Re: [PATCH] perf_event: fix slow and broken cgroup context switch code

From: Peter Zijlstra <peterz@infradead.org>
To: Stephane Eranian <eranian@google.com>
Cc: linux-kernel@vger.kernel.org, mingo@elte.hu,
	ming.m.lin@intel.com, Oleg Nesterov <oleg@redhat.com>,
	Jiri Olsa <jolsa@redhat.com>
Subject: Re: [PATCH] perf_event: fix slow and broken cgroup context switch code
Date: Thu, 25 Aug 2011 16:20:20 +0200	[thread overview]
Message-ID: <1314282021.27911.12.camel@twins> (raw)
In-Reply-To: <20110825135803.GA4697@quad>

On Thu, 2011-08-25 at 15:58 +0200, Stephane Eranian wrote:
> +static inline void perf_event_task_sched_out(struct task_struct
> *prev,
> +                                            struct task_struct *next)
>  {
>         perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, NULL, 0);
>  
> -       __perf_event_task_sched_out(task, next);
> +       if (static_branch(&perf_sched_events))
> +               __perf_event_task_sched_out(prev, next);
>  } 

Right, so the reason we removed the static branch from there is

  lkml.kernel.org/r/20110324164436.GC1930@jolsa.brq.redhat.com

now I think the series 075e0b0085 to 64ce312618e should have cured that
problem, and adding the static_branch() is now safe again. But there's
no mention of any of this in the Changelog.

Also, the adding back of the static_branch() is mostly unrelated to the
rest of the patch, which I shall now stare at :-)