From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnaldo Carvalho de Melo Subject: Re: [perfmon2] updating perf tool Date: Mon, 14 Feb 2011 17:44:28 -0200 Message-ID: <20110214194428.GE21384@ghostprotocols.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-yx0-f174.google.com ([209.85.213.174]:35125 "EHLO mail-yx0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751280Ab1BNTod (ORCPT ); Mon, 14 Feb 2011 14:44:33 -0500 Received: by yxt3 with SMTP id 3so2222055yxt.19 for ; Mon, 14 Feb 2011 11:44:32 -0800 (PST) Content-Disposition: inline In-Reply-To: Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Arun Sharma Cc: Can Hankendi , linux-perf-users@vger.kernel.org Em Fri, Feb 11, 2011 at 12:31:11PM -0800, Arun Sharma escreveu: > On Fri, Feb 11, 2011 at 12:21 PM, Can Hankendi w= rote: > > Thanks for the reply.=A0 I've tried perf record -C 0 -e cycles slee= p 1 but i > > received an error: Couldn't record kernel reference relocation symb= ol. Is it > > possible to create per core report by using perf stat -C 0,1 so tha= t I can > > have individual stats for each core? >=20 > # ./perf stat -e cycles -A -a -- sleep 1 >=20 > Performance counter stats for 'sleep 1': >=20 > CPU0 2,145,862,116 cycles > CPU1 2,146,131,064 cycles > CPU2 2,146,021,047 cycles > [...] >=20 > and you can grep out only the CPUs you're looking for. >=20 > If you're looking to do something more complicated that really > requires hacking up a script, I wrapped the perf_event_open syscall i= n > python via swig. >=20 > sys.py --cpulist >=20 > may be interesting. >=20 > http://perfmon2.git.sourceforge.net/git/gitweb.cgi?p=3Dperfmon2/libpf= m4;a=3Dblob;f=3Dpython/sys.py;h=3De3a44f24174b979694e2588592f759010b776= 21e;hb=3Df70e032b5b69cf5aa6f7e0da746b1817c64341e9 Interesting, have you seen the python binding in tip/perf/core? - Arnaldo