = 0;
+		return;
+	}
+	*iter = bio;
+	*seg = bio->bi_idx;
+}
+
+static void iter_bio_next(struct bio **bio_iter, int *seg)
+{
+	if (*bio_iter == NULL)
+		return;
+
+	BUG_ON(*seg >= (*bio_iter)->bi_vcnt);
+
+	(*seg)++;
+	if (*seg == (*bio_iter)->bi_vcnt)
+		init_bio_iter((*bio_iter)->bi_next, bio_iter, seg);
+}
+#endif
+
 /*
  * Write as much message data payload as we can.  If we finish, queue
  * up the footer.
@@ -726,21 +764,46 @@
 	size_t len;
 	int crc = con->msgr->nocrc;
 	int ret;
+	int total_max_write;
+	int in_trail = 0;
+	size_t trail_len = (msg->trail ? msg->trail->length : 0);
 
 	dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n",
 	     con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages,
 	     con->out_msg_pos.page_pos);
 
-	while (con->out_msg_pos.page < con->out_msg->nr_pages) {
+#ifdef CONFIG_BLOCK
+	if (msg->bio && !msg->bio_iter)
+		init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg);
+#endif
+
+	while (data_len > con->out_msg_pos.data_pos) {
 		struct page *page = NULL;
 		void *kaddr = NULL;
+		int max_write = PAGE_SIZE;
+		int page_shift = 0;
+
+		total_max_write = data_len - trail_len -
+			con->out_msg_pos.data_pos;
 
 		/*
 		 * if we are calculating the data crc (the default), we need
 		 * to map the page.  if our pages[] has been revoked, use the
 		 * zero page.
 		 */
-		if (msg->pages) {
+
+		/* have we reached the trail part of the data? */
+		if (con->out_msg_pos.data_pos >= data_len - trail_len) {
+			in_trail = 1;
+
+			total_max_write = data_len - con->out_msg_pos.data_pos;
+
+			page = list_first_entry(&msg->trail->head,
+						struct page, lru);
+			if (crc)
+				kaddr = kmap(page);
+			max_write = PAGE_SIZE;
+		} else if (msg->pages) {
 			page = msg->pages[con->out_msg_pos.page];
 			if (crc)
 				kaddr = kmap(page);
@@ -749,13 +812,25 @@
 						struct page, lru);
 			if (crc)
 				kaddr = kmap(page);
+#ifdef CONFIG_BLOCK
+		} else if (msg->bio) {
+			struct bio_vec *bv;
+
+			bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg);
+			page = bv->bv_page;
+			page_shift = bv->bv_offset;
+			if (crc)
+				kaddr = kmap(page) + page_shift;
+			max_write = bv->bv_len;
+#endif
 		} else {
 			page = con->msgr->zero_page;
 			if (crc)
 				kaddr = page_address(con->msgr->zero_page);
 		}
-		len = min((int)(PAGE_SIZE - con->out_msg_pos.page_pos),
-			  (int)(data_len - con->out_msg_pos.data_pos));
+		len = min_t(int, max_write - con->out_msg_pos.page_pos,
+			    total_max_write);
+
 		if (crc && !con->out_msg_pos.did_page_crc) {
 			void *base = kaddr + con->out_msg_pos.page_pos;
 			u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc);
@@ -765,13 +840,14 @@
 				cpu_to_le32(crc32c(tmpcrc, base, len));
 			con->out_msg_pos.did_page_crc = 1;
 		}
-
 		ret = kernel_sendpage(con->sock, page,
-				      con->out_msg_pos.page_pos, len,
+				      con->out_msg_pos.page_pos + page_shift,
+				      len,
 				      MSG_DONTWAIT | MSG_NOSIGNAL |
 				      MSG_MORE);
 
-		if (crc && (msg->pages || msg->pagelist))
+		if (crc &&
+		    (msg->pages || msg->pagelist || msg->bio || in_trail))
 			kunmap(page);
 
 		if (ret <= 0)
@@ -783,9 +859,16 @@
 			con->out_msg_pos.page_pos = 0;
 			con->out_msg_pos.page++;
 			con->out_msg_pos.did_page_crc = 0;
-			if (msg->pagelist)
+			if (in_trail)
+				list_move_tail(&page->lru,
+					       &msg->trail->head);
+			else if (msg->pagelist)
 				list_move_tail(&page->lru,
 					       &msg->pagelist->head);
+#ifdef CONFIG_BLOCK
+			else if (msg->bio)
+				iter_bio_next(&msg->bio_iter, &msg->bio_seg);
+#endif
 		}
 	}
 
@@ -938,7 +1021,7 @@
 {
 	if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) {
 		pr_err("connect to %s got bad banner\n",
-		       pr_addr(&con->peer_addr.in_addr));
+		       ceph_pr_addr(&con->peer_addr.in_addr));
 		con->error_msg = "protocol error, bad banner";
 		return -1;
 	}
@@ -1041,7 +1124,7 @@
 
 		addr_set_port(ss, port);
 
-		dout("parse_ips got %s\n", pr_addr(ss));
+		dout("parse_ips got %s\n", ceph_pr_addr(ss));
 
 		if (p == end)
 			break;
@@ -1061,6 +1144,7 @@
 	pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
 	return -EINVAL;
 }
+EXPORT_SYMBOL(ceph_parse_ips);
 
 static int process_banner(struct ceph_connection *con)
 {
@@ -1082,9 +1166,9 @@
 	    !(addr_is_blank(&con->actual_peer_addr.in_addr) &&
 	      con->actual_peer_addr.nonce == con->peer_addr.nonce)) {
 		pr_warning("wrong peer, want %s/%d, got %s/%d\n",
-			   pr_addr(&con->peer_addr.in_addr),
+			   ceph_pr_addr(&con->peer_addr.in_addr),
 			   (int)le32_to_cpu(con->peer_addr.nonce),
-			   pr_addr(&con->actual_peer_addr.in_addr),
+			   ceph_pr_addr(&con->actual_peer_addr.in_addr),
 			   (int)le32_to_cpu(con->actual_peer_addr.nonce));
 		con->error_msg = "wrong peer at address";
 		return -1;
@@ -1102,7 +1186,7 @@
 		addr_set_port(&con->msgr->inst.addr.in_addr, port);
 		encode_my_addr(con->msgr);
 		dout("process_banner learned my addr is %s\n",
-		     pr_addr(&con->msgr->inst.addr.in_addr));
+		     ceph_pr_addr(&con->msgr->inst.addr.in_addr));
 	}
 
 	set_bit(NEGOTIATING, &con->state);
@@ -1123,8 +1207,8 @@
 
 static int process_connect(struct ceph_connection *con)
 {
-	u64 sup_feat = CEPH_FEATURE_SUPPORTED;
-	u64 req_feat = CEPH_FEATURE_REQUIRED;
+	u64 sup_feat = con->msgr->supported_features;
+	u64 req_feat = con->msgr->required_features;
 	u64 server_feat = le64_to_cpu(con->in_reply.features);
 
 	dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
@@ -1134,7 +1218,7 @@
 		pr_err("%s%lld %s feature set mismatch,"
 		       " my %llx < server's %llx, missing %llx\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
 		       sup_feat, server_feat, server_feat & ~sup_feat);
 		con->error_msg = "missing required protocol features";
 		fail_protocol(con);
@@ -1144,7 +1228,7 @@
 		pr_err("%s%lld %s protocol version mismatch,"
 		       " my %d != server's %d\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
 		       le32_to_cpu(con->out_connect.protocol_version),
 		       le32_to_cpu(con->in_reply.protocol_version));
 		con->error_msg = "protocol version mismatch";
@@ -1178,7 +1262,7 @@
 		     le32_to_cpu(con->in_connect.connect_seq));
 		pr_err("%s%lld %s connection reset\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr));
+		       ceph_pr_addr(&con->peer_addr.in_addr));
 		reset_connection(con);
 		prepare_write_connect(con->msgr, con, 0);
 		prepare_read_connect(con);
@@ -1223,7 +1307,7 @@
 			pr_err("%s%lld %s protocol feature mismatch,"
 			       " my required %llx > server's %llx, need %llx\n",
 			       ENTITY_NAME(con->peer_name),
-			       pr_addr(&con->peer_addr.in_addr),
+			       ceph_pr_addr(&con->peer_addr.in_addr),
 			       req_feat, server_feat, req_feat & ~server_feat);
 			con->error_msg = "missing required protocol features";
 			fail_protocol(con);
@@ -1305,8 +1389,7 @@
 					struct kvec *section,
 					unsigned int sec_len, u32 *crc)
 {
-	int left;
-	int ret;
+	int ret, left;
 
 	BUG_ON(!section);
 
@@ -1329,13 +1412,83 @@
 static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
 				struct ceph_msg_header *hdr,
 				int *skip);
+
+
+static int read_partial_message_pages(struct ceph_connection *con,
+				      struct page **pages,
+				      unsigned data_len, int datacrc)
+{
+	void *p;
+	int ret;
+	int left;
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
+	/* (page) data */
+	BUG_ON(pages == NULL);
+	p = kmap(pages[con->in_msg_pos.page]);
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(pages[con->in_msg_pos.page]);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == PAGE_SIZE) {
+		con->in_msg_pos.page_pos = 0;
+		con->in_msg_pos.page++;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_BLOCK
+static int read_partial_message_bio(struct ceph_connection *con,
+				    struct bio **bio_iter, int *bio_seg,
+				    unsigned data_len, int datacrc)
+{
+	struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg);
+	void *p;
+	int ret, left;
+
+	if (IS_ERR(bv))
+		return PTR_ERR(bv);
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(bv->bv_len - con->in_msg_pos.page_pos));
+
+	p = kmap(bv->bv_page) + bv->bv_offset;
+
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(bv->bv_page);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == bv->bv_len) {
+		con->in_msg_pos.page_pos = 0;
+		iter_bio_next(bio_iter, bio_seg);
+	}
+
+	return ret;
+}
+#endif
+
 /*
  * read (part of) a message.
  */
 static int read_partial_message(struct ceph_connection *con)
 {
 	struct ceph_msg *m = con->in_msg;
-	void *p;
 	int ret;
 	int to, left;
 	unsigned front_len, middle_len, data_len, data_off;
@@ -1381,7 +1534,7 @@
 	if ((s64)seq - (s64)con->in_seq < 1) {
 		pr_info("skipping %s%lld %s seq %lld, expected %lld\n",
 			ENTITY_NAME(con->peer_name),
-			pr_addr(&con->peer_addr.in_addr),
+			ceph_pr_addr(&con->peer_addr.in_addr),
 			seq, con->in_seq + 1);
 		con->in_base_pos = -front_len - middle_len - data_len -
 			sizeof(m->footer);
@@ -1422,7 +1575,10 @@
 			m->middle->vec.iov_len = 0;
 
 		con->in_msg_pos.page = 0;
-		con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
+		if (m->pages)
+			con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
+		else
+			con->in_msg_pos.page_pos = 0;
 		con->in_msg_pos.data_pos = 0;
 	}
 
@@ -1440,27 +1596,29 @@
 		if (ret <= 0)
 			return ret;
 	}
+#ifdef CONFIG_BLOCK
+	if (m->bio && !m->bio_iter)
+		init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg);
+#endif
 
 	/* (page) data */
 	while (con->in_msg_pos.data_pos < data_len) {
-		left = min((int)(data_len - con->in_msg_pos.data_pos),
-			   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
-		BUG_ON(m->pages == NULL);
-		p = kmap(m->pages[con->in_msg_pos.page]);
-		ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
-				       left);
-		if (ret > 0 && datacrc)
-			con->in_data_crc =
-				crc32c(con->in_data_crc,
-					  p + con->in_msg_pos.page_pos, ret);
-		kunmap(m->pages[con->in_msg_pos.page]);
-		if (ret <= 0)
-			return ret;
-		con->in_msg_pos.data_pos += ret;
-		con->in_msg_pos.page_pos += ret;
-		if (con->in_msg_pos.page_pos == PAGE_SIZE) {
-			con->in_msg_pos.page_pos = 0;
-			con->in_msg_pos.page++;
+		if (m->pages) {
+			ret = read_partial_message_pages(con, m->pages,
+						 data_len, datacrc);
+			if (ret <= 0)
+				return ret;
+#ifdef CONFIG_BLOCK
+		} else if (m->bio) {
+
+			ret = read_partial_message_bio(con,
+						 &m->bio_iter, &m->bio_seg,
+						 data_len, datacrc);
+			if (ret <= 0)
+				return ret;
+#endif
+		} else {
+			BUG_ON(1);
 		}
 	}
 
@@ -1874,9 +2032,9 @@
 static void ceph_fault(struct ceph_connection *con)
 {
 	pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
-	       pr_addr(&con->peer_addr.in_addr), con->error_msg);
+	       ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
 	dout("fault %p state %lu to peer %s\n",
-	     con, con->state, pr_addr(&con->peer_addr.in_addr));
+	     con, con->state, ceph_pr_addr(&con->peer_addr.in_addr));
 
 	if (test_bit(LOSSYTX, &con->state)) {
 		dout("fault on LOSSYTX channel\n");
@@ -1936,7 +2094,9 @@
 /*
  * create a new messenger instance
  */
-struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
+struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr,
+					     u32 supported_features,
+					     u32 required_features)
 {
 	struct ceph_messenger *msgr;
 
@@ -1944,6 +2104,9 @@
 	if (msgr == NULL)
 		return ERR_PTR(-ENOMEM);
 
+	msgr->supported_features = supported_features;
+	msgr->required_features = required_features;
+
 	spin_lock_init(&msgr->global_seq_lock);
 
 	/* the zero page is needed if a request is "canceled" while the message
@@ -1966,6 +2129,7 @@
 	dout("messenger_create %p\n", msgr);
 	return msgr;
 }
+EXPORT_SYMBOL(ceph_messenger_create);
 
 void ceph_messenger_destroy(struct ceph_messenger *msgr)
 {
@@ -1975,6 +2139,7 @@
 	kfree(msgr);
 	dout("destroyed messenger %p\n", msgr);
 }
+EXPORT_SYMBOL(ceph_messenger_destroy);
 
 /*
  * Queue up an outgoing message on the given connection.
@@ -2011,6 +2176,7 @@
 	if (test_and_set_bit(WRITE_PENDING, &con->state) == 0)
 		queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_send);
 
 /*
  * Revoke a message that was previously queued for send
@@ -2076,6 +2242,7 @@
 	    test_and_set_bit(WRITE_PENDING, &con->state) == 0)
 		queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_keepalive);
 
 
 /*
@@ -2136,6 +2303,10 @@
 	m->nr_pages = 0;
 	m->pages = NULL;
 	m->pagelist = NULL;
+	m->bio = NULL;
+	m->bio_iter = NULL;
+	m->bio_seg = 0;
+	m->trail = NULL;
 
 	dout("ceph_msg_new %p front %d\n", m, front_len);
 	return m;
@@ -2146,6 +2317,7 @@
 	pr_err("msg_new can't create type %d front %d\n", type, front_len);
 	return NULL;
 }
+EXPORT_SYMBOL(ceph_msg_new);
 
 /*
  * Allocate "middle" portion of a message, if it is needed and wasn't
@@ -2250,11 +2422,14 @@
 		m->pagelist = NULL;
 	}
 
+	m->trail = NULL;
+
 	if (m->pool)
 		ceph_msgpool_put(m->pool, m);
 	else
 		ceph_msg_kfree(m);
 }
+EXPORT_SYMBOL(ceph_msg_last_put);
 
 void ceph_msg_dump(struct ceph_msg *msg)
 {
@@ -2275,3 +2450,4 @@
 		       DUMP_PREFIX_OFFSET, 16, 1,
 		       &msg->footer, sizeof(msg->footer), true);
 }
+EXPORT_SYMBOL(ceph_msg_dump);
diff --git a/fs/ceph/mon_client.c b/net/ceph/mon_client.c
similarity index 94%
rename from fs/ceph/mon_client.c
rename to net/ceph/mon_client.c
index b2a5a3e..8a07939 100644
--- a/fs/ceph/mon_client.c
+++ b/net/ceph/mon_client.c
@@ -1,14 +1,16 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/random.h>
 #include <linux/sched.h>
 
-#include "mon_client.h"
-#include "super.h"
-#include "auth.h"
-#include "decode.h"
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/decode.h>
+
+#include <linux/ceph/auth.h>
 
 /*
  * Interact with Ceph monitor cluster.  Handle requests for new map
@@ -74,7 +76,7 @@
 	     m->num_mon);
 	for (i = 0; i < m->num_mon; i++)
 		dout("monmap_decode  mon%d is %s\n", i,
-		     pr_addr(&m->mon_inst[i].addr.in_addr));
+		     ceph_pr_addr(&m->mon_inst[i].addr.in_addr));
 	return m;
 
 bad:
@@ -191,30 +193,33 @@
 		struct ceph_msg *msg = monc->m_subscribe;
 		struct ceph_mon_subscribe_item *i;
 		void *p, *end;
+		int num;
 
 		p = msg->front.iov_base;
 		end = p + msg->front_max;
 
-		dout("__send_subscribe to 'mdsmap' %u+\n",
-		     (unsigned)monc->have_mdsmap);
+		num = 1 + !!monc->want_next_osdmap + !!monc->want_mdsmap;
+		ceph_encode_32(&p, num);
+
 		if (monc->want_next_osdmap) {
 			dout("__send_subscribe to 'osdmap' %u\n",
 			     (unsigned)monc->have_osdmap);
-			ceph_encode_32(&p, 3);
 			ceph_encode_string(&p, end, "osdmap", 6);
 			i = p;
 			i->have = cpu_to_le64(monc->have_osdmap);
 			i->onetime = 1;
 			p += sizeof(*i);
 			monc->want_next_osdmap = 2;  /* requested */
-		} else {
-			ceph_encode_32(&p, 2);
 		}
-		ceph_encode_string(&p, end, "mdsmap", 6);
-		i = p;
-		i->have = cpu_to_le64(monc->have_mdsmap);
-		i->onetime = 0;
-		p += sizeof(*i);
+		if (monc->want_mdsmap) {
+			dout("__send_subscribe to 'mdsmap' %u+\n",
+			     (unsigned)monc->have_mdsmap);
+			ceph_encode_string(&p, end, "mdsmap", 6);
+			i = p;
+			i->have = cpu_to_le64(monc->have_mdsmap);
+			i->onetime = 0;
+			p += sizeof(*i);
+		}
 		ceph_encode_string(&p, end, "monmap", 6);
 		i = p;
 		i->have = 0;
@@ -243,7 +248,8 @@
 	mutex_lock(&monc->mutex);
 	if (monc->hunting) {
 		pr_info("mon%d %s session established\n",
-			monc->cur_mon, pr_addr(&monc->con->peer_addr.in_addr));
+			monc->cur_mon,
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
 		monc->hunting = false;
 	}
 	dout("handle_subscribe_ack after %d seconds\n", seconds);
@@ -266,6 +272,7 @@
 	mutex_unlock(&monc->mutex);
 	return 0;
 }
+EXPORT_SYMBOL(ceph_monc_got_mdsmap);
 
 int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 got)
 {
@@ -310,6 +317,7 @@
 	mutex_unlock(&monc->mutex);
 	return 0;
 }
+EXPORT_SYMBOL(ceph_monc_open_session);
 
 /*
  * The monitor responds with mount ack indicate mount success.  The
@@ -540,6 +548,7 @@
 	kref_put(&req->kref, release_generic_request);
 	return err;
 }
+EXPORT_SYMBOL(ceph_monc_do_statfs);
 
 /*
  * pool ops
@@ -651,6 +660,7 @@
 				   pool, 0, (char *)snapid, sizeof(*snapid));
 
 }
+EXPORT_SYMBOL(ceph_monc_create_snapid);
 
 int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
 			    u32 pool, u64 snapid)
@@ -708,9 +718,9 @@
  */
 static int build_initial_monmap(struct ceph_mon_client *monc)
 {
-	struct ceph_mount_args *args = monc->client->mount_args;
-	struct ceph_entity_addr *mon_addr = args->mon_addr;
-	int num_mon = args->num_mon;
+	struct ceph_options *opt = monc->client->options;
+	struct ceph_entity_addr *mon_addr = opt->mon_addr;
+	int num_mon = opt->num_mon;
 	int i;
 
 	/* build initial monmap */
@@ -728,11 +738,6 @@
 	}
 	monc->monmap->num_mon = num_mon;
 	monc->have_fsid = false;
-
-	/* release addr memory */
-	kfree(args->mon_addr);
-	args->mon_addr = NULL;
-	args->num_mon = 0;
 	return 0;
 }
 
@@ -753,8 +758,8 @@
 	monc->con = NULL;
 
 	/* authentication */
-	monc->auth = ceph_auth_init(cl->mount_args->name,
-				    cl->mount_args->secret);
+	monc->auth = ceph_auth_init(cl->options->name,
+				    cl->options->secret);
 	if (IS_ERR(monc->auth))
 		return PTR_ERR(monc->auth);
 	monc->auth->want_keys =
@@ -808,6 +813,7 @@
 out:
 	return err;
 }
+EXPORT_SYMBOL(ceph_monc_init);
 
 void ceph_monc_stop(struct ceph_mon_client *monc)
 {
@@ -832,6 +838,7 @@
 
 	kfree(monc->monmap);
 }
+EXPORT_SYMBOL(ceph_monc_stop);
 
 static void handle_auth_reply(struct ceph_mon_client *monc,
 			      struct ceph_msg *msg)
@@ -889,6 +896,7 @@
 	mutex_unlock(&monc->mutex);
 	return ret;
 }
+EXPORT_SYMBOL(ceph_monc_validate_auth);
 
 /*
  * handle incoming message
@@ -922,15 +930,16 @@
 		ceph_monc_handle_map(monc, msg);
 		break;
 
-	case CEPH_MSG_MDS_MAP:
-		ceph_mdsc_handle_map(&monc->client->mdsc, msg);
-		break;
-
 	case CEPH_MSG_OSD_MAP:
 		ceph_osdc_handle_map(&monc->client->osdc, msg);
 		break;
 
 	default:
+		/* can the chained handler handle it? */
+		if (monc->client->extra_mon_dispatch &&
+		    monc->client->extra_mon_dispatch(monc->client, msg) == 0)
+			break;
+			
 		pr_err("received unknown message type %d %s\n", type,
 		       ceph_msg_type_name(type));
 	}
@@ -994,7 +1003,7 @@
 	if (monc->con && !monc->hunting)
 		pr_info("mon%d %s session lost, "
 			"hunting for new mon\n", monc->cur_mon,
-			pr_addr(&monc->con->peer_addr.in_addr));
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
 
 	__close_session(monc);
 	if (!monc->hunting) {
diff --git a/fs/ceph/msgpool.c b/net/ceph/msgpool.c
similarity index 95%
rename from fs/ceph/msgpool.c
rename to net/ceph/msgpool.c
index dd65a64..d5f2d97 100644
--- a/fs/ceph/msgpool.c
+++ b/net/ceph/msgpool.c
@@ -1,11 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/sched.h>
 #include <linux/types.h>
 #include <linux/vmalloc.h>
 
-#include "msgpool.h"
+#include <linux/ceph/msgpool.h>
 
 static void *alloc_fn(gfp_t gfp_mask, void *arg)
 {
diff --git a/fs/ceph/osd_client.c b/net/ceph/osd_client.c
similarity index 84%
rename from fs/ceph/osd_client.c
rename to net/ceph/osd_client.c
index dfced1d..7939199 100644
--- a/fs/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -1,17 +1,22 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/err.h>
 #include <linux/highmem.h>
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
+#ifdef CONFIG_BLOCK
+#include <linux/bio.h>
+#endif
 
-#include "super.h"
-#include "osd_client.h"
-#include "messenger.h"
-#include "decode.h"
-#include "auth.h"
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/osd_client.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/pagelist.h>
 
 #define OSD_OP_FRONT_LEN	4096
 #define OSD_OPREPLY_FRONT_LEN	512
@@ -22,6 +27,59 @@
 
 static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd);
 
+static int op_needs_trail(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+	case CEPH_OSD_OP_CALL:
+		return 1;
+	default:
+		return 0;
+	}
+}
+
+static int op_has_extent(int op)
+{
+	return (op == CEPH_OSD_OP_READ ||
+		op == CEPH_OSD_OP_WRITE);
+}
+
+void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
+{
+	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
+	u64 orig_len = *plen;
+	u64 objoff, objlen;    /* extent in object */
+
+	reqhead->snapid = cpu_to_le64(snapid);
+
+	/* object extent? */
+	ceph_calc_file_object_mapping(layout, off, plen, bno,
+				      &objoff, &objlen);
+	if (*plen < orig_len)
+		dout(" skipping last %llu, final file extent %llu~%llu\n",
+		     orig_len - *plen, off, *plen);
+
+	if (op_has_extent(op->op)) {
+		op->extent.offset = objoff;
+		op->extent.length = objlen;
+	}
+	req->r_num_pages = calc_pages_for(off, *plen);
+	if (op->op == CEPH_OSD_OP_WRITE)
+		op->payload_len = *plen;
+
+	dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
+	     *bno, objoff, objlen, req->r_num_pages);
+
+}
+EXPORT_SYMBOL(ceph_calc_raw_layout);
+
 /*
  * Implement client access to distributed object storage cluster.
  *
@@ -48,34 +106,19 @@
  * fill osd op in request message.
  */
 static void calc_layout(struct ceph_osd_client *osdc,
-			struct ceph_vino vino, struct ceph_file_layout *layout,
+			struct ceph_vino vino,
+			struct ceph_file_layout *layout,
 			u64 off, u64 *plen,
-			struct ceph_osd_request *req)
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
 {
-	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
-	struct ceph_osd_op *op = (void *)(reqhead + 1);
-	u64 orig_len = *plen;
-	u64 objoff, objlen;    /* extent in object */
 	u64 bno;
 
-	reqhead->snapid = cpu_to_le64(vino.snap);
-
-	/* object extent? */
-	ceph_calc_file_object_mapping(layout, off, plen, &bno,
-				      &objoff, &objlen);
-	if (*plen < orig_len)
-		dout(" skipping last %llu, final file extent %llu~%llu\n",
-		     orig_len - *plen, off, *plen);
+	ceph_calc_raw_layout(osdc, layout, vino.snap, off,
+			     plen, &bno, req, op);
 
 	sprintf(req->r_oid, "%llx.%08llx", vino.ino, bno);
 	req->r_oid_len = strlen(req->r_oid);
-
-	op->extent.offset = cpu_to_le64(objoff);
-	op->extent.length = cpu_to_le64(objlen);
-	req->r_num_pages = calc_pages_for(off, *plen);
-
-	dout("calc_layout %s (%d) %llu~%llu (%d pages)\n",
-	     req->r_oid, req->r_oid_len, objoff, objlen, req->r_num_pages);
 }
 
 /*
@@ -101,12 +144,259 @@
 	if (req->r_own_pages)
 		ceph_release_page_vector(req->r_pages,
 					 req->r_num_pages);
+#ifdef CONFIG_BLOCK
+	if (req->r_bio)
+		bio_put(req->r_bio);
+#endif
 	ceph_put_snap_context(req->r_snapc);
+	if (req->r_trail) {
+		ceph_pagelist_release(req->r_trail);
+		kfree(req->r_trail);
+	}
 	if (req->r_mempool)
 		mempool_free(req, req->r_osdc->req_mempool);
 	else
 		kfree(req);
 }
+EXPORT_SYMBOL(ceph_osdc_release_request);
+
+static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail)
+{
+	int i = 0;
+
+	if (needs_trail)
+		*needs_trail = 0;
+	while (ops[i].op) {
+		if (needs_trail && op_needs_trail(ops[i].op))
+			*needs_trail = 1;
+		i++;
+	}
+
+	return i;
+}
+
+struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio)
+{
+	struct ceph_osd_request *req;
+	struct ceph_msg *msg;
+	int needs_trail;
+	int num_op = get_num_ops(ops, &needs_trail);
+	size_t msg_size = sizeof(struct ceph_osd_request_head);
+
+	msg_size += num_op*sizeof(struct ceph_osd_op);
+
+	if (use_mempool) {
+		req = mempool_alloc(osdc->req_mempool, gfp_flags);
+		memset(req, 0, sizeof(*req));
+	} else {
+		req = kzalloc(sizeof(*req), gfp_flags);
+	}
+	if (req == NULL)
+		return NULL;
+
+	req->r_osdc = osdc;
+	req->r_mempool = use_mempool;
+
+	kref_init(&req->r_kref);
+	init_completion(&req->r_completion);
+	init_completion(&req->r_safe_completion);
+	INIT_LIST_HEAD(&req->r_unsafe_item);
+	req->r_flags = flags;
+
+	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
+
+	/* create reply message */
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
+				   OSD_OPREPLY_FRONT_LEN, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+	req->r_reply = msg;
+
+	/* allocate space for the trailing data */
+	if (needs_trail) {
+		req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags);
+		if (!req->r_trail) {
+			ceph_osdc_put_request(req);
+			return NULL;
+		}
+		ceph_pagelist_init(req->r_trail);
+	}
+	/* create request message; allow space for oid */
+	msg_size += 40;
+	if (snapc)
+		msg_size += sizeof(u64) * snapc->num_snaps;
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+
+	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
+	memset(msg->front.iov_base, 0, msg->front.iov_len);
+
+	req->r_request = msg;
+	req->r_pages = pages;
+#ifdef CONFIG_BLOCK
+	if (bio) {
+		req->r_bio = bio;
+		bio_get(req->r_bio);
+	}
+#endif
+
+	return req;
+}
+EXPORT_SYMBOL(ceph_osdc_alloc_request);
+
+static void osd_req_encode_op(struct ceph_osd_request *req,
+			      struct ceph_osd_op *dst,
+			      struct ceph_osd_req_op *src)
+{
+	dst->op = cpu_to_le16(src->op);
+
+	switch (dst->op) {
+	case CEPH_OSD_OP_READ:
+	case CEPH_OSD_OP_WRITE:
+		dst->extent.offset =
+			cpu_to_le64(src->extent.offset);
+		dst->extent.length =
+			cpu_to_le64(src->extent.length);
+		dst->extent.truncate_size =
+			cpu_to_le64(src->extent.truncate_size);
+		dst->extent.truncate_seq =
+			cpu_to_le32(src->extent.truncate_seq);
+		break;
+
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+		BUG_ON(!req->r_trail);
+
+		dst->xattr.name_len = cpu_to_le32(src->xattr.name_len);
+		dst->xattr.value_len = cpu_to_le32(src->xattr.value_len);
+		dst->xattr.cmp_op = src->xattr.cmp_op;
+		dst->xattr.cmp_mode = src->xattr.cmp_mode;
+		ceph_pagelist_append(req->r_trail, src->xattr.name,
+				     src->xattr.name_len);
+		ceph_pagelist_append(req->r_trail, src->xattr.val,
+				     src->xattr.value_len);
+		break;
+	case CEPH_OSD_OP_CALL:
+		BUG_ON(!req->r_trail);
+
+		dst->cls.class_len = src->cls.class_len;
+		dst->cls.method_len = src->cls.method_len;
+		dst->cls.indata_len = cpu_to_le32(src->cls.indata_len);
+
+		ceph_pagelist_append(req->r_trail, src->cls.class_name,
+				     src->cls.class_len);
+		ceph_pagelist_append(req->r_trail, src->cls.method_name,
+				     src->cls.method_len);
+		ceph_pagelist_append(req->r_trail, src->cls.indata,
+				     src->cls.indata_len);
+		break;
+	case CEPH_OSD_OP_ROLLBACK:
+		dst->snap.snapid = cpu_to_le64(src->snap.snapid);
+		break;
+	case CEPH_OSD_OP_STARTSYNC:
+		break;
+	default:
+		pr_err("unrecognized osd opcode %d\n", dst->op);
+		WARN_ON(1);
+		break;
+	}
+	dst->payload_len = cpu_to_le32(src->payload_len);
+}
+
+/*
+ * build new request AND message
+ *
+ */
+void ceph_osdc_build_request(struct ceph_osd_request *req,
+			     u64 off, u64 *plen,
+			     struct ceph_osd_req_op *src_ops,
+			     struct ceph_snap_context *snapc,
+			     struct timespec *mtime,
+			     const char *oid,
+			     int oid_len)
+{
+	struct ceph_msg *msg = req->r_request;
+	struct ceph_osd_request_head *head;
+	struct ceph_osd_req_op *src_op;
+	struct ceph_osd_op *op;
+	void *p;
+	int num_op = get_num_ops(src_ops, NULL);
+	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
+	int flags = req->r_flags;
+	u64 data_len = 0;
+	int i;
+
+	head = msg->front.iov_base;
+	op = (void *)(head + 1);
+	p = (void *)(op + num_op);
+
+	req->r_snapc = ceph_get_snap_context(snapc);
+
+	head->client_inc = cpu_to_le32(1); /* always, for now. */
+	head->flags = cpu_to_le32(flags);
+	if (flags & CEPH_OSD_FLAG_WRITE)
+		ceph_encode_timespec(&head->mtime, mtime);
+	head->num_ops = cpu_to_le16(num_op);
+
+
+	/* fill in oid */
+	head->object_len = cpu_to_le32(oid_len);
+	memcpy(p, oid, oid_len);
+	p += oid_len;
+
+	src_op = src_ops;
+	while (src_op->op) {
+		osd_req_encode_op(req, op, src_op);
+		src_op++;
+		op++;
+	}
+
+	if (req->r_trail)
+		data_len += req->r_trail->length;
+
+	if (snapc) {
+		head->snap_seq = cpu_to_le64(snapc->seq);
+		head->num_snaps = cpu_to_le32(snapc->num_snaps);
+		for (i = 0; i < snapc->num_snaps; i++) {
+			put_unaligned_le64(snapc->snaps[i], p);
+			p += sizeof(u64);
+		}
+	}
+
+	if (flags & CEPH_OSD_FLAG_WRITE) {
+		req->r_request->hdr.data_off = cpu_to_le16(off);
+		req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len);
+	} else if (data_len) {
+		req->r_request->hdr.data_off = 0;
+		req->r_request->hdr.data_len = cpu_to_le32(data_len);
+	}
+
+	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
+	msg_size = p - msg->front.iov_base;
+	msg->front.iov_len = msg_size;
+	msg->hdr.front_len = cpu_to_le32(msg_size);
+	return;
+}
+EXPORT_SYMBOL(ceph_osdc_build_request);
 
 /*
  * build new request AND message, calculate layout, and adjust file
@@ -131,110 +421,40 @@
 					       struct timespec *mtime,
 					       bool use_mempool, int num_reply)
 {
+	struct ceph_osd_req_op ops[3];
 	struct ceph_osd_request *req;
-	struct ceph_msg *msg;
-	struct ceph_osd_request_head *head;
-	struct ceph_osd_op *op;
-	void *p;
-	int num_op = 1 + do_sync;
-	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
-	int i;
 
-	if (use_mempool) {
-		req = mempool_alloc(osdc->req_mempool, GFP_NOFS);
-		memset(req, 0, sizeof(*req));
-	} else {
-		req = kzalloc(sizeof(*req), GFP_NOFS);
-	}
-	if (req == NULL)
-		return NULL;
-
-	req->r_osdc = osdc;
-	req->r_mempool = use_mempool;
-	kref_init(&req->r_kref);
-	init_completion(&req->r_completion);
-	init_completion(&req->r_safe_completion);
-	INIT_LIST_HEAD(&req->r_unsafe_item);
-	req->r_flags = flags;
-
-	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
-
-	/* create reply message */
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
-				   OSD_OPREPLY_FRONT_LEN, GFP_NOFS);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-	req->r_reply = msg;
-
-	/* create request message; allow space for oid */
-	msg_size += 40;
-	if (snapc)
-		msg_size += sizeof(u64) * snapc->num_snaps;
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, GFP_NOFS);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
-	memset(msg->front.iov_base, 0, msg->front.iov_len);
-	head = msg->front.iov_base;
-	op = (void *)(head + 1);
-	p = (void *)(op + num_op);
-
-	req->r_request = msg;
-	req->r_snapc = ceph_get_snap_context(snapc);
-
-	head->client_inc = cpu_to_le32(1); /* always, for now. */
-	head->flags = cpu_to_le32(flags);
-	if (flags & CEPH_OSD_FLAG_WRITE)
-		ceph_encode_timespec(&head->mtime, mtime);
-	head->num_ops = cpu_to_le16(num_op);
-	op->op = cpu_to_le16(opcode);
-
-	/* calculate max write size */
-	calc_layout(osdc, vino, layout, off, plen, req);
-	req->r_file_layout = *layout;  /* keep a copy */
-
-	if (flags & CEPH_OSD_FLAG_WRITE) {
-		req->r_request->hdr.data_off = cpu_to_le16(off);
-		req->r_request->hdr.data_len = cpu_to_le32(*plen);
-		op->payload_len = cpu_to_le32(*plen);
-	}
-	op->extent.truncate_size = cpu_to_le64(truncate_size);
-	op->extent.truncate_seq = cpu_to_le32(truncate_seq);
-
-	/* fill in oid */
-	head->object_len = cpu_to_le32(req->r_oid_len);
-	memcpy(p, req->r_oid, req->r_oid_len);
-	p += req->r_oid_len;
+	ops[0].op = opcode;
+	ops[0].extent.truncate_seq = truncate_seq;
+	ops[0].extent.truncate_size = truncate_size;
+	ops[0].payload_len = 0;
 
 	if (do_sync) {
-		op++;
-		op->op = cpu_to_le16(CEPH_OSD_OP_STARTSYNC);
-	}
-	if (snapc) {
-		head->snap_seq = cpu_to_le64(snapc->seq);
-		head->num_snaps = cpu_to_le32(snapc->num_snaps);
-		for (i = 0; i < snapc->num_snaps; i++) {
-			put_unaligned_le64(snapc->snaps[i], p);
-			p += sizeof(u64);
-		}
-	}
+		ops[1].op = CEPH_OSD_OP_STARTSYNC;
+		ops[1].payload_len = 0;
+		ops[2].op = 0;
+	} else
+		ops[1].op = 0;
 
-	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
-	msg_size = p - msg->front.iov_base;
-	msg->front.iov_len = msg_size;
-	msg->hdr.front_len = cpu_to_le32(msg_size);
+	req = ceph_osdc_alloc_request(osdc, flags,
+					 snapc, ops,
+					 use_mempool,
+					 GFP_NOFS, NULL, NULL);
+	if (IS_ERR(req))
+		return req;
+
+	/* calculate max write size */
+	calc_layout(osdc, vino, layout, off, plen, req, ops);
+	req->r_file_layout = *layout;  /* keep a copy */
+
+	ceph_osdc_build_request(req, off, plen, ops,
+				snapc,
+				mtime,
+				req->r_oid, req->r_oid_len);
+
 	return req;
 }
+EXPORT_SYMBOL(ceph_osdc_new_request);
 
 /*
  * We keep osd requests in an rbtree, sorted by ->r_tid.
@@ -389,7 +609,7 @@
 	dout("__move_osd_to_lru %p\n", osd);
 	BUG_ON(!list_empty(&osd->o_osd_lru));
 	list_add_tail(&osd->o_osd_lru, &osdc->osd_lru);
-	osd->lru_ttl = jiffies + osdc->client->mount_args->osd_idle_ttl * HZ;
+	osd->lru_ttl = jiffies + osdc->client->options->osd_idle_ttl * HZ;
 }
 
 static void __remove_osd_from_lru(struct ceph_osd *osd)
@@ -483,7 +703,7 @@
 static void __schedule_osd_timeout(struct ceph_osd_client *osdc)
 {
 	schedule_delayed_work(&osdc->timeout_work,
-			osdc->client->mount_args->osd_keepalive_timeout * HZ);
+			osdc->client->options->osd_keepalive_timeout * HZ);
 }
 
 static void __cancel_osd_timeout(struct ceph_osd_client *osdc)
@@ -549,7 +769,7 @@
  */
 static void __cancel_request(struct ceph_osd_request *req)
 {
-	if (req->r_sent) {
+	if (req->r_sent && req->r_osd) {
 		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
 		req->r_sent = 0;
 	}
@@ -684,9 +904,9 @@
 		container_of(work, struct ceph_osd_client, timeout_work.work);
 	struct ceph_osd_request *req, *last_req = NULL;
 	struct ceph_osd *osd;
-	unsigned long timeout = osdc->client->mount_args->osd_timeout * HZ;
+	unsigned long timeout = osdc->client->options->osd_timeout * HZ;
 	unsigned long keepalive =
-		osdc->client->mount_args->osd_keepalive_timeout * HZ;
+		osdc->client->options->osd_keepalive_timeout * HZ;
 	unsigned long last_stamp = 0;
 	struct rb_node *p;
 	struct list_head slow_osds;
@@ -773,7 +993,7 @@
 		container_of(work, struct ceph_osd_client,
 			     osds_timeout_work.work);
 	unsigned long delay =
-		osdc->client->mount_args->osd_idle_ttl * HZ >> 2;
+		osdc->client->options->osd_idle_ttl * HZ >> 2;
 
 	dout("osds timeout\n");
 	down_read(&osdc->map_sem);
@@ -1104,6 +1324,10 @@
 
 	req->r_request->pages = req->r_pages;
 	req->r_request->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+	req->r_request->bio = req->r_bio;
+#endif
+	req->r_request->trail = req->r_trail;
 
 	register_request(osdc, req);
 
@@ -1131,6 +1355,7 @@
 	up_read(&osdc->map_sem);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_start_request);
 
 /*
  * wait for a request to complete
@@ -1153,6 +1378,7 @@
 	dout("wait_request tid %llu result %d\n", req->r_tid, req->r_result);
 	return req->r_result;
 }
+EXPORT_SYMBOL(ceph_osdc_wait_request);
 
 /*
  * sync - wait for all in-flight requests to flush.  avoid starvation.
@@ -1186,6 +1412,7 @@
 	mutex_unlock(&osdc->request_mutex);
 	dout("sync done (thru tid %llu)\n", last_tid);
 }
+EXPORT_SYMBOL(ceph_osdc_sync);
 
 /*
  * init, shutdown
@@ -1211,7 +1438,7 @@
 	INIT_DELAYED_WORK(&osdc->osds_timeout_work, handle_osds_timeout);
 
 	schedule_delayed_work(&osdc->osds_timeout_work,
-	   round_jiffies_relative(osdc->client->mount_args->osd_idle_ttl * HZ));
+	   round_jiffies_relative(osdc->client->options->osd_idle_ttl * HZ));
 
 	err = -ENOMEM;
 	osdc->req_mempool = mempool_create_kmalloc_pool(10,
@@ -1237,6 +1464,7 @@
 out:
 	return err;
 }
+EXPORT_SYMBOL(ceph_osdc_init);
 
 void ceph_osdc_stop(struct ceph_osd_client *osdc)
 {
@@ -1251,6 +1479,7 @@
 	ceph_msgpool_destroy(&osdc->msgpool_op);
 	ceph_msgpool_destroy(&osdc->msgpool_op_reply);
 }
+EXPORT_SYMBOL(ceph_osdc_stop);
 
 /*
  * Read some contiguous pages.  If we cross a stripe boundary, shorten
@@ -1288,6 +1517,7 @@
 	dout("readpages result %d\n", rc);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_readpages);
 
 /*
  * do a synchronous write on N pages
@@ -1330,6 +1560,7 @@
 	dout("writepages result %d\n", rc);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_writepages);
 
 /*
  * handle incoming message
@@ -1420,6 +1651,9 @@
 		}
 		m->pages = req->r_pages;
 		m->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+		m->bio = req->r_bio;
+#endif
 	}
 	*skip = 0;
 	req->r_con_filling_msg = ceph_con_get(con);
diff --git a/fs/ceph/osdmap.c b/net/ceph/osdmap.c
similarity index 97%
rename from fs/ceph/osdmap.c
rename to net/ceph/osdmap.c
index e31f118..d73f3f6 100644
--- a/fs/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -1,14 +1,15 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/slab.h>
 #include <asm/div64.h>
 
-#include "super.h"
-#include "osdmap.h"
-#include "crush/hash.h"
-#include "crush/mapper.h"
-#include "decode.h"
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/osdmap.h>
+#include <linux/ceph/decode.h>
+#include <linux/crush/hash.h>
+#include <linux/crush/mapper.h>
 
 char *ceph_osdmap_state_str(char *str, int len, int state)
 {
@@ -417,6 +418,20 @@
 	return NULL;
 }
 
+int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name)
+{
+	struct rb_node *rbp;
+
+	for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) {
+		struct ceph_pg_pool_info *pi =
+			rb_entry(rbp, struct ceph_pg_pool_info, node);
+		if (pi->name && strcmp(pi->name, name) == 0)
+			return pi->id;
+	}
+	return -ENOENT;
+}
+EXPORT_SYMBOL(ceph_pg_poolid_by_name);
+
 static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
 {
 	rb_erase(&pi->node, root);
@@ -966,6 +981,7 @@
 
 	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
 }
+EXPORT_SYMBOL(ceph_calc_file_object_mapping);
 
 /*
  * calculate an object layout (i.e. pgid) from an oid,
@@ -1011,6 +1027,7 @@
 	ol->ol_stripe_unit = fl->fl_object_stripe_unit;
 	return 0;
 }
+EXPORT_SYMBOL(ceph_calc_object_layout);
 
 /*
  * Calculate raw osd vector for the given pgid.  Return pointer to osd
@@ -1108,3 +1125,4 @@
 			return osds[i];
 	return -1;
 }
+EXPORT_SYMBOL(ceph_calc_pg_primary);
diff --git a/net/ceph/pagelist.c b/net/ceph/pagelist.c
new file mode 100644
index 0000000..13cb409
--- /dev/null
+++ b/net/ceph/pagelist.c
@@ -0,0 +1,154 @@
+
+#include <linux/module.h>
+#include <linux/gfp.h>
+#include <linux/pagemap.h>
+#include <linux/highmem.h>
+#include <linux/ceph/pagelist.h>
+
+static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
+{
+	if (pl->mapped_tail) {
+		struct page *page = list_entry(pl->head.prev, struct page, lru);
+		kunmap(page);
+		pl->mapped_tail = NULL;
+	}
+}
+
+int ceph_pagelist_release(struct ceph_pagelist *pl)
+{
+	ceph_pagelist_unmap_tail(pl);
+	while (!list_empty(&pl->head)) {
+		struct page *page = list_first_entry(&pl->head, struct page,
+						     lru);
+		list_del(&page->lru);
+		__free_page(page);
+	}
+	ceph_pagelist_free_reserve(pl);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_release);
+
+static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
+{
+	struct page *page;
+
+	if (!pl->num_pages_free) {
+		page = __page_cache_alloc(GFP_NOFS);
+	} else {
+		page = list_first_entry(&pl->free_list, struct page, lru);
+		list_del(&page->lru);
+		--pl->num_pages_free;
+	}
+	if (!page)
+		return -ENOMEM;
+	pl->room += PAGE_SIZE;
+	ceph_pagelist_unmap_tail(pl);
+	list_add_tail(&page->lru, &pl->head);
+	pl->mapped_tail = kmap(page);
+	return 0;
+}
+
+int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len)
+{
+	while (pl->room < len) {
+		size_t bit = pl->room;
+		int ret;
+
+		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
+		       buf, bit);
+		pl->length += bit;
+		pl->room -= bit;
+		buf += bit;
+		len -= bit;
+		ret = ceph_pagelist_addpage(pl);
+		if (ret)
+			return ret;
+	}
+
+	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
+	pl->length += len;
+	pl->room -= len;
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_append);
+
+/**
+ * Allocate enough pages for a pagelist to append the given amount
+ * of data without without allocating.
+ * Returns: 0 on success, -ENOMEM on error.
+ */
+int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space)
+{
+	if (space <= pl->room)
+		return 0;
+	space -= pl->room;
+	space = (space + PAGE_SIZE - 1) >> PAGE_SHIFT;   /* conv to num pages */
+
+	while (space > pl->num_pages_free) {
+		struct page *page = __page_cache_alloc(GFP_NOFS);
+		if (!page)
+			return -ENOMEM;
+		list_add_tail(&page->lru, &pl->free_list);
+		++pl->num_pages_free;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_reserve);
+
+/**
+ * Free any pages that have been preallocated.
+ */
+int ceph_pagelist_free_reserve(struct ceph_pagelist *pl)
+{
+	while (!list_empty(&pl->free_list)) {
+		struct page *page = list_first_entry(&pl->free_list,
+						     struct page, lru);
+		list_del(&page->lru);
+		__free_page(page);
+		--pl->num_pages_free;
+	}
+	BUG_ON(pl->num_pages_free);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_free_reserve);
+
+/**
+ * Create a truncation point.
+ */
+void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+			      struct ceph_pagelist_cursor *c)
+{
+	c->pl = pl;
+	c->page_lru = pl->head.prev;
+	c->room = pl->room;
+}
+EXPORT_SYMBOL(ceph_pagelist_set_cursor);
+
+/**
+ * Truncate a pagelist to the given point. Move extra pages to reserve.
+ * This won't sleep.
+ * Returns: 0 on success,
+ *          -EINVAL if the pagelist doesn't match the trunc point pagelist
+ */
+int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+			   struct ceph_pagelist_cursor *c)
+{
+	struct page *page;
+
+	if (pl != c->pl)
+		return -EINVAL;
+	ceph_pagelist_unmap_tail(pl);
+	while (pl->head.prev != c->page_lru) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		list_del(&page->lru);                /* remove from pagelist */
+		list_add_tail(&page->lru, &pl->free_list); /* add to reserve */
+		++pl->num_pages_free;
+	}
+	pl->room = c->room;
+	if (!list_empty(&pl->head)) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		pl->mapped_tail = kmap(page);
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_truncate);
diff --git a/net/ceph/pagevec.c b/net/ceph/pagevec.c
new file mode 100644
index 0000000..54caf06
--- /dev/null
+++ b/net/ceph/pagevec.c
@@ -0,0 +1,223 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/file.h>
+#include <linux/namei.h>
+#include <linux/writeback.h>
+
+#include <linux/ceph/libceph.h>
+
+/*
+ * build a vector of user pages
+ */
+struct page **ceph_get_direct_page_vector(const char __user *data,
+						 int num_pages,
+						 loff_t off, size_t len)
+{
+	struct page **pages;
+	int rc;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+
+	down_read(&current->mm->mmap_sem);
+	rc = get_user_pages(current, current->mm, (unsigned long)data,
+			    num_pages, 0, 0, pages, NULL);
+	up_read(&current->mm->mmap_sem);
+	if (rc < 0)
+		goto fail;
+	return pages;
+
+fail:
+	kfree(pages);
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL(ceph_get_direct_page_vector);
+
+void ceph_put_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		put_page(pages[i]);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_put_page_vector);
+
+void ceph_release_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		__free_pages(pages[i], 0);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_release_page_vector);
+
+/*
+ * allocate a vector new pages
+ */
+struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
+{
+	struct page **pages;
+	int i;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, flags);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+	for (i = 0; i < num_pages; i++) {
+		pages[i] = __page_cache_alloc(flags);
+		if (pages[i] == NULL) {
+			ceph_release_page_vector(pages, i);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+	return pages;
+}
+EXPORT_SYMBOL(ceph_alloc_page_vector);
+
+/*
+ * copy user data into a page vector
+ */
+int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, PAGE_CACHE_SIZE-po, left);
+		bad = copy_from_user(page_address(pages[i]) + po, data, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		po += l - bad;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_user_to_page_vector);
+
+int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(page_address(pages[i]) + po, data, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_to_page_vector);
+
+int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(data, page_address(pages[i]) + po, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_from_page_vector);
+
+/*
+ * copy user data from a page vector into a user pointer
+ */
+int ceph_copy_page_vector_to_user(struct page **pages,
+					 char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, left, PAGE_CACHE_SIZE-po);
+		bad = copy_to_user(data, page_address(pages[i]) + po, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		if (po) {
+			po += l - bad;
+			if (po == PAGE_CACHE_SIZE)
+				po = 0;
+		}
+		i++;
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_page_vector_to_user);
+
+/*
+ * Zero an extent within a page vector.  Offset is relative to the
+ * start of the first page.
+ */
+void ceph_zero_page_vector_range(int off, int len, struct page **pages)
+{
+	int i = off >> PAGE_CACHE_SHIFT;
+
+	off &= ~PAGE_CACHE_MASK;
+
+	dout("zero_page_vector_page %u~%u\n", off, len);
+
+	/* leading partial page? */
+	if (off) {
+		int end = min((int)PAGE_CACHE_SIZE, off + len);
+		dout("zeroing %d %p head from %d\n", i, pages[i],
+		     (int)off);
+		zero_user_segment(pages[i], off, end);
+		len -= (end - off);
+		i++;
+	}
+	while (len >= PAGE_CACHE_SIZE) {
+		dout("zeroing %d %p len=%d\n", i, pages[i], len);
+		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
+		len -= PAGE_CACHE_SIZE;
+		i++;
+	}
+	/* trailing partial page? */
+	if (len) {
+		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
+		zero_user_segment(pages[i], 0, len);
+	}
+}
+EXPORT_SYMBOL(ceph_zero_page_vector_range);
+
diff --git a/net/core/datagram.c b/net/core/datagram.c
index 251997a..282806b 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -243,6 +243,7 @@
 	unlock_sock_fast(sk, slow);
 
 	/* skb is now orphaned, can be freed outside of locked section */
+	trace_kfree_skb(skb, skb_free_datagram_locked);
 	__kfree_skb(skb);
 }
 EXPORT_SYMBOL(skb_free_datagram_locked);
diff --git a/net/core/dev.c b/net/core/dev.c
index 660dd41..7ec85e2 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -128,6 +128,8 @@
 #include <linux/jhash.h>
 #include <linux/random.h>
 #include <trace/events/napi.h>
+#include <trace/events/net.h>
+#include <trace/events/skb.h>
 #include <linux/pci.h>
 
 #include "net-sysfs.h"
@@ -1978,6 +1980,7 @@
 		}
 
 		rc = ops->ndo_start_xmit(skb, dev);
+		trace_net_dev_xmit(skb, rc);
 		if (rc == NETDEV_TX_OK)
 			txq_trans_update(txq);
 		return rc;
@@ -1998,6 +2001,7 @@
 			skb_dst_drop(nskb);
 
 		rc = ops->ndo_start_xmit(nskb, dev);
+		trace_net_dev_xmit(nskb, rc);
 		if (unlikely(rc != NETDEV_TX_OK)) {
 			if (rc & ~NETDEV_TX_MASK)
 				goto out_kfree_gso_skb;
@@ -2186,6 +2190,7 @@
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS);
 #endif
+	trace_net_dev_queue(skb);
 	if (q->enqueue) {
 		rc = __dev_xmit_skb(skb, q, dev, txq);
 		goto out;
@@ -2512,6 +2517,7 @@
 	if (netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
+	trace_netif_rx(skb);
 #ifdef CONFIG_RPS
 	{
 		struct rps_dev_flow voidflow, *rflow = &voidflow;
@@ -2571,6 +2577,7 @@
 			clist = clist->next;
 
 			WARN_ON(atomic_read(&skb->users));
+			trace_kfree_skb(skb, net_tx_action);
 			__kfree_skb(skb);
 		}
 	}
@@ -2828,6 +2835,7 @@
 	if (!netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
+	trace_netif_receive_skb(skb);
 	if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb))
 		return NET_RX_SUCCESS;
 
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 7a85367..8451ab4 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -348,7 +348,7 @@
 	if (info.cmd == ETHTOOL_GRXCLSRLALL) {
 		if (info.rule_cnt > 0) {
 			if (info.rule_cnt <= KMALLOC_MAX_SIZE / sizeof(u32))
-				rule_buf = kmalloc(info.rule_cnt * sizeof(u32),
+				rule_buf = kzalloc(info.rule_cnt * sizeof(u32),
 						   GFP_USER);
 			if (!rule_buf)
 				return -ENOMEM;
@@ -397,7 +397,7 @@
 	    (KMALLOC_MAX_SIZE - sizeof(*indir)) / sizeof(*indir->ring_index))
 		return -ENOMEM;
 	full_size = sizeof(*indir) + sizeof(*indir->ring_index) * table_size;
-	indir = kmalloc(full_size, GFP_USER);
+	indir = kzalloc(full_size, GFP_USER);
 	if (!indir)
 		return -ENOMEM;
 
@@ -538,7 +538,7 @@
 
 	gstrings.len = ret;
 
-	data = kmalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER);
+	data = kzalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 
@@ -775,7 +775,7 @@
 	if (regs.len > reglen)
 		regs.len = reglen;
 
-	regbuf = kmalloc(reglen, GFP_USER);
+	regbuf = kzalloc(reglen, GFP_USER);
 	if (!regbuf)
 		return -ENOMEM;
 
diff --git a/net/core/net-traces.c b/net/core/net-traces.c
index afa6380..7f1bb2a 100644
--- a/net/core/net-traces.c
+++ b/net/core/net-traces.c
@@ -26,6 +26,7 @@
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/skb.h>
+#include <trace/events/net.h>
 #include <trace/events/napi.h>
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index c83b421..56ba3c4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -466,6 +466,7 @@
 		smp_rmb();
 	else if (likely(!atomic_dec_and_test(&skb->users)))
 		return;
+	trace_consume_skb(skb);
 	__kfree_skb(skb);
 }
 EXPORT_SYMBOL(consume_skb);
diff --git a/net/core/sock.c b/net/core/sock.c
index ef30e9d..7d99e13 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1078,8 +1078,11 @@
 #ifdef CONFIG_CGROUPS
 void sock_update_classid(struct sock *sk)
 {
-	u32 classid = task_cls_classid(current);
+	u32 classid;
 
+	rcu_read_lock();  /* doing current task, which cannot vanish. */
+	classid = task_cls_classid(current);
+	rcu_read_unlock();
 	if (classid && classid != sk->sk_classid)
 		sk->sk_classid = classid;
 }
diff --git a/net/core/stream.c b/net/core/stream.c
index d959e0f..f5df85d 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -141,10 +141,10 @@
 
 		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
 		sk->sk_write_pending++;
-		sk_wait_event(sk, &current_timeo, !sk->sk_err &&
-						  !(sk->sk_shutdown & SEND_SHUTDOWN) &&
-						  sk_stream_memory_free(sk) &&
-						  vm_wait);
+		sk_wait_event(sk, &current_timeo, sk->sk_err ||
+						  (sk->sk_shutdown & SEND_SHUTDOWN) ||
+						  (sk_stream_memory_free(sk) &&
+						  !vm_wait));
 		sk->sk_write_pending--;
 
 		if (vm_wait) {
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 72380a3..7cd7760 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -413,7 +413,7 @@
 	  If unsure, say Y.
 
 config INET_LRO
-	bool "Large Receive Offload (ipv4/tcp)"
+	tristate "Large Receive Offload (ipv4/tcp)"
 	default y
 	---help---
 	  Support for Large Receive Offload (ipv4/tcp).
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 1fdcacd..2a4bb76 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -834,7 +834,7 @@
 	int			mark = 0;
 
 
-	if (len == 8 || IGMP_V2_SEEN(in_dev)) {
+	if (len == 8) {
 		if (ih->code == 0) {
 			/* Alas, old v1 router presents here. */
 
@@ -856,6 +856,18 @@
 		igmpv3_clear_delrec(in_dev);
 	} else if (len < 12) {
 		return;	/* ignore bogus packet; freed by caller */
+	} else if (IGMP_V1_SEEN(in_dev)) {
+		/* This is a v3 query with v1 queriers present */
+		max_delay = IGMP_Query_Response_Interval;
+		group = 0;
+	} else if (IGMP_V2_SEEN(in_dev)) {
+		/* this is a v3 query with v2 queriers present;
+		 * Interpretation of the max_delay code is problematic here.
+		 * A real v2 host would use ih_code directly, while v3 has a
+		 * different encoding. We use the v3 encoding as more likely
+		 * to be intended in a v3 query.
+		 */
+		max_delay = IGMPV3_MRC(ih3->code)*(HZ/IGMP_TIMER_SCALE);
 	} else { /* v3 */
 		if (!pskb_may_pull(skb, sizeof(struct igmpv3_query)))
 			return;
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
index 244f7cb..37f8adb 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
@@ -11,6 +11,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/percpu.h>
+#include <linux/security.h>
 #include <net/net_namespace.h>
 
 #include <linux/netfilter.h>
@@ -87,6 +88,29 @@
 	rcu_read_unlock();
 }
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct)
+{
+	int ret;
+	u32 len;
+	char *secctx;
+
+	ret = security_secid_to_secctx(ct->secmark, &secctx, &len);
+	if (ret)
+		return ret;
+
+	ret = seq_printf(s, "secctx=%s ", secctx);
+
+	security_release_secctx(secctx, len);
+	return ret;
+}
+#else
+static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct)
+{
+	return 0;
+}
+#endif
+
 static int ct_seq_show(struct seq_file *s, void *v)
 {
 	struct nf_conntrack_tuple_hash *hash = v;
@@ -148,10 +172,8 @@
 		goto release;
 #endif
 
-#ifdef CONFIG_NF_CONNTRACK_SECMARK
-	if (seq_printf(s, "secmark=%u ", ct->secmark))
+	if (ct_show_secctx(s, ct))
 		goto release;
-#endif
 
 	if (seq_printf(s, "use=%u\n", atomic_read(&ct->ct_general.use)))
 		goto release;
diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c
index 8c8632d..957c924 100644
--- a/net/ipv4/netfilter/nf_nat_core.c
+++ b/net/ipv4/netfilter/nf_nat_core.c
@@ -38,7 +38,7 @@
 static struct nf_conntrack_l3proto *l3proto __read_mostly;
 
 #define MAX_IP_NAT_PROTO 256
-static const struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO]
+static const struct nf_nat_protocol __rcu *nf_nat_protos[MAX_IP_NAT_PROTO]
 						__read_mostly;
 
 static inline const struct nf_nat_protocol *
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8323136..a275c6e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1556,14 +1556,13 @@
  *	i.e. Path MTU discovery
  */
 
-void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
-			struct net_device *dev, u32 pmtu)
+static void rt6_do_pmtu_disc(struct in6_addr *daddr, struct in6_addr *saddr,
+			     struct net *net, u32 pmtu, int ifindex)
 {
 	struct rt6_info *rt, *nrt;
-	struct net *net = dev_net(dev);
 	int allfrag = 0;
 
-	rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0);
+	rt = rt6_lookup(net, daddr, saddr, ifindex, 0);
 	if (rt == NULL)
 		return;
 
@@ -1631,6 +1630,27 @@
 	dst_release(&rt->dst);
 }
 
+void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
+			struct net_device *dev, u32 pmtu)
+{
+	struct net *net = dev_net(dev);
+
+	/*
+	 * RFC 1981 states that a node "MUST reduce the size of the packets it
+	 * is sending along the path" that caused the Packet Too Big message.
+	 * Since it's not possible in the general case to determine which
+	 * interface was used to send the original packet, we update the MTU
+	 * on the interface that will be used to send future packets. We also
+	 * update the MTU on the interface that received the Packet Too Big in
+	 * case the original packet was forced out that interface with
+	 * SO_BINDTODEVICE or similar. This is the next best thing to the
+	 * correct behaviour, which would be to update the MTU on all
+	 * interfaces.
+	 */
+	rt6_do_pmtu_disc(daddr, saddr, net, pmtu, 0);
+	rt6_do_pmtu_disc(daddr, saddr, net, pmtu, dev->ifindex);
+}
+
 /*
  *	Misc support functions
  */
diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index c893f23..8f23401 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -175,6 +175,8 @@
 
 	set_bit(HT_AGG_STATE_STOPPING, &tid_tx->state);
 
+	del_timer_sync(&tid_tx->addba_resp_timer);
+
 	/*
 	 * After this packets are no longer handed right through
 	 * to the driver but are put onto tid_tx->pending instead,
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 10caec5..34da679 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -377,7 +377,7 @@
 				skb2 = skb_clone(skb, GFP_ATOMIC);
 				if (skb2) {
 					skb2->dev = prev_dev;
-					netif_receive_skb(skb2);
+					netif_rx(skb2);
 				}
 			}
 
@@ -386,7 +386,7 @@
 	}
 	if (prev_dev) {
 		skb->dev = prev_dev;
-		netif_receive_skb(skb);
+		netif_rx(skb);
 		skb = NULL;
 	}
 	rcu_read_unlock();
diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 78b505d..fdaec7d 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -27,7 +27,7 @@
 
 static DEFINE_MUTEX(afinfo_mutex);
 
-const struct nf_afinfo *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly;
+const struct nf_afinfo __rcu *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly;
 EXPORT_SYMBOL(nf_afinfo);
 
 int nf_register_afinfo(const struct nf_afinfo *afinfo)
diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c
index cdcc764..5702de3 100644
--- a/net/netfilter/nf_conntrack_ecache.c
+++ b/net/netfilter/nf_conntrack_ecache.c
@@ -26,10 +26,10 @@
 
 static DEFINE_MUTEX(nf_ct_ecache_mutex);
 
-struct nf_ct_event_notifier *nf_conntrack_event_cb __read_mostly;
+struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb __read_mostly;
 EXPORT_SYMBOL_GPL(nf_conntrack_event_cb);
 
-struct nf_exp_event_notifier *nf_expect_event_cb __read_mostly;
+struct nf_exp_event_notifier __rcu *nf_expect_event_cb __read_mostly;
 EXPORT_SYMBOL_GPL(nf_expect_event_cb);
 
 /* deliver cached events and clear cache entry - must be called with locally
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index 8d9e4c9..bd82450 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -16,7 +16,7 @@
 #include <linux/skbuff.h>
 #include <net/netfilter/nf_conntrack_extend.h>
 
-static struct nf_ct_ext_type *nf_ct_ext_types[NF_CT_EXT_NUM];
+static struct nf_ct_ext_type __rcu *nf_ct_ext_types[NF_CT_EXT_NUM];
 static DEFINE_MUTEX(nf_ct_ext_type_mutex);
 
 void __nf_ct_ext_destroy(struct nf_conn *ct)
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 5bae1cd..146476c 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -22,6 +22,7 @@
 #include <linux/rculist_nulls.h>
 #include <linux/types.h>
 #include <linux/timer.h>
+#include <linux/security.h>
 #include <linux/skbuff.h>
 #include <linux/errno.h>
 #include <linux/netlink.h>
@@ -245,16 +246,31 @@
 
 #ifdef CONFIG_NF_CONNTRACK_SECMARK
 static inline int
-ctnetlink_dump_secmark(struct sk_buff *skb, const struct nf_conn *ct)
+ctnetlink_dump_secctx(struct sk_buff *skb, const struct nf_conn *ct)
 {
-	NLA_PUT_BE32(skb, CTA_SECMARK, htonl(ct->secmark));
-	return 0;
+	struct nlattr *nest_secctx;
+	int len, ret;
+	char *secctx;
 
+	ret = security_secid_to_secctx(ct->secmark, &secctx, &len);
+	if (ret)
+		return ret;
+
+	ret = -1;
+	nest_secctx = nla_nest_start(skb, CTA_SECCTX | NLA_F_NESTED);
+	if (!nest_secctx)
+		goto nla_put_failure;
+
+	NLA_PUT_STRING(skb, CTA_SECCTX_NAME, secctx);
+	nla_nest_end(skb, nest_secctx);
+
+	ret = 0;
 nla_put_failure:
-	return -1;
+	security_release_secctx(secctx, len);
+	return ret;
 }
 #else
-#define ctnetlink_dump_secmark(a, b) (0)
+#define ctnetlink_dump_secctx(a, b) (0)
 #endif
 
 #define master_tuple(ct) &(ct->master->tuplehash[IP_CT_DIR_ORIGINAL].tuple)
@@ -391,7 +407,7 @@
 	    ctnetlink_dump_protoinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_helpinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_mark(skb, ct) < 0 ||
-	    ctnetlink_dump_secmark(skb, ct) < 0 ||
+	    ctnetlink_dump_secctx(skb, ct) < 0 ||
 	    ctnetlink_dump_id(skb, ct) < 0 ||
 	    ctnetlink_dump_use(skb, ct) < 0 ||
 	    ctnetlink_dump_master(skb, ct) < 0 ||
@@ -437,6 +453,17 @@
 	       ;
 }
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static int ctnetlink_nlmsg_secctx_size(const struct nf_conn *ct)
+{
+	int len;
+
+	security_secid_to_secctx(ct->secmark, NULL, &len);
+
+	return sizeof(char) * len;
+}
+#endif
+
 static inline size_t
 ctnetlink_nlmsg_size(const struct nf_conn *ct)
 {
@@ -453,7 +480,8 @@
 	       + nla_total_size(0) /* CTA_HELP */
 	       + nla_total_size(NF_CT_HELPER_NAME_LEN) /* CTA_HELP_NAME */
 #ifdef CONFIG_NF_CONNTRACK_SECMARK
-	       + nla_total_size(sizeof(u_int32_t)) /* CTA_SECMARK */
+	       + nla_total_size(0) /* CTA_SECCTX */
+	       + nla_total_size(ctnetlink_nlmsg_secctx_size(ct)) /* CTA_SECCTX_NAME */
 #endif
 #ifdef CONFIG_NF_NAT_NEEDED
 	       + 2 * nla_total_size(0) /* CTA_NAT_SEQ_ADJ_ORIG|REPL */
@@ -556,7 +584,7 @@
 
 #ifdef CONFIG_NF_CONNTRACK_SECMARK
 		if ((events & (1 << IPCT_SECMARK) || ct->secmark)
-		    && ctnetlink_dump_secmark(skb, ct) < 0)
+		    && ctnetlink_dump_secctx(skb, ct) < 0)
 			goto nla_put_failure;
 #endif
 
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c
index 5886ba1..ed6d929 100644
--- a/net/netfilter/nf_conntrack_proto.c
+++ b/net/netfilter/nf_conntrack_proto.c
@@ -28,8 +28,8 @@
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_core.h>
 
-static struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly;
-struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX] __read_mostly;
+static struct nf_conntrack_l4proto __rcu **nf_ct_protos[PF_MAX] __read_mostly;
+struct nf_conntrack_l3proto __rcu *nf_ct_l3protos[AF_MAX] __read_mostly;
 EXPORT_SYMBOL_GPL(nf_ct_l3protos);
 
 static DEFINE_MUTEX(nf_ct_proto_mutex);
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index eb973fc..0fb6570 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -15,6 +15,7 @@
 #include <linux/seq_file.h>
 #include <linux/percpu.h>
 #include <linux/netdevice.h>
+#include <linux/security.h>
 #include <net/net_namespace.h>
 #ifdef CONFIG_SYSCTL
 #include <linux/sysctl.h>
@@ -108,6 +109,29 @@
 	rcu_read_unlock();
 }
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct)
+{
+	int ret;
+	u32 len;
+	char *secctx;
+
+	ret = security_secid_to_secctx(ct->secmark, &secctx, &len);
+	if (ret)
+		return ret;
+
+	ret = seq_printf(s, "secctx=%s ", secctx);
+
+	security_release_secctx(secctx, len);
+	return ret;
+}
+#else
+static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct)
+{
+	return 0;
+}
+#endif
+
 /* return 0 on success, 1 in case of error */
 static int ct_seq_show(struct seq_file *s, void *v)
 {
@@ -168,10 +192,8 @@
 		goto release;
 #endif
 
-#ifdef CONFIG_NF_CONNTRACK_SECMARK
-	if (seq_printf(s, "secmark=%u ", ct->secmark))
+	if (ct_show_secctx(s, ct))
 		goto release;
-#endif
 
 #ifdef CONFIG_NF_CONNTRACK_ZONES
 	if (seq_printf(s, "zone=%u ", nf_ct_zone(ct)))
diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index 7df37fd..b07393e 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -16,7 +16,7 @@
 #define NF_LOG_PREFIXLEN		128
 #define NFLOGGER_NAME_LEN		64
 
-static const struct nf_logger *nf_loggers[NFPROTO_NUMPROTO] __read_mostly;
+static const struct nf_logger __rcu *nf_loggers[NFPROTO_NUMPROTO] __read_mostly;
 static struct list_head nf_loggers_l[NFPROTO_NUMPROTO] __read_mostly;
 static DEFINE_MUTEX(nf_log_mutex);
 
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 78b3cf9c..74aebed 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -18,7 +18,7 @@
  * long term mutex.  The handler must provide an an outfn() to accept packets
  * for queueing and must reinject all packets it receives, no matter what.
  */
-static const struct nf_queue_handler *queue_handler[NFPROTO_NUMPROTO] __read_mostly;
+static const struct nf_queue_handler __rcu *queue_handler[NFPROTO_NUMPROTO] __read_mostly;
 
 static DEFINE_MUTEX(queue_handler_mutex);
 
diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c
index 0cb6053..782e519 100644
--- a/net/netfilter/xt_CT.c
+++ b/net/netfilter/xt_CT.c
@@ -9,7 +9,6 @@
 #include <linux/module.h>
 #include <linux/gfp.h>
 #include <linux/skbuff.h>
-#include <linux/selinux.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv6/ip6_tables.h>
 #include <linux/netfilter/x_tables.h>
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index 23b2d6c..9faf5e0 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -14,8 +14,8 @@
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 #include <linux/module.h>
+#include <linux/security.h>
 #include <linux/skbuff.h>
-#include <linux/selinux.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_SECMARK.h>
 
@@ -39,9 +39,8 @@
 
 	switch (mode) {
 	case SECMARK_MODE_SEL:
-		secmark = info->u.sel.selsid;
+		secmark = info->secid;
 		break;
-
 	default:
 		BUG();
 	}
@@ -50,33 +49,33 @@
 	return XT_CONTINUE;
 }
 
-static int checkentry_selinux(struct xt_secmark_target_info *info)
+static int checkentry_lsm(struct xt_secmark_target_info *info)
 {
 	int err;
-	struct xt_secmark_target_selinux_info *sel = &info->u.sel;
 
-	sel->selctx[SECMARK_SELCTX_MAX - 1] = '\0';
+	info->secctx[SECMARK_SECCTX_MAX - 1] = '\0';
+	info->secid = 0;
 
-	err = selinux_string_to_sid(sel->selctx, &sel->selsid);
+	err = security_secctx_to_secid(info->secctx, strlen(info->secctx),
+				       &info->secid);
 	if (err) {
 		if (err == -EINVAL)
-			pr_info("invalid SELinux context \'%s\'\n",
-				sel->selctx);
+			pr_info("invalid security context \'%s\'\n", info->secctx);
 		return err;
 	}
 
-	if (!sel->selsid) {
-		pr_info("unable to map SELinux context \'%s\'\n", sel->selctx);
+	if (!info->secid) {
+		pr_info("unable to map security context \'%s\'\n", info->secctx);
 		return -ENOENT;
 	}
 
-	err = selinux_secmark_relabel_packet_permission(sel->selsid);
+	err = security_secmark_relabel_packet(info->secid);
 	if (err) {
 		pr_info("unable to obtain relabeling permission\n");
 		return err;
 	}
 
-	selinux_secmark_refcount_inc();
+	security_secmark_refcount_inc();
 	return 0;
 }
 
@@ -100,16 +99,16 @@
 
 	switch (info->mode) {
 	case SECMARK_MODE_SEL:
-		err = checkentry_selinux(info);
-		if (err <= 0)
-			return err;
 		break;
-
 	default:
 		pr_info("invalid mode: %hu\n", info->mode);
 		return -EINVAL;
 	}
 
+	err = checkentry_lsm(info);
+	if (err)
+		return err;
+
 	if (!mode)
 		mode = info->mode;
 	return 0;
@@ -119,7 +118,7 @@
 {
 	switch (mode) {
 	case SECMARK_MODE_SEL:
-		selinux_secmark_refcount_dec();
+		security_secmark_refcount_dec();
 	}
 }
 
diff --git a/net/rds/page.c b/net/rds/page.c
index 595a952..1dfbfea 100644
--- a/net/rds/page.c
+++ b/net/rds/page.c
@@ -57,30 +57,17 @@
 	unsigned long ret;
 	void *addr;
 
-	if (to_user)
+	addr = kmap(page);
+	if (to_user) {
 		rds_stats_add(s_copy_to_user, bytes);
-	else
+		ret = copy_to_user(ptr, addr + offset, bytes);
+	} else {
 		rds_stats_add(s_copy_from_user, bytes);
-
-	addr = kmap_atomic(page, KM_USER0);
-	if (to_user)
-		ret = __copy_to_user_inatomic(ptr, addr + offset, bytes);
-	else
-		ret = __copy_from_user_inatomic(addr + offset, ptr, bytes);
-	kunmap_atomic(addr, KM_USER0);
-
-	if (ret) {
-		addr = kmap(page);
-		if (to_user)
-			ret = copy_to_user(ptr, addr + offset, bytes);
-		else
-			ret = copy_from_user(addr + offset, ptr, bytes);
-		kunmap(page);
-		if (ret)
-			return -EFAULT;
+		ret = copy_from_user(addr + offset, ptr, bytes);
 	}
+	kunmap(page);
 
-	return 0;
+	return ret ? -EFAULT : 0;
 }
 EXPORT_SYMBOL_GPL(rds_page_copy_user);
 
diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c
index 78ef2c5..37dff78 100644
--- a/net/sched/cls_cgroup.c
+++ b/net/sched/cls_cgroup.c
@@ -123,7 +123,7 @@
 	 * calls by looking at the number of nested bh disable calls because
 	 * softirqs always disables bh.
 	 */
-	if (softirq_count() != SOFTIRQ_OFFSET) {
+	if (in_serving_softirq()) {
 		/* If there is an sk_classid we'll use that. */
 		if (!skb->sk)
 			return -1;
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 7416a5c..b0c2a82 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -137,7 +137,7 @@
 			int toff = off + key->off + (off2 & key->offmask);
 			__be32 *data, _data;
 
-			if (skb_headroom(skb) + toff < 0)
+			if (skb_headroom(skb) + toff > INT_MAX)
 				goto out;
 
 			data = skb_header_pointer(skb, toff, 4, &_data);
diff --git a/net/sctp/auth.c b/net/sctp/auth.c
index 8636639..ddbbf7c 100644
--- a/net/sctp/auth.c
+++ b/net/sctp/auth.c
@@ -543,16 +543,20 @@
 		id = ntohs(hmacs->hmac_ids[i]);
 
 		/* Check the id is in the supported range */
-		if (id > SCTP_AUTH_HMAC_ID_MAX)
+		if (id > SCTP_AUTH_HMAC_ID_MAX) {
+			id = 0;
 			continue;
+		}
 
 		/* See is we support the id.  Supported IDs have name and
 		 * length fields set, so that we can allocated and use
 		 * them.  We can safely just check for name, for without the
 		 * name, we can't allocate the TFM.
 		 */
-		if (!sctp_hmac_list[id].hmac_name)
+		if (!sctp_hmac_list[id].hmac_name) {
+			id = 0;
 			continue;
+		}
 
 		break;
 	}
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index ca44917..fbb7077 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -916,6 +916,11 @@
 	/* Walk through the addrs buffer and count the number of addresses. */
 	addr_buf = kaddrs;
 	while (walk_size < addrs_size) {
+		if (walk_size + sizeof(sa_family_t) > addrs_size) {
+			kfree(kaddrs);
+			return -EINVAL;
+		}
+
 		sa_addr = (struct sockaddr *)addr_buf;
 		af = sctp_get_af_specific(sa_addr->sa_family);
 
@@ -1002,9 +1007,13 @@
 	/* Walk through the addrs buffer and count the number of addresses. */
 	addr_buf = kaddrs;
 	while (walk_size < addrs_size) {
+		if (walk_size + sizeof(sa_family_t) > addrs_size) {
+			err = -EINVAL;
+			goto out_free;
+		}
+
 		sa_addr = (union sctp_addr *)addr_buf;
 		af = sctp_get_af_specific(sa_addr->sa.sa_family);
-		port = ntohs(sa_addr->v4.sin_port);
 
 		/* If the address family is not supported or if this address
 		 * causes the address buffer to overflow return EINVAL.
@@ -1014,6 +1023,8 @@
 			goto out_free;
 		}
 
+		port = ntohs(sa_addr->v4.sin_port);
+
 		/* Save current address so we can work with it */
 		memcpy(&to, sa_addr, af->sockaddr_len);
 
diff --git a/scripts/Makefile b/scripts/Makefile
index 842dbc2..2e08810 100644
--- a/scripts/Makefile
+++ b/scripts/Makefile
@@ -11,6 +11,7 @@
 hostprogs-$(CONFIG_LOGO)         += pnmtologo
 hostprogs-$(CONFIG_VT)           += conmakehash
 hostprogs-$(CONFIG_IKCONFIG)     += bin2c
+hostprogs-$(BUILD_C_RECORDMCOUNT) += recordmcount
 
 always		:= $(hostprogs-y) $(hostprogs-m)
 
diff --git a/scripts/Makefile.build b/scripts/Makefile.build
index a1a5cf9..843bd4f 100644
--- a/scripts/Makefile.build
+++ b/scripts/Makefile.build
@@ -209,12 +209,22 @@
 endif
 
 ifdef CONFIG_FTRACE_MCOUNT_RECORD
+ifdef BUILD_C_RECORDMCOUNT
+# Due to recursion, we must skip empty.o.
+# The empty.o file is created in the make process in order to determine
+#  the target endianness and word size. It is made before all other C
+#  files, including recordmcount.
+cmd_record_mcount = if [ $(@) != "scripts/mod/empty.o" ]; then			\
+			$(objtree)/scripts/recordmcount "$(@)";			\
+		    fi;
+else
 cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \
 	"$(if $(CONFIG_CPU_BIG_ENDIAN),big,little)" \
 	"$(if $(CONFIG_64BIT),64,32)" \
 	"$(OBJDUMP)" "$(OBJCOPY)" "$(CC)" "$(LD)" "$(NM)" "$(RM)" "$(MV)" \
 	"$(if $(part-of-module),1,0)" "$(@)";
 endif
+endif
 
 define rule_cc_o_c
 	$(call echo-cmd,checksrc) $(cmd_checksrc)			  \
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 54fd1b7..7bfcf1a 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -101,14 +101,6 @@
 modname_flags  = $(if $(filter 1,$(words $(modname))),\
                  -D"KBUILD_MODNAME=KBUILD_STR($(call name-fix,$(modname)))")
 
-#hash values
-ifdef CONFIG_DYNAMIC_DEBUG
-debug_flags = -D"DEBUG_HASH=$(shell ./scripts/basic/hash djb2 $(@D)$(modname))"\
-              -D"DEBUG_HASH2=$(shell ./scripts/basic/hash r5 $(@D)$(modname))"
-else
-debug_flags =
-endif
-
 orig_c_flags   = $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(KBUILD_SUBDIR_CCFLAGS) \
                  $(ccflags-y) $(CFLAGS_$(basetarget).o)
 _c_flags       = $(filter-out $(CFLAGS_REMOVE_$(basetarget).o), $(orig_c_flags))
@@ -152,8 +144,7 @@
 
 c_flags        = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(LINUXINCLUDE)     \
 		 $(__c_flags) $(modkern_cflags)                           \
-		 -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags) \
-		  $(debug_flags)
+		 -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags)
 
 a_flags        = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(LINUXINCLUDE)     \
 		 $(__a_flags) $(modkern_aflags)
diff --git a/scripts/basic/Makefile b/scripts/basic/Makefile
index 0955995..4c324a1 100644
--- a/scripts/basic/Makefile
+++ b/scripts/basic/Makefile
@@ -9,7 +9,7 @@
 # fixdep: 	 Used to generate dependency information during build process
 # docproc:	 Used in Documentation/DocBook
 
-hostprogs-y	:= fixdep docproc hash
+hostprogs-y	:= fixdep docproc
 always		:= $(hostprogs-y)
 
 # fixdep is needed to compile other host programs
diff --git a/scripts/basic/hash.c b/scripts/basic/hash.c
deleted file mode 100644
index 2ef5d3f..0000000
--- a/scripts/basic/hash.c
+++ /dev/null
@@ -1,64 +0,0 @@
-/*
- * Copyright (C) 2008 Red Hat, Inc., Jason Baron <jbaron@redhat.com>
- *
- */
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-
-#define DYNAMIC_DEBUG_HASH_BITS 6
-
-static const char *program;
-
-static void usage(void)
-{
-	printf("Usage: %s <djb2|r5> <modname>\n", program);
-	exit(1);
-}
-
-/* djb2 hashing algorithm by Dan Bernstein. From:
- * http://www.cse.yorku.ca/~oz/hash.html
- */
-
-static unsigned int djb2_hash(char *str)
-{
-	unsigned long hash = 5381;
-	int c;
-
-	c = *str;
-	while (c) {
-		hash = ((hash << 5) + hash) + c;
-		c = *++str;
-	}
-	return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1));
-}
-
-static unsigned int r5_hash(char *str)
-{
-	unsigned long hash = 0;
-	int c;
-
-	c = *str;
-	while (c) {
-		hash = (hash + (c << 4) + (c >> 4)) * 11;
-		c = *++str;
-	}
-	return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1));
-}
-
-int main(int argc, char *argv[])
-{
-	program = argv[0];
-
-	if (argc != 3)
-		usage();
-	if (!strcmp(argv[1], "djb2"))
-		printf("%d\n", djb2_hash(argv[2]));
-	else if (!strcmp(argv[1], "r5"))
-		printf("%d\n", r5_hash(argv[2]));
-	else
-		usage();
-	exit(0);
-}
-
diff --git a/scripts/gcc-goto.sh b/scripts/gcc-goto.sh
new file mode 100644
index 0000000..520d16b
--- /dev/null
+++ b/scripts/gcc-goto.sh
@@ -0,0 +1,5 @@
+#!/bin/sh
+# Test for gcc 'asm goto' suport
+# Copyright (C) 2010, Jason Baron <jbaron@redhat.com>
+
+echo "int main(void) { entry: asm goto (\"\"::::entry); return 0; }" | $@ -x c - -c -o /dev/null >/dev/null 2>&1 && echo "y"
diff --git a/scripts/kconfig/conf.c b/scripts/kconfig/conf.c
index 5b7c86e..7ef429c 100644
--- a/scripts/kconfig/conf.c
+++ b/scripts/kconfig/conf.c
@@ -427,7 +427,7 @@
 				if (sym->name && !sym_is_choice_value(sym)) {
 					printf("CONFIG_%s\n", sym->name);
 				}
-			} else {
+			} else if (input_mode != oldnoconfig) {
 				if (!conf_cnt++)
 					printf(_("*\n* Restart config...\n*\n"));
 				rootEntry = menu_get_parent_menu(menu);
diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h
index 6ee2e4f..170459c 100644
--- a/scripts/kconfig/expr.h
+++ b/scripts/kconfig/expr.h
@@ -165,7 +165,6 @@
 	struct symbol *sym;
 	struct property *prompt;
 	struct expr *dep;
-	struct expr *dir_dep;
 	unsigned int flags;
 	char *help;
 	struct file *file;
diff --git a/scripts/kconfig/menu.c b/scripts/kconfig/menu.c
index 4fb5902..edda8b4 100644
--- a/scripts/kconfig/menu.c
+++ b/scripts/kconfig/menu.c
@@ -107,7 +107,6 @@
 void menu_add_dep(struct expr *dep)
 {
 	current_entry->dep = expr_alloc_and(current_entry->dep, menu_check_dep(dep));
-	current_entry->dir_dep = current_entry->dep;
 }
 
 void menu_set_type(int type)
@@ -291,10 +290,6 @@
 		for (menu = parent->list; menu; menu = menu->next)
 			menu_finalize(menu);
 	} else if (sym) {
-		/* ignore inherited dependencies for dir_dep */
-		sym->dir_dep.expr = expr_transform(expr_copy(parent->dir_dep));
-		sym->dir_dep.expr = expr_eliminate_dups(sym->dir_dep.expr);
-
 		basedep = parent->prompt ? parent->prompt->visible.expr : NULL;
 		basedep = expr_trans_compare(basedep, E_UNEQUAL, &symbol_no);
 		basedep = expr_eliminate_dups(expr_transform(basedep));
@@ -325,6 +320,8 @@
 			parent->next = last_menu->next;
 			last_menu->next = NULL;
 		}
+
+		sym->dir_dep.expr = parent->dep;
 	}
 	for (menu = parent->list; menu; menu = menu->next) {
 		if (sym && sym_is_choice(sym) &&
diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c
index 943712c..1f8b305 100644
--- a/scripts/kconfig/symbol.c
+++ b/scripts/kconfig/symbol.c
@@ -350,6 +350,7 @@
 				}
 			}
 		calc_newval:
+#if 0
 			if (sym->dir_dep.tri == no && sym->rev_dep.tri != no) {
 				fprintf(stderr, "warning: (");
 				expr_fprint(sym->rev_dep.expr, stderr);
@@ -358,6 +359,7 @@
 				expr_fprint(sym->dir_dep.expr, stderr);
 				fprintf(stderr, ")\n");
 			}
+#endif
 			newval.tri = EXPR_OR(newval.tri, sym->rev_dep.tri);
 		}
 		if (newval.tri == mod && sym_get_type(sym) == S_BOOLEAN)
diff --git a/scripts/recordmcount.c b/scripts/recordmcount.c
new file mode 100644
index 0000000..26e1271
--- /dev/null
+++ b/scripts/recordmcount.c
@@ -0,0 +1,363 @@
+/*
+ * recordmcount.c: construct a table of the locations of calls to 'mcount'
+ * so that ftrace can find them quickly.
+ * Copyright 2009 John F. Reiser <jreiser@BitWagon.com>.  All rights reserved.
+ * Licensed under the GNU General Public License, version 2 (GPLv2).
+ *
+ * Restructured to fit Linux format, as well as other updates:
+ *  Copyright 2010 Steven Rostedt <srostedt@redhat.com>, Red Hat Inc.
+ */
+
+/*
+ * Strategy: alter the .o file in-place.
+ *
+ * Append a new STRTAB that has the new section names, followed by a new array
+ * ElfXX_Shdr[] that has the new section headers, followed by the section
+ * contents for __mcount_loc and its relocations.  The old shstrtab strings,
+ * and the old ElfXX_Shdr[] array, remain as "garbage" (commonly, a couple
+ * kilobytes.)  Subsequent processing by /bin/ld (or the kernel module loader)
+ * will ignore the garbage regions, because they are not designated by the
+ * new .e_shoff nor the new ElfXX_Shdr[].  [In order to remove the garbage,
+ * then use "ld -r" to create a new file that omits the garbage.]
+ */
+
+#include <sys/types.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <elf.h>
+#include <fcntl.h>
+#include <setjmp.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+static int fd_map;	/* File descriptor for file being modified. */
+static int mmap_failed; /* Boolean flag. */
+static void *ehdr_curr; /* current ElfXX_Ehdr *  for resource cleanup */
+static char gpfx;	/* prefix for global symbol name (sometimes '_') */
+static struct stat sb;	/* Remember .st_size, etc. */
+static jmp_buf jmpenv;	/* setjmp/longjmp per-file error escape */
+
+/* setjmp() return values */
+enum {
+	SJ_SETJMP = 0,  /* hardwired first return */
+	SJ_FAIL,
+	SJ_SUCCEED
+};
+
+/* Per-file resource cleanup when multiple files. */
+static void
+cleanup(void)
+{
+	if (!mmap_failed)
+		munmap(ehdr_curr, sb.st_size);
+	else
+		free(ehdr_curr);
+	close(fd_map);
+}
+
+static void __attribute__((noreturn))
+fail_file(void)
+{
+	cleanup();
+	longjmp(jmpenv, SJ_FAIL);
+}
+
+static void __attribute__((noreturn))
+succeed_file(void)
+{
+	cleanup();
+	longjmp(jmpenv, SJ_SUCCEED);
+}
+
+/* ulseek, uread, ...:  Check return value for errors. */
+
+static off_t
+ulseek(int const fd, off_t const offset, int const whence)
+{
+	off_t const w = lseek(fd, offset, whence);
+	if ((off_t)-1 == w) {
+		perror("lseek");
+		fail_file();
+	}
+	return w;
+}
+
+static size_t
+uread(int const fd, void *const buf, size_t const count)
+{
+	size_t const n = read(fd, buf, count);
+	if (n != count) {
+		perror("read");
+		fail_file();
+	}
+	return n;
+}
+
+static size_t
+uwrite(int const fd, void const *const buf, size_t const count)
+{
+	size_t const n = write(fd, buf, count);
+	if (n != count) {
+		perror("write");
+		fail_file();
+	}
+	return n;
+}
+
+static void *
+umalloc(size_t size)
+{
+	void *const addr = malloc(size);
+	if (0 == addr) {
+		fprintf(stderr, "malloc failed: %zu bytes\n", size);
+		fail_file();
+	}
+	return addr;
+}
+
+/*
+ * Get the whole file as a programming convenience in order to avoid
+ * malloc+lseek+read+free of many pieces.  If successful, then mmap
+ * avoids copying unused pieces; else just read the whole file.
+ * Open for both read and write; new info will be appended to the file.
+ * Use MAP_PRIVATE so that a few changes to the in-memory ElfXX_Ehdr
+ * do not propagate to the file until an explicit overwrite at the last.
+ * This preserves most aspects of consistency (all except .st_size)
+ * for simultaneous readers of the file while we are appending to it.
+ * However, multiple writers still are bad.  We choose not to use
+ * locking because it is expensive and the use case of kernel build
+ * makes multiple writers unlikely.
+ */
+static void *mmap_file(char const *fname)
+{
+	void *addr;
+
+	fd_map = open(fname, O_RDWR);
+	if (0 > fd_map || 0 > fstat(fd_map, &sb)) {
+		perror(fname);
+		fail_file();
+	}
+	if (!S_ISREG(sb.st_mode)) {
+		fprintf(stderr, "not a regular file: %s\n", fname);
+		fail_file();
+	}
+	addr = mmap(0, sb.st_size, PROT_READ|PROT_WRITE, MAP_PRIVATE,
+		    fd_map, 0);
+	mmap_failed = 0;
+	if (MAP_FAILED == addr) {
+		mmap_failed = 1;
+		addr = umalloc(sb.st_size);
+		uread(fd_map, addr, sb.st_size);
+	}
+	return addr;
+}
+
+/* w8rev, w8nat, ...: Handle endianness. */
+
+static uint64_t w8rev(uint64_t const x)
+{
+	return   ((0xff & (x >> (0 * 8))) << (7 * 8))
+	       | ((0xff & (x >> (1 * 8))) << (6 * 8))
+	       | ((0xff & (x >> (2 * 8))) << (5 * 8))
+	       | ((0xff & (x >> (3 * 8))) << (4 * 8))
+	       | ((0xff & (x >> (4 * 8))) << (3 * 8))
+	       | ((0xff & (x >> (5 * 8))) << (2 * 8))
+	       | ((0xff & (x >> (6 * 8))) << (1 * 8))
+	       | ((0xff & (x >> (7 * 8))) << (0 * 8));
+}
+
+static uint32_t w4rev(uint32_t const x)
+{
+	return   ((0xff & (x >> (0 * 8))) << (3 * 8))
+	       | ((0xff & (x >> (1 * 8))) << (2 * 8))
+	       | ((0xff & (x >> (2 * 8))) << (1 * 8))
+	       | ((0xff & (x >> (3 * 8))) << (0 * 8));
+}
+
+static uint32_t w2rev(uint16_t const x)
+{
+	return   ((0xff & (x >> (0 * 8))) << (1 * 8))
+	       | ((0xff & (x >> (1 * 8))) << (0 * 8));
+}
+
+static uint64_t w8nat(uint64_t const x)
+{
+	return x;
+}
+
+static uint32_t w4nat(uint32_t const x)
+{
+	return x;
+}
+
+static uint32_t w2nat(uint16_t const x)
+{
+	return x;
+}
+
+static uint64_t (*w8)(uint64_t);
+static uint32_t (*w)(uint32_t);
+static uint32_t (*w2)(uint16_t);
+
+/* Names of the sections that could contain calls to mcount. */
+static int
+is_mcounted_section_name(char const *const txtname)
+{
+	return 0 == strcmp(".text",          txtname) ||
+		0 == strcmp(".sched.text",    txtname) ||
+		0 == strcmp(".spinlock.text", txtname) ||
+		0 == strcmp(".irqentry.text", txtname) ||
+		0 == strcmp(".text.unlikely", txtname);
+}
+
+/* 32 bit and 64 bit are very similar */
+#include "recordmcount.h"
+#define RECORD_MCOUNT_64
+#include "recordmcount.h"
+
+static void
+do_file(char const *const fname)
+{
+	Elf32_Ehdr *const ehdr = mmap_file(fname);
+	unsigned int reltype = 0;
+
+	ehdr_curr = ehdr;
+	w = w4nat;
+	w2 = w2nat;
+	w8 = w8nat;
+	switch (ehdr->e_ident[EI_DATA]) {
+		static unsigned int const endian = 1;
+	default: {
+		fprintf(stderr, "unrecognized ELF data encoding %d: %s\n",
+			ehdr->e_ident[EI_DATA], fname);
+		fail_file();
+	} break;
+	case ELFDATA2LSB: {
+		if (1 != *(unsigned char const *)&endian) {
+			/* main() is big endian, file.o is little endian. */
+			w = w4rev;
+			w2 = w2rev;
+			w8 = w8rev;
+		}
+	} break;
+	case ELFDATA2MSB: {
+		if (0 != *(unsigned char const *)&endian) {
+			/* main() is little endian, file.o is big endian. */
+			w = w4rev;
+			w2 = w2rev;
+			w8 = w8rev;
+		}
+	} break;
+	}  /* end switch */
+	if (0 != memcmp(ELFMAG, ehdr->e_ident, SELFMAG)
+	||  ET_REL != w2(ehdr->e_type)
+	||  EV_CURRENT != ehdr->e_ident[EI_VERSION]) {
+		fprintf(stderr, "unrecognized ET_REL file %s\n", fname);
+		fail_file();
+	}
+
+	gpfx = 0;
+	switch (w2(ehdr->e_machine)) {
+	default: {
+		fprintf(stderr, "unrecognized e_machine %d %s\n",
+			w2(ehdr->e_machine), fname);
+		fail_file();
+	} break;
+	case EM_386:	 reltype = R_386_32;                   break;
+	case EM_ARM:	 reltype = R_ARM_ABS32;                break;
+	case EM_IA_64:	 reltype = R_IA64_IMM64;   gpfx = '_'; break;
+	case EM_PPC:	 reltype = R_PPC_ADDR32;   gpfx = '_'; break;
+	case EM_PPC64:	 reltype = R_PPC64_ADDR64; gpfx = '_'; break;
+	case EM_S390:    /* reltype: e_class    */ gpfx = '_'; break;
+	case EM_SH:	 reltype = R_SH_DIR32;                 break;
+	case EM_SPARCV9: reltype = R_SPARC_64;     gpfx = '_'; break;
+	case EM_X86_64:	 reltype = R_X86_64_64;                break;
+	}  /* end switch */
+
+	switch (ehdr->e_ident[EI_CLASS]) {
+	default: {
+		fprintf(stderr, "unrecognized ELF class %d %s\n",
+			ehdr->e_ident[EI_CLASS], fname);
+		fail_file();
+	} break;
+	case ELFCLASS32: {
+		if (sizeof(Elf32_Ehdr) != w2(ehdr->e_ehsize)
+		||  sizeof(Elf32_Shdr) != w2(ehdr->e_shentsize)) {
+			fprintf(stderr,
+				"unrecognized ET_REL file: %s\n", fname);
+			fail_file();
+		}
+		if (EM_S390 == w2(ehdr->e_machine))
+			reltype = R_390_32;
+		do32(ehdr, fname, reltype);
+	} break;
+	case ELFCLASS64: {
+		Elf64_Ehdr *const ghdr = (Elf64_Ehdr *)ehdr;
+		if (sizeof(Elf64_Ehdr) != w2(ghdr->e_ehsize)
+		||  sizeof(Elf64_Shdr) != w2(ghdr->e_shentsize)) {
+			fprintf(stderr,
+				"unrecognized ET_REL file: %s\n", fname);
+			fail_file();
+		}
+		if (EM_S390 == w2(ghdr->e_machine))
+			reltype = R_390_64;
+		do64(ghdr, fname, reltype);
+	} break;
+	}  /* end switch */
+
+	cleanup();
+}
+
+int
+main(int argc, char const *argv[])
+{
+	const char ftrace[] = "kernel/trace/ftrace.o";
+	int ftrace_size = sizeof(ftrace) - 1;
+	int n_error = 0;  /* gcc-4.3.0 false positive complaint */
+
+	if (argc <= 1) {
+		fprintf(stderr, "usage: recordmcount file.o...\n");
+		return 0;
+	}
+
+	/* Process each file in turn, allowing deep failure. */
+	for (--argc, ++argv; 0 < argc; --argc, ++argv) {
+		int const sjval = setjmp(jmpenv);
+		int len;
+
+		/*
+		 * The file kernel/trace/ftrace.o references the mcount
+		 * function but does not call it. Since ftrace.o should
+		 * not be traced anyway, we just skip it.
+		 */
+		len = strlen(argv[0]);
+		if (len >= ftrace_size &&
+		    strcmp(argv[0] + (len - ftrace_size), ftrace) == 0)
+			continue;
+
+		switch (sjval) {
+		default: {
+			fprintf(stderr, "internal error: %s\n", argv[0]);
+			exit(1);
+		} break;
+		case SJ_SETJMP: {  /* normal sequence */
+			/* Avoid problems if early cleanup() */
+			fd_map = -1;
+			ehdr_curr = NULL;
+			mmap_failed = 1;
+			do_file(argv[0]);
+		} break;
+		case SJ_FAIL: {  /* error in do_file or below */
+			++n_error;
+		} break;
+		case SJ_SUCCEED: {  /* premature success */
+			/* do nothing */
+		} break;
+		}  /* end switch */
+	}
+	return !!n_error;
+}
+
+
diff --git a/scripts/recordmcount.h b/scripts/recordmcount.h
new file mode 100644
index 0000000..7f39d09
--- /dev/null
+++ b/scripts/recordmcount.h
@@ -0,0 +1,366 @@
+/*
+ * recordmcount.h
+ *
+ * This code was taken out of recordmcount.c written by
+ * Copyright 2009 John F. Reiser <jreiser@BitWagon.com>.  All rights reserved.
+ *
+ * The original code had the same algorithms for both 32bit
+ * and 64bit ELF files, but the code was duplicated to support
+ * the difference in structures that were used. This
+ * file creates a macro of everything that is different between
+ * the 64 and 32 bit code, such that by including this header
+ * twice we can create both sets of functions by including this
+ * header once with RECORD_MCOUNT_64 undefined, and again with
+ * it defined.
+ *
+ * This conversion to macros was done by:
+ * Copyright 2010 Steven Rostedt <srostedt@redhat.com>, Red Hat Inc.
+ *
+ * Licensed under the GNU General Public License, version 2 (GPLv2).
+ */
+#undef append_func
+#undef sift_rel_mcount
+#undef find_secsym_ndx
+#undef __has_rel_mcount
+#undef has_rel_mcount
+#undef tot_relsize
+#undef do_func
+#undef Elf_Ehdr
+#undef Elf_Shdr
+#undef Elf_Rel
+#undef Elf_Rela
+#undef Elf_Sym
+#undef ELF_R_SYM
+#undef ELF_R_INFO
+#undef ELF_ST_BIND
+#undef uint_t
+#undef _w
+#undef _align
+#undef _size
+
+#ifdef RECORD_MCOUNT_64
+# define append_func		append64
+# define sift_rel_mcount	sift64_rel_mcount
+# define find_secsym_ndx	find64_secsym_ndx
+# define __has_rel_mcount	__has64_rel_mcount
+# define has_rel_mcount		has64_rel_mcount
+# define tot_relsize		tot64_relsize
+# define do_func		do64
+# define Elf_Ehdr		Elf64_Ehdr
+# define Elf_Shdr		Elf64_Shdr
+# define Elf_Rel		Elf64_Rel
+# define Elf_Rela		Elf64_Rela
+# define Elf_Sym		Elf64_Sym
+# define ELF_R_SYM		ELF64_R_SYM
+# define ELF_R_INFO		ELF64_R_INFO
+# define ELF_ST_BIND		ELF64_ST_BIND
+# define uint_t			uint64_t
+# define _w			w8
+# define _align			7u
+# define _size			8
+#else
+# define append_func		append32
+# define sift_rel_mcount	sift32_rel_mcount
+# define find_secsym_ndx	find32_secsym_ndx
+# define __has_rel_mcount	__has32_rel_mcount
+# define has_rel_mcount		has32_rel_mcount
+# define tot_relsize		tot32_relsize
+# define do_func		do32
+# define Elf_Ehdr		Elf32_Ehdr
+# define Elf_Shdr		Elf32_Shdr
+# define Elf_Rel		Elf32_Rel
+# define Elf_Rela		Elf32_Rela
+# define Elf_Sym		Elf32_Sym
+# define ELF_R_SYM		ELF32_R_SYM
+# define ELF_R_INFO		ELF32_R_INFO
+# define ELF_ST_BIND		ELF32_ST_BIND
+# define uint_t			uint32_t
+# define _w			w
+# define _align			3u
+# define _size			4
+#endif
+
+/* Append the new shstrtab, Elf_Shdr[], __mcount_loc and its relocations. */
+static void append_func(Elf_Ehdr *const ehdr,
+			Elf_Shdr *const shstr,
+			uint_t const *const mloc0,
+			uint_t const *const mlocp,
+			Elf_Rel const *const mrel0,
+			Elf_Rel const *const mrelp,
+			unsigned int const rel_entsize,
+			unsigned int const symsec_sh_link)
+{
+	/* Begin constructing output file */
+	Elf_Shdr mcsec;
+	char const *mc_name = (sizeof(Elf_Rela) == rel_entsize)
+		? ".rela__mcount_loc"
+		:  ".rel__mcount_loc";
+	unsigned const old_shnum = w2(ehdr->e_shnum);
+	uint_t const old_shoff = _w(ehdr->e_shoff);
+	uint_t const old_shstr_sh_size   = _w(shstr->sh_size);
+	uint_t const old_shstr_sh_offset = _w(shstr->sh_offset);
+	uint_t t = 1 + strlen(mc_name) + _w(shstr->sh_size);
+	uint_t new_e_shoff;
+
+	shstr->sh_size = _w(t);
+	shstr->sh_offset = _w(sb.st_size);
+	t += sb.st_size;
+	t += (_align & -t);  /* word-byte align */
+	new_e_shoff = t;
+
+	/* body for new shstrtab */
+	ulseek(fd_map, sb.st_size, SEEK_SET);
+	uwrite(fd_map, old_shstr_sh_offset + (void *)ehdr, old_shstr_sh_size);
+	uwrite(fd_map, mc_name, 1 + strlen(mc_name));
+
+	/* old(modified) Elf_Shdr table, word-byte aligned */
+	ulseek(fd_map, t, SEEK_SET);
+	t += sizeof(Elf_Shdr) * old_shnum;
+	uwrite(fd_map, old_shoff + (void *)ehdr,
+	       sizeof(Elf_Shdr) * old_shnum);
+
+	/* new sections __mcount_loc and .rel__mcount_loc */
+	t += 2*sizeof(mcsec);
+	mcsec.sh_name = w((sizeof(Elf_Rela) == rel_entsize) + strlen(".rel")
+		+ old_shstr_sh_size);
+	mcsec.sh_type = w(SHT_PROGBITS);
+	mcsec.sh_flags = _w(SHF_ALLOC);
+	mcsec.sh_addr = 0;
+	mcsec.sh_offset = _w(t);
+	mcsec.sh_size = _w((void *)mlocp - (void *)mloc0);
+	mcsec.sh_link = 0;
+	mcsec.sh_info = 0;
+	mcsec.sh_addralign = _w(_size);
+	mcsec.sh_entsize = _w(_size);
+	uwrite(fd_map, &mcsec, sizeof(mcsec));
+
+	mcsec.sh_name = w(old_shstr_sh_size);
+	mcsec.sh_type = (sizeof(Elf_Rela) == rel_entsize)
+		? w(SHT_RELA)
+		: w(SHT_REL);
+	mcsec.sh_flags = 0;
+	mcsec.sh_addr = 0;
+	mcsec.sh_offset = _w((void *)mlocp - (void *)mloc0 + t);
+	mcsec.sh_size   = _w((void *)mrelp - (void *)mrel0);
+	mcsec.sh_link = w(symsec_sh_link);
+	mcsec.sh_info = w(old_shnum);
+	mcsec.sh_addralign = _w(_size);
+	mcsec.sh_entsize = _w(rel_entsize);
+	uwrite(fd_map, &mcsec, sizeof(mcsec));
+
+	uwrite(fd_map, mloc0, (void *)mlocp - (void *)mloc0);
+	uwrite(fd_map, mrel0, (void *)mrelp - (void *)mrel0);
+
+	ehdr->e_shoff = _w(new_e_shoff);
+	ehdr->e_shnum = w2(2 + w2(ehdr->e_shnum));  /* {.rel,}__mcount_loc */
+	ulseek(fd_map, 0, SEEK_SET);
+	uwrite(fd_map, ehdr, sizeof(*ehdr));
+}
+
+
+/*
+ * Look at the relocations in order to find the calls to mcount.
+ * Accumulate the section offsets that are found, and their relocation info,
+ * onto the end of the existing arrays.
+ */
+static uint_t *sift_rel_mcount(uint_t *mlocp,
+			       unsigned const offbase,
+			       Elf_Rel **const mrelpp,
+			       Elf_Shdr const *const relhdr,
+			       Elf_Ehdr const *const ehdr,
+			       unsigned const recsym,
+			       uint_t const recval,
+			       unsigned const reltype)
+{
+	uint_t *const mloc0 = mlocp;
+	Elf_Rel *mrelp = *mrelpp;
+	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
+		+ (void *)ehdr);
+	unsigned const symsec_sh_link = w(relhdr->sh_link);
+	Elf_Shdr const *const symsec = &shdr0[symsec_sh_link];
+	Elf_Sym const *const sym0 = (Elf_Sym const *)(_w(symsec->sh_offset)
+		+ (void *)ehdr);
+
+	Elf_Shdr const *const strsec = &shdr0[w(symsec->sh_link)];
+	char const *const str0 = (char const *)(_w(strsec->sh_offset)
+		+ (void *)ehdr);
+
+	Elf_Rel const *const rel0 = (Elf_Rel const *)(_w(relhdr->sh_offset)
+		+ (void *)ehdr);
+	unsigned rel_entsize = _w(relhdr->sh_entsize);
+	unsigned const nrel = _w(relhdr->sh_size) / rel_entsize;
+	Elf_Rel const *relp = rel0;
+
+	unsigned mcountsym = 0;
+	unsigned t;
+
+	for (t = nrel; t; --t) {
+		if (!mcountsym) {
+			Elf_Sym const *const symp =
+				&sym0[ELF_R_SYM(_w(relp->r_info))];
+			char const *symname = &str0[w(symp->st_name)];
+
+			if ('.' == symname[0])
+				++symname;  /* ppc64 hack */
+			if (0 == strcmp((('_' == gpfx) ? "_mcount" : "mcount"),
+					symname))
+				mcountsym = ELF_R_SYM(_w(relp->r_info));
+		}
+
+		if (mcountsym == ELF_R_SYM(_w(relp->r_info))) {
+			uint_t const addend = _w(_w(relp->r_offset) - recval);
+
+			mrelp->r_offset = _w(offbase
+				+ ((void *)mlocp - (void *)mloc0));
+			mrelp->r_info = _w(ELF_R_INFO(recsym, reltype));
+			if (sizeof(Elf_Rela) == rel_entsize) {
+				((Elf_Rela *)mrelp)->r_addend = addend;
+				*mlocp++ = 0;
+			} else
+				*mlocp++ = addend;
+
+			mrelp = (Elf_Rel *)(rel_entsize + (void *)mrelp);
+		}
+		relp = (Elf_Rel const *)(rel_entsize + (void *)relp);
+	}
+	*mrelpp = mrelp;
+	return mlocp;
+}
+
+
+/*
+ * Find a symbol in the given section, to be used as the base for relocating
+ * the table of offsets of calls to mcount.  A local or global symbol suffices,
+ * but avoid a Weak symbol because it may be overridden; the change in value
+ * would invalidate the relocations of the offsets of the calls to mcount.
+ * Often the found symbol will be the unnamed local symbol generated by
+ * GNU 'as' for the start of each section.  For example:
+ *    Num:    Value  Size Type    Bind   Vis      Ndx Name
+ *      2: 00000000     0 SECTION LOCAL  DEFAULT    1
+ */
+static unsigned find_secsym_ndx(unsigned const txtndx,
+				char const *const txtname,
+				uint_t *const recvalp,
+				Elf_Shdr const *const symhdr,
+				Elf_Ehdr const *const ehdr)
+{
+	Elf_Sym const *const sym0 = (Elf_Sym const *)(_w(symhdr->sh_offset)
+		+ (void *)ehdr);
+	unsigned const nsym = _w(symhdr->sh_size) / _w(symhdr->sh_entsize);
+	Elf_Sym const *symp;
+	unsigned t;
+
+	for (symp = sym0, t = nsym; t; --t, ++symp) {
+		unsigned int const st_bind = ELF_ST_BIND(symp->st_info);
+
+		if (txtndx == w2(symp->st_shndx)
+			/* avoid STB_WEAK */
+		    && (STB_LOCAL == st_bind || STB_GLOBAL == st_bind)) {
+			*recvalp = _w(symp->st_value);
+			return symp - sym0;
+		}
+	}
+	fprintf(stderr, "Cannot find symbol for section %d: %s.\n",
+		txtndx, txtname);
+	fail_file();
+}
+
+
+/* Evade ISO C restriction: no declaration after statement in has_rel_mcount. */
+static char const *
+__has_rel_mcount(Elf_Shdr const *const relhdr,  /* is SHT_REL or SHT_RELA */
+		 Elf_Shdr const *const shdr0,
+		 char const *const shstrtab,
+		 char const *const fname)
+{
+	/* .sh_info depends on .sh_type == SHT_REL[,A] */
+	Elf_Shdr const *const txthdr = &shdr0[w(relhdr->sh_info)];
+	char const *const txtname = &shstrtab[w(txthdr->sh_name)];
+
+	if (0 == strcmp("__mcount_loc", txtname)) {
+		fprintf(stderr, "warning: __mcount_loc already exists: %s\n",
+			fname);
+		succeed_file();
+	}
+	if (SHT_PROGBITS != w(txthdr->sh_type) ||
+	    !is_mcounted_section_name(txtname))
+		return NULL;
+	return txtname;
+}
+
+static char const *has_rel_mcount(Elf_Shdr const *const relhdr,
+				  Elf_Shdr const *const shdr0,
+				  char const *const shstrtab,
+				  char const *const fname)
+{
+	if (SHT_REL  != w(relhdr->sh_type) && SHT_RELA != w(relhdr->sh_type))
+		return NULL;
+	return __has_rel_mcount(relhdr, shdr0, shstrtab, fname);
+}
+
+
+static unsigned tot_relsize(Elf_Shdr const *const shdr0,
+			    unsigned nhdr,
+			    const char *const shstrtab,
+			    const char *const fname)
+{
+	unsigned totrelsz = 0;
+	Elf_Shdr const *shdrp = shdr0;
+
+	for (; nhdr; --nhdr, ++shdrp) {
+		if (has_rel_mcount(shdrp, shdr0, shstrtab, fname))
+			totrelsz += _w(shdrp->sh_size);
+	}
+	return totrelsz;
+}
+
+
+/* Overall supervision for Elf32 ET_REL file. */
+static void
+do_func(Elf_Ehdr *const ehdr, char const *const fname, unsigned const reltype)
+{
+	Elf_Shdr *const shdr0 = (Elf_Shdr *)(_w(ehdr->e_shoff)
+		+ (void *)ehdr);
+	unsigned const nhdr = w2(ehdr->e_shnum);
+	Elf_Shdr *const shstr = &shdr0[w2(ehdr->e_shstrndx)];
+	char const *const shstrtab = (char const *)(_w(shstr->sh_offset)
+		+ (void *)ehdr);
+
+	Elf_Shdr const *relhdr;
+	unsigned k;
+
+	/* Upper bound on space: assume all relevant relocs are for mcount. */
+	unsigned const totrelsz = tot_relsize(shdr0, nhdr, shstrtab, fname);
+	Elf_Rel *const mrel0 = umalloc(totrelsz);
+	Elf_Rel *      mrelp = mrel0;
+
+	/* 2*sizeof(address) <= sizeof(Elf_Rel) */
+	uint_t *const mloc0 = umalloc(totrelsz>>1);
+	uint_t *      mlocp = mloc0;
+
+	unsigned rel_entsize = 0;
+	unsigned symsec_sh_link = 0;
+
+	for (relhdr = shdr0, k = nhdr; k; --k, ++relhdr) {
+		char const *const txtname = has_rel_mcount(relhdr, shdr0,
+			shstrtab, fname);
+		if (txtname) {
+			uint_t recval = 0;
+			unsigned const recsym = find_secsym_ndx(
+				w(relhdr->sh_info), txtname, &recval,
+				&shdr0[symsec_sh_link = w(relhdr->sh_link)],
+				ehdr);
+
+			rel_entsize = _w(relhdr->sh_entsize);
+			mlocp = sift_rel_mcount(mlocp,
+				(void *)mlocp - (void *)mloc0, &mrelp,
+				relhdr, ehdr, recsym, recval, reltype);
+		}
+	}
+	if (mloc0 != mlocp) {
+		append_func(ehdr, shstr, mloc0, mlocp, mrel0, mrelp,
+			    rel_entsize, symsec_sh_link);
+	}
+	free(mrel0);
+	free(mloc0);
+}
diff --git a/security/apparmor/.gitignore b/security/apparmor/.gitignore
index 0a0a99f..4d995ae 100644
--- a/security/apparmor/.gitignore
+++ b/security/apparmor/.gitignore
@@ -3,3 +3,4 @@
 #
 af_names.h
 capability_names.h
+rlim_names.h
diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c
index 7320331..544ff58 100644
--- a/security/apparmor/apparmorfs.c
+++ b/security/apparmor/apparmorfs.c
@@ -29,7 +29,7 @@
  * aa_simple_write_to_buffer - common routine for getting policy from user
  * @op: operation doing the user buffer copy
  * @userbuf: user buffer to copy data from  (NOT NULL)
- * @alloc_size: size of user buffer
+ * @alloc_size: size of user buffer (REQUIRES: @alloc_size >= @copy_size)
  * @copy_size: size of data to copy from user buffer
  * @pos: position write is at in the file (NOT NULL)
  *
@@ -42,6 +42,8 @@
 {
 	char *data;
 
+	BUG_ON(copy_size > alloc_size);
+
 	if (*pos != 0)
 		/* only writes from pos 0, that is complete writes */
 		return ERR_PTR(-ESPIPE);
diff --git a/security/capability.c b/security/capability.c
index 95a6599..30ae00f 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -677,7 +677,18 @@
 {
 }
 
+static int cap_secmark_relabel_packet(u32 secid)
+{
+	return 0;
+}
 
+static void cap_secmark_refcount_inc(void)
+{
+}
+
+static void cap_secmark_refcount_dec(void)
+{
+}
 
 static void cap_req_classify_flow(const struct request_sock *req,
 				  struct flowi *fl)
@@ -777,7 +788,8 @@
 
 static int cap_secctx_to_secid(const char *secdata, u32 seclen, u32 *secid)
 {
-	return -EOPNOTSUPP;
+	*secid = 0;
+	return 0;
 }
 
 static void cap_release_secctx(char *secdata, u32 seclen)
@@ -1018,6 +1030,9 @@
 	set_to_cap_if_null(ops, inet_conn_request);
 	set_to_cap_if_null(ops, inet_csk_clone);
 	set_to_cap_if_null(ops, inet_conn_established);
+	set_to_cap_if_null(ops, secmark_relabel_packet);
+	set_to_cap_if_null(ops, secmark_refcount_inc);
+	set_to_cap_if_null(ops, secmark_refcount_dec);
 	set_to_cap_if_null(ops, req_classify_flow);
 	set_to_cap_if_null(ops, tun_dev_create);
 	set_to_cap_if_null(ops, tun_dev_post_create);
diff --git a/security/commoncap.c b/security/commoncap.c
index 9d172e6..5e632b4 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -719,14 +719,11 @@
 /**
  * cap_task_setscheduler - Detemine if scheduler policy change is permitted
  * @p: The task to affect
- * @policy: The policy to effect
- * @lp: The parameters to the scheduling policy
  *
  * Detemine if the requested scheduler policy change is permitted for the
  * specified task, returning 0 if permission is granted, -ve if denied.
  */
-int cap_task_setscheduler(struct task_struct *p, int policy,
-			   struct sched_param *lp)
+int cap_task_setscheduler(struct task_struct *p)
 {
 	return cap_safe_nice(p);
 }
diff --git a/security/security.c b/security/security.c
index c53949f..b50f472 100644
--- a/security/security.c
+++ b/security/security.c
@@ -89,20 +89,12 @@
  * Return true if:
  *	-The passed LSM is the one chosen by user at boot time,
  *	-or the passed LSM is configured as the default and the user did not
- *	 choose an alternate LSM at boot time,
- *	-or there is no default LSM set and the user didn't specify a
- *	 specific LSM and we're the first to ask for registration permission,
- *	-or the passed LSM is currently loaded.
+ *	 choose an alternate LSM at boot time.
  * Otherwise, return false.
  */
 int __init security_module_enable(struct security_operations *ops)
 {
-	if (!*chosen_lsm)
-		strncpy(chosen_lsm, ops->name, SECURITY_NAME_MAX);
-	else if (strncmp(ops->name, chosen_lsm, SECURITY_NAME_MAX))
-		return 0;
-
-	return 1;
+	return !strcmp(ops->name, chosen_lsm);
 }
 
 /**
@@ -786,10 +778,9 @@
 	return security_ops->task_setrlimit(p, resource, new_rlim);
 }
 
-int security_task_setscheduler(struct task_struct *p,
-				int policy, struct sched_param *lp)
+int security_task_setscheduler(struct task_struct *p)
 {
-	return security_ops->task_setscheduler(p, policy, lp);
+	return security_ops->task_setscheduler(p);
 }
 
 int security_task_getscheduler(struct task_struct *p)
@@ -1145,6 +1136,24 @@
 	security_ops->inet_conn_established(sk, skb);
 }
 
+int security_secmark_relabel_packet(u32 secid)
+{
+	return security_ops->secmark_relabel_packet(secid);
+}
+EXPORT_SYMBOL(security_secmark_relabel_packet);
+
+void security_secmark_refcount_inc(void)
+{
+	security_ops->secmark_refcount_inc();
+}
+EXPORT_SYMBOL(security_secmark_refcount_inc);
+
+void security_secmark_refcount_dec(void)
+{
+	security_ops->secmark_refcount_dec();
+}
+EXPORT_SYMBOL(security_secmark_refcount_dec);
+
 int security_tun_dev_create(void)
 {
 	return security_ops->tun_dev_create();
diff --git a/security/selinux/Makefile b/security/selinux/Makefile
index 58d80f3..ad5cd76 100644
--- a/security/selinux/Makefile
+++ b/security/selinux/Makefile
@@ -2,25 +2,20 @@
 # Makefile for building the SELinux module as part of the kernel tree.
 #
 
-obj-$(CONFIG_SECURITY_SELINUX) := selinux.o ss/
+obj-$(CONFIG_SECURITY_SELINUX) := selinux.o
 
-selinux-y := avc.o \
-	     hooks.o \
-	     selinuxfs.o \
-	     netlink.o \
-	     nlmsgtab.o \
-	     netif.o \
-	     netnode.o \
-	     netport.o \
-	     exports.o
+selinux-y := avc.o hooks.o selinuxfs.o netlink.o nlmsgtab.o netif.o \
+	     netnode.o netport.o exports.o \
+	     ss/ebitmap.o ss/hashtab.o ss/symtab.o ss/sidtab.o ss/avtab.o \
+	     ss/policydb.o ss/services.o ss/conditional.o ss/mls.o ss/status.o
 
 selinux-$(CONFIG_SECURITY_NETWORK_XFRM) += xfrm.o
 
 selinux-$(CONFIG_NETLABEL) += netlabel.o
 
-EXTRA_CFLAGS += -Isecurity/selinux -Isecurity/selinux/include
+ccflags-y := -Isecurity/selinux -Isecurity/selinux/include
 
-$(obj)/avc.o: $(obj)/flask.h
+$(addprefix $(obj)/,$(selinux-y)): $(obj)/flask.h
 
 quiet_cmd_flask = GEN     $(obj)/flask.h $(obj)/av_permissions.h
       cmd_flask = scripts/selinux/genheaders/genheaders $(obj)/flask.h $(obj)/av_permissions.h
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index c0a454a..9066438 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -11,58 +11,9 @@
  * it under the terms of the GNU General Public License version 2,
  * as published by the Free Software Foundation.
  */
-#include <linux/types.h>
-#include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/selinux.h>
-#include <linux/fs.h>
-#include <linux/ipc.h>
-#include <asm/atomic.h>
 
 #include "security.h"
-#include "objsec.h"
-
-/* SECMARK reference count */
-extern atomic_t selinux_secmark_refcount;
-
-int selinux_string_to_sid(char *str, u32 *sid)
-{
-	if (selinux_enabled)
-		return security_context_to_sid(str, strlen(str), sid);
-	else {
-		*sid = 0;
-		return 0;
-	}
-}
-EXPORT_SYMBOL_GPL(selinux_string_to_sid);
-
-int selinux_secmark_relabel_packet_permission(u32 sid)
-{
-	if (selinux_enabled) {
-		const struct task_security_struct *__tsec;
-		u32 tsid;
-
-		__tsec = current_security();
-		tsid = __tsec->sid;
-
-		return avc_has_perm(tsid, sid, SECCLASS_PACKET,
-				    PACKET__RELABELTO, NULL);
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_relabel_packet_permission);
-
-void selinux_secmark_refcount_inc(void)
-{
-	atomic_inc(&selinux_secmark_refcount);
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_refcount_inc);
-
-void selinux_secmark_refcount_dec(void)
-{
-	atomic_dec(&selinux_secmark_refcount);
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_refcount_dec);
 
 bool selinux_is_enabled(void)
 {
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 4796ddd..d9154cf 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3354,11 +3354,11 @@
 	return 0;
 }
 
-static int selinux_task_setscheduler(struct task_struct *p, int policy, struct sched_param *lp)
+static int selinux_task_setscheduler(struct task_struct *p)
 {
 	int rc;
 
-	rc = cap_task_setscheduler(p, policy, lp);
+	rc = cap_task_setscheduler(p);
 	if (rc)
 		return rc;
 
@@ -4279,6 +4279,27 @@
 	selinux_skb_peerlbl_sid(skb, family, &sksec->peer_sid);
 }
 
+static int selinux_secmark_relabel_packet(u32 sid)
+{
+	const struct task_security_struct *__tsec;
+	u32 tsid;
+
+	__tsec = current_security();
+	tsid = __tsec->sid;
+
+	return avc_has_perm(tsid, sid, SECCLASS_PACKET, PACKET__RELABELTO, NULL);
+}
+
+static void selinux_secmark_refcount_inc(void)
+{
+	atomic_inc(&selinux_secmark_refcount);
+}
+
+static void selinux_secmark_refcount_dec(void)
+{
+	atomic_dec(&selinux_secmark_refcount);
+}
+
 static void selinux_req_classify_flow(const struct request_sock *req,
 				      struct flowi *fl)
 {
@@ -5533,6 +5554,9 @@
 	.inet_conn_request =		selinux_inet_conn_request,
 	.inet_csk_clone =		selinux_inet_csk_clone,
 	.inet_conn_established =	selinux_inet_conn_established,
+	.secmark_relabel_packet =	selinux_secmark_relabel_packet,
+	.secmark_refcount_inc =		selinux_secmark_refcount_inc,
+	.secmark_refcount_dec =		selinux_secmark_refcount_dec,
 	.req_classify_flow =		selinux_req_classify_flow,
 	.tun_dev_create =		selinux_tun_dev_create,
 	.tun_dev_post_create = 		selinux_tun_dev_post_create,
diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h
index b4c9eb4..8858d2b 100644
--- a/security/selinux/include/classmap.h
+++ b/security/selinux/include/classmap.h
@@ -17,7 +17,7 @@
 	  { "compute_av", "compute_create", "compute_member",
 	    "check_context", "load_policy", "compute_relabel",
 	    "compute_user", "setenforce", "setbool", "setsecparam",
-	    "setcheckreqprot", NULL } },
+	    "setcheckreqprot", "read_policy", NULL } },
 	{ "process",
 	  { "fork", "transition", "sigchld", "sigkill",
 	    "sigstop", "signull", "signal", "ptrace", "getsched", "setsched",
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 1f7c249..671273e 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -9,6 +9,7 @@
 #define _SELINUX_SECURITY_H_
 
 #include <linux/magic.h>
+#include <linux/types.h>
 #include "flask.h"
 
 #define SECSID_NULL			0x00000000 /* unspecified SID */
@@ -82,6 +83,8 @@
 int security_mls_enabled(void);
 
 int security_load_policy(void *data, size_t len);
+int security_read_policy(void **data, ssize_t *len);
+size_t security_policydb_len(void);
 
 int security_policycap_supported(unsigned int req_cap);
 
@@ -191,5 +194,25 @@
 
 const char *security_get_initial_sid_context(u32 sid);
 
+/*
+ * status notifier using mmap interface
+ */
+extern struct page *selinux_kernel_status_page(void);
+
+#define SELINUX_KERNEL_STATUS_VERSION	1
+struct selinux_kernel_status {
+	u32	version;	/* version number of thie structure */
+	u32	sequence;	/* sequence number of seqlock logic */
+	u32	enforcing;	/* current setting of enforcing mode */
+	u32	policyload;	/* times of policy reloaded */
+	u32	deny_unknown;	/* current setting of deny_unknown */
+	/*
+	 * The version > 0 supports above members.
+	 */
+} __attribute__((packed));
+
+extern void selinux_status_update_setenforce(int enforcing);
+extern void selinux_status_update_policyload(int seqno);
+
 #endif /* _SELINUX_SECURITY_H_ */
 
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index 79a1bb6..87e0556 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -68,6 +68,8 @@
 static struct dentry *class_dir;
 static unsigned long last_class_ino;
 
+static char policy_opened;
+
 /* global data for policy capabilities */
 static struct dentry *policycap_dir;
 
@@ -110,6 +112,8 @@
 	SEL_COMPAT_NET,	/* whether to use old compat network packet controls */
 	SEL_REJECT_UNKNOWN, /* export unknown reject handling to userspace */
 	SEL_DENY_UNKNOWN, /* export unknown deny handling to userspace */
+	SEL_STATUS,	/* export current status using mmap() */
+	SEL_POLICY,	/* allow userspace to read the in kernel policy */
 	SEL_INO_NEXT,	/* The next inode number to use */
 };
 
@@ -171,6 +175,7 @@
 		if (selinux_enforcing)
 			avc_ss_reset(0);
 		selnl_notify_setenforce(selinux_enforcing);
+		selinux_status_update_setenforce(selinux_enforcing);
 	}
 	length = count;
 out:
@@ -205,6 +210,59 @@
 	.llseek		= generic_file_llseek,
 };
 
+static int sel_open_handle_status(struct inode *inode, struct file *filp)
+{
+	struct page    *status = selinux_kernel_status_page();
+
+	if (!status)
+		return -ENOMEM;
+
+	filp->private_data = status;
+
+	return 0;
+}
+
+static ssize_t sel_read_handle_status(struct file *filp, char __user *buf,
+				      size_t count, loff_t *ppos)
+{
+	struct page    *status = filp->private_data;
+
+	BUG_ON(!status);
+
+	return simple_read_from_buffer(buf, count, ppos,
+				       page_address(status),
+				       sizeof(struct selinux_kernel_status));
+}
+
+static int sel_mmap_handle_status(struct file *filp,
+				  struct vm_area_struct *vma)
+{
+	struct page    *status = filp->private_data;
+	unsigned long	size = vma->vm_end - vma->vm_start;
+
+	BUG_ON(!status);
+
+	/* only allows one page from the head */
+	if (vma->vm_pgoff > 0 || size != PAGE_SIZE)
+		return -EIO;
+	/* disallow writable mapping */
+	if (vma->vm_flags & VM_WRITE)
+		return -EPERM;
+	/* disallow mprotect() turns it into writable */
+	vma->vm_flags &= ~VM_MAYWRITE;
+
+	return remap_pfn_range(vma, vma->vm_start,
+			       page_to_pfn(status),
+			       size, vma->vm_page_prot);
+}
+
+static const struct file_operations sel_handle_status_ops = {
+	.open		= sel_open_handle_status,
+	.read		= sel_read_handle_status,
+	.mmap		= sel_mmap_handle_status,
+	.llseek		= generic_file_llseek,
+};
+
 #ifdef CONFIG_SECURITY_SELINUX_DISABLE
 static ssize_t sel_write_disable(struct file *file, const char __user *buf,
 				 size_t count, loff_t *ppos)
@@ -296,6 +354,141 @@
 	.llseek		= generic_file_llseek,
 };
 
+struct policy_load_memory {
+	size_t len;
+	void *data;
+};
+
+static int sel_open_policy(struct inode *inode, struct file *filp)
+{
+	struct policy_load_memory *plm = NULL;
+	int rc;
+
+	BUG_ON(filp->private_data);
+
+	mutex_lock(&sel_mutex);
+
+	rc = task_has_security(current, SECURITY__READ_POLICY);
+	if (rc)
+		goto err;
+
+	rc = -EBUSY;
+	if (policy_opened)
+		goto err;
+
+	rc = -ENOMEM;
+	plm = kzalloc(sizeof(*plm), GFP_KERNEL);
+	if (!plm)
+		goto err;
+
+	if (i_size_read(inode) != security_policydb_len()) {
+		mutex_lock(&inode->i_mutex);
+		i_size_write(inode, security_policydb_len());
+		mutex_unlock(&inode->i_mutex);
+	}
+
+	rc = security_read_policy(&plm->data, &plm->len);
+	if (rc)
+		goto err;
+
+	policy_opened = 1;
+
+	filp->private_data = plm;
+
+	mutex_unlock(&sel_mutex);
+
+	return 0;
+err:
+	mutex_unlock(&sel_mutex);
+
+	if (plm)
+		vfree(plm->data);
+	kfree(plm);
+	return rc;
+}
+
+static int sel_release_policy(struct inode *inode, struct file *filp)
+{
+	struct policy_load_memory *plm = filp->private_data;
+
+	BUG_ON(!plm);
+
+	policy_opened = 0;
+
+	vfree(plm->data);
+	kfree(plm);
+
+	return 0;
+}
+
+static ssize_t sel_read_policy(struct file *filp, char __user *buf,
+			       size_t count, loff_t *ppos)
+{
+	struct policy_load_memory *plm = filp->private_data;
+	int ret;
+
+	mutex_lock(&sel_mutex);
+
+	ret = task_has_security(current, SECURITY__READ_POLICY);
+	if (ret)
+		goto out;
+
+	ret = simple_read_from_buffer(buf, count, ppos, plm->data, plm->len);
+out:
+	mutex_unlock(&sel_mutex);
+	return ret;
+}
+
+static int sel_mmap_policy_fault(struct vm_area_struct *vma,
+				 struct vm_fault *vmf)
+{
+	struct policy_load_memory *plm = vma->vm_file->private_data;
+	unsigned long offset;
+	struct page *page;
+
+	if (vmf->flags & (FAULT_FLAG_MKWRITE | FAULT_FLAG_WRITE))
+		return VM_FAULT_SIGBUS;
+
+	offset = vmf->pgoff << PAGE_SHIFT;
+	if (offset >= roundup(plm->len, PAGE_SIZE))
+		return VM_FAULT_SIGBUS;
+
+	page = vmalloc_to_page(plm->data + offset);
+	get_page(page);
+
+	vmf->page = page;
+
+	return 0;
+}
+
+static struct vm_operations_struct sel_mmap_policy_ops = {
+	.fault = sel_mmap_policy_fault,
+	.page_mkwrite = sel_mmap_policy_fault,
+};
+
+int sel_mmap_policy(struct file *filp, struct vm_area_struct *vma)
+{
+	if (vma->vm_flags & VM_SHARED) {
+		/* do not allow mprotect to make mapping writable */
+		vma->vm_flags &= ~VM_MAYWRITE;
+
+		if (vma->vm_flags & VM_WRITE)
+			return -EACCES;
+	}
+
+	vma->vm_flags |= VM_RESERVED;
+	vma->vm_ops = &sel_mmap_policy_ops;
+
+	return 0;
+}
+
+static const struct file_operations sel_policy_ops = {
+	.open		= sel_open_policy,
+	.read		= sel_read_policy,
+	.mmap		= sel_mmap_policy,
+	.release	= sel_release_policy,
+};
+
 static ssize_t sel_write_load(struct file *file, const char __user *buf,
 			      size_t count, loff_t *ppos)
 
@@ -1612,6 +1805,8 @@
 		[SEL_CHECKREQPROT] = {"checkreqprot", &sel_checkreqprot_ops, S_IRUGO|S_IWUSR},
 		[SEL_REJECT_UNKNOWN] = {"reject_unknown", &sel_handle_unknown_ops, S_IRUGO},
 		[SEL_DENY_UNKNOWN] = {"deny_unknown", &sel_handle_unknown_ops, S_IRUGO},
+		[SEL_STATUS] = {"status", &sel_handle_status_ops, S_IRUGO},
+		[SEL_POLICY] = {"policy", &sel_policy_ops, S_IRUSR},
 		/* last one */ {""}
 	};
 	ret = simple_fill_super(sb, SELINUX_MAGIC, selinux_files);
diff --git a/security/selinux/ss/Makefile b/security/selinux/ss/Makefile
deleted file mode 100644
index 15d4e62..0000000
--- a/security/selinux/ss/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-#
-# Makefile for building the SELinux security server as part of the kernel tree.
-#
-
-EXTRA_CFLAGS += -Isecurity/selinux -Isecurity/selinux/include
-obj-y := ss.o
-
-ss-y := ebitmap.o hashtab.o symtab.o sidtab.o avtab.o policydb.o services.o conditional.o mls.o
-
diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c
index 929480c..a3dd9fa 100644
--- a/security/selinux/ss/avtab.c
+++ b/security/selinux/ss/avtab.c
@@ -266,8 +266,8 @@
 	if (shift > 2)
 		shift = shift - 2;
 	nslot = 1 << shift;
-	if (nslot > MAX_AVTAB_SIZE)
-		nslot = MAX_AVTAB_SIZE;
+	if (nslot > MAX_AVTAB_HASH_BUCKETS)
+		nslot = MAX_AVTAB_HASH_BUCKETS;
 	mask = nslot - 1;
 
 	h->htable = kcalloc(nslot, sizeof(*(h->htable)), GFP_KERNEL);
@@ -501,6 +501,48 @@
 	goto out;
 }
 
+int avtab_write_item(struct policydb *p, struct avtab_node *cur, void *fp)
+{
+	__le16 buf16[4];
+	__le32 buf32[1];
+	int rc;
+
+	buf16[0] = cpu_to_le16(cur->key.source_type);
+	buf16[1] = cpu_to_le16(cur->key.target_type);
+	buf16[2] = cpu_to_le16(cur->key.target_class);
+	buf16[3] = cpu_to_le16(cur->key.specified);
+	rc = put_entry(buf16, sizeof(u16), 4, fp);
+	if (rc)
+		return rc;
+	buf32[0] = cpu_to_le32(cur->datum.data);
+	rc = put_entry(buf32, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+	return 0;
+}
+
+int avtab_write(struct policydb *p, struct avtab *a, void *fp)
+{
+	unsigned int i;
+	int rc = 0;
+	struct avtab_node *cur;
+	__le32 buf[1];
+
+	buf[0] = cpu_to_le32(a->nel);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < a->nslot; i++) {
+		for (cur = a->htable[i]; cur; cur = cur->next) {
+			rc = avtab_write_item(p, cur, fp);
+			if (rc)
+				return rc;
+		}
+	}
+
+	return rc;
+}
 void avtab_cache_init(void)
 {
 	avtab_node_cachep = kmem_cache_create("avtab_node",
diff --git a/security/selinux/ss/avtab.h b/security/selinux/ss/avtab.h
index cd4f734..dff0c75 100644
--- a/security/selinux/ss/avtab.h
+++ b/security/selinux/ss/avtab.h
@@ -71,6 +71,8 @@
 		    void *p);
 
 int avtab_read(struct avtab *a, void *fp, struct policydb *pol);
+int avtab_write_item(struct policydb *p, struct avtab_node *cur, void *fp);
+int avtab_write(struct policydb *p, struct avtab *a, void *fp);
 
 struct avtab_node *avtab_insert_nonunique(struct avtab *h, struct avtab_key *key,
 					  struct avtab_datum *datum);
@@ -85,7 +87,6 @@
 #define MAX_AVTAB_HASH_BITS 11
 #define MAX_AVTAB_HASH_BUCKETS (1 << MAX_AVTAB_HASH_BITS)
 #define MAX_AVTAB_HASH_MASK (MAX_AVTAB_HASH_BUCKETS-1)
-#define MAX_AVTAB_SIZE MAX_AVTAB_HASH_BUCKETS
 
 #endif	/* _SS_AVTAB_H_ */
 
diff --git a/security/selinux/ss/conditional.c b/security/selinux/ss/conditional.c
index c91e150..655fe1c 100644
--- a/security/selinux/ss/conditional.c
+++ b/security/selinux/ss/conditional.c
@@ -490,6 +490,129 @@
 	return rc;
 }
 
+int cond_write_bool(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct cond_bool_datum *booldatum = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	__le32 buf[3];
+	u32 len;
+	int rc;
+
+	len = strlen(key);
+	buf[0] = cpu_to_le32(booldatum->value);
+	buf[1] = cpu_to_le32(booldatum->state);
+	buf[2] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 3, fp);
+	if (rc)
+		return rc;
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+	return 0;
+}
+
+/*
+ * cond_write_cond_av_list doesn't write out the av_list nodes.
+ * Instead it writes out the key/value pairs from the avtab. This
+ * is necessary because there is no way to uniquely identifying rules
+ * in the avtab so it is not possible to associate individual rules
+ * in the avtab with a conditional without saving them as part of
+ * the conditional. This means that the avtab with the conditional
+ * rules will not be saved but will be rebuilt on policy load.
+ */
+static int cond_write_av_list(struct policydb *p,
+			      struct cond_av_list *list, struct policy_file *fp)
+{
+	__le32 buf[1];
+	struct cond_av_list *cur_list;
+	u32 len;
+	int rc;
+
+	len = 0;
+	for (cur_list = list; cur_list != NULL; cur_list = cur_list->next)
+		len++;
+
+	buf[0] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	if (len == 0)
+		return 0;
+
+	for (cur_list = list; cur_list != NULL; cur_list = cur_list->next) {
+		rc = avtab_write_item(p, cur_list->node, fp);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
+
+int cond_write_node(struct policydb *p, struct cond_node *node,
+		    struct policy_file *fp)
+{
+	struct cond_expr *cur_expr;
+	__le32 buf[2];
+	int rc;
+	u32 len = 0;
+
+	buf[0] = cpu_to_le32(node->cur_state);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	for (cur_expr = node->expr; cur_expr != NULL; cur_expr = cur_expr->next)
+		len++;
+
+	buf[0] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	for (cur_expr = node->expr; cur_expr != NULL; cur_expr = cur_expr->next) {
+		buf[0] = cpu_to_le32(cur_expr->expr_type);
+		buf[1] = cpu_to_le32(cur_expr->bool);
+		rc = put_entry(buf, sizeof(u32), 2, fp);
+		if (rc)
+			return rc;
+	}
+
+	rc = cond_write_av_list(p, node->true_list, fp);
+	if (rc)
+		return rc;
+	rc = cond_write_av_list(p, node->false_list, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+int cond_write_list(struct policydb *p, struct cond_node *list, void *fp)
+{
+	struct cond_node *cur;
+	u32 len;
+	__le32 buf[1];
+	int rc;
+
+	len = 0;
+	for (cur = list; cur != NULL; cur = cur->next)
+		len++;
+	buf[0] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	for (cur = list; cur != NULL; cur = cur->next) {
+		rc = cond_write_node(p, cur, fp);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
 /* Determine whether additional permissions are granted by the conditional
  * av table, and if so, add them to the result
  */
diff --git a/security/selinux/ss/conditional.h b/security/selinux/ss/conditional.h
index 53ddb01..3f209c6 100644
--- a/security/selinux/ss/conditional.h
+++ b/security/selinux/ss/conditional.h
@@ -69,6 +69,8 @@
 
 int cond_read_bool(struct policydb *p, struct hashtab *h, void *fp);
 int cond_read_list(struct policydb *p, void *fp);
+int cond_write_bool(void *key, void *datum, void *ptr);
+int cond_write_list(struct policydb *p, struct cond_node *list, void *fp);
 
 void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decision *avd);
 
diff --git a/security/selinux/ss/ebitmap.c b/security/selinux/ss/ebitmap.c
index 04b6145..d42951f 100644
--- a/security/selinux/ss/ebitmap.c
+++ b/security/selinux/ss/ebitmap.c
@@ -22,6 +22,8 @@
 #include "ebitmap.h"
 #include "policydb.h"
 
+#define BITS_PER_U64	(sizeof(u64) * 8)
+
 int ebitmap_cmp(struct ebitmap *e1, struct ebitmap *e2)
 {
 	struct ebitmap_node *n1, *n2;
@@ -363,10 +365,10 @@
 	e->highbit = le32_to_cpu(buf[1]);
 	count = le32_to_cpu(buf[2]);
 
-	if (mapunit != sizeof(u64) * 8) {
+	if (mapunit != BITS_PER_U64) {
 		printk(KERN_ERR "SELinux: ebitmap: map size %u does not "
 		       "match my size %Zd (high bit was %d)\n",
-		       mapunit, sizeof(u64) * 8, e->highbit);
+		       mapunit, BITS_PER_U64, e->highbit);
 		goto bad;
 	}
 
@@ -446,3 +448,78 @@
 	ebitmap_destroy(e);
 	goto out;
 }
+
+int ebitmap_write(struct ebitmap *e, void *fp)
+{
+	struct ebitmap_node *n;
+	u32 count;
+	__le32 buf[3];
+	u64 map;
+	int bit, last_bit, last_startbit, rc;
+
+	buf[0] = cpu_to_le32(BITS_PER_U64);
+
+	count = 0;
+	last_bit = 0;
+	last_startbit = -1;
+	ebitmap_for_each_positive_bit(e, n, bit) {
+		if (rounddown(bit, (int)BITS_PER_U64) > last_startbit) {
+			count++;
+			last_startbit = rounddown(bit, BITS_PER_U64);
+		}
+		last_bit = roundup(bit + 1, BITS_PER_U64);
+	}
+	buf[1] = cpu_to_le32(last_bit);
+	buf[2] = cpu_to_le32(count);
+
+	rc = put_entry(buf, sizeof(u32), 3, fp);
+	if (rc)
+		return rc;
+
+	map = 0;
+	last_startbit = INT_MIN;
+	ebitmap_for_each_positive_bit(e, n, bit) {
+		if (rounddown(bit, (int)BITS_PER_U64) > last_startbit) {
+			__le64 buf64[1];
+
+			/* this is the very first bit */
+			if (!map) {
+				last_startbit = rounddown(bit, BITS_PER_U64);
+				map = (u64)1 << (bit - last_startbit);
+				continue;
+			}
+
+			/* write the last node */
+			buf[0] = cpu_to_le32(last_startbit);
+			rc = put_entry(buf, sizeof(u32), 1, fp);
+			if (rc)
+				return rc;
+
+			buf64[0] = cpu_to_le64(map);
+			rc = put_entry(buf64, sizeof(u64), 1, fp);
+			if (rc)
+				return rc;
+
+			/* set up for the next node */
+			map = 0;
+			last_startbit = rounddown(bit, BITS_PER_U64);
+		}
+		map |= (u64)1 << (bit - last_startbit);
+	}
+	/* write the last node */
+	if (map) {
+		__le64 buf64[1];
+
+		/* write the last node */
+		buf[0] = cpu_to_le32(last_startbit);
+		rc = put_entry(buf, sizeof(u32), 1, fp);
+		if (rc)
+			return rc;
+
+		buf64[0] = cpu_to_le64(map);
+		rc = put_entry(buf64, sizeof(u64), 1, fp);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
diff --git a/security/selinux/ss/ebitmap.h b/security/selinux/ss/ebitmap.h
index f283b43..1f4e93c 100644
--- a/security/selinux/ss/ebitmap.h
+++ b/security/selinux/ss/ebitmap.h
@@ -123,6 +123,7 @@
 int ebitmap_set_bit(struct ebitmap *e, unsigned long bit, int value);
 void ebitmap_destroy(struct ebitmap *e);
 int ebitmap_read(struct ebitmap *e, void *fp);
+int ebitmap_write(struct ebitmap *e, void *fp);
 
 #ifdef CONFIG_NETLABEL
 int ebitmap_netlbl_export(struct ebitmap *ebmap,
diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c
index 3a29704..94f630d 100644
--- a/security/selinux/ss/policydb.c
+++ b/security/selinux/ss/policydb.c
@@ -37,6 +37,7 @@
 #include "policydb.h"
 #include "conditional.h"
 #include "mls.h"
+#include "services.h"
 
 #define _DEBUG_HASHES
 
@@ -185,9 +186,19 @@
 static int rangetr_cmp(struct hashtab *h, const void *k1, const void *k2)
 {
 	const struct range_trans *key1 = k1, *key2 = k2;
-	return (key1->source_type != key2->source_type ||
-		key1->target_type != key2->target_type ||
-		key1->target_class != key2->target_class);
+	int v;
+
+	v = key1->source_type - key2->source_type;
+	if (v)
+		return v;
+
+	v = key1->target_type - key2->target_type;
+	if (v)
+		return v;
+
+	v = key1->target_class - key2->target_class;
+
+	return v;
 }
 
 /*
@@ -1624,11 +1635,11 @@
 
 static int type_bounds_sanity_check(void *key, void *datum, void *datap)
 {
-	struct type_datum *upper, *type;
+	struct type_datum *upper;
 	struct policydb *p = datap;
 	int depth = 0;
 
-	upper = type = datum;
+	upper = datum;
 	while (upper->bounds) {
 		if (++depth == POLICYDB_BOUNDS_MAXDEPTH) {
 			printk(KERN_ERR "SELinux: type %s: "
@@ -2306,3 +2317,843 @@
 	policydb_destroy(p);
 	goto out;
 }
+
+/*
+ * Write a MLS level structure to a policydb binary
+ * representation file.
+ */
+static int mls_write_level(struct mls_level *l, void *fp)
+{
+	__le32 buf[1];
+	int rc;
+
+	buf[0] = cpu_to_le32(l->sens);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	rc = ebitmap_write(&l->cat, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+/*
+ * Write a MLS range structure to a policydb binary
+ * representation file.
+ */
+static int mls_write_range_helper(struct mls_range *r, void *fp)
+{
+	__le32 buf[3];
+	size_t items;
+	int rc, eq;
+
+	eq = mls_level_eq(&r->level[1], &r->level[0]);
+
+	if (eq)
+		items = 2;
+	else
+		items = 3;
+	buf[0] = cpu_to_le32(items-1);
+	buf[1] = cpu_to_le32(r->level[0].sens);
+	if (!eq)
+		buf[2] = cpu_to_le32(r->level[1].sens);
+
+	BUG_ON(items > (sizeof(buf)/sizeof(buf[0])));
+
+	rc = put_entry(buf, sizeof(u32), items, fp);
+	if (rc)
+		return rc;
+
+	rc = ebitmap_write(&r->level[0].cat, fp);
+	if (rc)
+		return rc;
+	if (!eq) {
+		rc = ebitmap_write(&r->level[1].cat, fp);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
+
+static int sens_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct level_datum *levdatum = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	__le32 buf[2];
+	size_t len;
+	int rc;
+
+	len = strlen(key);
+	buf[0] = cpu_to_le32(len);
+	buf[1] = cpu_to_le32(levdatum->isalias);
+	rc = put_entry(buf, sizeof(u32), 2, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	rc = mls_write_level(levdatum->level, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int cat_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct cat_datum *catdatum = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	__le32 buf[3];
+	size_t len;
+	int rc;
+
+	len = strlen(key);
+	buf[0] = cpu_to_le32(len);
+	buf[1] = cpu_to_le32(catdatum->value);
+	buf[2] = cpu_to_le32(catdatum->isalias);
+	rc = put_entry(buf, sizeof(u32), 3, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int role_trans_write(struct role_trans *r, void *fp)
+{
+	struct role_trans *tr;
+	u32 buf[3];
+	size_t nel;
+	int rc;
+
+	nel = 0;
+	for (tr = r; tr; tr = tr->next)
+		nel++;
+	buf[0] = cpu_to_le32(nel);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+	for (tr = r; tr; tr = tr->next) {
+		buf[0] = cpu_to_le32(tr->role);
+		buf[1] = cpu_to_le32(tr->type);
+		buf[2] = cpu_to_le32(tr->new_role);
+		rc = put_entry(buf, sizeof(u32), 3, fp);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
+
+static int role_allow_write(struct role_allow *r, void *fp)
+{
+	struct role_allow *ra;
+	u32 buf[2];
+	size_t nel;
+	int rc;
+
+	nel = 0;
+	for (ra = r; ra; ra = ra->next)
+		nel++;
+	buf[0] = cpu_to_le32(nel);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+	for (ra = r; ra; ra = ra->next) {
+		buf[0] = cpu_to_le32(ra->role);
+		buf[1] = cpu_to_le32(ra->new_role);
+		rc = put_entry(buf, sizeof(u32), 2, fp);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+/*
+ * Write a security context structure
+ * to a policydb binary representation file.
+ */
+static int context_write(struct policydb *p, struct context *c,
+			 void *fp)
+{
+	int rc;
+	__le32 buf[3];
+
+	buf[0] = cpu_to_le32(c->user);
+	buf[1] = cpu_to_le32(c->role);
+	buf[2] = cpu_to_le32(c->type);
+
+	rc = put_entry(buf, sizeof(u32), 3, fp);
+	if (rc)
+		return rc;
+
+	rc = mls_write_range_helper(&c->range, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+/*
+ * The following *_write functions are used to
+ * write the symbol data to a policy database
+ * binary representation file.
+ */
+
+static int perm_write(void *vkey, void *datum, void *fp)
+{
+	char *key = vkey;
+	struct perm_datum *perdatum = datum;
+	__le32 buf[2];
+	size_t len;
+	int rc;
+
+	len = strlen(key);
+	buf[0] = cpu_to_le32(len);
+	buf[1] = cpu_to_le32(perdatum->value);
+	rc = put_entry(buf, sizeof(u32), 2, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int common_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct common_datum *comdatum = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	__le32 buf[4];
+	size_t len;
+	int rc;
+
+	len = strlen(key);
+	buf[0] = cpu_to_le32(len);
+	buf[1] = cpu_to_le32(comdatum->value);
+	buf[2] = cpu_to_le32(comdatum->permissions.nprim);
+	buf[3] = cpu_to_le32(comdatum->permissions.table->nel);
+	rc = put_entry(buf, sizeof(u32), 4, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	rc = hashtab_map(comdatum->permissions.table, perm_write, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int write_cons_helper(struct policydb *p, struct constraint_node *node,
+			     void *fp)
+{
+	struct constraint_node *c;
+	struct constraint_expr *e;
+	__le32 buf[3];
+	u32 nel;
+	int rc;
+
+	for (c = node; c; c = c->next) {
+		nel = 0;
+		for (e = c->expr; e; e = e->next)
+			nel++;
+		buf[0] = cpu_to_le32(c->permissions);
+		buf[1] = cpu_to_le32(nel);
+		rc = put_entry(buf, sizeof(u32), 2, fp);
+		if (rc)
+			return rc;
+		for (e = c->expr; e; e = e->next) {
+			buf[0] = cpu_to_le32(e->expr_type);
+			buf[1] = cpu_to_le32(e->attr);
+			buf[2] = cpu_to_le32(e->op);
+			rc = put_entry(buf, sizeof(u32), 3, fp);
+			if (rc)
+				return rc;
+
+			switch (e->expr_type) {
+			case CEXPR_NAMES:
+				rc = ebitmap_write(&e->names, fp);
+				if (rc)
+					return rc;
+				break;
+			default:
+				break;
+			}
+		}
+	}
+
+	return 0;
+}
+
+static int class_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct class_datum *cladatum = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	struct policydb *p = pd->p;
+	struct constraint_node *c;
+	__le32 buf[6];
+	u32 ncons;
+	size_t len, len2;
+	int rc;
+
+	len = strlen(key);
+	if (cladatum->comkey)
+		len2 = strlen(cladatum->comkey);
+	else
+		len2 = 0;
+
+	ncons = 0;
+	for (c = cladatum->constraints; c; c = c->next)
+		ncons++;
+
+	buf[0] = cpu_to_le32(len);
+	buf[1] = cpu_to_le32(len2);
+	buf[2] = cpu_to_le32(cladatum->value);
+	buf[3] = cpu_to_le32(cladatum->permissions.nprim);
+	if (cladatum->permissions.table)
+		buf[4] = cpu_to_le32(cladatum->permissions.table->nel);
+	else
+		buf[4] = 0;
+	buf[5] = cpu_to_le32(ncons);
+	rc = put_entry(buf, sizeof(u32), 6, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	if (cladatum->comkey) {
+		rc = put_entry(cladatum->comkey, 1, len2, fp);
+		if (rc)
+			return rc;
+	}
+
+	rc = hashtab_map(cladatum->permissions.table, perm_write, fp);
+	if (rc)
+		return rc;
+
+	rc = write_cons_helper(p, cladatum->constraints, fp);
+	if (rc)
+		return rc;
+
+	/* write out the validatetrans rule */
+	ncons = 0;
+	for (c = cladatum->validatetrans; c; c = c->next)
+		ncons++;
+
+	buf[0] = cpu_to_le32(ncons);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	rc = write_cons_helper(p, cladatum->validatetrans, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int role_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct role_datum *role = datum;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	struct policydb *p = pd->p;
+	__le32 buf[3];
+	size_t items, len;
+	int rc;
+
+	len = strlen(key);
+	items = 0;
+	buf[items++] = cpu_to_le32(len);
+	buf[items++] = cpu_to_le32(role->value);
+	if (p->policyvers >= POLICYDB_VERSION_BOUNDARY)
+		buf[items++] = cpu_to_le32(role->bounds);
+
+	BUG_ON(items > (sizeof(buf)/sizeof(buf[0])));
+
+	rc = put_entry(buf, sizeof(u32), items, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	rc = ebitmap_write(&role->dominates, fp);
+	if (rc)
+		return rc;
+
+	rc = ebitmap_write(&role->types, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int type_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct type_datum *typdatum = datum;
+	struct policy_data *pd = ptr;
+	struct policydb *p = pd->p;
+	void *fp = pd->fp;
+	__le32 buf[4];
+	int rc;
+	size_t items, len;
+
+	len = strlen(key);
+	items = 0;
+	buf[items++] = cpu_to_le32(len);
+	buf[items++] = cpu_to_le32(typdatum->value);
+	if (p->policyvers >= POLICYDB_VERSION_BOUNDARY) {
+		u32 properties = 0;
+
+		if (typdatum->primary)
+			properties |= TYPEDATUM_PROPERTY_PRIMARY;
+
+		if (typdatum->attribute)
+			properties |= TYPEDATUM_PROPERTY_ATTRIBUTE;
+
+		buf[items++] = cpu_to_le32(properties);
+		buf[items++] = cpu_to_le32(typdatum->bounds);
+	} else {
+		buf[items++] = cpu_to_le32(typdatum->primary);
+	}
+	BUG_ON(items > (sizeof(buf) / sizeof(buf[0])));
+	rc = put_entry(buf, sizeof(u32), items, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int user_write(void *vkey, void *datum, void *ptr)
+{
+	char *key = vkey;
+	struct user_datum *usrdatum = datum;
+	struct policy_data *pd = ptr;
+	struct policydb *p = pd->p;
+	void *fp = pd->fp;
+	__le32 buf[3];
+	size_t items, len;
+	int rc;
+
+	len = strlen(key);
+	items = 0;
+	buf[items++] = cpu_to_le32(len);
+	buf[items++] = cpu_to_le32(usrdatum->value);
+	if (p->policyvers >= POLICYDB_VERSION_BOUNDARY)
+		buf[items++] = cpu_to_le32(usrdatum->bounds);
+	BUG_ON(items > (sizeof(buf) / sizeof(buf[0])));
+	rc = put_entry(buf, sizeof(u32), items, fp);
+	if (rc)
+		return rc;
+
+	rc = put_entry(key, 1, len, fp);
+	if (rc)
+		return rc;
+
+	rc = ebitmap_write(&usrdatum->roles, fp);
+	if (rc)
+		return rc;
+
+	rc = mls_write_range_helper(&usrdatum->range, fp);
+	if (rc)
+		return rc;
+
+	rc = mls_write_level(&usrdatum->dfltlevel, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int (*write_f[SYM_NUM]) (void *key, void *datum,
+				void *datap) =
+{
+	common_write,
+	class_write,
+	role_write,
+	type_write,
+	user_write,
+	cond_write_bool,
+	sens_write,
+	cat_write,
+};
+
+static int ocontext_write(struct policydb *p, struct policydb_compat_info *info,
+			  void *fp)
+{
+	unsigned int i, j, rc;
+	size_t nel, len;
+	__le32 buf[3];
+	u32 nodebuf[8];
+	struct ocontext *c;
+	for (i = 0; i < info->ocon_num; i++) {
+		nel = 0;
+		for (c = p->ocontexts[i]; c; c = c->next)
+			nel++;
+		buf[0] = cpu_to_le32(nel);
+		rc = put_entry(buf, sizeof(u32), 1, fp);
+		if (rc)
+			return rc;
+		for (c = p->ocontexts[i]; c; c = c->next) {
+			switch (i) {
+			case OCON_ISID:
+				buf[0] = cpu_to_le32(c->sid[0]);
+				rc = put_entry(buf, sizeof(u32), 1, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				break;
+			case OCON_FS:
+			case OCON_NETIF:
+				len = strlen(c->u.name);
+				buf[0] = cpu_to_le32(len);
+				rc = put_entry(buf, sizeof(u32), 1, fp);
+				if (rc)
+					return rc;
+				rc = put_entry(c->u.name, 1, len, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[1], fp);
+				if (rc)
+					return rc;
+				break;
+			case OCON_PORT:
+				buf[0] = cpu_to_le32(c->u.port.protocol);
+				buf[1] = cpu_to_le32(c->u.port.low_port);
+				buf[2] = cpu_to_le32(c->u.port.high_port);
+				rc = put_entry(buf, sizeof(u32), 3, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				break;
+			case OCON_NODE:
+				nodebuf[0] = c->u.node.addr; /* network order */
+				nodebuf[1] = c->u.node.mask; /* network order */
+				rc = put_entry(nodebuf, sizeof(u32), 2, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				break;
+			case OCON_FSUSE:
+				buf[0] = cpu_to_le32(c->v.behavior);
+				len = strlen(c->u.name);
+				buf[1] = cpu_to_le32(len);
+				rc = put_entry(buf, sizeof(u32), 2, fp);
+				if (rc)
+					return rc;
+				rc = put_entry(c->u.name, 1, len, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				break;
+			case OCON_NODE6:
+				for (j = 0; j < 4; j++)
+					nodebuf[j] = c->u.node6.addr[j]; /* network order */
+				for (j = 0; j < 4; j++)
+					nodebuf[j + 4] = c->u.node6.mask[j]; /* network order */
+				rc = put_entry(nodebuf, sizeof(u32), 8, fp);
+				if (rc)
+					return rc;
+				rc = context_write(p, &c->context[0], fp);
+				if (rc)
+					return rc;
+				break;
+			}
+		}
+	}
+	return 0;
+}
+
+static int genfs_write(struct policydb *p, void *fp)
+{
+	struct genfs *genfs;
+	struct ocontext *c;
+	size_t len;
+	__le32 buf[1];
+	int rc;
+
+	len = 0;
+	for (genfs = p->genfs; genfs; genfs = genfs->next)
+		len++;
+	buf[0] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+	for (genfs = p->genfs; genfs; genfs = genfs->next) {
+		len = strlen(genfs->fstype);
+		buf[0] = cpu_to_le32(len);
+		rc = put_entry(buf, sizeof(u32), 1, fp);
+		if (rc)
+			return rc;
+		rc = put_entry(genfs->fstype, 1, len, fp);
+		if (rc)
+			return rc;
+		len = 0;
+		for (c = genfs->head; c; c = c->next)
+			len++;
+		buf[0] = cpu_to_le32(len);
+		rc = put_entry(buf, sizeof(u32), 1, fp);
+		if (rc)
+			return rc;
+		for (c = genfs->head; c; c = c->next) {
+			len = strlen(c->u.name);
+			buf[0] = cpu_to_le32(len);
+			rc = put_entry(buf, sizeof(u32), 1, fp);
+			if (rc)
+				return rc;
+			rc = put_entry(c->u.name, 1, len, fp);
+			if (rc)
+				return rc;
+			buf[0] = cpu_to_le32(c->v.sclass);
+			rc = put_entry(buf, sizeof(u32), 1, fp);
+			if (rc)
+				return rc;
+			rc = context_write(p, &c->context[0], fp);
+			if (rc)
+				return rc;
+		}
+	}
+	return 0;
+}
+
+static int range_count(void *key, void *data, void *ptr)
+{
+	int *cnt = ptr;
+	*cnt = *cnt + 1;
+
+	return 0;
+}
+
+static int range_write_helper(void *key, void *data, void *ptr)
+{
+	__le32 buf[2];
+	struct range_trans *rt = key;
+	struct mls_range *r = data;
+	struct policy_data *pd = ptr;
+	void *fp = pd->fp;
+	struct policydb *p = pd->p;
+	int rc;
+
+	buf[0] = cpu_to_le32(rt->source_type);
+	buf[1] = cpu_to_le32(rt->target_type);
+	rc = put_entry(buf, sizeof(u32), 2, fp);
+	if (rc)
+		return rc;
+	if (p->policyvers >= POLICYDB_VERSION_RANGETRANS) {
+		buf[0] = cpu_to_le32(rt->target_class);
+		rc = put_entry(buf, sizeof(u32), 1, fp);
+		if (rc)
+			return rc;
+	}
+	rc = mls_write_range_helper(r, fp);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int range_write(struct policydb *p, void *fp)
+{
+	size_t nel;
+	__le32 buf[1];
+	int rc;
+	struct policy_data pd;
+
+	pd.p = p;
+	pd.fp = fp;
+
+	/* count the number of entries in the hashtab */
+	nel = 0;
+	rc = hashtab_map(p->range_tr, range_count, &nel);
+	if (rc)
+		return rc;
+
+	buf[0] = cpu_to_le32(nel);
+	rc = put_entry(buf, sizeof(u32), 1, fp);
+	if (rc)
+		return rc;
+
+	/* actually write all of the entries */
+	rc = hashtab_map(p->range_tr, range_write_helper, &pd);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+/*
+ * Write the configuration data in a policy database
+ * structure to a policy database binary representation
+ * file.
+ */
+int policydb_write(struct policydb *p, void *fp)
+{
+	unsigned int i, num_syms;
+	int rc;
+	__le32 buf[4];
+	u32 config;
+	size_t len;
+	struct policydb_compat_info *info;
+
+	/*
+	 * refuse to write policy older than compressed avtab
+	 * to simplify the writer.  There are other tests dropped
+	 * since we assume this throughout the writer code.  Be
+	 * careful if you ever try to remove this restriction
+	 */
+	if (p->policyvers < POLICYDB_VERSION_AVTAB) {
+		printk(KERN_ERR "SELinux: refusing to write policy version %d."
+		       "  Because it is less than version %d\n", p->policyvers,
+		       POLICYDB_VERSION_AVTAB);
+		return -EINVAL;
+	}
+
+	config = 0;
+	if (p->mls_enabled)
+		config |= POLICYDB_CONFIG_MLS;
+
+	if (p->reject_unknown)
+		config |= REJECT_UNKNOWN;
+	if (p->allow_unknown)
+		config |= ALLOW_UNKNOWN;
+
+	/* Write the magic number and string identifiers. */
+	buf[0] = cpu_to_le32(POLICYDB_MAGIC);
+	len = strlen(POLICYDB_STRING);
+	buf[1] = cpu_to_le32(len);
+	rc = put_entry(buf, sizeof(u32), 2, fp);
+	if (rc)
+		return rc;
+	rc = put_entry(POLICYDB_STRING, 1, len, fp);
+	if (rc)
+		return rc;
+
+	/* Write the version, config, and table sizes. */
+	info = policydb_lookup_compat(p->policyvers);
+	if (!info) {
+		printk(KERN_ERR "SELinux: compatibility lookup failed for policy "
+		    "version %d", p->policyvers);
+		return rc;
+	}
+
+	buf[0] = cpu_to_le32(p->policyvers);
+	buf[1] = cpu_to_le32(config);
+	buf[2] = cpu_to_le32(info->sym_num);
+	buf[3] = cpu_to_le32(info->ocon_num);
+
+	rc = put_entry(buf, sizeof(u32), 4, fp);
+	if (rc)
+		return rc;
+
+	if (p->policyvers >= POLICYDB_VERSION_POLCAP) {
+		rc = ebitmap_write(&p->policycaps, fp);
+		if (rc)
+			return rc;
+	}
+
+	if (p->policyvers >= POLICYDB_VERSION_PERMISSIVE) {
+		rc = ebitmap_write(&p->permissive_map, fp);
+		if (rc)
+			return rc;
+	}
+
+	num_syms = info->sym_num;
+	for (i = 0; i < num_syms; i++) {
+		struct policy_data pd;
+
+		pd.fp = fp;
+		pd.p = p;
+
+		buf[0] = cpu_to_le32(p->symtab[i].nprim);
+		buf[1] = cpu_to_le32(p->symtab[i].table->nel);
+
+		rc = put_entry(buf, sizeof(u32), 2, fp);
+		if (rc)
+			return rc;
+		rc = hashtab_map(p->symtab[i].table, write_f[i], &pd);
+		if (rc)
+			return rc;
+	}
+
+	rc = avtab_write(p, &p->te_avtab, fp);
+	if (rc)
+		return rc;
+
+	rc = cond_write_list(p, p->cond_list, fp);
+	if (rc)
+		return rc;
+
+	rc = role_trans_write(p->role_tr, fp);
+	if (rc)
+		return rc;
+
+	rc = role_allow_write(p->role_allow, fp);
+	if (rc)
+		return rc;
+
+	rc = ocontext_write(p, info, fp);
+	if (rc)
+		return rc;
+
+	rc = genfs_write(p, fp);
+	if (rc)
+		return rc;
+
+	rc = range_write(p, fp);
+	if (rc)
+		return rc;
+
+	for (i = 0; i < p->p_types.nprim; i++) {
+		struct ebitmap *e = flex_array_get(p->type_attr_map_array, i);
+
+		BUG_ON(!e);
+		rc = ebitmap_write(e, fp);
+		if (rc)
+			return rc;
+	}
+
+	return 0;
+}
diff --git a/security/selinux/ss/policydb.h b/security/selinux/ss/policydb.h
index 310e944..95d3d7d 100644
--- a/security/selinux/ss/policydb.h
+++ b/security/selinux/ss/policydb.h
@@ -254,6 +254,9 @@
 
 	struct ebitmap permissive_map;
 
+	/* length of this policy when it was loaded */
+	size_t len;
+
 	unsigned int policyvers;
 
 	unsigned int reject_unknown : 1;
@@ -270,6 +273,7 @@
 extern int policydb_type_isvalid(struct policydb *p, unsigned int type);
 extern int policydb_role_isvalid(struct policydb *p, unsigned int role);
 extern int policydb_read(struct policydb *p, void *fp);
+extern int policydb_write(struct policydb *p, void *fp);
 
 #define PERM_SYMTAB_SIZE 32
 
@@ -290,6 +294,11 @@
 	size_t len;
 };
 
+struct policy_data {
+	struct policydb *p;
+	void *fp;
+};
+
 static inline int next_entry(void *buf, struct policy_file *fp, size_t bytes)
 {
 	if (bytes > fp->len)
@@ -301,6 +310,17 @@
 	return 0;
 }
 
+static inline int put_entry(void *buf, size_t bytes, int num, struct policy_file *fp)
+{
+	size_t len = bytes * num;
+
+	memcpy(fp->data, buf, len);
+	fp->data += len;
+	fp->len -= len;
+
+	return 0;
+}
+
 extern u16 string_to_security_class(struct policydb *p, const char *name);
 extern u32 string_to_av_perm(struct policydb *p, u16 tclass, const char *name);
 
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 9ea2fec..223c1ff 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -51,6 +51,7 @@
 #include <linux/mutex.h>
 #include <linux/selinux.h>
 #include <linux/flex_array.h>
+#include <linux/vmalloc.h>
 #include <net/netlabel.h>
 
 #include "flask.h"
@@ -991,7 +992,8 @@
 {
 	char *scontextp;
 
-	*scontext = NULL;
+	if (scontext)
+		*scontext = NULL;
 	*scontext_len = 0;
 
 	if (context->len) {
@@ -1008,6 +1010,9 @@
 	*scontext_len += strlen(policydb.p_type_val_to_name[context->type - 1]) + 1;
 	*scontext_len += mls_compute_context_len(context);
 
+	if (!scontext)
+		return 0;
+
 	/* Allocate space for the context; caller must free this space. */
 	scontextp = kmalloc(*scontext_len, GFP_ATOMIC);
 	if (!scontextp)
@@ -1047,7 +1052,8 @@
 	struct context *context;
 	int rc = 0;
 
-	*scontext = NULL;
+	if (scontext)
+		*scontext = NULL;
 	*scontext_len  = 0;
 
 	if (!ss_initialized) {
@@ -1055,6 +1061,8 @@
 			char *scontextp;
 
 			*scontext_len = strlen(initial_sid_to_string[sid]) + 1;
+			if (!scontext)
+				goto out;
 			scontextp = kmalloc(*scontext_len, GFP_ATOMIC);
 			if (!scontextp) {
 				rc = -ENOMEM;
@@ -1769,6 +1777,7 @@
 			return rc;
 		}
 
+		policydb.len = len;
 		rc = selinux_set_mapping(&policydb, secclass_map,
 					 &current_mapping,
 					 &current_mapping_size);
@@ -1791,6 +1800,7 @@
 		selinux_complete_init();
 		avc_ss_reset(seqno);
 		selnl_notify_policyload(seqno);
+		selinux_status_update_policyload(seqno);
 		selinux_netlbl_cache_invalidate();
 		selinux_xfrm_notify_policyload();
 		return 0;
@@ -1804,6 +1814,7 @@
 	if (rc)
 		return rc;
 
+	newpolicydb.len = len;
 	/* If switching between different policy types, log MLS status */
 	if (policydb.mls_enabled && !newpolicydb.mls_enabled)
 		printk(KERN_INFO "SELinux: Disabling MLS support...\n");
@@ -1870,6 +1881,7 @@
 
 	avc_ss_reset(seqno);
 	selnl_notify_policyload(seqno);
+	selinux_status_update_policyload(seqno);
 	selinux_netlbl_cache_invalidate();
 	selinux_xfrm_notify_policyload();
 
@@ -1883,6 +1895,17 @@
 
 }
 
+size_t security_policydb_len(void)
+{
+	size_t len;
+
+	read_lock(&policy_rwlock);
+	len = policydb.len;
+	read_unlock(&policy_rwlock);
+
+	return len;
+}
+
 /**
  * security_port_sid - Obtain the SID for a port.
  * @protocol: protocol number
@@ -2374,6 +2397,7 @@
 	if (!rc) {
 		avc_ss_reset(seqno);
 		selnl_notify_policyload(seqno);
+		selinux_status_update_policyload(seqno);
 		selinux_xfrm_notify_policyload();
 	}
 	return rc;
@@ -3129,3 +3153,38 @@
 	return rc;
 }
 #endif /* CONFIG_NETLABEL */
+
+/**
+ * security_read_policy - read the policy.
+ * @data: binary policy data
+ * @len: length of data in bytes
+ *
+ */
+int security_read_policy(void **data, ssize_t *len)
+{
+	int rc;
+	struct policy_file fp;
+
+	if (!ss_initialized)
+		return -EINVAL;
+
+	*len = security_policydb_len();
+
+	*data = vmalloc_user(*len);
+	if (!*data)
+		return -ENOMEM;
+
+	fp.data = *data;
+	fp.len = *len;
+
+	read_lock(&policy_rwlock);
+	rc = policydb_write(&policydb, &fp);
+	read_unlock(&policy_rwlock);
+
+	if (rc)
+		return rc;
+
+	*len = (unsigned long)fp.data - (unsigned long)*data;
+	return 0;
+
+}
diff --git a/security/selinux/ss/status.c b/security/selinux/ss/status.c
new file mode 100644
index 0000000..d982365
--- /dev/null
+++ b/security/selinux/ss/status.c
@@ -0,0 +1,126 @@
+/*
+ * mmap based event notifications for SELinux
+ *
+ * Author: KaiGai Kohei <kaigai@ak.jp.nec.com>
+ *
+ * Copyright (C) 2010 NEC corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2,
+ * as published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/mm.h>
+#include <linux/mutex.h>
+#include "avc.h"
+#include "services.h"
+
+/*
+ * The selinux_status_page shall be exposed to userspace applications
+ * using mmap interface on /selinux/status.
+ * It enables to notify applications a few events that will cause reset
+ * of userspace access vector without context switching.
+ *
+ * The selinux_kernel_status structure on the head of status page is
+ * protected from concurrent accesses using seqlock logic, so userspace
+ * application should reference the status page according to the seqlock
+ * logic.
+ *
+ * Typically, application checks status->sequence at the head of access
+ * control routine. If it is odd-number, kernel is updating the status,
+ * so please wait for a moment. If it is changed from the last sequence
+ * number, it means something happen, so application will reset userspace
+ * avc, if needed.
+ * In most cases, application shall confirm the kernel status is not
+ * changed without any system call invocations.
+ */
+static struct page *selinux_status_page;
+static DEFINE_MUTEX(selinux_status_lock);
+
+/*
+ * selinux_kernel_status_page
+ *
+ * It returns a reference to selinux_status_page. If the status page is
+ * not allocated yet, it also tries to allocate it at the first time.
+ */
+struct page *selinux_kernel_status_page(void)
+{
+	struct selinux_kernel_status   *status;
+	struct page		       *result = NULL;
+
+	mutex_lock(&selinux_status_lock);
+	if (!selinux_status_page) {
+		selinux_status_page = alloc_page(GFP_KERNEL|__GFP_ZERO);
+
+		if (selinux_status_page) {
+			status = page_address(selinux_status_page);
+
+			status->version = SELINUX_KERNEL_STATUS_VERSION;
+			status->sequence = 0;
+			status->enforcing = selinux_enforcing;
+			/*
+			 * NOTE: the next policyload event shall set
+			 * a positive value on the status->policyload,
+			 * although it may not be 1, but never zero.
+			 * So, application can know it was updated.
+			 */
+			status->policyload = 0;
+			status->deny_unknown = !security_get_allow_unknown();
+		}
+	}
+	result = selinux_status_page;
+	mutex_unlock(&selinux_status_lock);
+
+	return result;
+}
+
+/*
+ * selinux_status_update_setenforce
+ *
+ * It updates status of the current enforcing/permissive mode.
+ */
+void selinux_status_update_setenforce(int enforcing)
+{
+	struct selinux_kernel_status   *status;
+
+	mutex_lock(&selinux_status_lock);
+	if (selinux_status_page) {
+		status = page_address(selinux_status_page);
+
+		status->sequence++;
+		smp_wmb();
+
+		status->enforcing = enforcing;
+
+		smp_wmb();
+		status->sequence++;
+	}
+	mutex_unlock(&selinux_status_lock);
+}
+
+/*
+ * selinux_status_update_policyload
+ *
+ * It updates status of the times of policy reloaded, and current
+ * setting of deny_unknown.
+ */
+void selinux_status_update_policyload(int seqno)
+{
+	struct selinux_kernel_status   *status;
+
+	mutex_lock(&selinux_status_lock);
+	if (selinux_status_page) {
+		status = page_address(selinux_status_page);
+
+		status->sequence++;
+		smp_wmb();
+
+		status->policyload = seqno;
+		status->deny_unknown = !security_get_allow_unknown();
+
+		smp_wmb();
+		status->sequence++;
+	}
+	mutex_unlock(&selinux_status_lock);
+}
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index c448d57..bc39f40 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -1281,12 +1281,11 @@
  *
  * Return 0 if read access is permitted
  */
-static int smack_task_setscheduler(struct task_struct *p, int policy,
-				   struct sched_param *lp)
+static int smack_task_setscheduler(struct task_struct *p)
 {
 	int rc;
 
-	rc = cap_task_setscheduler(p, policy, lp);
+	rc = cap_task_setscheduler(p);
 	if (rc == 0)
 		rc = smk_curacc_on_task(p, MAY_WRITE);
 	return rc;
@@ -3005,7 +3004,8 @@
 {
 	char *sp = smack_from_secid(secid);
 
-	*secdata = sp;
+	if (secdata)
+		*secdata = sp;
 	*seclen = strlen(sp);
 	return 0;
 }
diff --git a/security/tomoyo/common.c b/security/tomoyo/common.c
index c668b44..7556315 100644
--- a/security/tomoyo/common.c
+++ b/security/tomoyo/common.c
@@ -768,8 +768,10 @@
 		return true; /* Do nothing if open(O_WRONLY). */
 	memset(&head->r, 0, sizeof(head->r));
 	head->r.print_this_domain_only = true;
-	head->r.eof = !domain;
-	head->r.domain = &domain->list;
+	if (domain)
+		head->r.domain = &domain->list;
+	else
+		head->r.eof = 1;
 	tomoyo_io_printf(head, "# select %s\n", data);
 	if (domain && domain->is_deleted)
 		tomoyo_io_printf(head, "# This is a deleted domain.\n");
@@ -2051,13 +2053,22 @@
 		const u8 profile = domain->profile;
 		if (tomoyo_profile_ptr[profile])
 			continue;
+		printk(KERN_ERR "You need to define profile %u before using it.\n",
+		       profile);
+		printk(KERN_ERR "Please see http://tomoyo.sourceforge.jp/2.3/ "
+		       "for more information.\n");
 		panic("Profile %u (used by '%s') not defined.\n",
 		      profile, domain->domainname->name);
 	}
 	tomoyo_read_unlock(idx);
-	if (tomoyo_profile_version != 20090903)
+	if (tomoyo_profile_version != 20090903) {
+		printk(KERN_ERR "You need to install userland programs for "
+		       "TOMOYO 2.3 and initialize policy configuration.\n");
+		printk(KERN_ERR "Please see http://tomoyo.sourceforge.jp/2.3/ "
+		       "for more information.\n");
 		panic("Profile version %u is not supported.\n",
 		      tomoyo_profile_version);
+	}
 	printk(KERN_INFO "TOMOYO: 2.3.0\n");
 	printk(KERN_INFO "Mandatory Access Control activated.\n");
 }
diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c
index a7868ad..cbbed0d 100644
--- a/sound/core/rawmidi.c
+++ b/sound/core/rawmidi.c
@@ -535,13 +535,15 @@
 {
 	struct snd_rawmidi_file *rfile;
 	struct snd_rawmidi *rmidi;
+	struct module *module;
 
 	rfile = file->private_data;
 	rmidi = rfile->rmidi;
 	rawmidi_release_priv(rfile);
 	kfree(rfile);
+	module = rmidi->card->module;
 	snd_card_file_remove(rmidi->card, file);
-	module_put(rmidi->card->module);
+	module_put(module);
 	return 0;
 }
 
diff --git a/sound/oss/soundcard.c b/sound/oss/soundcard.c
index 92aa762..07f803e 100644
--- a/sound/oss/soundcard.c
+++ b/sound/oss/soundcard.c
@@ -391,11 +391,11 @@
 	case SND_DEV_DSP:
 	case SND_DEV_DSP16:
 	case SND_DEV_AUDIO:
-		return audio_ioctl(dev, file, cmd, p);
+		ret = audio_ioctl(dev, file, cmd, p);
 		break;
 
 	case SND_DEV_MIDIN:
-		return MIDIbuf_ioctl(dev, file, cmd, p);
+		ret = MIDIbuf_ioctl(dev, file, cmd, p);
 		break;
 
 	}
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 95148e5..c16c5ba 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -1747,6 +1747,8 @@
 		      "HP dv6", STAC_HP_DV5),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x3061,
 		      "HP dv6", STAC_HP_DV5), /* HP dv6-1110ax */
+	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x363e,
+		      "HP DV6", STAC_HP_DV5),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xfff0, 0x7010,
 		      "HP", STAC_HP_DV5),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x0233,
diff --git a/sound/pcmcia/pdaudiocf/pdaudiocf.c b/sound/pcmcia/pdaudiocf/pdaudiocf.c
index 7ab9174..8cc4733 100644
--- a/sound/pcmcia/pdaudiocf/pdaudiocf.c
+++ b/sound/pcmcia/pdaudiocf/pdaudiocf.c
@@ -142,10 +142,9 @@
 	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	link->resource[0]->end = 16;
 
-	link->conf.Attributes = CONF_ENABLE_IRQ | CONF_ENABLE_PULSE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-	link->conf.ConfigIndex = 1;
-	link->conf.Present = PRESENT_OPTION;
+	link->config_flags = CONF_ENABLE_IRQ | CONF_ENABLE_PULSE_IRQ;
+	link->config_index = 1;
+	link->config_regs = PRESENT_OPTION;
 
 	return pdacf_config(link);
 }
@@ -217,7 +216,8 @@
 	int ret;
 
 	snd_printdd(KERN_DEBUG "pdacf_config called\n");
-	link->conf.ConfigIndex = 0x5;
+	link->config_index = 0x5;
+	link->config_flags |= CONF_ENABLE_IRQ | CONF_ENABLE_PULSE_IRQ;
 
 	ret = pcmcia_request_io(link);
 	if (ret)
@@ -227,7 +227,7 @@
 	if (ret)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -287,9 +287,7 @@
 
 static struct pcmcia_driver pdacf_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "snd-pdaudiocf",
-	},
+	.name		= "snd-pdaudiocf",
 	.probe		= snd_pdacf_probe,
 	.remove		= snd_pdacf_detach,
 	.id_table	= snd_pdacf_ids,
diff --git a/sound/pcmcia/pdaudiocf/pdaudiocf.h b/sound/pcmcia/pdaudiocf/pdaudiocf.h
index 5cc3e45..bd26e09 100644
--- a/sound/pcmcia/pdaudiocf/pdaudiocf.h
+++ b/sound/pcmcia/pdaudiocf/pdaudiocf.h
@@ -24,7 +24,6 @@
 #include <sound/pcm.h>
 #include <asm/io.h>
 #include <linux/interrupt.h>
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
diff --git a/sound/pcmcia/vx/vxpocket.c b/sound/pcmcia/vx/vxpocket.c
index a6edfc3..80000d6 100644
--- a/sound/pcmcia/vx/vxpocket.c
+++ b/sound/pcmcia/vx/vxpocket.c
@@ -2,7 +2,7 @@
  * Driver for Digigram VXpocket V2/440 soundcards
  *
  * Copyright (c) 2002 by Takashi Iwai <tiwai@suse.de>
- *
+
  *   This program is free software; you can redistribute it and/or modify
  *   it under the terms of the GNU General Public License as published by
  *   the Free Software Foundation; either version 2 of the License, or
@@ -162,10 +162,9 @@
 	link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;
 	link->resource[0]->end = 16;
 
-	link->conf.Attributes = CONF_ENABLE_IRQ;
-	link->conf.IntType = INT_MEMORY_AND_IO;
-	link->conf.ConfigIndex = 1;
-	link->conf.Present = PRESENT_OPTION;
+	link->config_flags |= CONF_ENABLE_IRQ;
+	link->config_index = 1;
+	link->config_regs = PRESENT_OPTION;
 
 	*chip_ret = vxp;
 	return 0;
@@ -234,7 +233,7 @@
 	if (ret)
 		goto failed;
 
-	ret = pcmcia_request_configuration(link, &link->conf);
+	ret = pcmcia_enable_device(link);
 	if (ret)
 		goto failed;
 
@@ -359,9 +358,7 @@
 
 static struct pcmcia_driver vxp_cs_driver = {
 	.owner		= THIS_MODULE,
-	.drv		= {
-		.name	= "snd-vxpocket",
-	},
+	.name		= "snd-vxpocket",
 	.probe		= vxpocket_probe,
 	.remove		= vxpocket_detach,
 	.id_table	= vxp_ids,
diff --git a/sound/pcmcia/vx/vxpocket.h b/sound/pcmcia/vx/vxpocket.h
index d911066..13d658c 100644
--- a/sound/pcmcia/vx/vxpocket.h
+++ b/sound/pcmcia/vx/vxpocket.h
@@ -23,7 +23,6 @@
 
 #include <sound/vx_core.h>
 
-#include <pcmcia/cs.h>
 #include <pcmcia/cistpl.h>
 #include <pcmcia/ds.h>
 
diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
index 5164a65..b2c6330 100644
--- a/tools/perf/Documentation/perf-annotate.txt
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -8,7 +8,7 @@
 SYNOPSIS
 --------
 [verse]
-'perf annotate' [-i <file> | --input=file] symbol_name
+'perf annotate' [-i <file> | --input=file] [symbol_name]
 
 DESCRIPTION
 -----------
@@ -24,6 +24,13 @@
 --input=::
         Input file name. (default: perf.data)
 
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface Use of --tui requires a tty, if one is not
+	present, as when piping to other commands, the stdio interface is
+	used. This interfaces starts by centering on the line with more
+	samples, TAB/UNTAB cycles thru the lines with more samples.
+
 SEE ALSO
 --------
-linkperf:perf-record[1]
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index abfabe9..12052c9 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -65,6 +65,13 @@
 		 the tree is considered as a new profiled object. +
 	Default: fractal,0.5.
 
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface, that is integrated with annotate and allows
+        zooming into DSOs or threads, among other features. Use of --tui
+	requires a tty, if one is not present, as when piping to other
+	commands, the stdio interface is used.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1]
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 1950e19..d1db0f6 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -313,6 +313,9 @@
 
 SCRIPT_SH += perf-archive.sh
 
+grep-libs = $(filter -l%,$(1))
+strip-libs = $(filter-out -l%,$(1))
+
 #
 # No Perl scripts right now:
 #
@@ -588,14 +591,17 @@
 ifdef NO_LIBPERL
 	BASIC_CFLAGS += -DNO_LIBPERL
 else
-	PERL_EMBED_LDOPTS = `perl -MExtUtils::Embed -e ldopts 2>/dev/null`
+       PERL_EMBED_LDOPTS = $(shell perl -MExtUtils::Embed -e ldopts 2>/dev/null)
+       PERL_EMBED_LDFLAGS = $(call strip-libs,$(PERL_EMBED_LDOPTS))
+       PERL_EMBED_LIBADD = $(call grep-libs,$(PERL_EMBED_LDOPTS))
 	PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null`
 	FLAGS_PERL_EMBED=$(PERL_EMBED_CCOPTS) $(PERL_EMBED_LDOPTS)
 
 	ifneq ($(call try-cc,$(SOURCE_PERL_EMBED),$(FLAGS_PERL_EMBED)),y)
 		BASIC_CFLAGS += -DNO_LIBPERL
 	else
-		ALL_LDFLAGS += $(PERL_EMBED_LDOPTS)
+               ALL_LDFLAGS += $(PERL_EMBED_LDFLAGS)
+               EXTLIBS += $(PERL_EMBED_LIBADD)
 		LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o
 		LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o
 	endif
@@ -604,13 +610,16 @@
 ifdef NO_LIBPYTHON
 	BASIC_CFLAGS += -DNO_LIBPYTHON
 else
-	PYTHON_EMBED_LDOPTS = `python-config --ldflags 2>/dev/null`
+       PYTHON_EMBED_LDOPTS = $(shell python-config --ldflags 2>/dev/null)
+       PYTHON_EMBED_LDFLAGS = $(call strip-libs,$(PYTHON_EMBED_LDOPTS))
+       PYTHON_EMBED_LIBADD = $(call grep-libs,$(PYTHON_EMBED_LDOPTS))
 	PYTHON_EMBED_CCOPTS = `python-config --cflags 2>/dev/null`
 	FLAGS_PYTHON_EMBED=$(PYTHON_EMBED_CCOPTS) $(PYTHON_EMBED_LDOPTS)
 	ifneq ($(call try-cc,$(SOURCE_PYTHON_EMBED),$(FLAGS_PYTHON_EMBED)),y)
 		BASIC_CFLAGS += -DNO_LIBPYTHON
 	else
-		ALL_LDFLAGS += $(PYTHON_EMBED_LDOPTS)
+               ALL_LDFLAGS += $(PYTHON_EMBED_LDFLAGS)
+               EXTLIBS += $(PYTHON_EMBED_LIBADD)
 		LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o
 		LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o
 	endif
@@ -653,6 +662,15 @@
 	endif
 endif
 
+
+ifdef NO_STRLCPY
+	BASIC_CFLAGS += -DNO_STRLCPY
+else
+	ifneq ($(call try-cc,$(SOURCE_STRLCPY),),y)
+		BASIC_CFLAGS += -DNO_STRLCPY
+	endif
+endif
+
 ifndef CC_LD_DYNPATH
 	ifdef NO_R_TO_GCC_LINKER
 		# Some gcc does not accept and pass -R to the linker to specify
@@ -910,8 +928,8 @@
 		$(ALL_CFLAGS) -c $(filter %.c,$^) -o $@
 
 $(OUTPUT)perf$X: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS)
-	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(OUTPUT)perf.o \
-		$(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS)
+	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) $(ALL_LDFLAGS) $(OUTPUT)perf.o \
+               $(BUILTIN_OBJS) $(LIBS) -o $@
 
 $(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 1478dc6..6d5604d8d 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -28,7 +28,7 @@
 
 static char		const *input_name = "perf.data";
 
-static bool		force;
+static bool		force, use_tui, use_stdio;
 
 static bool		full_paths;
 
@@ -321,7 +321,7 @@
 
 static void hists__find_annotations(struct hists *self)
 {
-	struct rb_node *first = rb_first(&self->entries), *nd = first;
+	struct rb_node *nd = rb_first(&self->entries), *next;
 	int key = KEY_RIGHT;
 
 	while (nd) {
@@ -343,20 +343,19 @@
 
 		if (use_browser > 0) {
 			key = hist_entry__tui_annotate(he);
-			if (is_exit_key(key))
-				break;
 			switch (key) {
 			case KEY_RIGHT:
-			case '\t':
-				nd = rb_next(nd);
+				next = rb_next(nd);
 				break;
 			case KEY_LEFT:
-				if (nd == first)
-					continue;
-				nd = rb_prev(nd);
-			default:
+				next = rb_prev(nd);
 				break;
+			default:
+				return;
 			}
+
+			if (next != NULL)
+				nd = next;
 		} else {
 			hist_entry__tty_annotate(he);
 			nd = rb_next(nd);
@@ -428,6 +427,8 @@
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
 	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
 		   "file", "vmlinux pathname"),
 	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
@@ -443,6 +444,11 @@
 {
 	argc = parse_options(argc, argv, options, annotate_usage, 0);
 
+	if (use_stdio)
+		use_browser = 0;
+	else if (use_tui)
+		use_browser = 1;
+
 	setup_browser();
 
 	symbol_conf.priv_size = sizeof(struct sym_priv);
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 55fc1f4..5de405d 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -32,7 +32,7 @@
 
 static char		const *input_name = "perf.data";
 
-static bool		force;
+static bool		force, use_tui, use_stdio;
 static bool		hide_unresolved;
 static bool		dont_use_callchains;
 
@@ -107,7 +107,8 @@
 		goto out_free_syms;
 	err = 0;
 	if (symbol_conf.use_callchain) {
-		err = append_chain(he->callchain, data->callchain, syms, data->period);
+		err = callchain_append(he->callchain, data->callchain, syms,
+				       data->period);
 		if (err)
 			goto out_free_syms;
 	}
@@ -450,6 +451,8 @@
 		    "Show per-thread event counters"),
 	OPT_STRING(0, "pretty", &pretty_printing_style, "key",
 		   "pretty printing style key: normal raw"),
+	OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
 	OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization,
@@ -482,8 +485,15 @@
 {
 	argc = parse_options(argc, argv, options, report_usage, 0);
 
+	if (use_stdio)
+		use_browser = 0;
+	else if (use_tui)
+		use_browser = 1;
+
 	if (strcmp(input_name, "-") != 0)
 		setup_browser();
+	else
+		use_browser = 0;
 	/*
 	 * Only in the newt browser we are doing integrated annotation,
 	 * so don't allocate extra space that won't be used in the stdio
diff --git a/tools/perf/feature-tests.mak b/tools/perf/feature-tests.mak
index 7a7b608..b253db6 100644
--- a/tools/perf/feature-tests.mak
+++ b/tools/perf/feature-tests.mak
@@ -110,6 +110,17 @@
 }
 endef
 
+define SOURCE_STRLCPY
+#include <stdlib.h>
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+
+int main(void)
+{
+	strlcpy(NULL, NULL, 0);
+	return 0;
+}
+endef
+
 # try-cc
 # Usage: option = $(call try-cc, source-to-build, cc-options)
 try-cc = $(shell sh -c						  \
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index ef7aa0a..95aaf56 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -73,6 +73,18 @@
 #define cpu_relax()	asm volatile("":::"memory")
 #endif
 
+#ifdef __mips__
+#include "../../arch/mips/include/asm/unistd.h"
+#define rmb()		asm volatile(					\
+				".set	mips2\n\t"			\
+				"sync\n\t"				\
+				".set	mips0"				\
+				: /* no output */			\
+				: /* no input */			\
+				: "memory")
+#define cpu_relax()	asm volatile("" ::: "memory")
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
diff --git a/tools/perf/scripts/python/bin/netdev-times-record b/tools/perf/scripts/python/bin/netdev-times-record
new file mode 100644
index 0000000..d931a82
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+perf record -a -e net:net_dev_xmit -e net:net_dev_queue		\
+		-e net:netif_receive_skb -e net:netif_rx		\
+		-e skb:consume_skb -e skb:kfree_skb			\
+		-e skb:skb_copy_datagram_iovec -e napi:napi_poll	\
+		-e irq:irq_handler_entry -e irq:irq_handler_exit	\
+		-e irq:softirq_entry -e irq:softirq_exit		\
+		-e irq:softirq_raise $@
diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report
new file mode 100644
index 0000000..c3d0a63
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-report
@@ -0,0 +1,5 @@
+#!/bin/bash
+# description: display a process of packet and processing time
+# args: [tx] [rx] [dev=] [debug]
+
+perf trace -s ~/libexec/perf-core/scripts/python/netdev-times.py $@
diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py
new file mode 100644
index 0000000..9aa0a32
--- /dev/null
+++ b/tools/perf/scripts/python/netdev-times.py
@@ -0,0 +1,464 @@
+# Display a process of packets and processed time.
+# It helps us to investigate networking or network device.
+#
+# options
+# tx: show only tx chart
+# rx: show only rx chart
+# dev=: show only thing related to specified device
+# debug: work with debug mode. It shows buffer status.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+all_event_list = []; # insert all tracepoint event related with this script
+irq_dic = {}; # key is cpu and value is a list which stacks irqs
+              # which raise NET_RX softirq
+net_rx_dic = {}; # key is cpu and value include time of NET_RX softirq-entry
+		 # and a list which stacks receive
+receive_hunk_list = []; # a list which include a sequence of receive events
+rx_skb_list = []; # received packet list for matching
+		       # skb_copy_datagram_iovec
+
+buffer_budget = 65536; # the budget of rx_skb_list, tx_queue_list and
+		       # tx_xmit_list
+of_count_rx_skb_list = 0; # overflow count
+
+tx_queue_list = []; # list of packets which pass through dev_queue_xmit
+of_count_tx_queue_list = 0; # overflow count
+
+tx_xmit_list = [];  # list of packets which pass through dev_hard_start_xmit
+of_count_tx_xmit_list = 0; # overflow count
+
+tx_free_list = [];  # list of packets which is freed
+
+# options
+show_tx = 0;
+show_rx = 0;
+dev = 0; # store a name of device specified by option "dev="
+debug = 0;
+
+# indices of event_info tuple
+EINFO_IDX_NAME=   0
+EINFO_IDX_CONTEXT=1
+EINFO_IDX_CPU=    2
+EINFO_IDX_TIME=   3
+EINFO_IDX_PID=    4
+EINFO_IDX_COMM=   5
+
+# Calculate a time interval(msec) from src(nsec) to dst(nsec)
+def diff_msec(src, dst):
+	return (dst - src) / 1000000.0
+
+# Display a process of transmitting a packet
+def print_transmit(hunk):
+	if dev != 0 and hunk['dev'].find(dev) < 0:
+		return
+	print "%7s %5d %6d.%06dsec %12.3fmsec      %12.3fmsec" % \
+		(hunk['dev'], hunk['len'],
+		nsecs_secs(hunk['queue_t']),
+		nsecs_nsecs(hunk['queue_t'])/1000,
+		diff_msec(hunk['queue_t'], hunk['xmit_t']),
+		diff_msec(hunk['xmit_t'], hunk['free_t']))
+
+# Format for displaying rx packet processing
+PF_IRQ_ENTRY= "  irq_entry(+%.3fmsec irq=%d:%s)"
+PF_SOFT_ENTRY="  softirq_entry(+%.3fmsec)"
+PF_NAPI_POLL= "  napi_poll_exit(+%.3fmsec %s)"
+PF_JOINT=     "         |"
+PF_WJOINT=    "         |            |"
+PF_NET_RECV=  "         |---netif_receive_skb(+%.3fmsec skb=%x len=%d)"
+PF_NET_RX=    "         |---netif_rx(+%.3fmsec skb=%x)"
+PF_CPY_DGRAM= "         |      skb_copy_datagram_iovec(+%.3fmsec %d:%s)"
+PF_KFREE_SKB= "         |      kfree_skb(+%.3fmsec location=%x)"
+PF_CONS_SKB=  "         |      consume_skb(+%.3fmsec)"
+
+# Display a process of received packets and interrputs associated with
+# a NET_RX softirq
+def print_receive(hunk):
+	show_hunk = 0
+	irq_list = hunk['irq_list']
+	cpu = irq_list[0]['cpu']
+	base_t = irq_list[0]['irq_ent_t']
+	# check if this hunk should be showed
+	if dev != 0:
+		for i in range(len(irq_list)):
+			if irq_list[i]['name'].find(dev) >= 0:
+				show_hunk = 1
+				break
+	else:
+		show_hunk = 1
+	if show_hunk == 0:
+		return
+
+	print "%d.%06dsec cpu=%d" % \
+		(nsecs_secs(base_t), nsecs_nsecs(base_t)/1000, cpu)
+	for i in range(len(irq_list)):
+		print PF_IRQ_ENTRY % \
+			(diff_msec(base_t, irq_list[i]['irq_ent_t']),
+			irq_list[i]['irq'], irq_list[i]['name'])
+		print PF_JOINT
+		irq_event_list = irq_list[i]['event_list']
+		for j in range(len(irq_event_list)):
+			irq_event = irq_event_list[j]
+			if irq_event['event'] == 'netif_rx':
+				print PF_NET_RX % \
+					(diff_msec(base_t, irq_event['time']),
+					irq_event['skbaddr'])
+				print PF_JOINT
+	print PF_SOFT_ENTRY % \
+		diff_msec(base_t, hunk['sirq_ent_t'])
+	print PF_JOINT
+	event_list = hunk['event_list']
+	for i in range(len(event_list)):
+		event = event_list[i]
+		if event['event_name'] == 'napi_poll':
+			print PF_NAPI_POLL % \
+			    (diff_msec(base_t, event['event_t']), event['dev'])
+			if i == len(event_list) - 1:
+				print ""
+			else:
+				print PF_JOINT
+		else:
+			print PF_NET_RECV % \
+			    (diff_msec(base_t, event['event_t']), event['skbaddr'],
+				event['len'])
+			if 'comm' in event.keys():
+				print PF_WJOINT
+				print PF_CPY_DGRAM % \
+					(diff_msec(base_t, event['comm_t']),
+					event['pid'], event['comm'])
+			elif 'handle' in event.keys():
+				print PF_WJOINT
+				if event['handle'] == "kfree_skb":
+					print PF_KFREE_SKB % \
+						(diff_msec(base_t,
+						event['comm_t']),
+						event['location'])
+				elif event['handle'] == "consume_skb":
+					print PF_CONS_SKB % \
+						diff_msec(base_t,
+							event['comm_t'])
+			print PF_JOINT
+
+def trace_begin():
+	global show_tx
+	global show_rx
+	global dev
+	global debug
+
+	for i in range(len(sys.argv)):
+		if i == 0:
+			continue
+		arg = sys.argv[i]
+		if arg == 'tx':
+			show_tx = 1
+		elif arg =='rx':
+			show_rx = 1
+		elif arg.find('dev=',0, 4) >= 0:
+			dev = arg[4:]
+		elif arg == 'debug':
+			debug = 1
+	if show_tx == 0  and show_rx == 0:
+		show_tx = 1
+		show_rx = 1
+
+def trace_end():
+	# order all events in time
+	all_event_list.sort(lambda a,b :cmp(a[EINFO_IDX_TIME],
+					    b[EINFO_IDX_TIME]))
+	# process all events
+	for i in range(len(all_event_list)):
+		event_info = all_event_list[i]
+		name = event_info[EINFO_IDX_NAME]
+		if name == 'irq__softirq_exit':
+			handle_irq_softirq_exit(event_info)
+		elif name == 'irq__softirq_entry':
+			handle_irq_softirq_entry(event_info)
+		elif name == 'irq__softirq_raise':
+			handle_irq_softirq_raise(event_info)
+		elif name == 'irq__irq_handler_entry':
+			handle_irq_handler_entry(event_info)
+		elif name == 'irq__irq_handler_exit':
+			handle_irq_handler_exit(event_info)
+		elif name == 'napi__napi_poll':
+			handle_napi_poll(event_info)
+		elif name == 'net__netif_receive_skb':
+			handle_netif_receive_skb(event_info)
+		elif name == 'net__netif_rx':
+			handle_netif_rx(event_info)
+		elif name == 'skb__skb_copy_datagram_iovec':
+			handle_skb_copy_datagram_iovec(event_info)
+		elif name == 'net__net_dev_queue':
+			handle_net_dev_queue(event_info)
+		elif name == 'net__net_dev_xmit':
+			handle_net_dev_xmit(event_info)
+		elif name == 'skb__kfree_skb':
+			handle_kfree_skb(event_info)
+		elif name == 'skb__consume_skb':
+			handle_consume_skb(event_info)
+	# display receive hunks
+	if show_rx:
+		for i in range(len(receive_hunk_list)):
+			print_receive(receive_hunk_list[i])
+	# display transmit hunks
+	if show_tx:
+		print "   dev    len      Qdisc        " \
+			"       netdevice             free"
+		for i in range(len(tx_free_list)):
+			print_transmit(tx_free_list[i])
+	if debug:
+		print "debug buffer status"
+		print "----------------------------"
+		print "xmit Qdisc:remain:%d overflow:%d" % \
+			(len(tx_queue_list), of_count_tx_queue_list)
+		print "xmit netdevice:remain:%d overflow:%d" % \
+			(len(tx_xmit_list), of_count_tx_xmit_list)
+		print "receive:remain:%d overflow:%d" % \
+			(len(rx_skb_list), of_count_rx_skb_list)
+
+# called from perf, when it finds a correspoinding event
+def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm,
+			irq, irq_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			irq, irq_name)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, irq, ret):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret)
+	all_event_list.append(event_info)
+
+def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, napi, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			napi, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, skblen, rc, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, rc ,dev_name)
+	all_event_list.append(event_info)
+
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, protocol, location):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, protocol, location)
+	all_event_list.append(event_info)
+
+def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr)
+	all_event_list.append(event_info)
+
+def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm,
+	skbaddr, skblen):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen)
+	all_event_list.append(event_info)
+
+def handle_irq_handler_entry(event_info):
+	(name, context, cpu, time, pid, comm, irq, irq_name) = event_info
+	if cpu not in irq_dic.keys():
+		irq_dic[cpu] = []
+	irq_record = {'irq':irq, 'name':irq_name, 'cpu':cpu, 'irq_ent_t':time}
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_handler_exit(event_info):
+	(name, context, cpu, time, pid, comm, irq, ret) = event_info
+	if cpu not in irq_dic.keys():
+		return
+	irq_record = irq_dic[cpu].pop()
+	if irq != irq_record['irq']:
+		return
+	irq_record.update({'irq_ext_t':time})
+	# if an irq doesn't include NET_RX softirq, drop.
+	if 'event_list' in irq_record.keys():
+		irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_raise(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'sirq_raise'})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_entry(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	net_rx_dic[cpu] = {'sirq_ent_t':time, 'event_list':[]}
+
+def handle_irq_softirq_exit(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	irq_list = []
+	event_list = 0
+	if cpu in irq_dic.keys():
+		irq_list = irq_dic[cpu]
+		del irq_dic[cpu]
+	if cpu in net_rx_dic.keys():
+		sirq_ent_t = net_rx_dic[cpu]['sirq_ent_t']
+		event_list = net_rx_dic[cpu]['event_list']
+		del net_rx_dic[cpu]
+	if irq_list == [] or event_list == 0:
+		return
+	rec_data = {'sirq_ent_t':sirq_ent_t, 'sirq_ext_t':time,
+		    'irq_list':irq_list, 'event_list':event_list}
+	# merge information realted to a NET_RX softirq
+	receive_hunk_list.append(rec_data)
+
+def handle_napi_poll(event_info):
+	(name, context, cpu, time, pid, comm, napi, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		event_list = net_rx_dic[cpu]['event_list']
+		rec_data = {'event_name':'napi_poll',
+				'dev':dev_name, 'event_t':time}
+		event_list.append(rec_data)
+
+def handle_netif_rx(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'netif_rx',
+		'skbaddr':skbaddr, 'skblen':skblen, 'dev_name':dev_name})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_netif_receive_skb(event_info):
+	global of_count_rx_skb_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		rec_data = {'event_name':'netif_receive_skb',
+			    'event_t':time, 'skbaddr':skbaddr, 'len':skblen}
+		event_list = net_rx_dic[cpu]['event_list']
+		event_list.append(rec_data)
+		rx_skb_list.insert(0, rec_data)
+		if len(rx_skb_list) > buffer_budget:
+			rx_skb_list.pop()
+			of_count_rx_skb_list += 1
+
+def handle_net_dev_queue(event_info):
+	global of_count_tx_queue_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	skb = {'dev':dev_name, 'skbaddr':skbaddr, 'len':skblen, 'queue_t':time}
+	tx_queue_list.insert(0, skb)
+	if len(tx_queue_list) > buffer_budget:
+		tx_queue_list.pop()
+		of_count_tx_queue_list += 1
+
+def handle_net_dev_xmit(event_info):
+	global of_count_tx_xmit_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, rc, dev_name) = event_info
+	if rc == 0: # NETDEV_TX_OK
+		for i in range(len(tx_queue_list)):
+			skb = tx_queue_list[i]
+			if skb['skbaddr'] == skbaddr:
+				skb['xmit_t'] = time
+				tx_xmit_list.insert(0, skb)
+				del tx_queue_list[i]
+				if len(tx_xmit_list) > buffer_budget:
+					tx_xmit_list.pop()
+					of_count_tx_xmit_list += 1
+				return
+
+def handle_kfree_skb(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, protocol, location) = event_info
+	for i in range(len(tx_queue_list)):
+		skb = tx_queue_list[i]
+		if skb['skbaddr'] == skbaddr:
+			del tx_queue_list[i]
+			return
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if rec_data['skbaddr'] == skbaddr:
+			rec_data.update({'handle':"kfree_skb",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
+
+def handle_consume_skb(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr) = event_info
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+
+def handle_skb_copy_datagram_iovec(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr, skblen) = event_info
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if skbaddr == rec_data['skbaddr']:
+			rec_data.update({'handle':"skb_copy_datagram_iovec",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
index 27e9ebe..a772979 100644
--- a/tools/perf/util/cache.h
+++ b/tools/perf/util/cache.h
@@ -82,6 +82,8 @@
 extern char *perf_pathdup(const char *fmt, ...)
 	__attribute__((format (printf, 1, 2)));
 
+#ifdef NO_STRLCPY
 extern size_t strlcpy(char *dest, const char *src, size_t size);
+#endif
 
 #endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index f231f43..e12d539 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -28,6 +28,9 @@
 #define chain_for_each_child(child, parent)	\
 	list_for_each_entry(child, &parent->children, brothers)
 
+#define chain_for_each_child_safe(child, next, parent)	\
+	list_for_each_entry_safe(child, next, &parent->children, brothers)
+
 static void
 rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
 		    enum chain_mode mode)
@@ -86,10 +89,10 @@
  * sort them by hit
  */
 static void
-sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
+sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
 		u64 min_hit, struct callchain_param *param __used)
 {
-	__sort_chain_flat(rb_root, node, min_hit);
+	__sort_chain_flat(rb_root, &root->node, min_hit);
 }
 
 static void __sort_chain_graph_abs(struct callchain_node *node,
@@ -108,11 +111,11 @@
 }
 
 static void
-sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_node *chain_root,
+sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
 		     u64 min_hit, struct callchain_param *param __used)
 {
-	__sort_chain_graph_abs(chain_root, min_hit);
-	rb_root->rb_node = chain_root->rb_root.rb_node;
+	__sort_chain_graph_abs(&chain_root->node, min_hit);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
 }
 
 static void __sort_chain_graph_rel(struct callchain_node *node,
@@ -133,11 +136,11 @@
 }
 
 static void
-sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root,
+sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
 		     u64 min_hit __used, struct callchain_param *param)
 {
-	__sort_chain_graph_rel(chain_root, param->min_percent / 100.0);
-	rb_root->rb_node = chain_root->rb_root.rb_node;
+	__sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
 }
 
 int register_callchain_param(struct callchain_param *param)
@@ -284,19 +287,18 @@
 }
 
 static int
-__append_chain(struct callchain_node *root, struct resolved_chain *chain,
-	       unsigned int start, u64 period);
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+	     unsigned int start, u64 period);
 
 static void
-__append_chain_children(struct callchain_node *root,
-			struct resolved_chain *chain,
-			unsigned int start, u64 period)
+append_chain_children(struct callchain_node *root, struct resolved_chain *chain,
+		      unsigned int start, u64 period)
 {
 	struct callchain_node *rnode;
 
 	/* lookup in childrens */
 	chain_for_each_child(rnode, root) {
-		unsigned int ret = __append_chain(rnode, chain, start, period);
+		unsigned int ret = append_chain(rnode, chain, start, period);
 
 		if (!ret)
 			goto inc_children_hit;
@@ -309,8 +311,8 @@
 }
 
 static int
-__append_chain(struct callchain_node *root, struct resolved_chain *chain,
-	       unsigned int start, u64 period)
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+	     unsigned int start, u64 period)
 {
 	struct callchain_list *cnode;
 	unsigned int i = start;
@@ -357,7 +359,7 @@
 	}
 
 	/* We match the node and still have a part remaining */
-	__append_chain_children(root, chain, i, period);
+	append_chain_children(root, chain, i, period);
 
 	return 0;
 }
@@ -380,8 +382,8 @@
 }
 
 
-int append_chain(struct callchain_node *root, struct ip_callchain *chain,
-		 struct map_symbol *syms, u64 period)
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+		     struct map_symbol *syms, u64 period)
 {
 	struct resolved_chain *filtered;
 
@@ -398,9 +400,65 @@
 	if (!filtered->nr)
 		goto end;
 
-	__append_chain_children(root, filtered, 0, period);
+	append_chain_children(&root->node, filtered, 0, period);
+
+	if (filtered->nr > root->max_depth)
+		root->max_depth = filtered->nr;
 end:
 	free(filtered);
 
 	return 0;
 }
+
+static int
+merge_chain_branch(struct callchain_node *dst, struct callchain_node *src,
+		   struct resolved_chain *chain)
+{
+	struct callchain_node *child, *next_child;
+	struct callchain_list *list, *next_list;
+	int old_pos = chain->nr;
+	int err = 0;
+
+	list_for_each_entry_safe(list, next_list, &src->val, list) {
+		chain->ips[chain->nr].ip = list->ip;
+		chain->ips[chain->nr].ms = list->ms;
+		chain->nr++;
+		list_del(&list->list);
+		free(list);
+	}
+
+	if (src->hit)
+		append_chain_children(dst, chain, 0, src->hit);
+
+	chain_for_each_child_safe(child, next_child, src) {
+		err = merge_chain_branch(dst, child, chain);
+		if (err)
+			break;
+
+		list_del(&child->brothers);
+		free(child);
+	}
+
+	chain->nr = old_pos;
+
+	return err;
+}
+
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src)
+{
+	struct resolved_chain *chain;
+	int err;
+
+	chain = malloc(sizeof(*chain) +
+		       src->max_depth * sizeof(struct resolved_ip));
+	if (!chain)
+		return -ENOMEM;
+
+	chain->nr = 0;
+
+	err = merge_chain_branch(&dst->node, &src->node, chain);
+
+	free(chain);
+
+	return err;
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 6de4313..c15fb8c 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -26,9 +26,14 @@
 	u64			children_hit;
 };
 
+struct callchain_root {
+	u64			max_depth;
+	struct callchain_node	node;
+};
+
 struct callchain_param;
 
-typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_node *,
+typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *,
 				 u64, struct callchain_param *);
 
 struct callchain_param {
@@ -44,15 +49,16 @@
 	struct list_head	list;
 };
 
-static inline void callchain_init(struct callchain_node *node)
+static inline void callchain_init(struct callchain_root *root)
 {
-	INIT_LIST_HEAD(&node->brothers);
-	INIT_LIST_HEAD(&node->children);
-	INIT_LIST_HEAD(&node->val);
+	INIT_LIST_HEAD(&root->node.brothers);
+	INIT_LIST_HEAD(&root->node.children);
+	INIT_LIST_HEAD(&root->node.val);
 
-	node->children_hit = 0;
-	node->parent = NULL;
-	node->hit = 0;
+	root->node.parent = NULL;
+	root->node.hit = 0;
+	root->node.children_hit = 0;
+	root->max_depth = 0;
 }
 
 static inline u64 cumul_hits(struct callchain_node *node)
@@ -61,8 +67,9 @@
 }
 
 int register_callchain_param(struct callchain_param *param);
-int append_chain(struct callchain_node *root, struct ip_callchain *chain,
-		 struct map_symbol *syms, u64 period);
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+		     struct map_symbol *syms, u64 period);
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src);
 
 bool ip_callchain__valid(struct ip_callchain *chain, const event_t *event);
 #endif	/* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index be22ae6..2022e87 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -87,7 +87,7 @@
 
 static struct hist_entry *hist_entry__new(struct hist_entry *template)
 {
-	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_node) : 0;
+	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0;
 	struct hist_entry *self = malloc(sizeof(*self) + callchain_size);
 
 	if (self != NULL) {
@@ -226,6 +226,8 @@
 
 		if (!cmp) {
 			iter->period += he->period;
+			if (symbol_conf.use_callchain)
+				callchain_merge(iter->callchain, he->callchain);
 			hist_entry__free(he);
 			return false;
 		}
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
index 58a470d..bd74977 100644
--- a/tools/perf/util/path.c
+++ b/tools/perf/util/path.c
@@ -22,6 +22,7 @@
 	return ".";
 }
 
+#ifdef NO_STRLCPY
 size_t strlcpy(char *dest, const char *src, size_t size)
 {
 	size_t ret = strlen(src);
@@ -33,7 +34,7 @@
 	}
 	return ret;
 }
-
+#endif
 
 static char *get_pathname(void)
 {
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index 46e531d..0b91053 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -70,7 +70,7 @@
 		struct hist_entry *pair;
 		struct rb_root	  sorted_chain;
 	};
-	struct callchain_node	callchain[0];
+	struct callchain_root	callchain[0];
 };
 
 enum sort_type {
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index b2f5ae9..b39f499 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -388,6 +388,20 @@
 	return fprintf(fp, "%s", sbuild_id);
 }
 
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+	struct symbol_name_rb_node *pos;
+
+	for (nd = rb_first(&self->symbol_names[type]); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct symbol_name_rb_node, rb_node);
+		fprintf(fp, "%s\n", pos->sym.name);
+	}
+
+	return ret;
+}
+
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
 {
 	struct rb_node *nd;
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ea95c27..038f220 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -182,6 +182,7 @@
 size_t machines__fprintf_dsos_buildid(struct rb_root *self, FILE *fp, bool with_hits);
 
 size_t dso__fprintf_buildid(struct dso *self, FILE *fp);
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp);
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp);
 
 enum dso_origin {
diff --git a/tools/perf/util/ui/browser.c b/tools/perf/util/ui/browser.c
index 66f2d58..6d0df80 100644
--- a/tools/perf/util/ui/browser.c
+++ b/tools/perf/util/ui/browser.c
@@ -1,16 +1,6 @@
-#define _GNU_SOURCE
-#include <stdio.h>
-#undef _GNU_SOURCE
-/*
- * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks
- * the build if it isn't defined. Use the equivalent one that glibc
- * has on features.h.
- */
-#include <features.h>
-#ifndef HAVE_LONG_LONG
-#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG
-#endif
 #include <slang.h>
+#include "libslang.h"
+#include <linux/compiler.h>
 #include <linux/list.h>
 #include <linux/rbtree.h>
 #include <stdlib.h>
@@ -19,17 +9,9 @@
 #include "helpline.h"
 #include "../color.h"
 #include "../util.h"
+#include <stdio.h>
 
-#if SLANG_VERSION < 20104
-#define sltt_set_color(obj, name, fg, bg) \
-	SLtt_set_color(obj,(char *)name, (char *)fg, (char *)bg)
-#else
-#define sltt_set_color SLtt_set_color
-#endif
-
-newtComponent newt_form__new(void);
-
-int ui_browser__percent_color(double percent, bool current)
+static int ui_browser__percent_color(double percent, bool current)
 {
 	if (current)
 		return HE_COLORSET_SELECTED;
@@ -40,6 +22,23 @@
 	return HE_COLORSET_NORMAL;
 }
 
+void ui_browser__set_color(struct ui_browser *self __used, int color)
+{
+	SLsmg_set_color(color);
+}
+
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current)
+{
+	 int color = ui_browser__percent_color(percent, current);
+	 ui_browser__set_color(self, color);
+}
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x)
+{
+	SLsmg_gotorc(self->y + y, self->x + x);
+}
+
 void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence)
 {
 	struct list_head *head = self->entries;
@@ -111,7 +110,7 @@
 	nd = self->top;
 
 	while (nd != NULL) {
-		SLsmg_gotorc(self->y + row, self->x);
+		ui_browser__gotorc(self, row, 0);
 		self->write(self, nd, row);
 		if (++row == self->height)
 			break;
@@ -131,13 +130,10 @@
 	int cols, rows;
 	newtGetScreenSize(&cols, &rows);
 
-	if (self->width > cols - 4)
-		self->width = cols - 4;
-	self->height = rows - 5;
-	if (self->height > self->nr_entries)
-		self->height = self->nr_entries;
-	self->y  = (rows - self->height) / 2;
-	self->x = (cols - self->width) / 2;
+	self->width = cols - 1;
+	self->height = rows - 2;
+	self->y = 1;
+	self->x = 0;
 }
 
 void ui_browser__reset_index(struct ui_browser *self)
@@ -146,34 +142,48 @@
 	self->seek(self, 0, SEEK_SET);
 }
 
+void ui_browser__add_exit_key(struct ui_browser *self, int key)
+{
+	newtFormAddHotKey(self->form, key);
+}
+
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[])
+{
+	int i = 0;
+
+	while (keys[i] && i < 64) {
+		ui_browser__add_exit_key(self, keys[i]);
+		++i;
+	}
+}
+
 int ui_browser__show(struct ui_browser *self, const char *title,
 		     const char *helpline, ...)
 {
 	va_list ap;
+	int keys[] = { NEWT_KEY_UP, NEWT_KEY_DOWN, NEWT_KEY_PGUP,
+		       NEWT_KEY_PGDN, NEWT_KEY_HOME, NEWT_KEY_END, ' ',
+		       NEWT_KEY_LEFT, NEWT_KEY_ESCAPE, 'q', CTRL('c'), 0 };
 
-	if (self->form != NULL) {
+	if (self->form != NULL)
 		newtFormDestroy(self->form);
-		newtPopWindow();
-	}
+
 	ui_browser__refresh_dimensions(self);
-	newtCenteredWindow(self->width, self->height, title);
-	self->form = newt_form__new();
+	self->form = newtForm(NULL, NULL, 0);
 	if (self->form == NULL)
 		return -1;
 
-	self->sb = newtVerticalScrollbar(self->width, 0, self->height,
+	self->sb = newtVerticalScrollbar(self->width, 1, self->height,
 					 HE_COLORSET_NORMAL,
 					 HE_COLORSET_SELECTED);
 	if (self->sb == NULL)
 		return -1;
 
-	newtFormAddHotKey(self->form, NEWT_KEY_UP);
-	newtFormAddHotKey(self->form, NEWT_KEY_DOWN);
-	newtFormAddHotKey(self->form, NEWT_KEY_PGUP);
-	newtFormAddHotKey(self->form, NEWT_KEY_PGDN);
-	newtFormAddHotKey(self->form, NEWT_KEY_HOME);
-	newtFormAddHotKey(self->form, NEWT_KEY_END);
-	newtFormAddHotKey(self->form, ' ');
+	SLsmg_gotorc(0, 0);
+	ui_browser__set_color(self, NEWT_COLORSET_ROOT);
+	slsmg_write_nstring(title, self->width);
+
+	ui_browser__add_exit_keys(self, keys);
 	newtFormAddComponent(self->form, self->sb);
 
 	va_start(ap, helpline);
@@ -185,7 +195,6 @@
 void ui_browser__hide(struct ui_browser *self)
 {
 	newtFormDestroy(self->form);
-	newtPopWindow();
 	self->form = NULL;
 	ui_helpline__pop();
 }
@@ -196,28 +205,28 @@
 
 	newtScrollbarSet(self->sb, self->index, self->nr_entries - 1);
 	row = self->refresh(self);
-	SLsmg_set_color(HE_COLORSET_NORMAL);
+	ui_browser__set_color(self, HE_COLORSET_NORMAL);
 	SLsmg_fill_region(self->y + row, self->x,
 			  self->height - row, self->width, ' ');
 
 	return 0;
 }
 
-int ui_browser__run(struct ui_browser *self, struct newtExitStruct *es)
+int ui_browser__run(struct ui_browser *self)
 {
+	struct newtExitStruct es;
+
 	if (ui_browser__refresh(self) < 0)
 		return -1;
 
 	while (1) {
 		off_t offset;
 
-		newtFormRun(self->form, es);
+		newtFormRun(self->form, &es);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
+		if (es.reason != NEWT_EXIT_HOTKEY)
 			break;
-		if (is_exit_key(es->u.key))
-			return es->u.key;
-		switch (es->u.key) {
+		switch (es.u.key) {
 		case NEWT_KEY_DOWN:
 			if (self->index == self->nr_entries - 1)
 				break;
@@ -274,12 +283,12 @@
 			self->seek(self, -offset, SEEK_END);
 			break;
 		default:
-			return es->u.key;
+			return es.u.key;
 		}
 		if (ui_browser__refresh(self) < 0)
 			return -1;
 	}
-	return 0;
+	return -1;
 }
 
 unsigned int ui_browser__list_head_refresh(struct ui_browser *self)
@@ -294,7 +303,7 @@
 	pos = self->top;
 
 	list_for_each_from(pos, head) {
-		SLsmg_gotorc(self->y + row, self->x);
+		ui_browser__gotorc(self, row, 0);
 		self->write(self, pos, row);
 		if (++row == self->height)
 			break;
diff --git a/tools/perf/util/ui/browser.h b/tools/perf/util/ui/browser.h
index 0b9f829..0dc7e4d 100644
--- a/tools/perf/util/ui/browser.h
+++ b/tools/perf/util/ui/browser.h
@@ -25,16 +25,21 @@
 };
 
 
-int ui_browser__percent_color(double percent, bool current);
+void ui_browser__set_color(struct ui_browser *self, int color);
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current);
 bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row);
 void ui_browser__refresh_dimensions(struct ui_browser *self);
 void ui_browser__reset_index(struct ui_browser *self);
 
+void ui_browser__gotorc(struct ui_browser *self, int y, int x);
+void ui_browser__add_exit_key(struct ui_browser *self, int key);
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[]);
 int ui_browser__show(struct ui_browser *self, const char *title,
 		     const char *helpline, ...);
 void ui_browser__hide(struct ui_browser *self);
 int ui_browser__refresh(struct ui_browser *self);
-int ui_browser__run(struct ui_browser *self, struct newtExitStruct *es);
+int ui_browser__run(struct ui_browser *self);
 
 void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence);
 unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self);
diff --git a/tools/perf/util/ui/browsers/annotate.c b/tools/perf/util/ui/browsers/annotate.c
index a90273e..82b78f9 100644
--- a/tools/perf/util/ui/browsers/annotate.c
+++ b/tools/perf/util/ui/browsers/annotate.c
@@ -40,14 +40,12 @@
 
 	if (ol->offset != -1) {
 		struct objdump_line_rb_node *olrb = objdump_line__rb(ol);
-		int color = ui_browser__percent_color(olrb->percent, current_entry);
-		SLsmg_set_color(color);
+		ui_browser__set_percent_color(self, olrb->percent, current_entry);
 		slsmg_printf(" %7.2f ", olrb->percent);
 		if (!current_entry)
-			SLsmg_set_color(HE_COLORSET_CODE);
+			ui_browser__set_color(self, HE_COLORSET_CODE);
 	} else {
-		int color = ui_browser__percent_color(0, current_entry);
-		SLsmg_set_color(color);
+		ui_browser__set_percent_color(self, 0, current_entry);
 		slsmg_write_nstring(" ", 9);
 	}
 
@@ -135,32 +133,31 @@
 	self->curr_hot = nd;
 }
 
-static int annotate_browser__run(struct annotate_browser *self,
-				 struct newtExitStruct *es)
+static int annotate_browser__run(struct annotate_browser *self)
 {
 	struct rb_node *nd;
 	struct hist_entry *he = self->b.priv;
+	int key;
 
 	if (ui_browser__show(&self->b, he->ms.sym->name,
-			     "<- or ESC: exit, TAB/shift+TAB: cycle thru samples") < 0)
+			     "<-, -> or ESC: exit, TAB/shift+TAB: cycle thru samples") < 0)
 		return -1;
-
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_RIGHT);
+	/*
+	 * To allow builtin-annotate to cycle thru multiple symbols by
+	 * examining the exit key for this function.
+	 */
+	ui_browser__add_exit_key(&self->b, NEWT_KEY_RIGHT);
 
 	nd = self->curr_hot;
 	if (nd) {
-		newtFormAddHotKey(self->b.form, NEWT_KEY_TAB);
-		newtFormAddHotKey(self->b.form, NEWT_KEY_UNTAB);
+		int tabs[] = { NEWT_KEY_TAB, NEWT_KEY_UNTAB, 0 };
+		ui_browser__add_exit_keys(&self->b, tabs);
 	}
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-
-		switch (es->u.key) {
+		switch (key) {
 		case NEWT_KEY_TAB:
 			nd = rb_prev(nd);
 			if (nd == NULL)
@@ -179,12 +176,11 @@
 	}
 out:
 	ui_browser__hide(&self->b);
-	return es->u.key;
+	return key;
 }
 
 int hist_entry__tui_annotate(struct hist_entry *self)
 {
-	struct newtExitStruct es;
 	struct objdump_line *pos, *n;
 	struct objdump_line_rb_node *rbpos;
 	LIST_HEAD(head);
@@ -232,7 +228,7 @@
 		annotate_browser__set_top(&browser, browser.curr_hot);
 
 	browser.b.width += 18; /* Percentage */
-	ret = annotate_browser__run(&browser, &es);
+	ret = annotate_browser__run(&browser);
 	list_for_each_entry_safe(pos, n, &head, node) {
 		list_del(&pos->node);
 		objdump_line__free(pos);
diff --git a/tools/perf/util/ui/browsers/hists.c b/tools/perf/util/ui/browsers/hists.c
index 6866aa4..ebda8c3 100644
--- a/tools/perf/util/ui/browsers/hists.c
+++ b/tools/perf/util/ui/browsers/hists.c
@@ -58,6 +58,11 @@
 	return map_symbol__folded(&self->ms);
 }
 
+static void map_symbol__set_folding(struct map_symbol *self, bool unfold)
+{
+	self->unfolded = unfold ? self->has_children : false;
+}
+
 static int callchain_node__count_rows_rb_tree(struct callchain_node *self)
 {
 	int n = 0;
@@ -129,16 +134,16 @@
 	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
 		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
 		struct callchain_list *chain;
-		int first = true;
+		bool first = true;
 
 		list_for_each_entry(chain, &child->val, list) {
 			if (first) {
 				first = false;
 				chain->ms.has_children = chain->list.next != &child->val ||
-							 rb_first(&child->rb_root) != NULL;
+							 !RB_EMPTY_ROOT(&child->rb_root);
 			} else
 				chain->ms.has_children = chain->list.next == &child->val &&
-							 rb_first(&child->rb_root) != NULL;
+							 !RB_EMPTY_ROOT(&child->rb_root);
 		}
 
 		callchain_node__init_have_children_rb_tree(child);
@@ -150,7 +155,7 @@
 	struct callchain_list *chain;
 
 	list_for_each_entry(chain, &self->val, list)
-		chain->ms.has_children = rb_first(&self->rb_root) != NULL;
+		chain->ms.has_children = !RB_EMPTY_ROOT(&self->rb_root);
 
 	callchain_node__init_have_children_rb_tree(self);
 }
@@ -168,6 +173,7 @@
 static void hist_entry__init_have_children(struct hist_entry *self)
 {
 	if (!self->init_have_children) {
+		self->ms.has_children = !RB_EMPTY_ROOT(&self->sorted_chain);
 		callchain__init_have_children(&self->sorted_chain);
 		self->init_have_children = true;
 	}
@@ -195,43 +201,114 @@
 	return false;
 }
 
-static int hist_browser__run(struct hist_browser *self, const char *title,
-			     struct newtExitStruct *es)
+static int callchain_node__set_folding_rb_tree(struct callchain_node *self, bool unfold)
 {
-	char str[256], unit;
-	unsigned long nr_events = self->hists->stats.nr_events[PERF_RECORD_SAMPLE];
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool has_children = false;
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			map_symbol__set_folding(&chain->ms, unfold);
+			has_children = chain->ms.has_children;
+		}
+
+		if (has_children)
+			n += callchain_node__set_folding_rb_tree(child, unfold);
+	}
+
+	return n;
+}
+
+static int callchain_node__set_folding(struct callchain_node *node, bool unfold)
+{
+	struct callchain_list *chain;
+	bool has_children = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		map_symbol__set_folding(&chain->ms, unfold);
+		has_children = chain->ms.has_children;
+	}
+
+	if (has_children)
+		n += callchain_node__set_folding_rb_tree(node, unfold);
+
+	return n;
+}
+
+static int callchain__set_folding(struct rb_root *chain, bool unfold)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__set_folding(node, unfold);
+	}
+
+	return n;
+}
+
+static void hist_entry__set_folding(struct hist_entry *self, bool unfold)
+{
+	hist_entry__init_have_children(self);
+	map_symbol__set_folding(&self->ms, unfold);
+
+	if (self->ms.has_children) {
+		int n = callchain__set_folding(&self->sorted_chain, unfold);
+		self->nr_rows = unfold ? n : 0;
+	} else
+		self->nr_rows = 0;
+}
+
+static void hists__set_folding(struct hists *self, bool unfold)
+{
+	struct rb_node *nd;
+
+	self->nr_entries = 0;
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+		hist_entry__set_folding(he, unfold);
+		self->nr_entries += 1 + he->nr_rows;
+	}
+}
+
+static void hist_browser__set_folding(struct hist_browser *self, bool unfold)
+{
+	hists__set_folding(self->hists, unfold);
+	self->b.nr_entries = self->hists->nr_entries;
+	/* Go to the start, we may be way after valid entries after a collapse */
+	ui_browser__reset_index(&self->b);
+}
+
+static int hist_browser__run(struct hist_browser *self, const char *title)
+{
+	int key;
+	int exit_keys[] = { 'a', '?', 'h', 'C', 'd', 'D', 'E', 't',
+			    NEWT_KEY_ENTER, NEWT_KEY_RIGHT, NEWT_KEY_LEFT, 0, };
 
 	self->b.entries = &self->hists->entries;
 	self->b.nr_entries = self->hists->nr_entries;
 
 	hist_browser__refresh_dimensions(self);
 
-	nr_events = convert_unit(nr_events, &unit);
-	snprintf(str, sizeof(str), "Events: %lu%c                            ",
-		 nr_events, unit);
-	newtDrawRootText(0, 0, str);
-
 	if (ui_browser__show(&self->b, title,
 			     "Press '?' for help on key bindings") < 0)
 		return -1;
 
-	newtFormAddHotKey(self->b.form, 'a');
-	newtFormAddHotKey(self->b.form, '?');
-	newtFormAddHotKey(self->b.form, 'h');
-	newtFormAddHotKey(self->b.form, 'd');
-	newtFormAddHotKey(self->b.form, 'D');
-	newtFormAddHotKey(self->b.form, 't');
-
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_RIGHT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_ENTER);
+	ui_browser__add_exit_keys(&self->b, exit_keys);
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-		switch (es->u.key) {
+		switch (key) {
 		case 'D': { /* Debug */
 			static int seq;
 			struct hist_entry *h = rb_entry(self->b.top,
@@ -245,18 +322,26 @@
 					   self->b.top_idx,
 					   h->row_offset, h->nr_rows);
 		}
-			continue;
+			break;
+		case 'C':
+			/* Collapse the whole world. */
+			hist_browser__set_folding(self, false);
+			break;
+		case 'E':
+			/* Expand the whole world. */
+			hist_browser__set_folding(self, true);
+			break;
 		case NEWT_KEY_ENTER:
 			if (hist_browser__toggle_fold(self))
 				break;
 			/* fall thru */
 		default:
-			return 0;
+			goto out;
 		}
 	}
-
+out:
 	ui_browser__hide(&self->b);
-	return 0;
+	return key;
 }
 
 static char *callchain_list__sym_name(struct callchain_list *self,
@@ -306,15 +391,10 @@
 			int color;
 			bool was_first = first;
 
-			if (first) {
+			if (first)
 				first = false;
-				chain->ms.has_children = chain->list.next != &child->val ||
-							 rb_first(&child->rb_root) != NULL;
-			} else {
+			else
 				extra_offset = LEVEL_OFFSET_STEP;
-				chain->ms.has_children = chain->list.next == &child->val &&
-							 rb_first(&child->rb_root) != NULL;
-			}
 
 			folded_sign = callchain_list__folded(chain);
 			if (*row_offset != 0) {
@@ -341,8 +421,8 @@
 				*is_current_entry = true;
 			}
 
-			SLsmg_set_color(color);
-			SLsmg_gotorc(self->b.y + row, self->b.x);
+			ui_browser__set_color(&self->b, color);
+			ui_browser__gotorc(&self->b, row, 0);
 			slsmg_write_nstring(" ", offset + extra_offset);
 			slsmg_printf("%c ", folded_sign);
 			slsmg_write_nstring(str, width);
@@ -384,12 +464,7 @@
 	list_for_each_entry(chain, &node->val, list) {
 		char ipstr[BITS_PER_LONG / 4 + 1], *s;
 		int color;
-		/*
-		 * FIXME: This should be moved to somewhere else,
-		 * probably when the callchain is created, so as not to
-		 * traverse it all over again
-		 */
-		chain->ms.has_children = rb_first(&node->rb_root) != NULL;
+
 		folded_sign = callchain_list__folded(chain);
 
 		if (*row_offset != 0) {
@@ -405,8 +480,8 @@
 		}
 
 		s = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
-		SLsmg_gotorc(self->b.y + row, self->b.x);
-		SLsmg_set_color(color);
+		ui_browser__gotorc(&self->b, row, 0);
+		ui_browser__set_color(&self->b, color);
 		slsmg_write_nstring(" ", offset);
 		slsmg_printf("%c ", folded_sign);
 		slsmg_write_nstring(s, width - 2);
@@ -465,7 +540,7 @@
 	}
 
 	if (symbol_conf.use_callchain) {
-		entry->ms.has_children = !RB_EMPTY_ROOT(&entry->sorted_chain);
+		hist_entry__init_have_children(entry);
 		folded_sign = hist_entry__folded(entry);
 	}
 
@@ -484,8 +559,8 @@
 				color = HE_COLORSET_NORMAL;
 		}
 
-		SLsmg_set_color(color);
-		SLsmg_gotorc(self->b.y + row, self->b.x);
+		ui_browser__set_color(&self->b, color);
+		ui_browser__gotorc(&self->b, row, 0);
 		if (symbol_conf.use_callchain) {
 			slsmg_printf("%c ", folded_sign);
 			width -= 2;
@@ -687,8 +762,6 @@
 
 static void hist_browser__delete(struct hist_browser *self)
 {
-	newtFormDestroy(self->b.form);
-	newtPopWindow();
 	free(self);
 }
 
@@ -702,21 +775,26 @@
 	return self->he_selection->thread;
 }
 
-static int hist_browser__title(char *bf, size_t size, const char *ev_name,
-			       const struct dso *dso, const struct thread *thread)
+static int hists__browser_title(struct hists *self, char *bf, size_t size,
+				const char *ev_name, const struct dso *dso,
+				const struct thread *thread)
 {
-	int printed = 0;
+	char unit;
+	int printed;
+	unsigned long nr_events = self->stats.nr_events[PERF_RECORD_SAMPLE];
+
+	nr_events = convert_unit(nr_events, &unit);
+	printed = snprintf(bf, size, "Events: %lu%c %s", nr_events, unit, ev_name);
 
 	if (thread)
 		printed += snprintf(bf + printed, size - printed,
-				    "Thread: %s(%d)",
-				    (thread->comm_set ?  thread->comm : ""),
+				    ", Thread: %s(%d)",
+				    (thread->comm_set ? thread->comm : ""),
 				    thread->pid);
 	if (dso)
 		printed += snprintf(bf + printed, size - printed,
-				    "%sDSO: %s", thread ? " " : "",
-				    dso->short_name);
-	return printed ?: snprintf(bf, size, "Event: %s", ev_name);
+				    ", DSO: %s", dso->short_name);
+	return printed;
 }
 
 int hists__browse(struct hists *self, const char *helpline, const char *ev_name)
@@ -725,7 +803,6 @@
 	struct pstack *fstack;
 	const struct thread *thread_filter = NULL;
 	const struct dso *dso_filter = NULL;
-	struct newtExitStruct es;
 	char msg[160];
 	int key = -1;
 
@@ -738,9 +815,8 @@
 
 	ui_helpline__push(helpline);
 
-	hist_browser__title(msg, sizeof(msg), ev_name,
-			    dso_filter, thread_filter);
-
+	hists__browser_title(self, msg, sizeof(msg), ev_name,
+			     dso_filter, thread_filter);
 	while (1) {
 		const struct thread *thread;
 		const struct dso *dso;
@@ -749,70 +825,63 @@
 		    annotate = -2, zoom_dso = -2, zoom_thread = -2,
 		    browse_map = -2;
 
-		if (hist_browser__run(browser, msg, &es))
-			break;
+		key = hist_browser__run(browser, msg);
 
 		thread = hist_browser__selected_thread(browser);
 		dso = browser->selection->map ? browser->selection->map->dso : NULL;
 
-		if (es.reason == NEWT_EXIT_HOTKEY) {
-			key = es.u.key;
-
-			switch (key) {
-			case NEWT_KEY_F1:
-				goto do_help;
-			case NEWT_KEY_TAB:
-			case NEWT_KEY_UNTAB:
-				/*
-				 * Exit the browser, let hists__browser_tree
-				 * go to the next or previous
-				 */
-				goto out_free_stack;
-			default:;
-			}
-
-			switch (key) {
-			case 'a':
-				if (browser->selection->map == NULL ||
-				    browser->selection->map->dso->annotate_warned)
-					continue;
-				goto do_annotate;
-			case 'd':
-				goto zoom_dso;
-			case 't':
-				goto zoom_thread;
-			case 'h':
-			case '?':
-do_help:
-				ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
-						"<-        Zoom out\n"
-						"a         Annotate current symbol\n"
-						"h/?/F1    Show this window\n"
-						"d         Zoom into current DSO\n"
-						"t         Zoom into current Thread\n"
-						"q/CTRL+C  Exit browser");
+		switch (key) {
+		case NEWT_KEY_TAB:
+		case NEWT_KEY_UNTAB:
+			/*
+			 * Exit the browser, let hists__browser_tree
+			 * go to the next or previous
+			 */
+			goto out_free_stack;
+		case 'a':
+			if (browser->selection->map == NULL &&
+			    browser->selection->map->dso->annotate_warned)
 				continue;
-			default:;
-			}
-			if (is_exit_key(key)) {
-				if (key == NEWT_KEY_ESCAPE &&
-				    !ui__dialog_yesno("Do you really want to exit?"))
-					continue;
-				break;
-			}
+			goto do_annotate;
+		case 'd':
+			goto zoom_dso;
+		case 't':
+			goto zoom_thread;
+		case NEWT_KEY_F1:
+		case 'h':
+		case '?':
+			ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
+					"<-        Zoom out\n"
+					"a         Annotate current symbol\n"
+					"h/?/F1    Show this window\n"
+					"C         Collapse all callchains\n"
+					"E         Expand all callchains\n"
+					"d         Zoom into current DSO\n"
+					"t         Zoom into current Thread\n"
+					"q/CTRL+C  Exit browser");
+			continue;
+		case NEWT_KEY_ENTER:
+		case NEWT_KEY_RIGHT:
+			/* menu */
+			break;
+		case NEWT_KEY_LEFT: {
+			const void *top;
 
-			if (es.u.key == NEWT_KEY_LEFT) {
-				const void *top;
-
-				if (pstack__empty(fstack))
-					continue;
-				top = pstack__pop(fstack);
-				if (top == &dso_filter)
-					goto zoom_out_dso;
-				if (top == &thread_filter)
-					goto zoom_out_thread;
+			if (pstack__empty(fstack))
 				continue;
-			}
+			top = pstack__pop(fstack);
+			if (top == &dso_filter)
+				goto zoom_out_dso;
+			if (top == &thread_filter)
+				goto zoom_out_thread;
+			continue;
+		}
+		case NEWT_KEY_ESCAPE:
+			if (!ui__dialog_yesno("Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		default:
+			goto out_free_stack;
 		}
 
 		if (browser->selection->sym != NULL &&
@@ -885,8 +954,8 @@
 				pstack__push(fstack, &dso_filter);
 			}
 			hists__filter_by_dso(self, dso_filter);
-			hist_browser__title(msg, sizeof(msg), ev_name,
-					    dso_filter, thread_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
 			hist_browser__reset(browser);
 		} else if (choice == zoom_thread) {
 zoom_thread:
@@ -903,8 +972,8 @@
 				pstack__push(fstack, &thread_filter);
 			}
 			hists__filter_by_thread(self, thread_filter);
-			hist_browser__title(msg, sizeof(msg), ev_name,
-					    dso_filter, thread_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
 			hist_browser__reset(browser);
 		}
 	}
@@ -925,10 +994,6 @@
 		const char *ev_name = __event_name(hists->type, hists->config);
 
 		key = hists__browse(hists, help, ev_name);
-
-		if (is_exit_key(key))
-			break;
-
 		switch (key) {
 		case NEWT_KEY_TAB:
 			next = rb_next(nd);
@@ -940,7 +1005,7 @@
 				continue;
 			nd = rb_prev(nd);
 		default:
-			break;
+			return key;
 		}
 	}
 
diff --git a/tools/perf/util/ui/browsers/map.c b/tools/perf/util/ui/browsers/map.c
index 142b825..e35437d 100644
--- a/tools/perf/util/ui/browsers/map.c
+++ b/tools/perf/util/ui/browsers/map.c
@@ -1,6 +1,5 @@
 #include "../libslang.h"
 #include <elf.h>
-#include <newt.h>
 #include <sys/ttydefaults.h>
 #include <ctype.h>
 #include <string.h>
@@ -47,7 +46,6 @@
 struct map_browser {
 	struct ui_browser b;
 	struct map	  *map;
-	u16		  namelen;
 	u8		  addrlen;
 };
 
@@ -56,14 +54,16 @@
 	struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
 	struct map_browser *mb = container_of(self, struct map_browser, b);
 	bool current_entry = ui_browser__is_current_entry(self, row);
-	int color = ui_browser__percent_color(0, current_entry);
+	int width;
 
-	SLsmg_set_color(color);
+	ui_browser__set_percent_color(self, 0, current_entry);
 	slsmg_printf("%*llx %*llx %c ",
 		     mb->addrlen, sym->start, mb->addrlen, sym->end,
 		     sym->binding == STB_GLOBAL ? 'g' :
 		     sym->binding == STB_LOCAL  ? 'l' : 'w');
-	slsmg_write_nstring(sym->name, mb->namelen);
+	width = self->width - ((mb->addrlen * 2) + 4);
+	if (width > 0)
+		slsmg_write_nstring(sym->name, width);
 }
 
 /* FIXME uber-kludgy, see comment on cmd_report... */
@@ -98,31 +98,29 @@
 	return 0;
 }
 
-static int map_browser__run(struct map_browser *self, struct newtExitStruct *es)
+static int map_browser__run(struct map_browser *self)
 {
+	int key;
+
 	if (ui_browser__show(&self->b, self->map->dso->long_name,
 			     "Press <- or ESC to exit, %s / to search",
 			     verbose ? "" : "restart with -v to use") < 0)
 		return -1;
 
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_ENTER);
 	if (verbose)
-		newtFormAddHotKey(self->b.form, '/');
+		ui_browser__add_exit_key(&self->b, '/');
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-		if (verbose && es->u.key == '/')
+		if (verbose && key == '/')
 			map_browser__search(self);
 		else
 			break;
 	}
 
 	ui_browser__hide(&self->b);
-	return 0;
+	return key;
 }
 
 int map__browse(struct map *self)
@@ -136,7 +134,6 @@
 		},
 		.map = self,
 	};
-	struct newtExitStruct es;
 	struct rb_node *nd;
 	char tmp[BITS_PER_LONG / 4];
 	u64 maxaddr = 0;
@@ -144,8 +141,6 @@
 	for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) {
 		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
 
-		if (mb.namelen < pos->namelen)
-			mb.namelen = pos->namelen;
 		if (maxaddr < pos->end)
 			maxaddr = pos->end;
 		if (verbose) {
@@ -156,6 +151,5 @@
 	}
 
 	mb.addrlen = snprintf(tmp, sizeof(tmp), "%llx", maxaddr);
-	mb.b.width += mb.addrlen * 2 + 4 + mb.namelen;
-	return map_browser__run(&mb, &es);
+	return map_browser__run(&mb);
 }
diff --git a/tools/perf/util/ui/util.c b/tools/perf/util/ui/util.c
index 04600e2..9706d9d 100644
--- a/tools/perf/util/ui/util.c
+++ b/tools/perf/util/ui/util.c
@@ -11,8 +11,6 @@
 #include "helpline.h"
 #include "util.h"
 
-newtComponent newt_form__new(void);
-
 static void newt_form__set_exit_keys(newtComponent self)
 {
 	newtFormAddHotKey(self, NEWT_KEY_LEFT);
@@ -22,7 +20,7 @@
 	newtFormAddHotKey(self, CTRL('c'));
 }
 
-newtComponent newt_form__new(void)
+static newtComponent newt_form__new(void)
 {
 	newtComponent self = newtForm(NULL, NULL, 0);
 	if (self)
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index f380fed..7562707 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -266,19 +266,6 @@
 bool strlazymatch(const char *str, const char *pat);
 unsigned long convert_unit(unsigned long value, char *unit);
 
-#ifndef ESC
-#define ESC 27
-#endif
-
-static inline bool is_exit_key(int key)
-{
-	char up;
-	if (key == CTRL('c') || key == ESC)
-		return true;
-	up = toupper(key);
-	return up == 'Q';
-}
-
 #define _STR(x) #x
 #define STR(x) _STR(x)