From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263894AbTGGLSd (ORCPT ); Mon, 7 Jul 2003 07:18:33 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264904AbTGGLSd (ORCPT ); Mon, 7 Jul 2003 07:18:33 -0400 Received: from dyn-ctb-210-9-243-115.webone.com.au ([210.9.243.115]:15110 "EHLO chimp.local.net") by vger.kernel.org with ESMTP id S263894AbTGGLSc (ORCPT ); Mon, 7 Jul 2003 07:18:32 -0400 Message-ID: <3F095A65.4070100@cyberone.com.au> Date: Mon, 07 Jul 2003 21:32:53 +1000 From: Nick Piggin User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3.1) Gecko/20030618 Debian/1.3.1-3 X-Accept-Language: en MIME-Version: 1.0 To: Con Kolivas CC: linux-kernel Subject: Process scheduler fairness bug (feature?) Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi Con, In recent testing unrelated to your interactivity stuff, I have found the following with 74-mm2, although I don't think its due to your interactivity stuff. I don't have a real workload that is bothered by this btw. Just wondering if its fixable, or there is a reason for it. On UP, 2 processes of same priority. One is doing an infinite loop of nothing, the other doing an infinite loop of fork+waiting for children which count to a million then exit (do a bit of work). The non forking CPU hog gets 75% of the cpu.