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 S965446AbcBQTer (ORCPT <rfc822;w@1wt.eu>);
	Wed, 17 Feb 2016 14:34:47 -0500
Received: from mx1.redhat.com ([209.132.183.28]:46882 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S965237AbcBQTep (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Wed, 17 Feb 2016 14:34:45 -0500
Date: Wed, 17 Feb 2016 20:34:42 +0100
From: Oleg Nesterov <oleg@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Jiri Olsa <jolsa@redhat.com>, Pratyush Anand <panand@redhat.com>,
        Jeff Bastian <jbastian@redhat.com>,
        Michael Petlan <mpetlan@redhat.com>, linux-kernel@vger.kernel.org,
        arm-devel@redhat.com
Subject: Re: perf: wrong event->count report (Was: perf basic-test-aarch64
 failures)
Message-ID: <20160217193442.GB14068@redhat.com>
References: <20160211093047.GA12115@dhcp-0-82.del.redhat.com>
 <20160215070620.GA23850@dhcppc3.redhat.com>
 <20160217033416.GA7371@redhat.com>
 <20160217093539.GC10042@krava.brq.redhat.com>
 <20160217102555.GZ6357@twins.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160217102555.GZ6357@twins.programming.kicks-ass.net>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 02/17, Peter Zijlstra wrote:
>
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -3173,6 +3173,10 @@ static void perf_event_enable_on_exec(in
>
>  	cpuctx = __get_cpu_context(ctx);
>  	perf_ctx_lock(cpuctx, ctx);
> +
> +	update_context_time(ctx);
> +	update_cgrp_time_from_cpuctx(cpuctx);
> +

Even if I don't really understand this change I agree, probably we need to update
the counters for enable_on_exec events somehow. But I don't see how this change can
make total_time_running == total_time_enabled.

And probably this is fine. In fact I do not understand why they should be equal.
I inserted printf() into perf_counts_values__scale() to verify that /usr/bin/perf
always sees count->run < count->ena even if I do, say,

	# perf stat -x- -e module:module_load -a sleep 1

	0--module:module_load-1007397333-100.00

the kernel reports run=1007397333 ena=1007397573. Close but not equal. Again, I
do not think this is wrong, I am only saying that I don't understand this logic
because this always means "scaled" for perf.

Nevermind, I never looked into tools/perf before, most probably I simply do not
understand it at all.

Oleg.