Re: [RFC v2 5/7] perf: add a pmu capability for "exclusive" events

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, Robert Richter <rric@kernel.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Mike Galbraith <efault@gmx.de>, Paul Mackerras <paulus@samba.org>,
	Stephane Eranian <eranian@google.com>,
	Andi Kleen <ak@linux.intel.com>
Subject: Re: [RFC v2 5/7] perf: add a pmu capability for "exclusive" events
Date: Wed, 25 Jun 2014 14:12:05 +0300	[thread overview]
Message-ID: <87ionpxmyi.fsf@ashishki-desk.ger.corp.intel.com> (raw)
In-Reply-To: <20140624171819.GQ3213@twins.programming.kicks-ass.net>

Peter Zijlstra <peterz@infradead.org> writes:

> On Wed, Jun 11, 2014 at 06:41:48PM +0300, Alexander Shishkin wrote:
>> +static bool exclusive_event_ok(struct perf_event *event,
>> +			      struct perf_event_context *ctx)
>> +{
>> +	struct perf_event *iter_event;
>> +
>> +	if (!(event->pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE))
>> +		return true;
>> +
>> +	list_for_each_entry(iter_event, &ctx->event_list, event_entry) {
>> +		if (exclusive_event_match(iter_event, event))
>> +			return false;
>> +	}
>
> This list iteration needs either rcu or ctx->lock or ctx->mutex, and the
> two callsites below don't have either afaict.

True, this needs to be done before perf_install_in_context() under the
same ctx->mutex. Thanks!

Regards,
--
Alex