From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264231AbUHCGv4 (ORCPT ); Tue, 3 Aug 2004 02:51:56 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265091AbUHCGv4 (ORCPT ); Tue, 3 Aug 2004 02:51:56 -0400 Received: from colin2.muc.de ([193.149.48.15]:1042 "HELO colin2.muc.de") by vger.kernel.org with SMTP id S264231AbUHCGvz (ORCPT ); Tue, 3 Aug 2004 02:51:55 -0400 Date: 3 Aug 2004 08:51:54 +0200 Date: Tue, 3 Aug 2004 08:51:54 +0200 From: Andi Kleen To: Andrew Morton Cc: Peter Williams , linux-kernel@vger.kernel.org Subject: Re: [PATCH] V-3.0 Single Priority Array O(1) CPU Scheduler Evaluation Message-ID: <20040803065154.GA85840@muc.de> References: <2oEEn-197-9@gated-at.bofh.it> <410EDBF5.40205@bigpond.net.au> <20040802205332.3413cd6d.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040802205332.3413cd6d.akpm@osdl.org> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > But let me re-repeat again that CPU scheduler problems tend to take a > _long_ time to turn up - you make some change and two months later some > person with a weird workload on expensive hardware hits a nasty corner > case. So I do think that we'd have to hit a nasty problem with the current > scheduler to go making deep changes. How about just simplifying the code? Both Con's and Peter's code look a lot simpler compared to the stock scheduler and are easier to understand. If they don't work significantly worse I think that would be a strong argument to move to one of them. -Andi