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 S1752537AbbFZIkZ (ORCPT <rfc822;w@1wt.eu>);
	Fri, 26 Jun 2015 04:40:25 -0400
Received: from e28smtp07.in.ibm.com ([122.248.162.7]:38714 "EHLO
	e28smtp07.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751815AbbFZIkF (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 26 Jun 2015 04:40:05 -0400
X-Helo: d28dlp01.in.ibm.com
X-MailFrom: naveen.n.rao@linux.vnet.ibm.com
X-RcptTo: linux-kernel@vger.kernel.org
Date: Fri, 26 Jun 2015 14:09:33 +0530
From: "Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Balbir Singh <bsingharora@gmail.com>, Ingo Molnar <mingo@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] proc/schedstat: Expose /proc/<pid>/schedstat if delay
 accounting is enabled
Message-ID: <20150626083933.GB15883@naverao1-tp.in.ibm.com>
References: <1432537954-26665-1-git-send-email-naveen.n.rao@linux.vnet.ibm.com>
 <CAKTCnzm5A6n=jvskHqUmpdq7yynzKqwuVFJL-ia=3PmgDeBV-w@mail.gmail.com>
 <CAM_iQpW8+pHOhdnZS7iwE987Qc1AqX0qCRYPvEZx5Xc8Ep+92w@mail.gmail.com>
 <20150625182707.GA15883@naverao1-tp.in.ibm.com>
 <CAM_iQpXd6_1Ry1WxfMUbjnwhm=7teM8NHsgutw_qHvgjwU78TA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAM_iQpXd6_1Ry1WxfMUbjnwhm=7teM8NHsgutw_qHvgjwU78TA@mail.gmail.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
X-TM-AS-MML: disable
X-Content-Scanned: Fidelis XPS MAILER
x-cbid: 15062608-0025-0000-0000-000005861798
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 2015/06/25 11:40AM, Cong Wang wrote:
> On Thu, Jun 25, 2015 at 11:27 AM, Naveen N. Rao
> <naveen.n.rao@linux.vnet.ibm.com> wrote:
> >
> > The primary issue with CONFIG_SCHEDSTATS is the (slight) additional
> > overhead it introduces, per Kconfig. Due to this, it is not enabled by
> > default by some of the distro kernels.
> >
> 
> The overhead is same for CONFIG_TASK_DELAY_ACCT since the
> core sched info code is protected by CONFIG_TASK_DELAY_ACCT
> or CONFIG_TASK_DELAY_ACCT?

CONFIG_SCHEDSTATS enables a lot more scheduler statistics 
(/proc/schedstat) than just the task delay accounting enabled by 
CONFIG_TASK_DELAY_ACCT. So, no - the overhead is not the same.

- Naveen