Skip to content

Commit

Permalink
perf, x86: Avoid double disable on throttle vs ioctl(PERF_IOC_DISABLE)
Browse files Browse the repository at this point in the history
Calling ioctl(PERF_EVENT_IOC_DISABLE) on a thottled counter would result
in a double disable, cure this by using x86_pmu_{start,stop} for
throttle/unthrottle and teach x86_pmu_stop() to check ->active_mask.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: paulus@samba.org
Cc: eranian@google.com
Cc: robert.richter@amd.com
Cc: fweisbec@gmail.com
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
  • Loading branch information
Peter Zijlstra authored and Ingo Molnar committed Mar 10, 2010
1 parent c08053e commit 71e2d28
Show file tree
Hide file tree
Showing 2 changed files with 7 additions and 15 deletions.
20 changes: 6 additions & 14 deletions arch/x86/kernel/cpu/perf_event.c
Original file line number Diff line number Diff line change
Expand Up @@ -983,14 +983,8 @@ static int x86_pmu_start(struct perf_event *event)

static void x86_pmu_unthrottle(struct perf_event *event)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;

if (WARN_ON_ONCE(hwc->idx >= X86_PMC_IDX_MAX ||
cpuc->events[hwc->idx] != event))
return;

x86_pmu.enable(event);
int ret = x86_pmu_start(event);
WARN_ON_ONCE(ret);
}

void perf_event_print_debug(void)
Expand Down Expand Up @@ -1050,11 +1044,9 @@ static void x86_pmu_stop(struct perf_event *event)
struct hw_perf_event *hwc = &event->hw;
int idx = hwc->idx;

/*
* Must be done before we disable, otherwise the nmi handler
* could reenable again:
*/
__clear_bit(idx, cpuc->active_mask);
if (!__test_and_clear_bit(idx, cpuc->active_mask))
return;

x86_pmu.disable(event);

/*
Expand Down Expand Up @@ -1123,7 +1115,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
continue;

if (perf_event_overflow(event, 1, &data, regs))
x86_pmu.disable(event);
x86_pmu_stop(event);
}

if (handled)
Expand Down
2 changes: 1 addition & 1 deletion arch/x86/kernel/cpu/perf_event_intel.c
Original file line number Diff line number Diff line change
Expand Up @@ -774,7 +774,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs)
data.period = event->hw.last_period;

if (perf_event_overflow(event, 1, &data, regs))
intel_pmu_disable_event(event);
x86_pmu_stop(event);
}

intel_pmu_ack_status(ack);
Expand Down

0 comments on commit 71e2d28

Please sign in to comment.