Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752927Ab0D1HwY (ORCPT ); Wed, 28 Apr 2010 03:52:24 -0400 Received: from mail-pz0-f204.google.com ([209.85.222.204]:60427 "EHLO mail-pz0-f204.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751743Ab0D1HwW convert rfc822-to-8bit (ORCPT ); Wed, 28 Apr 2010 03:52:22 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; b=BcYfPqP/YX+ic5Iu6lAQ0reGKzyigvDSf1PHomMwRkZY4maNQy1KxJWWE5P1vPg2BO MhoqZtP4DVJ7yB+io9WevSNW98BSC+IVhKK2wy90Ax+4Ldx1BkY59WYp/njhM55bjjjz ESNf4YCxWV79qeE+9iJuefRSy+XFwk9W6PGCc= MIME-Version: 1.0 In-Reply-To: References: <1272430986-20436-1-git-send-email-xiaosuo@gmail.com> From: Changli Gao Date: Wed, 28 Apr 2010 15:52:01 +0800 Message-ID: Subject: Re: [RFC] sched: implement the exclusive wait queue as a LIFO queue To: Xiaotian Feng Cc: Ingo Molnar , Alexander Viro , Andrew Morton , "Eric W. Biederman" , Davide Libenzi , Roland Dreier , Stefan Richter , Peter Zijlstra , "David S. Miller" , Eric Dumazet , Christoph Lameter , Andreas Herrmann , Thomas Gleixner , David Howells , Takashi Iwai , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1066 Lines: 27 On Wed, Apr 28, 2010 at 3:47 PM, Xiaotian Feng wrote: > On Wed, Apr 28, 2010 at 1:03 PM, Changli Gao wrote: >> implement the exclusive wait queue as a LIFO queue >> >> If the exclusive wait queue is also a LIFO queue as the normal wait queue, the >> process who goes to sleep recently, will be woke up first. As its memory is >> more likely in cache, we will get better performance. And when there are many >> processes waiting on a exclusive wait queue, some of them may not be woke up, >> if the others can handle the workload, and it will reduce the load of >> the scheduler. >> > > Starve some processes for performance? > Starve? Oh, No. If we don't need these processes, and we can do better without them, why we wake them up? -- Regards， Changli Gao(xiaosuo@gmail.com) -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/