perf, x86: P4 PMU - Don't forget to clear cpuc->active_mask on overflow
Cyrill Gorcunov [Thu, 21 Apr 2011 15:03:21 +0000 (11:03 -0400)]
It's not enough to simply disable event on overflow the
cpuc->active_mask should be cleared as well otherwise counter
may stall in "active" even in real being already disabled (which
potentially may lead to the situation that user may not use this
counter further).

Don pointed out that:

 " I also noticed this patch fixed some unknown NMIs
   on a P4 when I stressed the box".

Tested-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Acked-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Link: http://lkml.kernel.org/r/1303398203-2918-3-git-send-email-dzickus@redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

arch/x86/kernel/cpu/perf_event_p4.c

index c2520e1..d1f77e2 100644 (file)
@@ -947,7 +947,7 @@ static int p4_pmu_handle_irq(struct pt_regs *regs)
                if (!x86_perf_event_set_period(event))
                        continue;
                if (perf_event_overflow(event, 1, &data, regs))
-                       p4_pmu_disable_event(event);
+                       x86_pmu_stop(event, 0);
        }
 
        if (handled) {