From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2992695AbXCGWo4 (ORCPT ); Wed, 7 Mar 2007 17:44:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S2992693AbXCGWoz (ORCPT ); Wed, 7 Mar 2007 17:44:55 -0500 Received: from ns2.suse.de ([195.135.220.15]:34713 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2992730AbXCGWoy (ORCPT ); Wed, 7 Mar 2007 17:44:54 -0500 Date: Wed, 07 Mar 2007 23:44:52 +0100 Message-ID: From: Takashi Iwai To: Michal Piotrowski Cc: Andrew Morton , Randy Cushman , Jaroslav Kysela , alsa-devel@alsa-project.org, linux-kernel@vger.kernel.org, Linus Torvalds , Adrian Bunk Subject: Re: mm snapshot broken-out-2007-03-05-02-22.tar.gz uploaded In-Reply-To: <45EF3B2C.9060801@googlemail.com> References: <200703051024.l25AOIF3015951@shell0.pdx.osdl.net> <45EC2DEB.1050108@googlemail.com> <45EEA3C4.7000501@googlemail.com> <20070307104746.f08a2e84.akpm@linux-foundation.org> <6bffcb0e0703071141j42309702w793e4a76b1b23bfa@mail.gmail.com> <20070307115024.756c848f.akpm@linux-foundation.org> <45EF3B2C.9060801@googlemail.com> User-Agent: Wanderlust/2.12.0 (Your Wildest Dreams) SEMI/1.14.6 (Maruoka) FLIM/1.14.7 (=?ISO-8859-4?Q?Sanj=F2?=) APEL/10.6 MULE XEmacs/21.5 (beta27) (fiddleheads) (+CVS-20060704) (i386-suse-linux) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=ISO-8859-2 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org At Wed, 07 Mar 2007 23:22:36 +0100, Michal Piotrowski wrote: > > Hi Takashi, > > Takashi Iwai napisał(a): > > At Wed, 7 Mar 2007 11:50:24 -0800, > > Andrew Morton wrote: > >> On Wed, 7 Mar 2007 20:41:30 +0100 "Michal Piotrowski" wrote: > >> > >>>>> 831466f4ad2b5fe23dff77edbe6a7c244435e973 is first bad commit > >>>>> commit 831466f4ad2b5fe23dff77edbe6a7c244435e973 > >>>>> Author: Randy Cushman > >>>>> Date: Tue Dec 19 18:42:16 2006 +0100 > >>>>> > >>>>> [ALSA] ac97 - fix microphone and line_in selection logic > >>>>> > >>>>> This patch fixes the Microphone and LINE_IN select logic for > >>>>> Analog Devices surround codecs with shared jacks. The existing > >>>>> code can never utilize the shared jacks for Microphone and LINE_IN > >>>>> due to the reversed jack selection logic. The patched code > >>>>> correctly selects the shared jack for input if the 'Channel Mode' > >>>>> selector does not specify that the jack is to be used for output. > >>>>> Specifically, in '2ch' mode the Center/LFE jack is used for > >>>>> microphone input and the Surround jack is used for LINE_IN, > >>>>> in '4ch' mode the Center/LFE jack is used for microphone input > >>>>> and the Surround jack is used for output, and in '6ch' mode > >>>>> both jacks are used for output. > >>>>> > >>>>> Signed-off-by: Randy Cushman > >>>>> Signed-off-by: Takashi Iwai > >>>>> Signed-off-by: Jaroslav Kysela > >>>>> > >>>> Is 2.6.21-rc3 similarly broken? > >>> Yes. > >>> > >> OK. Can you please confirm that applying the below fixes 2.6.21-rc3? > >> > >> If so, I'll queue it up for application in a week or two if there is no > >> progress on this. > > > > [sorry for the late interrupt - I'm just back from a vacation] > > > > Could you postpone reverting for a while? > > I guess this possible regression is limited only to a certain type of > > ac97 codec chip. The patch applied there was a fix for many codec > > chips. > > > > Which codec chip are you using specifically? Please check > > /proc/asound/card0/codec97#0/* file(s). > > 0-0/0: Analog Devices AD1985 OK, that's good to know. > control.43 { > comment.access 'read write' > comment.type BOOLEAN > comment.count 1 > iface MIXER > name 'Spread Front to Surround and Center/LFE' > value true > } So, you're playing two-channel samples and get no output from surround and CLFE? Or, does it happen if you play a real 5.1 channel file? (for example, try speaker-test -c6) Also, it'd be helpful if you compare /proc/asound/card0/codec97#0/ac97#0-0+regs file before and after the patch. This reveals which register bits differ actually. thanks, Takashi