From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 12 Sep 2002 16:50:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 12 Sep 2002 16:50:34 -0400 Received: from ns.suse.de ([213.95.15.193]:47117 "EHLO Cantor.suse.de") by vger.kernel.org with ESMTP id ; Thu, 12 Sep 2002 16:50:32 -0400 To: Alan Cox Cc: rml@tech9.net (Robert Love), mingo@elte.hu, linux-kernel@vger.kernel.org Subject: Re: [PATCH] 2.4-ac task->cpu abstraction and optimization References: <15744.57073.2852.707839@kim.it.uu.se.suse.lists.linux.kernel> <200209122022.g8CKMJS15137@devserv.devel.redhat.com.suse.lists.linux.kernel> From: Andi Kleen Date: 12 Sep 2002 22:55:20 +0200 In-Reply-To: Alan Cox's message of "12 Sep 2002 22:28:54 +0200" Message-ID: X-Mailer: Gnus v5.7/Emacs 20.6 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Alan Cox writes: > > > also took a look at your patch -- looks good, you should submit it to > > > Marcelo... it cannot hurt for 2.4. > > > > I might do that, unless Alan plans on pushing the -ac sched.c stuff to > > Marcelo, in which case my patch would just confuse things. Alan? > > I'd like to see it in 2.4 base. Its really Marcelo's call. One imho major problem with the new scheduler is that its new sched_yield breaks programs like OpenOffice, who rely on the old sched_yield behaviour. With new scheduler and 2.5 yield OpenOffice can be completely starved just by a kernel compile because sched_yield kills all its time slices. I don't think a stable release should break programs in such a way. -Andi