From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753022AbbFZPIa (ORCPT ); Fri, 26 Jun 2015 11:08:30 -0400 Received: from mail.kernel.org ([198.145.29.136]:45868 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752457AbbFZPIX (ORCPT ); Fri, 26 Jun 2015 11:08:23 -0400 Date: Fri, 26 Jun 2015 12:08:18 -0300 From: Arnaldo Carvalho de Melo To: Adrian Hunter Cc: Ingo Molnar , linux-kernel@vger.kernel.org, Jiri Olsa , Stephane Eranian Subject: Re: [PATCH V6 08/17] perf tools: Add Intel PT support Message-ID: <20150626150818.GD3265@kernel.org> References: <20150622230005.GA8510@kernel.org> <5588FCCE.8090403@intel.com> <20150623151553.GD3489@kernel.org> <558C03FD.4020600@intel.com> <20150625134557.GB3253@kernel.org> <20150625235634.GA6633@kernel.org> <20150626000958.GB6633@kernel.org> <558CF5B4.70901@intel.com> <20150626134156.GA3265@kernel.org> <558D57FB.6090704@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <558D57FB.6090704@intel.com> X-Url: http://acmel.wordpress.com User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Em Fri, Jun 26, 2015 at 04:47:39PM +0300, Adrian Hunter escreveu: > On 26/06/15 16:41, Arnaldo Carvalho de Melo wrote: > >> Another thing, the intel_bts implementation does not support > >> "instructions" samples because there is no timing information to > >> use to create periodic samples. But callchains are added only > >> to "instructions" samples so there are no callchains in 'perf report' > >> for intel_bts. The call information is still available for > > > > Humm, so IOW, what you say is that we should refuse to run 'record' when > > asking for callchains and intel_bts? > > 'record' can record other events at the same time which can have callchains. > e.g. > > perf record -g --per-thread -e intel_bts//u,branch-misses:u ls Right, what I was trying to say is that there are combinations where one can ask for a callchain in record, it will act as if it did what was asked for but then when report runs, no callchain will be available. I guess we can check for that situation and warn the user. - Arnaldo