From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boszormenyi Zoltan Subject: Re: Radeon HDMI question Date: Wed, 06 Mar 2013 11:10:33 +0100 Message-ID: <51371619.8050206@pr.hu> References: <50F85279.4080607@pr.hu> <50F93A39.3080606@vodafone.de> <513638DD.90806@pr.hu> <51371156.2050909@vodafone.de> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2"; Format="flowed" Content-Transfer-Encoding: quoted-printable Return-path: Received: from mail.pr.hu (mail.pr.hu [87.242.0.5]) by gabe.freedesktop.org (Postfix) with ESMTP id 800CCE5CB6 for ; Wed, 6 Mar 2013 02:10:49 -0800 (PST) In-Reply-To: <51371156.2050909@vodafone.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org Errors-To: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org To: =?ISO-8859-2?Q?Christian_K=F6nig?= Cc: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org Hi, 2013-03-06 10:50 keltez=E9ssel, Christian K=F6nig =EDrta: > Hi, > > Am 05.03.2013 19:26, schrieb Boszormenyi Zoltan: >> OK, a new problem now with 5.1 sound, it can be reproduced. >> >> I still have the same setup as described above. >> I started watching a video. Initially, HDMI audio just works nicely. >> I had to do something else so I paused Xine. >> During this time, X went to DPMS blanking. >> The TV switches itself off to save power if it sees no input. >> So, after waking up the monitor from DPMS and turning on the >> TV again, I pressed play on the video. It continues playing but >> without the sound. >> >> It seems to be the same with Linux 3.8.1 and earlier kernels. > > Does that also happen with just plain stereo playback? If yes then this i= s a bug with = > restoring the correct audio settings after DPMS blanking. I will try to test it later, also with mplayer. > > If not (and that's what I think is more likely) than that's because DPMS = blanking causes = > a desync between xine and your TV. Unfortunately there isn't much we can = do about it = > (maybe expect disabling DPMS all together), Cause from the hardware point= of view it's = > just not possible to tell the audio codec to stop playback because the di= splay isn't = > turned on any more. > > Does it work when you close xine and restart it? If yes than it might be = worth a try to = > tell xine to close and reopen the audio device on pause/restart of playba= ck. Unfortunately, after HDMI audio goes away, restarting Xine doesn't help at = all. The audio driver is pulseaudio in Xine, so I can switch between the built-in audio with stereo speakers and the HDMI output and the speakers produce the expected sounds. Also, the vumeter in pavucontrol shows that sound is generated from Xine, no matter which hardware is selected and even when HDMI is selected but there is no sound on the TV after it went out. The sound only comes back when I reboot the computer. > >> >> There is nothing in dmesg, only: >> >> ALSA sound/pci/hda/hda_eld.c:337 HDMI: ELD buf size is 0, force 128 >> ALSA sound/pci/hda/hda_eld.c:356 HDMI: invalid ELD data byte 0 > > That's unrelated, the alsa driver just assumes the standard ELD informati= ons for the = > HDMI audio codec, but AMD hardware is actually not using it. > > The real solution to your problem is to finally implement 8 channel PCM a= udio, instead = > of just using this ugly 5.1 pass through workaround. But so far nobody ha= d time to do so. Thanks for the information. Best regards, Zolt=E1n B=F6sz=F6rm=E9nyi