From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758491AbcHaCfL (ORCPT ); Tue, 30 Aug 2016 22:35:11 -0400 Received: from mail-pf0-f196.google.com ([209.85.192.196]:34465 "EHLO mail-pf0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755532AbcHaCfI (ORCPT ); Tue, 30 Aug 2016 22:35:08 -0400 Date: Wed, 31 Aug 2016 11:35:00 +0900 From: Namhyung Kim To: Steven Rostedt Cc: LKML , Ingo Molnar , Josh Poimboeuf Subject: Re: [PATCH] ftrace: Access ret_stack->subtime only in the function profiler Message-ID: <20160831023500.GB26190@danjae.aot.lge.com> References: <20160829030518.5383-1-namhyung@kernel.org> <20160829160700.71dc249a@gandalf.local.home> <20160830013441.GA13062@sejong> <20160831021138.GA26190@danjae.aot.lge.com> <20160830221647.6a9716cc@gandalf.local.home> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20160830221647.6a9716cc@gandalf.local.home> User-Agent: Mutt/1.7.0 (2016-08-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Aug 30, 2016 at 10:16:47PM -0400, Steven Rostedt wrote: > On Wed, 31 Aug 2016 11:11:38 +0900 > Namhyung Kim wrote: > > > On Tue, Aug 30, 2016 at 10:34:41AM +0900, Namhyung Kim wrote: > > > Hi Steve, > > > > > > On Mon, Aug 29, 2016 at 04:07:00PM -0400, Steven Rostedt wrote: > > > > On Mon, 29 Aug 2016 12:05:18 +0900 > > > > Namhyung Kim wrote: > > > > > > > > > The subtime is used only for function profiler with function graph > > > > > tracer enabled. Move the definition of subtime under > > > > > CONFIG_FUNCTION_PROFILER to reduce the memory usage. Also move the > > > > > initialization of subtime into the graph entry callback. > > > > > > > > Hmm, I think documentation needs to be updated. Although it was never > > > > implemented, I believe I added the subtime to not only work with the > > > > profiler, but also with the normal tracing (to have the time of the > > > > internal functions subtracted from the upper level functions). But it > > > > appears that part was never implemented. > > > > > > > > I'm fine with the patch, or actually implementing what graph-time > > > > states in Documentation/ftrace.txt. If we take this patch, that comment > > > > needs to be made to only mention the profiler (and the option should > > > > only be shown when the profiler is enabled). > > > > > > Ah, missed the documentation part. To implement it in the normal > > > tracing, I think we need to add 'subtime' field to struct > > > ftrace_graph_ret which will increase disk size. Are you ok with this? > > > > On second thought, I think I can do it by just adding value of subtime > > to ftrace_graph_ret.calltime when graph-time is off. Then the > > calltime would not be the timestamp at function entry, but it seems > > not guaranteed due to the sleep-time anyway. Now I wonder why it > > doesn't have 'duration' in the ftrace_graph_ret instead of having > > calltime and rettime. > > > > As it hasn't worked, like forever, I'm thinking of nuking it. Nobody > seemed to have noticed. I haven't needed to use it, and apparently > nobody else has either. Why support a feature that nobody uses? > > I have used it for profiling, but not normal function graph tracing. > You can see the function times inside and do the logic post processing. > > Best bet is to just update the documentation to what the current code > does. Ok, will send v2. Thanks, Namhyung