xen: Implement sched_clock

Implement xen_sched_clock, which returns the number of ns the current
vcpu has been actually in an unstolen state (ie, running or blocked,
vs runnable-but-not-running, or offline) since boot.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Cc: john stultz <johnstul@us.ibm.com>

3 files changed, 28 insertions, 4 deletions

diff --git a/arch/i386/xen/enlighten.c b/arch/i386/xen/enlighten.c
index 9550ae3b1fb1..a9ba834295a2 100644
--- a/arch/i386/xen/enlighten.c
+++ b/arch/i386/xen/enlighten.c
@@ -683,7 +683,7 @@ static const struct paravirt_ops xen_paravirt_ops __initdata = {
         .set_wallclock = xen_set_wallclock,
         .get_wallclock = xen_get_wallclock,
         .get_cpu_khz = xen_cpu_khz,
-        .sched_clock = xen_clocksource_read,
+        .sched_clock = xen_sched_clock,
 
         .cpuid = xen_cpuid,
 
diff --git a/arch/i386/xen/time.c b/arch/i386/xen/time.c
index acbfd9969462..2aab44bec2a5 100644
--- a/arch/i386/xen/time.c
+++ b/arch/i386/xen/time.c
@@ -28,6 +28,8 @@
 #define TIMER_SLOP      100000
 #define NS_PER_TICK     (1000000000LL / HZ)
 
+static cycle_t xen_clocksource_read(void);
+
 /* These are perodically updated in shared_info, and then copied here. */
 struct shadow_time_info {
         u64 tsc_timestamp;     /* TSC at last update of time vals.  */
@@ -169,6 +171,29 @@ static void do_stolen_accounting(void)
         account_steal_time(idle_task(smp_processor_id()), ticks);
 }
 
+/*
+ * Xen sched_clock implementation.  Returns the number of unstolen
+ * nanoseconds, which is nanoseconds the VCPU spent in RUNNING+BLOCKED
+ * states.
+ */
+unsigned long long xen_sched_clock(void)
+{
+        struct vcpu_runstate_info state;
+        cycle_t now = xen_clocksource_read();
+        s64 offset;
+
+        get_runstate_snapshot(&state);
+
+        WARN_ON(state.state != RUNSTATE_running);
+
+        offset = now - state.state_entry_time;
+        if (offset < 0)
+                offset = 0;
+
+        return state.time[RUNSTATE_blocked] +
+                state.time[RUNSTATE_running] +
+                offset;
+}
 
 
 /* Get the CPU speed from Xen */
@@ -261,7 +286,7 @@ static u64 get_nsec_offset(struct shadow_time_info *shadow)
         return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift);
 }
 
-cycle_t xen_clocksource_read(void)
+static cycle_t xen_clocksource_read(void)
 {
         struct shadow_time_info *shadow = &get_cpu_var(shadow_time);
         cycle_t ret;
diff --git a/arch/i386/xen/xen-ops.h b/arch/i386/xen/xen-ops.h
index 54d98b52085e..7667abd390ec 100644
--- a/arch/i386/xen/xen-ops.h
+++ b/arch/i386/xen/xen-ops.h
@@ -2,7 +2,6 @@
 #define XEN_OPS_H
 
 #include <linux/init.h>
-#include <linux/clocksource.h>
 
 DECLARE_PER_CPU(struct vcpu_info *, xen_vcpu);
 DECLARE_PER_CPU(unsigned long, xen_cr3);
@@ -18,7 +17,7 @@ unsigned long xen_cpu_khz(void);
 void __init xen_time_init(void);
 unsigned long xen_get_wallclock(void);
 int xen_set_wallclock(unsigned long time);
-cycle_t xen_clocksource_read(void);
+unsigned long long xen_sched_clock(void);
 
 void xen_mark_init_mm_pinned(void);