From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933143AbbFIUJl (ORCPT ); Tue, 9 Jun 2015 16:09:41 -0400 Received: from mx1.redhat.com ([209.132.183.28]:52118 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753333AbbFIUJf (ORCPT ); Tue, 9 Jun 2015 16:09:35 -0400 Date: Tue, 9 Jun 2015 22:09:23 +0200 From: Jiri Olsa To: Arnaldo Carvalho de Melo Cc: Jiri Olsa , lkml , Peter Zijlstra , Paul Mackerras , David Ahern , Namhyung Kim , Ingo Molnar , Andi Kleen , Stephane Eranian Subject: Re: [PATCHv3] perf stat: Add id into perf_stat struct Message-ID: <20150609200922.GA7884@krava.redhat.com> References: <1433341559-31848-1-git-send-email-jolsa@kernel.org> <1433341559-31848-2-git-send-email-jolsa@kernel.org> <20150604135055.GB23625@krava.redhat.com> <20150608140340.GC3136@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150608140340.GC3136@kernel.org> 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 On Mon, Jun 08, 2015 at 11:03:40AM -0300, Arnaldo Carvalho de Melo wrote: > Em Thu, Jun 04, 2015 at 03:50:55PM +0200, Jiri Olsa escreveu: > > On Wed, Jun 03, 2015 at 04:25:51PM +0200, Jiri Olsa wrote: > > > We need fast way to identify evsel as transaction event > > > for shadow counters computation. Currently we are using > > > possition (in evlist) based way. > > > > > > Adding 'id' into 'struct perf_stat' so it can carry transaction > > > event ID and we can use it for shadow counters computations. > > > > > > Link: http://lkml.kernel.org/n/tip-9lnvjksibrs0flhkpix2qkm9@git.kernel.org > > > Signed-off-by: Jiri Olsa > > > --- > > > > using perf_evsel__name instead of evsel->name pointer > > It now passes 'perf test' on rhel7/fedora21 systems, but it fails on a > rhel6 system, I applied it, will test it further on the rhel6 system to > see what went wrong there, unless you do it first ;-) > > [root@sandy ~]# perf test 15 > 15: struct perf_event_attr setup :sh: line > 1: 5381 Aborted (core dumped) > PERF_TEST_ATTR=/tmp/tmp8ZBLwc /home/acme/bin/perf stat -o > /tmp/tmp8ZBLwc/perf.data --group -e cycles,instructions kill > /dev/null > 2>&1 > sh: line 1: 5438 Aborted (core dumped) > PERF_TEST_ATTR=/tmp/tmpnRDADB /home/acme/bin/perf stat -o > /tmp/tmpnRDADB/perf.data -e '{cycles,instructions}' kill > /dev/null > 2>&1 > Ok > [root@sandy ~]# uname -r > 2.6.32-504.8.1.el6.x86_64 > [root@sandy ~]# cat /etc/redhat-release > Red Hat Enterprise Linux Server release 6.6 (Santiago) > [root@sandy ~]# hum, cant reproduce [root@ibm-x3650m4-02 perf]# ./perf test 15 15: struct perf_event_attr setup : Ok the server is latest rhel6 kernel (scratch): [root@ibm-x3650m4-02 perf]# uname -a Linux ibm-x3650m4-02.lab.eng.brq.redhat.com 2.6.32-556.el6perf_ivb_haswell_1.x86_64 #1 SMP Wed Apr 29 06:00:24 EDT 2015 x86_64 x86_64 x86_64 GNU/Linux [root@ibm-x3650m4-02 perf]# cat /etc/redhat-release Red Hat Enterprise Linux Server release 6.6 (Santiago) jirka