Messages in this thread Patch in this message | | | Subject | Re: [PATCH 3/5] MIPS/Perf-events: Check event state in validate_event() | From | Will Deacon <> | Date | Fri, 19 Nov 2010 09:43:26 +0000 |
| |
Hi Deng-Cheng,
On Thu, 2010-11-18 at 06:56 +0000, Deng-Cheng Zhu wrote: > Ignore events that are not for this PMU or are in off/error state. > Sorry I didn't see this before, thanks for pointing out that you had included it for MIPS.
> Signed-off-by: Deng-Cheng Zhu <dengcheng.zhu@gmail.com> > --- > arch/mips/kernel/perf_event.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/arch/mips/kernel/perf_event.c b/arch/mips/kernel/perf_event.c > index 1ee44a3..9c6442a 100644 > --- a/arch/mips/kernel/perf_event.c > +++ b/arch/mips/kernel/perf_event.c > @@ -486,7 +486,7 @@ static int validate_event(struct cpu_hw_events *cpuc, > { > struct hw_perf_event fake_hwc = event->hw; > > - if (event->pmu && event->pmu != &pmu) > + if (event->pmu != &pmu || event->state <= PERF_EVENT_STATE_OFF) > return 0; > > return mipspmu->alloc_counter(cpuc, &fake_hwc) >= 0;
So this is the opposite of what we're doing on ARM. Our approach is to ignore events that are OFF (or in the ERROR state) or that belong to a different PMU. We do this by allowing them to *pass* validation (i.e. by returning 1 above). This means that we won't unconditionally fail a mixed event group.
x86 does something similar in the collect_events function.
Will
| |