Add a sched_clock paravirt_op
The tsc-based get_scheduled_cycles interface is not a good match for
Xen's runstate accounting, which reports everything in nanoseconds.
This patch replaces this interface with a sched_clock interface, which
matches both Xen and VMI's requirements.
In order to do this, we:
1. replace get_scheduled_cycles with sched_clock
2. hoist cycles_2_ns into a common header
3. update vmi accordingly
One thing to note: because sched_clock is implemented as a weak
function in kernel/sched.c, we must define a real function in order to
override this weak binding. This means the usual paravirt_ops
technique of using an inline function won't work in this case.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Dan Hecht <dhecht@vmware.com>
Cc: john stultz <johnstul@us.ibm.com>
diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h
index 690ada2..7df88be 100644
--- a/include/asm-i386/paravirt.h
+++ b/include/asm-i386/paravirt.h
@@ -118,7 +118,7 @@
u64 (*read_tsc)(void);
u64 (*read_pmc)(void);
- u64 (*get_scheduled_cycles)(void);
+ unsigned long long (*sched_clock)(void);
unsigned long (*get_cpu_khz)(void);
/* Segment descriptor handling */
@@ -566,7 +566,10 @@
#define rdtscll(val) (val = paravirt_read_tsc())
-#define get_scheduled_cycles(val) (val = paravirt_ops.get_scheduled_cycles())
+static inline unsigned long long paravirt_sched_clock(void)
+{
+ return PVOP_CALL0(unsigned long long, sched_clock);
+}
#define calculate_cpu_khz() (paravirt_ops.get_cpu_khz())
#define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
diff --git a/include/asm-i386/timer.h b/include/asm-i386/timer.h
index 153770e..51a713e 100644
--- a/include/asm-i386/timer.h
+++ b/include/asm-i386/timer.h
@@ -15,8 +15,38 @@
extern int recalibrate_cpu_khz(void);
#ifndef CONFIG_PARAVIRT
-#define get_scheduled_cycles(val) rdtscll(val)
#define calculate_cpu_khz() native_calculate_cpu_khz()
#endif
+/* Accellerators for sched_clock()
+ * convert from cycles(64bits) => nanoseconds (64bits)
+ * basic equation:
+ * ns = cycles / (freq / ns_per_sec)
+ * ns = cycles * (ns_per_sec / freq)
+ * ns = cycles * (10^9 / (cpu_khz * 10^3))
+ * ns = cycles * (10^6 / cpu_khz)
+ *
+ * Then we use scaling math (suggested by george@mvista.com) to get:
+ * ns = cycles * (10^6 * SC / cpu_khz) / SC
+ * ns = cycles * cyc2ns_scale / SC
+ *
+ * And since SC is a constant power of two, we can convert the div
+ * into a shift.
+ *
+ * We can use khz divisor instead of mhz to keep a better percision, since
+ * cyc2ns_scale is limited to 10^6 * 2^10, which fits in 32 bits.
+ * (mathieu.desnoyers@polymtl.ca)
+ *
+ * -johnstul@us.ibm.com "math is hard, lets go shopping!"
+ */
+extern unsigned long cyc2ns_scale __read_mostly;
+
+#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */
+
+static inline unsigned long long cycles_2_ns(unsigned long long cyc)
+{
+ return (cyc * cyc2ns_scale) >> CYC2NS_SCALE_FACTOR;
+}
+
+
#endif
diff --git a/include/asm-i386/vmi_time.h b/include/asm-i386/vmi_time.h
index 213930b..4781881 100644
--- a/include/asm-i386/vmi_time.h
+++ b/include/asm-i386/vmi_time.h
@@ -49,7 +49,7 @@
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_get_sched_cycles(void);
+extern unsigned long long vmi_sched_clock(void);
extern unsigned long vmi_cpu_khz(void);
#ifdef CONFIG_X86_LOCAL_APIC
