From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: ratio of progress of the counters Date: Fri, 01 Nov 2013 12:35:46 -0700 Message-ID: <871u2zykx9.fsf@tassilo.jf.intel.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mga09.intel.com ([134.134.136.24]:26374 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752343Ab3KATgP convert rfc822-to-8bit (ORCPT ); Fri, 1 Nov 2013 15:36:15 -0400 In-Reply-To: (=?utf-8?Q?=22Jos=C3=A9_Sim=C3=A3o=22's?= message of "Thu, 31 Oct 2013 16:39:20 +0000") Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: =?utf-8?B?Sm9zw6kgU2ltw6Nv?= Cc: linux-perf-users@vger.kernel.org Jos=C3=A9 Sim=C3=A3o writes: > Dear all, > > I'm trying to collect ASCII-based progress information regarding a se= t > of performance counters. "Progress" means, for example, the ratios of > values/second (e.g. page faults/seconds). > > I can get information in this form using 'perf stat' but 'stat' only > gives me the final value. I would like to collect, or calculate, > theses values from the running command and determine how these metric= s > evolved over time (i.e. faster or slower). Newer perf stat has a -I (interval) option for this. -Andi --=20 ak@linux.intel.com -- Speaking for myself only