]> www.infradead.org Git - users/jedix/linux-maple.git/commitdiff
perf: Cure event->pending_disable race
authorPeter Zijlstra <peterz@infradead.org>
Thu, 24 Mar 2016 11:14:52 +0000 (11:14 +0000)
committerChuck Anderson <chuck.anderson@oracle.com>
Thu, 26 May 2016 22:45:46 +0000 (15:45 -0700)
Orabug: 23331014

[ Upstream commit 28a967c3a2f99fa3b5f762f25cb2a319d933571b ]

Because event_sched_out() checks event->pending_disable _before_
actually disabling the event, it can happen that the event fires after
it checks but before it gets disabled.

This would leave event->pending_disable set and the queued irq_work
will try and process it.

However, if the event trigger was during schedule(), the event might
have been de-scheduled by the time the irq_work runs, and
perf_event_disable_local() will fail.

Fix this by checking event->pending_disable _after_ we call
event->pmu->del(). This depends on the latter being a compiler
barrier, such that the compiler does not lift the load and re-creates
the problem.

Tested-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: 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: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dvyukov@google.com
Cc: eranian@google.com
Cc: oleg@redhat.com
Cc: panand@redhat.com
Cc: sasha.levin@oracle.com
Cc: vince@deater.net
Link: http://lkml.kernel.org/r/20160224174948.040469884@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: He Kuang <hekuang@huawei.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
(cherry picked from commit 882f862db7f3509f055208b2e3e5bd263265a03b)

Signed-off-by: Dan Duval <dan.duval@oracle.com>
kernel/events/core.c

index 992b16aac559a661f912154a4c730905075db659..0eaf3f0b0974a70908fc86f6680b00c3491108ac 100644 (file)
@@ -1562,14 +1562,14 @@ event_sched_out(struct perf_event *event,
 
        perf_pmu_disable(event->pmu);
 
+       event->tstamp_stopped = tstamp;
+       event->pmu->del(event, 0);
+       event->oncpu = -1;
        event->state = PERF_EVENT_STATE_INACTIVE;
        if (event->pending_disable) {
                event->pending_disable = 0;
                event->state = PERF_EVENT_STATE_OFF;
        }
-       event->tstamp_stopped = tstamp;
-       event->pmu->del(event, 0);
-       event->oncpu = -1;
 
        if (!is_software_event(event))
                cpuctx->active_oncpu--;