Re: [PATCH 06/14] perf, core: always switch pmu specific data duringcontext switch
From: Stephane Eranian
Date: Wed Feb 05 2014 - 13:35:52 EST
On Wed, Feb 5, 2014 at 6:55 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> On Wed, Feb 05, 2014 at 06:19:27PM +0100, Stephane Eranian wrote:
>> On Fri, Jan 3, 2014 at 6:48 AM, Yan, Zheng <zheng.z.yan@xxxxxxxxx> wrote:
>> > If two tasks were both forked from the same parent task, Events in their perf
>> > task contexts can be the same. Perf core optimizes context switch oout in this
>> > case.
>> >
>> > Previous patch inroduces pmu specific data. The data is task specific, so we
>> > should switch the data even when context switch is optimized out.
>> >
>> Reviwed-by: Stephane Eranian <eranian@xxxxxxxxxx>
>
> You should look again.. that xchg() is an atomic op and a total waste of
> time since the assignment back onto ctx->task_ctx_data is non-atomic.
>
> Complete fail there.
>
I admit, it was not clear to me why the xchg().
>> > Signed-off-by: Yan, Zheng <zheng.z.yan@xxxxxxxxx>
>> > ---
>> > kernel/events/core.c | 2 ++
>> > 1 file changed, 2 insertions(+)
>> >
>> > diff --git a/kernel/events/core.c b/kernel/events/core.c
>> > index b6650ab..d6d8dea 100644
>> > --- a/kernel/events/core.c
>> > +++ b/kernel/events/core.c
>> > @@ -2319,6 +2319,8 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
>> > next->perf_event_ctxp[ctxn] = ctx;
>> > ctx->task = next;
>> > next_ctx->task = task;
>> > + ctx->task_ctx_data = xchg(&next_ctx->task_ctx_data,
>> > + ctx->task_ctx_data);
>> > do_switch = 0;
>> >
>> > perf_event_sync_stat(ctx, next_ctx);
>> > --
>> > 1.8.4.2
>> >
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/