From mboxrd@z Thu Jan  1 00:00:00 1970
From: Peter Zijlstra <peterz@infradead.org>
Subject: Re: Profiling sleep times?
Date: Sat, 15 Oct 2011 21:22:02 +0200
Message-ID: <1318706522.2664.8.camel@laptop>
References: <4E8A0F53.7020408@fb.com> <1317673027.20367.41.camel@twins>
	 <4E8A2EC5.1030901@fb.com> <1317717291.25926.13.camel@twins>
	 <4E8E2417.2000903@fb.com> <4E8FAB3A.2040801@gmail.com>
	 <4E933E65.1090207@fb.com> <20111012074113.GK18618@elte.hu>
	 <4E98A75E.3010000@fb.com>  <20111015170044.GB29782@elte.hu>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Return-path: <linux-perf-users-owner@vger.kernel.org>
Received: from merlin.infradead.org ([205.233.59.134]:49563 "EHLO
	merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1754626Ab1JOTY3 (ORCPT
	<rfc822;linux-perf-users@vger.kernel.org>);
	Sat, 15 Oct 2011 15:24:29 -0400
Received: from canuck.infradead.org ([2001:4978:20e::1])
	by merlin.infradead.org with esmtps (Exim 4.76 #1 (Red Hat Linux))
	id 1RF9qP-0004ke-0l
	for linux-perf-users@vger.kernel.org; Sat, 15 Oct 2011 19:24:29 +0000
Received: from j77219.upc-j.chello.nl ([24.132.77.219] helo=dyad.programming.kicks-ass.net)
	by canuck.infradead.org with esmtpsa (Exim 4.76 #1 (Red Hat Linux))
	id 1RF9qO-000482-OR
	for linux-perf-users@vger.kernel.org; Sat, 15 Oct 2011 19:24:28 +0000
In-Reply-To: <20111015170044.GB29782@elte.hu>
Sender: linux-perf-users-owner@vger.kernel.org
List-ID: <linux-perf-users.vger.kernel.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Arun Sharma <asharma@fb.com>, avagin@gmail.com, linux-perf-users@vger.kernel.org, acme@ghostprotocols.net, Stephane Eranian <eranian@google.com>, Frederic Weisbecker <fweisbec@gmail.com>

On Sat, 2011-10-15 at 19:00 +0200, Ingo Molnar wrote:
> * Arun Sharma <asharma@fb.com> wrote:
> 
> > On 10/12/11 12:41 AM, Ingo Molnar wrote:
> > >>Generally I'm happy with the direction of these patches. [...]
> > >
> > >Ok, will someone please send an updated series of all missing
> > >patches, with all acks included, for potential v3.2 pickup?
> > >
> > >Sleep time profiling is obviously a very hot feature ...
> > >
> > 
> > I acked two of Andrew's patches. The third one is the most 
> > important one. I haven't had a chance to review that carefully yet 
> > - although it's working well for me.
> > 
> > We still need to figure out what's the best way to munge all this 
> > data in user space to implement wall clock profiling.
> 
> Well, if 'perf report' can show sleep time ordered entries just fine 
> by default (if a proper perf record was done), which also works in 
> call graph recording mode then i'm a happy camper.
> 
> Sleep time should really just be a different notion of 'cost of the 
> function/callchain' and fit into the existing scheme, right?

The problem with andrew's patches is that it wrecks the callchain
semantics. The waittime tracepoint is in the wakeup path (and hence
generates the wakee's callchain) whereas they really want the callchain
of the woken task to show where it spend time.