From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757817Ab0ITX37 (ORCPT ); Mon, 20 Sep 2010 19:29:59 -0400 Received: from mga11.intel.com ([192.55.52.93]:55940 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754843Ab0ITX37 (ORCPT ); Mon, 20 Sep 2010 19:29:59 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.56,395,1280732400"; d="scan'208";a="839386493" Subject: Re: [patch]cfq-iosched: schedule dispatch for noidle queue From: Shaohua Li To: Vivek Goyal Cc: lkml , "jaxboe@fusionio.com" , "czoccolo@gmail.com" In-Reply-To: <20100920140242.GD6138@redhat.com> References: <1284972817.6598.19.camel@sli10-conroe.sh.intel.com> <20100920140242.GD6138@redhat.com> Content-Type: text/plain; charset="UTF-8" Date: Tue, 21 Sep 2010 07:29:56 +0800 Message-ID: <1285025396.1668.4.camel@shli-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2010-09-20 at 22:02 +0800, Vivek Goyal wrote: > On Mon, Sep 20, 2010 at 04:53:37PM +0800, Shaohua Li wrote: > > A queue is idle at cfq_dispatch_requests(), but it gets noidle later for > > different reasons. Unless other task explictly does unplug or all requests > > are drained, we will not deliever requests to the disk even cfq_arm_slice_timer > > doesn't make the queue idle. For example, cfq_should_idle() returns true > > because of service_tree->count == 1, and then other queues are added. > > Note, I didn't see obvious performance impacts so far with the patch, but > > just thought this could be a problem. > > > > Shaohua Li, > > What are those reasons when an idling queue becomes a non idle one > (without getting a new request in the queue)? OOps, I used to add a new reason for another patch but forgot to update the comment. So this is the only reason. Thanks, Shaohua