Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756679Ab1BYAOn (ORCPT ); Thu, 24 Feb 2011 19:14:43 -0500 Received: from relay2.sgi.com ([192.48.179.30]:33198 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756632Ab1BYAOm (ORCPT ); Thu, 24 Feb 2011 19:14:42 -0500 To: linux-kernel@vger.kernel.org Subject: [RFC] Performance Events 'Samples' report Cc: a.p.zijlstra@chello.nl, paulus@samba.org, mingo@elte.hu, acme@ghostprotocols.net Message-Id: From: Cliff Wickman Date: Thu, 24 Feb 2011 18:15:26 -0600 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1511 Lines: 39 From: Cliff Wickman The perf command's report prints clock periods under the heading of 'Samples'. Our applications people have been confused by this "Samples" heading. perf report -n # Overhead Samples Command Shared Object Symbol # ........ .......... ............... .................... ................................ # 99.41%1522136216246 pthreads pthreads [.] work_function 0.33% 4980298302 perf [kernel.kallsyms] [k] clear_page_c .... It would be more compact to print the number of samples. And that would be consistent with the heading. Below is the code that displays the periods. I don't see that the number of samples is available here, but probably easily could be. I presume that clock periods are derived from #samples * sample period. perf/util/hist.c 640 641 if (symbol_conf.show_nr_samples) { 642 if (sep) 643 ret += snprintf(s + ret, size - ret, "%c%" PRIu64, *sep, period); 644 else 645 ret += snprintf(s + ret, size - ret, "%11" PRIu64, period); 646 } 647 Signed-off-by: Cliff Wickman -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/