Merge branch 'for-ingo' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-sfi-2.6 into x86/apic
Merge reason: the SFI (Simple Firmware Interface) feature in the ACPI
tree needs this cleanup, pull it into the APIC branch as
well so that there's no interactions.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index bb7d479..586b7ad 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -183,6 +183,10 @@
}
#define x2apic_supported() (cpu_has_x2apic)
+static inline void x2apic_force_phys(void)
+{
+ x2apic_phys = 1;
+}
#else
static inline void check_x2apic(void)
{
@@ -194,6 +198,9 @@
{
return 0;
}
+static inline void x2apic_force_phys(void)
+{
+}
#define x2apic_preenabled 0
#define x2apic_supported() 0
diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h
index 7ddb36a..7386bfa 100644
--- a/arch/x86/include/asm/apicdef.h
+++ b/arch/x86/include/asm/apicdef.h
@@ -8,7 +8,8 @@
* Ingo Molnar <mingo@redhat.com>, 1999, 2000
*/
-#define APIC_DEFAULT_PHYS_BASE 0xfee00000
+#define IO_APIC_DEFAULT_PHYS_BASE 0xfec00000
+#define APIC_DEFAULT_PHYS_BASE 0xfee00000
#define APIC_ID 0x20
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 7e62d1e..67e929b 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1079,9 +1079,8 @@
* If MPS is present, it will handle them,
* otherwise the system will stay in PIC mode
*/
- if (acpi_disabled || acpi_noirq) {
+ if (acpi_disabled || acpi_noirq)
return -ENODEV;
- }
if (!cpu_has_apic)
return -ENODEV;
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 0a1c283..159740d 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -49,6 +49,7 @@
#include <asm/mtrr.h>
#include <asm/smp.h>
#include <asm/mce.h>
+#include <asm/kvm_para.h>
unsigned int num_processors;
@@ -1361,52 +1362,80 @@
}
#endif /* CONFIG_X86_X2APIC */
-void __init enable_IR_x2apic(void)
+int __init enable_IR(void)
{
#ifdef CONFIG_INTR_REMAP
- int ret;
- unsigned long flags;
- struct IO_APIC_route_entry **ioapic_entries = NULL;
-
- ret = dmar_table_init();
- if (ret) {
- pr_debug("dmar_table_init() failed with %d:\n", ret);
- goto ir_failed;
- }
-
if (!intr_remapping_supported()) {
pr_debug("intr-remapping not supported\n");
- goto ir_failed;
+ return 0;
}
-
if (!x2apic_preenabled && skip_ioapic_setup) {
pr_info("Skipped enabling intr-remap because of skipping "
"io-apic setup\n");
- return;
+ return 0;
}
+ if (enable_intr_remapping(x2apic_supported()))
+ return 0;
+
+ pr_info("Enabled Interrupt-remapping\n");
+
+ return 1;
+
+#endif
+ return 0;
+}
+
+void __init enable_IR_x2apic(void)
+{
+ unsigned long flags;
+ struct IO_APIC_route_entry **ioapic_entries = NULL;
+ int ret, x2apic_enabled = 0;
+ int dmar_table_init_ret = 0;
+
+#ifdef CONFIG_INTR_REMAP
+ dmar_table_init_ret = dmar_table_init();
+ if (dmar_table_init_ret)
+ pr_debug("dmar_table_init() failed with %d:\n",
+ dmar_table_init_ret);
+#endif
+
ioapic_entries = alloc_ioapic_entries();
if (!ioapic_entries) {
- pr_info("Allocate ioapic_entries failed: %d\n", ret);
- goto end;
+ pr_err("Allocate ioapic_entries failed\n");
+ goto out;
}
ret = save_IO_APIC_setup(ioapic_entries);
if (ret) {
pr_info("Saving IO-APIC state failed: %d\n", ret);
- goto end;
+ goto out;
}
local_irq_save(flags);
- mask_IO_APIC_setup(ioapic_entries);
mask_8259A();
+ mask_IO_APIC_setup(ioapic_entries);
- ret = enable_intr_remapping(x2apic_supported());
- if (ret)
- goto end_restore;
+ if (dmar_table_init_ret)
+ ret = 0;
+ else
+ ret = enable_IR();
- pr_info("Enabled Interrupt-remapping\n");
+ if (!ret) {
+ /* IR is required if there is APIC ID > 255 even when running
+ * under KVM
+ */
+ if (max_physical_apicid > 255 || !kvm_para_available())
+ goto nox2apic;
+ /*
+ * without IR all CPUs can be addressed by IOAPIC/MSI
+ * only in physical mode
+ */
+ x2apic_force_phys();
+ }
+
+ x2apic_enabled = 1;
if (x2apic_supported() && !x2apic_mode) {
x2apic_mode = 1;
@@ -1414,41 +1443,25 @@
pr_info("Enabled x2apic\n");
}
-end_restore:
- if (ret)
- /*
- * IR enabling failed
- */
+nox2apic:
+ if (!ret) /* IR enabling failed */
restore_IO_APIC_setup(ioapic_entries);
-
unmask_8259A();
local_irq_restore(flags);
-end:
+out:
if (ioapic_entries)
free_ioapic_entries(ioapic_entries);
- if (!ret)
+ if (x2apic_enabled)
return;
-ir_failed:
if (x2apic_preenabled)
- panic("x2apic enabled by bios. But IR enabling failed");
+ panic("x2apic: enabled by BIOS but kernel init failed.");
else if (cpu_has_x2apic)
- pr_info("Not enabling x2apic,Intr-remapping\n");
-#else
- if (!cpu_has_x2apic)
- return;
-
- if (x2apic_preenabled)
- panic("x2apic enabled prior OS handover,"
- " enable CONFIG_X86_X2APIC, CONFIG_INTR_REMAP");
-#endif
-
- return;
+ pr_info("Not enabling x2apic, Intr-remapping init failed.\n");
}
-
#ifdef CONFIG_X86_64
/*
* Detect and enable local APICs on non-SMP boards.
@@ -1549,8 +1562,6 @@
#ifdef CONFIG_X86_64
void __init early_init_lapic_mapping(void)
{
- unsigned long phys_addr;
-
/*
* If no local APIC can be found then go out
* : it means there is no mpatable and MADT
@@ -1558,11 +1569,9 @@
if (!smp_found_config)
return;
- phys_addr = mp_lapic_addr;
-
- set_fixmap_nocache(FIX_APIC_BASE, phys_addr);
+ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n",
- APIC_BASE, phys_addr);
+ APIC_BASE, mp_lapic_addr);
/*
* Fetch the APIC ID of the BSP in case we have a
@@ -1651,7 +1660,6 @@
APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) {
pr_err("BIOS bug, local APIC 0x%x not detected!...\n",
boot_cpu_physical_apicid);
- clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
return -1;
}
#endif
diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index 8952a58..89174f8 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -167,7 +167,7 @@
{
/* MPENTIUMIII */
if (boot_cpu_data.x86 == 6 &&
- (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11))
+ (boot_cpu_data.x86_model >= 7 && boot_cpu_data.x86_model <= 11))
return 1;
return 0;
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index a8c0232..3c8f9e7 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -66,6 +66,8 @@
#include <asm/apic.h>
#define __apicdebuginit(type) static type __init
+#define for_each_irq_pin(entry, head) \
+ for (entry = head; entry; entry = entry->next)
/*
* Is the SiS APIC rmw bug present ?
@@ -119,15 +121,6 @@
}
early_param("noapic", parse_noapic);
-struct irq_pin_list;
-
-/*
- * This is performance-critical, we want to do it O(1)
- *
- * the indexing order of this array favors 1:1 mappings
- * between pins and IRQs.
- */
-
struct irq_pin_list {
int apic, pin;
struct irq_pin_list *next;
@@ -142,6 +135,11 @@
return pin;
}
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * Most irqs are mapped 1:1 with pins.
+ */
struct irq_cfg {
struct irq_pin_list *irq_2_pin;
cpumask_var_t domain;
@@ -417,13 +415,10 @@
unsigned long flags;
spin_lock_irqsave(&ioapic_lock, flags);
- entry = cfg->irq_2_pin;
- for (;;) {
+ for_each_irq_pin(entry, cfg->irq_2_pin) {
unsigned int reg;
int pin;
- if (!entry)
- break;
pin = entry->pin;
reg = io_apic_read(entry->apic, 0x10 + pin*2);
/* Is the remote IRR bit set? */
@@ -431,9 +426,6 @@
spin_unlock_irqrestore(&ioapic_lock, flags);
return true;
}
- if (!entry->next)
- break;
- entry = entry->next;
}
spin_unlock_irqrestore(&ioapic_lock, flags);
@@ -501,72 +493,68 @@
* shared ISA-space IRQs, so we have to support them. We are super
* fast in the common case, and fast for shared ISA-space IRQs.
*/
-static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
+static int
+add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin)
{
- struct irq_pin_list *entry;
+ struct irq_pin_list **last, *entry;
- entry = cfg->irq_2_pin;
- if (!entry) {
- entry = get_one_free_irq_2_pin(node);
- if (!entry) {
- printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n",
- apic, pin);
- return;
- }
- cfg->irq_2_pin = entry;
- entry->apic = apic;
- entry->pin = pin;
- return;
- }
-
- while (entry->next) {
- /* not again, please */
+ /* don't allow duplicates */
+ last = &cfg->irq_2_pin;
+ for_each_irq_pin(entry, cfg->irq_2_pin) {
if (entry->apic == apic && entry->pin == pin)
- return;
-
- entry = entry->next;
+ return 0;
+ last = &entry->next;
}
- entry->next = get_one_free_irq_2_pin(node);
- entry = entry->next;
+ entry = get_one_free_irq_2_pin(node);
+ if (!entry) {
+ printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n",
+ node, apic, pin);
+ return -ENOMEM;
+ }
entry->apic = apic;
entry->pin = pin;
+
+ *last = entry;
+ return 0;
+}
+
+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))
+ panic("IO-APIC: failed to add irq-pin. Can not proceed\n");
}
/*
* Reroute an IRQ to a different pin.
*/
static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
- int oldapic, int oldpin,
- int newapic, int newpin)
+ int oldapic, int oldpin,
+ int newapic, int newpin)
{
- struct irq_pin_list *entry = cfg->irq_2_pin;
- int replaced = 0;
+ struct irq_pin_list *entry;
- while (entry) {
+ for_each_irq_pin(entry, cfg->irq_2_pin) {
if (entry->apic == oldapic && entry->pin == oldpin) {
entry->apic = newapic;
entry->pin = newpin;
- replaced = 1;
/* every one is different, right? */
- break;
+ return;
}
- entry = entry->next;
}
- /* why? call replace before add? */
- if (!replaced)
- add_pin_to_irq_node(cfg, node, newapic, newpin);
+ /* old apic/pin didn't exist, so just add new ones */
+ add_pin_to_irq_node(cfg, node, newapic, newpin);
}
-static inline void io_apic_modify_irq(struct irq_cfg *cfg,
- int mask_and, int mask_or,
- void (*final)(struct irq_pin_list *entry))
+static void io_apic_modify_irq(struct irq_cfg *cfg,
+ int mask_and, int mask_or,
+ void (*final)(struct irq_pin_list *entry))
{
int pin;
struct irq_pin_list *entry;
- for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
+ for_each_irq_pin(entry, cfg->irq_2_pin) {
unsigned int reg;
pin = entry->pin;
reg = io_apic_read(entry->apic, 0x10 + pin * 2);
@@ -583,7 +571,6 @@
io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
}
-#ifdef CONFIG_X86_64
static void io_apic_sync(struct irq_pin_list *entry)
{
/*
@@ -599,11 +586,6 @@
{
io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
}
-#else /* CONFIG_X86_32 */
-static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
-{
- io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL);
-}
static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg)
{
@@ -616,7 +598,6 @@
io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED,
IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
}
-#endif /* CONFIG_X86_32 */
static void mask_IO_APIC_irq_desc(struct irq_desc *desc)
{
@@ -1705,12 +1686,8 @@
if (!entry)
continue;
printk(KERN_DEBUG "IRQ%d ", irq);
- for (;;) {
+ for_each_irq_pin(entry, cfg->irq_2_pin)
printk("-> %d:%d", entry->apic, entry->pin);
- if (!entry->next)
- break;
- entry = entry->next;
- }
printk("\n");
}
@@ -2214,7 +2191,6 @@
return was_pending;
}
-#ifdef CONFIG_X86_64
static int ioapic_retrigger_irq(unsigned int irq)
{
@@ -2227,14 +2203,6 @@
return 1;
}
-#else
-static int ioapic_retrigger_irq(unsigned int irq)
-{
- apic->send_IPI_self(irq_cfg(irq)->vector);
-
- return 1;
-}
-#endif
/*
* Level and edge triggered IO-APIC interrupts need different handling,
@@ -2272,13 +2240,9 @@
struct irq_pin_list *entry;
u8 vector = cfg->vector;
- entry = cfg->irq_2_pin;
- for (;;) {
+ for_each_irq_pin(entry, cfg->irq_2_pin) {
unsigned int reg;
- if (!entry)
- break;
-
apic = entry->apic;
pin = entry->pin;
/*
@@ -2291,9 +2255,6 @@
reg &= ~IO_APIC_REDIR_VECTOR_MASK;
reg |= vector;
io_apic_modify(apic, 0x10 + pin*2, reg);
- if (!entry->next)
- break;
- entry = entry->next;
}
}
@@ -2518,11 +2479,8 @@
static void ack_apic_level(unsigned int irq)
{
struct irq_desc *desc = irq_to_desc(irq);
-
-#ifdef CONFIG_X86_32
unsigned long v;
int i;
-#endif
struct irq_cfg *cfg;
int do_unmask_irq = 0;
@@ -2535,31 +2493,28 @@
}
#endif
-#ifdef CONFIG_X86_32
/*
- * It appears there is an erratum which affects at least version 0x11
- * of I/O APIC (that's the 82093AA and cores integrated into various
- * chipsets). Under certain conditions a level-triggered interrupt is
- * erroneously delivered as edge-triggered one but the respective IRR
- * bit gets set nevertheless. As a result the I/O unit expects an EOI
- * message but it will never arrive and further interrupts are blocked
- * from the source. The exact reason is so far unknown, but the
- * phenomenon was observed when two consecutive interrupt requests
- * from a given source get delivered to the same CPU and the source is
- * temporarily disabled in between.
- *
- * A workaround is to simulate an EOI message manually. We achieve it
- * by setting the trigger mode to edge and then to level when the edge
- * trigger mode gets detected in the TMR of a local APIC for a
- * level-triggered interrupt. We mask the source for the time of the
- * operation to prevent an edge-triggered interrupt escaping meanwhile.
- * The idea is from Manfred Spraul. --macro
- */
+ * It appears there is an erratum which affects at least version 0x11
+ * of I/O APIC (that's the 82093AA and cores integrated into various
+ * chipsets). Under certain conditions a level-triggered interrupt is
+ * erroneously delivered as edge-triggered one but the respective IRR
+ * bit gets set nevertheless. As a result the I/O unit expects an EOI
+ * message but it will never arrive and further interrupts are blocked
+ * from the source. The exact reason is so far unknown, but the
+ * phenomenon was observed when two consecutive interrupt requests
+ * from a given source get delivered to the same CPU and the source is
+ * temporarily disabled in between.
+ *
+ * A workaround is to simulate an EOI message manually. We achieve it
+ * by setting the trigger mode to edge and then to level when the edge
+ * trigger mode gets detected in the TMR of a local APIC for a
+ * level-triggered interrupt. We mask the source for the time of the
+ * operation to prevent an edge-triggered interrupt escaping meanwhile.
+ * The idea is from Manfred Spraul. --macro
+ */
cfg = desc->chip_data;
i = cfg->vector;
-
v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
-#endif
/*
* We must acknowledge the irq before we move it or the acknowledge will
@@ -2601,7 +2556,7 @@
unmask_IO_APIC_irq_desc(desc);
}
-#ifdef CONFIG_X86_32
+ /* Tail end of version 0x11 I/O APIC bug workaround */
if (!(v & (1 << (i & 0x1f)))) {
atomic_inc(&irq_mis_count);
spin_lock(&ioapic_lock);
@@ -2609,26 +2564,15 @@
__unmask_and_level_IO_APIC_irq(cfg);
spin_unlock(&ioapic_lock);
}
-#endif
}
#ifdef CONFIG_INTR_REMAP
static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
{
- int apic, pin;
struct irq_pin_list *entry;
- entry = cfg->irq_2_pin;
- for (;;) {
-
- if (!entry)
- break;
-
- apic = entry->apic;
- pin = entry->pin;
- io_apic_eoi(apic, pin);
- entry = entry->next;
- }
+ for_each_irq_pin(entry, cfg->irq_2_pin)
+ io_apic_eoi(entry->apic, entry->pin);
}
static void
@@ -3244,8 +3188,7 @@
cfg = desc->chip_data;
dynamic_irq_cleanup(irq);
/* connect back irq_cfg */
- if (desc)
- desc->chip_data = cfg;
+ desc->chip_data = cfg;
free_irte(irq);
spin_lock_irqsave(&vector_lock, flags);
@@ -3915,7 +3858,11 @@
*/
if (irq >= NR_IRQS_LEGACY) {
cfg = desc->chip_data;
- add_pin_to_irq_node(cfg, node, ioapic, pin);
+ if (add_pin_to_irq_node_nopanic(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);
@@ -4127,7 +4074,7 @@
static struct resource *ioapic_resources;
-static struct resource * __init ioapic_setup_resources(void)
+static struct resource * __init ioapic_setup_resources(int nr_ioapics)
{
unsigned long n;
struct resource *res;
@@ -4143,15 +4090,13 @@
mem = alloc_bootmem(n);
res = (void *)mem;
- if (mem != NULL) {
- mem += sizeof(struct resource) * nr_ioapics;
+ mem += sizeof(struct resource) * nr_ioapics;
- for (i = 0; i < nr_ioapics; i++) {
- res[i].name = mem;
- res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
- sprintf(mem, "IOAPIC %u", i);
- mem += IOAPIC_RESOURCE_NAME_SIZE;
- }
+ for (i = 0; i < nr_ioapics; i++) {
+ res[i].name = mem;
+ res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+ sprintf(mem, "IOAPIC %u", i);
+ mem += IOAPIC_RESOURCE_NAME_SIZE;
}
ioapic_resources = res;
@@ -4165,7 +4110,7 @@
struct resource *ioapic_res;
int i;
- ioapic_res = ioapic_setup_resources();
+ ioapic_res = ioapic_setup_resources(nr_ioapics);
for (i = 0; i < nr_ioapics; i++) {
if (smp_found_config) {
ioapic_phys = mp_ioapics[i].apicaddr;
@@ -4194,11 +4139,9 @@
__fix_to_virt(idx), ioapic_phys);
idx++;
- if (ioapic_res != NULL) {
- ioapic_res->start = ioapic_phys;
- ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
- ioapic_res++;
- }
+ ioapic_res->start = ioapic_phys;
+ ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
+ ioapic_res++;
}
}
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index bc3e880..f3b1037 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -50,11 +50,11 @@
void __init default_setup_apic_routing(void)
{
#ifdef CONFIG_X86_X2APIC
- if (x2apic_mode && (apic != &apic_x2apic_phys &&
+ if (x2apic_mode
#ifdef CONFIG_X86_UV
- apic != &apic_x2apic_uv_x &&
+ && apic != &apic_x2apic_uv_x
#endif
- apic != &apic_x2apic_cluster)) {
+ ) {
if (x2apic_phys)
apic = &apic_x2apic_phys;
else
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 651c93b..fcd513b 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -482,11 +482,11 @@
MP_bus_info(&bus);
}
- ioapic.type = MP_IOAPIC;
- ioapic.apicid = 2;
- ioapic.apicver = mpc_default_type > 4 ? 0x10 : 0x01;
- ioapic.flags = MPC_APIC_USABLE;
- ioapic.apicaddr = 0xFEC00000;
+ ioapic.type = MP_IOAPIC;
+ ioapic.apicid = 2;
+ ioapic.apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+ ioapic.flags = MPC_APIC_USABLE;
+ ioapic.apicaddr = IO_APIC_DEFAULT_PHYS_BASE;
MP_ioapic_info(&ioapic);
/*
