perf events: Allow per-task-per-cpu counters (f4c4176f) · Commits · 戴 / test

In order to allow for per-task-per-cpu counters, useful for scalability when profiling task hierarchies, we allow installing events with event->cpu != -1 in task contexts. __perf_event_sched_in() already skips events where ->cpu mis-matches the current cpu, fix up __perf_install_in_context() and __perf_event_enable() to also respect this filter. This does lead to vary hard to interpret enabled/running times for such counters, but I don't see a simple solution for that. Signed-off-by:

Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: fweisbec@gmail.com Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <20091216165904.831451147@chello.nl> Signed-off-by:

Ingo Molnar <mingo@elte.hu>

kernel/perf_event.c

+7 −4

Original line number	Diff line number	Diff line
		@@ -782,6 +782,9 @@ static void __perf_install_in_context(void *info)

		add_event_to_ctx(event, ctx);

		if (event->cpu != -1 && event->cpu != smp_processor_id())
		goto unlock;

		/*
		* Don't put the event on if it is disabled or if
		* it is in a group and the group isn't on.
		@@ -925,6 +928,9 @@ static void __perf_event_enable(void *info)
		goto unlock;
		__perf_event_mark_enabled(event, ctx);

		if (event->cpu != -1 && event->cpu != smp_processor_id())
		goto unlock;

		/*
		* If the event is in a group and isn't the group leader,
		* then don't put it on unless the group is on.
		@@ -1595,10 +1601,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu)
		unsigned long flags;
		int err;

		/*
		* If cpu is not a wildcard then this is a percpu event:
		*/
		if (cpu != -1) {
		if (pid == -1 && cpu != -1) {
		/* Must be root to operate on a CPU event: */
		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
		return ERR_PTR(-EACCES);

Admin message