Commit e5289d4a authored by Peter Zijlstra's avatar Peter Zijlstra Committed by Ingo Molnar

perf_counter: Simplify and fix task migration counting

The task migrations counter was causing rare and hard to decypher
memory corruptions under load. After a day of debugging and bisection
we found that the problem was introduced with:

  3f731ca6: perf_counter: Fix cpu migration counter

Turning them off fixes the crashes. Incidentally, the whole
perf_counter_task_migration() logic can be done simpler as well,
by injecting a proper sw-counter event.

This cleanup also fixed the crashes. The precise failure mode is
not completely clear yet, but we are clearly not unhappy about
having a fix ;-)
Signed-off-by: default avatarPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
parent f5970550
...@@ -682,8 +682,6 @@ static inline void perf_counter_mmap(struct vm_area_struct *vma) ...@@ -682,8 +682,6 @@ static inline void perf_counter_mmap(struct vm_area_struct *vma)
extern void perf_counter_comm(struct task_struct *tsk); extern void perf_counter_comm(struct task_struct *tsk);
extern void perf_counter_fork(struct task_struct *tsk); extern void perf_counter_fork(struct task_struct *tsk);
extern void perf_counter_task_migration(struct task_struct *task, int cpu);
extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs);
extern int sysctl_perf_counter_paranoid; extern int sysctl_perf_counter_paranoid;
...@@ -724,8 +722,6 @@ static inline void perf_counter_mmap(struct vm_area_struct *vma) { } ...@@ -724,8 +722,6 @@ static inline void perf_counter_mmap(struct vm_area_struct *vma) { }
static inline void perf_counter_comm(struct task_struct *tsk) { } static inline void perf_counter_comm(struct task_struct *tsk) { }
static inline void perf_counter_fork(struct task_struct *tsk) { } static inline void perf_counter_fork(struct task_struct *tsk) { }
static inline void perf_counter_init(void) { } static inline void perf_counter_init(void) { }
static inline void perf_counter_task_migration(struct task_struct *task,
int cpu) { }
#endif #endif
#endif /* __KERNEL__ */ #endif /* __KERNEL__ */
......
...@@ -124,7 +124,7 @@ void perf_enable(void) ...@@ -124,7 +124,7 @@ void perf_enable(void)
static void get_ctx(struct perf_counter_context *ctx) static void get_ctx(struct perf_counter_context *ctx)
{ {
atomic_inc(&ctx->refcount); WARN_ON(!atomic_inc_not_zero(&ctx->refcount));
} }
static void free_ctx(struct rcu_head *head) static void free_ctx(struct rcu_head *head)
...@@ -3467,27 +3467,6 @@ static const struct pmu perf_ops_task_clock = { ...@@ -3467,27 +3467,6 @@ static const struct pmu perf_ops_task_clock = {
.read = task_clock_perf_counter_read, .read = task_clock_perf_counter_read,
}; };
/*
* Software counter: cpu migrations
*/
void perf_counter_task_migration(struct task_struct *task, int cpu)
{
struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
struct perf_counter_context *ctx;
perf_swcounter_ctx_event(&cpuctx->ctx, PERF_TYPE_SOFTWARE,
PERF_COUNT_SW_CPU_MIGRATIONS,
1, 1, NULL, 0);
ctx = perf_pin_task_context(task);
if (ctx) {
perf_swcounter_ctx_event(ctx, PERF_TYPE_SOFTWARE,
PERF_COUNT_SW_CPU_MIGRATIONS,
1, 1, NULL, 0);
perf_unpin_context(ctx);
}
}
#ifdef CONFIG_EVENT_PROFILE #ifdef CONFIG_EVENT_PROFILE
void perf_tpcounter_event(int event_id) void perf_tpcounter_event(int event_id)
{ {
......
...@@ -1978,7 +1978,8 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) ...@@ -1978,7 +1978,8 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
if (task_hot(p, old_rq->clock, NULL)) if (task_hot(p, old_rq->clock, NULL))
schedstat_inc(p, se.nr_forced2_migrations); schedstat_inc(p, se.nr_forced2_migrations);
#endif #endif
perf_counter_task_migration(p, new_cpu); perf_swcounter_event(PERF_COUNT_SW_CPU_MIGRATIONS,
1, 1, NULL, 0);
} }
p->se.vruntime -= old_cfsrq->min_vruntime - p->se.vruntime -= old_cfsrq->min_vruntime -
new_cfsrq->min_vruntime; new_cfsrq->min_vruntime;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment