perf: Fix the software context switch counter

Stephane noticed that because the perf_sw_event() call is inside the
perf_event_task_sched_out() call it won't get called unless we
have a per-task counter.

Reported-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 142e3d6042c721dc1176ba6d8f484bf7d09617c0..de2c41758e29e20d6a709303f21a6d8242696fa0 100644 (file)
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -909,20 +909,6 @@ 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 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);
@@ -1031,6 +1017,21 @@ have_event:
        __perf_sw_event(event_id, nr, nmi, regs, addr);
 }
 
+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)
+{
+       perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+
+       COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
+}
+
 extern void perf_event_mmap(struct vm_area_struct *vma);
 extern struct perf_guest_info_callbacks *perf_guest_cbs;
 extern int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks *callbacks);

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index f365dd8ef8b02a9b05fb9da95e664b03bc754d6f..eac7e3364335a7a3f94d902e69b9ed4eac4df74a 100644 (file)
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1287,8 +1287,6 @@ void __perf_event_task_sched_out(struct task_struct *task,
 {
        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);
 }