Re: [RFC v2 6/7] perf: add api for pmus to write to AUX space

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 6/7] perf: add api for pmus to write to AUX space
Date: Wed, 25 Jun 2014 14:24:51 +0300	[thread overview]
Message-ID: <87fvitxmd8.fsf@ashishki-desk.ger.corp.intel.com> (raw)
In-Reply-To: <20140624172723.GS3213@twins.programming.kicks-ass.net>

Peter Zijlstra <peterz@infradead.org> writes:

> On Wed, Jun 11, 2014 at 06:41:49PM +0300, Alexander Shishkin wrote:
>> +void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size,
>> +			 bool truncated)
>> +{
>> +	struct ring_buffer *rb = handle->rb;
>> +	unsigned long aux_head;
>> +
>> +	aux_head = local_read(&rb->aux_head);
>> +	local_add(size, &rb->aux_head);
>> +
>> +	rb->user_page->aux_head = local_read(&rb->aux_head);
>> +	smp_wmb();
>> +
>> +	perf_output_wakeup(handle);
>> +	handle->event = NULL;
>> +
>> +	ring_buffer_put(rb);
>> +}
>
> This thing is distinctly less complex than perf_output_put_handle().. ?

This one doesn't support nested writers, but there shouldn't be any
either. But I notice now that the barrier is misplaced. Wakeup watermark
is not yet taken care of, but will be. Anything else I'm missing?

Regards,
--
Alex