From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S270486AbTHLQGA (ORCPT ); Tue, 12 Aug 2003 12:06:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S270519AbTHLQGA (ORCPT ); Tue, 12 Aug 2003 12:06:00 -0400 Received: from 153.Red-213-4-13.pooles.rima-tde.net ([213.4.13.153]:59142 "EHLO small.felipe-alfaro.com") by vger.kernel.org with ESMTP id S270486AbTHLQF5 (ORCPT ); Tue, 12 Aug 2003 12:05:57 -0400 Subject: Re: Linux 2.6 doesn't like Rhythmbox From: Felipe Alfaro Solana To: in7y118@public.uni-hamburg.de Cc: LKML , rhythmbox-devel@gnome.org, gstreamer-devel@lists.sourceforge.net In-Reply-To: <1060699703.3f38fe37b8a1f@rzaixsrv6.rrz.uni-hamburg.de> References: <1060699703.3f38fe37b8a1f@rzaixsrv6.rrz.uni-hamburg.de> Content-Type: text/plain Message-Id: <1060704354.856.1.camel@teapot.felipe-alfaro.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.4 Date: Tue, 12 Aug 2003 18:05:54 +0200 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2003-08-12 at 16:48, in7y118@public.uni-hamburg.de wrote: > Let me explain how threads in Rhythmbox work: The main thread is used for the > GUI, other threads (mostly idle) take care of the library - reading out > artist/title/... tags and monitoring file changes so the playlists gets updated > automagically - and then there is a playback thread. This thread is spawned > when playback of a file starts (a new one for each file). It starts by > inspecting the file and constructing a pipeline depending on the file type (ogg > decoder vs mp3 decoder vs ...). This takes half a second, maybe less. After > that it proceeds to do read - decode - output to soundcard looping until the > song is done playing. > The priority according to top starts becoming worse from the beginning of > playback and gets worse during playback. > > > My question now is simple: Who shall I blame for this? Don't blame anyone still... There's still ongoing kernel scheduler work. Please, try the latest -mm patches on top of 2.6.0-test3. You will find them at ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches. Experiment with 2.6.0-test3-mm1 to see if it still shows the behaviour you described.