From mboxrd@z Thu Jan 1 00:00:00 1970 From: Namhyung Kim Subject: Re: Perf event for Wall-time based sampling? Date: Sat, 20 Sep 2014 00:50:57 +0900 Message-ID: <1411141857.1680.23.camel@leonhard> References: <2221771.b2oSN5LR6X@milian-kdab2> <874mw4m9pg.fsf@sejong.aot.lge.com> <20140919143340.GB32694@kernel.org> <38149799.gZAnD74qWP@milian-kdab2> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-pd0-f175.google.com ([209.85.192.175]:51817 "EHLO mail-pd0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755682AbaISPvD (ORCPT ); Fri, 19 Sep 2014 11:51:03 -0400 Received: by mail-pd0-f175.google.com with SMTP id z10so183184pdj.6 for ; Fri, 19 Sep 2014 08:51:03 -0700 (PDT) In-Reply-To: <38149799.gZAnD74qWP@milian-kdab2> Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Milian Wolff Cc: Arnaldo Carvalho de Melo , linux-perf-users , Ingo Molnar , Joseph Schuchart 2014-09-19 (=EA=B8=88), 16:53 +0200, Milian Wolff: > On Friday 19 September 2014 11:33:40 Arnaldo Carvalho de Melo wrote: > > Em Fri, Sep 19, 2014 at 02:59:55PM +0900, Namhyung Kim escreveu: > > > Hi Arnaldo and Millan, > > >=20 > > > On Thu, 18 Sep 2014 16:17:13 -0300, Arnaldo Carvalho de Melo wrot= e: > > > > Em Thu, Sep 18, 2014 at 06:37:47PM +0200, Milian Wolff escreveu= : > > > >> That would indeed be very welcome. There are multiple "default= s" in > > > >> perf which I find highly confusing. The --no-children above e.= g. > > > >> could/should probably be the default, no? Similar, I find it e= xtremely > > > >> irritating that `perf report -g`> >=20 > > > > It was, this is something we've actually been discussing recent= ly: the > > > > change that made --children be the default mode. That is why I = added > > > > Namhyung and Ingo to the CC list, so that they become aware of = more > > > > reaction to this change. > > >=20 > > > Yeah, we should rethink about changing the default now. Actually= I'm > > > okay with the change, Ingo what do you think? > > >=20 > > > >> defaults to `-g fractal` and not `-g graph`. > > > >>=20 > > > >> 100% foo > > > >>=20 > > > >> 70% bar > > > >> =20 > > > >> 70% asdf > > > >> 30% lalala > > > >> =20 > > > >> 30% baz > > > >>=20 > > > >> is much harder to interpret than > > > >>=20 > > > >> 100% foo > > > >>=20 > > > >> 70% bar > > > >> =20 > > > >> 49% asdf > > > >> 21% lalala > > > >> =20 > > > >> 30% baz > > >=20 > > > I also agree with you. :) > > >=20 > > > > But the question then is if this is configurable, if not that w= ould be a > > > > first step, i.e. making this possible via some ~/.perfconfig ch= ange. > > >=20 > > > Yes, we have record.call-graph and top.call-graph config options = now so > > > adding a new report.call-graph option should not be difficult. H= owever > > > I think it'd be better being call-graph.XXX as it can be applied = to all > >=20 > > No problem, with sourcing being supported in ~/.perfconfig, we can = have > > as many #include call-graph.XXX as needed, multiple levels of inclu= des > > and all. > >=20 > > > other subcommands transparently. > > >=20 > > > What about like below? > > >=20 > > > [call-graph] > > >=20 > > > mode =3D dwarf > > > dump-size =3D 8192 > > > print-type =3D fractal > > > order =3D callee > > > threshold =3D 0.5 > > > print-limit =3D 128 > > > sort-key =3D function > >=20 > > Milian, does this provide what you expect? How would we call this > > specific call-graph profile? >=20 > print-type should be graph, not fractal. Otherwise it sounds good to = me. But=20 > how would one use it? I tried putting it into ~/.perfconfig, but appa= rently my=20 > 3.16.2-1-ARCH Perf does not support this feature yet? How/when would = that=20 > config be used? As soon as one does "perf record -g" (for mode and du= mp-size)=20 > or "perf report" (for a perf.data with call graphs)? That would be ve= ry=20 > useful! This is a proposal for new config items which can be included into the ~/.perfconfig file. You can change it as you want later and it should apply to default behavior of perf record -g and perf report. Thanks, Namhyung=20