Commit 7bbba0eb authored Feb 15, 2017 by Peter Zijlstra Committed by Ingo Molnar Feb 24, 2017

perf/core: Fix perf_event_enable_on_exec() timekeeping (again)



Where commit:

  7fce2509 ("perf: Fix scaling vs.  perf_event_enable_on_exec()")

disabled the ctx-time a-priory, such that all events get enabled and
scheduled at the time point in time, there is one hole in that patch,
when no events do get enabled nothing re-enables the ctx-time.

Reported-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: 7fce2509 ("perf: Fix scaling vs.  perf_event_enable_on_exec()")
Signed-off-by: Ingo Molnar <mingo@kernel.org>

parent 279b5165

kernel/events/core.c

+2 −0

Original line number	Diff line number	Diff line
		@@ -3522,6 +3522,8 @@ static void perf_event_enable_on_exec(int ctxn)
		if (enabled) {
		clone_ctx = unclone_ctx(ctx);
		ctx_resched(cpuctx, ctx, event_type);
		} else {
		ctx_sched_in(ctx, cpuctx, EVENT_TIME, current);
		}
		perf_ctx_unlock(cpuctx, ctx);

Admin message