From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 10 Oct 2001 13:58:07 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 10 Oct 2001 13:57:58 -0400 Received: from atlante.atlas-iap.es ([194.224.1.3]:19468 "EHLO atlante.atlas-iap.es") by vger.kernel.org with ESMTP id ; Wed, 10 Oct 2001 13:57:53 -0400 From: "Ricardo Galli" To: Subject: 2.4.11 responsiveness and mtrr Date: Wed, 10 Oct 2001 19:59:39 +0200 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Importance: Normal Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, I've trying 2.4.10 and 2.4.11. The two problems I've reported about 2.4 before don't reporduce anymore: memory consumption: http://www.cs.helsinki.fi/linux/linux-kernel/2001-04/1578.html agressive swap/read once cache files (few weeks ago). Now, after listening hours of mp3 the memory usage keeps very low and there is no a byte swapped out (with KDE and konqueror running): gallir@linux:~$ free total used free shared buffers cached Mem: 513620 402880 110740 0 56232 317240 -/+ buffers/cache: 29408 484212 Swap: 503992 0 503992 But, the system responsiviness is worse than before, especially in two cases (on a PIII 1000): - The X-mouse pointer blocks for short periods and the audio is also interrupted when buffers are flushed to disk (using ReiserFS), it occurs also with the X-server reniced to -15. - mmap/blocking when reading files from the cd-driver (iso9660), very often noticeable when starting to play a new MP3 file (I believe noatun uses mmap). It seems a problem of mmap and/or read ahead buffers. I didn't notice this problem in pre-2.4.10 kernels, at least no so often. I also saw a strange errors that didn't appear before (may be related to closed nvidia modules, if so disregard it): ... NVRM: loading NVIDIA kernel module version 1.0-1541 mtrr: no MTRR for f0000000,2000000 found mtrr: no MTRR for f0000000,2000000 found mtrr: no MTRR for f0000000,2000000 found ... --ricardo http://m3d.uib.es/~gallir/