perfcounters: pull inherited counters
Change counter inheritance from a 'push' to a 'pull' model: instead of
child tasks pushing their final counts to the parent, reuse the wait4
infrastructure to pull counters as child tasks are exit-processed,
much like how cutime/cstime is collected.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/kernel/exit.c b/kernel/exit.c
index 244edfd..101b7ee 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -153,6 +153,9 @@
{
struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);
+#ifdef CONFIG_PERF_COUNTERS
+ WARN_ON_ONCE(!list_empty(&tsk->perf_counter_ctx.counter_list));
+#endif
trace_sched_process_free(tsk);
put_task_struct(tsk);
}
@@ -922,12 +925,6 @@
forget_original_parent(tsk);
exit_task_namespaces(tsk);
- /*
- * Flush inherited counters to the parent - before the parent
- * gets woken up by child-exit notifications.
- */
- perf_counter_exit_task(tsk);
-
write_lock_irq(&tasklist_lock);
if (group_dead)
kill_orphaned_pgrp(tsk->group_leader, NULL);
@@ -1122,12 +1119,6 @@
if (tsk->splice_pipe)
__free_pipe_info(tsk->splice_pipe);
- /*
- * These must happen late, after the PID is not
- * hashed anymore, but still at a point that may sleep:
- */
- perf_counter_exit_task(tsk);
-
preempt_disable();
/* causes final put_task_struct in finish_task_switch(). */
tsk->state = TASK_DEAD;
@@ -1371,6 +1362,12 @@
*/
read_unlock(&tasklist_lock);
+ /*
+ * Flush inherited counters to the parent - before the parent
+ * gets woken up by child-exit notifications.
+ */
+ perf_counter_exit_task(p);
+
retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0;
status = (p->signal->flags & SIGNAL_GROUP_EXIT)
? p->signal->group_exit_code : p->exit_code;
