Re: [PATCH 2/7] perf: Generalize task_function_call()ers

From: Alexander Shishkin <alexander.shishkin@linux.intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>,
	linux-kernel@vger.kernel.org, vince@deater.net,
	eranian@google.com, johannes@sipsolutions.net,
	Arnaldo Carvalho de Melo <acme@infradead.org>
Subject: Re: [PATCH 2/7] perf: Generalize task_function_call()ers
Date: Fri, 18 Dec 2015 19:41:54 +0200	[thread overview]
Message-ID: <8737uzy7q5.fsf@ashishki-desk.ger.corp.intel.com> (raw)
In-Reply-To: <20151218164704.GQ6344@twins.programming.kicks-ass.net>

Peter Zijlstra <peterz@infradead.org> writes:

> On Fri, Dec 18, 2015 at 05:07:34PM +0200, Alexander Shishkin wrote:
>> Peter Zijlstra <peterz@infradead.org> writes:
>> 
>> > @@ -2774,9 +2696,6 @@ static void perf_event_context_sched_in(
>> >  	 */
>> >  	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
>> >  
>> > -	if (ctx->nr_events)
>> > -		cpuctx->task_ctx = ctx;
>> > -
>> >  	perf_event_sched_in(cpuctx, cpuctx->task_ctx, task);
>> 
>> This then should probably become
>> 
>>   	perf_event_sched_in(cpuctx, ctx, task);
>> 
>> otherwise task contexts just don't get scheduled any more.
>
> Very good, thanks!
>
> Running with that triggers another WARN, I'll have a look at that later
> tonight.

Hmm, is it the one from event_function() where a task context is
scheduled, but is inactive and/or doesn't have events?

I think there's also more fun to be had there now that you've started
digging it up, I'll also poke at it more later today.

Regards,
--
Alex