From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1753386AbbC3Rdp (ORCPT <rfc822;w@1wt.eu>);
	Mon, 30 Mar 2015 13:33:45 -0400
Received: from mail-ie0-f172.google.com ([209.85.223.172]:36478 "EHLO
	mail-ie0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753044AbbC3Rdo (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Mon, 30 Mar 2015 13:33:44 -0400
Message-ID: <551988F6.8000501@gmail.com>
Date: Mon, 30 Mar 2015 11:33:42 -0600
From: David Ahern <dsahern@gmail.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Peter Zijlstra <peterz@infradead.org>,
        Stephane Eranian <eranian@google.com>
CC: Arnaldo Carvalho de Melo <acme@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>, Jiri Olsa <jolsa@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        LKML <linux-kernel@vger.kernel.org>,
        John Stultz <john.stultz@linaro.org>, "H. Peter Anvin" <hpa@zytor.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH] perf, record: Add clockid parameter
References: <tip-34f439278cef7b1177f8ce24f9fc81dfc6221d3b@git.kernel.org> <20150327143201.GG21418@twins.programming.kicks-ass.net> <55158F25.9040100@gmail.com> <20150327172059.GK23123@twins.programming.kicks-ass.net> <551594DD.7050705@gmail.com> <20150327201534.GD6291@redhat.com> <20150327215942.GL23123@twins.programming.kicks-ass.net> <CABPqkBR_ou8E01kRmDOBb72ngcdzBkfVPNUj71XWnsYE6i0Bdg@mail.gmail.com> <20150328075549.GG27490@worktop.programming.kicks-ass.net>
In-Reply-To: <20150328075549.GG27490@worktop.programming.kicks-ass.net>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 3/28/15 1:55 AM, Peter Zijlstra wrote:
> @@ -761,6 +762,11 @@ void perf_evsel__config(struct perf_evse
>   		attr->disabled = 0;
>   		attr->enable_on_exec = 0;
>   	}
> +
> +	if (opts->clockid >= 0) {
> +		attr->use_clockid = 1;
> +		attr->clockid = opts->clockid;
> +	}
>   }

One more: you need to set attr->clockid to -1 if use_clockid is not set 
so that the analysis side knows whether attr->clockid was used. 
Otherwise it defaults to 0 == CLOCK_REALTIME which is misleading.


diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 1abf6919b8a2..27679ab38511 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -766,7 +766,8 @@ void perf_evsel__config(struct perf_evsel *evsel, 
struct record_opts *opts)
     if (opts->clockid >= 0) {
         attr->use_clockid = 1;
         attr->clockid = opts->clockid;
-   }
+   } else
+       attr->clockid = -1;
  }


Dumping the setting in the header is useful as well:

diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index fb432153e2aa..40bc8d010fcb 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -1098,6 +1098,7 @@ static void print_event_desc(struct perf_header 
*ph, int fd, FILE *fp)
             }
             fprintf(fp, " }");
         }
+       fprintf(fp, ", clockid = %d", evsel->attr.clockid);

         fputc('\n', fp);
     }