From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933310Ab1ERPUk (ORCPT ); Wed, 18 May 2011 11:20:40 -0400 Received: from hrndva-omtalb.mail.rr.com ([71.74.56.125]:39240 "EHLO hrndva-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933164Ab1ERPUj (ORCPT ); Wed, 18 May 2011 11:20:39 -0400 X-Authority-Analysis: v=1.1 cv=ou1QuR4lBR9YeJgEH9ccYmbAdaWqVVq3lOvCKJtMpGM= c=1 sm=0 a=9MPN8-DmjYcA:10 a=5SG0PmZfjMsA:10 a=Q9fys5e9bTEA:10 a=OPBmh+XkhLl+Enan7BmTLg==:17 a=07d9gI8wAAAA:8 a=Qg6-wrgoqUaQF5U1cKEA:9 a=PUjeQqilurYA:10 a=OPBmh+XkhLl+Enan7BmTLg==:117 X-Cloudmark-Score: 0 X-Originating-IP: 67.242.120.143 Subject: Re: [PATCH] sched: change pull_rt_task() to decrease time waiting on runqueue From: Steven Rostedt To: Hillf Danton Cc: LKML , Mike Galbraith , Ingo Molnar , Peter Zijlstra , Yong Zhang In-Reply-To: References: <1305725244.26849.13.camel@gandalf.stny.rr.com> Content-Type: text/plain; charset="ISO-8859-15" Date: Wed, 18 May 2011 11:20:37 -0400 Message-ID: <1305732037.11267.7.camel@gandalf.stny.rr.com> Mime-Version: 1.0 X-Mailer: Evolution 2.32.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2011-05-18 at 22:54 +0800, Hillf Danton wrote: > In short, if there are pushable tasks and if there are RQs, > NOT LIMITED TO our RQ, in lower priority, tasks should be pushed to > RQs as many as we could. I understand what you are trying to do. But this change modifies a lot of assumptions. Please supply test cases that shows how this helps. Have a look at: http://lwn.net/Articles/425583/ Where I did a bit of work just to make sure my change to sched_rt.c was appropriate. Just coming up with scenarios may not be good enough. Seeing it in practice is worth much more. For example, you may be making the fast path slower. This may do what you expect, with a hit in performance. I'm not sure I like that idea. -- Steve