From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Subject: Re: [PATCH] Add inverted call graph report support to perf tool Date: Fri, 6 May 2011 10:54:38 +0200 Message-ID: <20110506085438.GA928@elte.hu> References: <20110307180619.GG1873@nowhere> <20110310024355.GG2533@nowhere> <20110311115657.GB1826@nowhere> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mx3.mail.elte.hu ([157.181.1.138]:54006 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754766Ab1EFIyn (ORCPT ); Fri, 6 May 2011 04:54:43 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Sam Liao Cc: Frederic Weisbecker , linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, acme@redhat.com, Peter Zijlstra * Sam Liao wrote: > > It depends if you select reverse callchain or not: > > > > $ perf report -s caller > > > > That will report main and kernel_thread as hists, and regular callee -> caller callchains. > > Hence under main hist, you'll a lot of callchain starting from random points and all > > ending in main! > > > > $ perf report -s caller -g caller > > > > That will report main and kernel_thread as hists, with callchains starting from > > main under main. > > > > It becomes interesting when you want more granularity with -s caller,dso if we bring a way > > to push forward the entrypoint one day. I suspect even more sorting combinations are > > going to be interesting. > > > > Thanks for clarification. I'll try to come up with patches as you talked. I'm wondering, is there any progress with this feature? Having sysprof-alike reverse ordering for call chains is a really powerful form of visualization IMO. Thanks, Ingo