From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S274863AbTHPQRl (ORCPT ); Sat, 16 Aug 2003 12:17:41 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S274866AbTHPQRl (ORCPT ); Sat, 16 Aug 2003 12:17:41 -0400 Received: from 153.Red-213-4-13.pooles.rima-tde.net ([213.4.13.153]:15877 "EHLO small.felipe-alfaro.com") by vger.kernel.org with ESMTP id S274863AbTHPQRk (ORCPT ); Sat, 16 Aug 2003 12:17:40 -0400 Subject: O16.2int scheduler interactivity improvements From: Felipe Alfaro Solana To: LKML Cc: Con Kolivas Content-Type: text/plain Message-Id: <1061050658.891.7.camel@teapot.felipe-alfaro.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.4 Date: Sat, 16 Aug 2003 18:17:39 +0200 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, all! I've been testing 2.6.0-test3-mm2 plus Con scheduler patches up to O16.2int. I must say I feel it keeps improving. For my workloads (usually interactive X applications and sound) O16.2int makes X feels very smooth, even under heavy load (while true; do a=2; done). I recommend everyone out there who is (somewhat) idle to give O16.2int a spin. The more people testing this, the more likely we are able to get a robust and fair scheduler. Sometimes, we feel the scheduler just doesn't behave as we would like (and I've felt that way many times), but that's no reason to give up. Please, test the latest OXXint patches and keep your reports coming :-) We know we can do it better... Thanks!