http://bugs.winehq.org/show_bug.cgi?id=32706
Bug #: 32706 Summary: No sound with foobar2000: Unknown error code (0x80070057) Product: Wine Version: unspecified Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: winehq@supradigital.org Classification: Unclassified
After the wine upgrade to wine-1.5.21 I have exactly the problem described at bug 23568 (which was that instead of sound one would get an error message)
None of the suggested method worked (changing windows version, deleting prefix, re-installing foobar).
I made sure an output device is selected but for all devices but the null device I get the elsewhere mentioned error message:
"Unrecoverable playback error: Unknown error code (0x80070057)"
I am using arch linux on a 64bit system with OSS as my output. Sound does work in winecfg when pressing the Test button.
Maybe of importance: When I press play in foobar, I here a short "noise" in the speakers.
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #1 from Marenz winehq@supradigital.org 2013-01-11 19:05:46 CST --- I just noticed that it was not the update 1.5.20 -> 21 that broke it but that the last version I had that worked was wine-1.5.16
http://bugs.winehq.org/show_bug.cgi?id=32706
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
--- Comment #2 from Austin English austinenglish@gmail.com 2013-01-11 21:50:08 CST --- (In reply to comment #1)
I just noticed that it was not the update 1.5.20 -> 21 that broke it but that the last version I had that worked was wine-1.5.16
Please run a regression test: http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #3 from Marenz winehq@supradigital.org 2013-01-11 23:22:13 CST --- I just finished a git bisect and the commit that introduced this problem for me was this one:
dc0cef3d3b68ae10c80c13846265a9317eecdfd3 is the first bad commit commit dc0cef3d3b68ae10c80c13846265a9317eecdfd3 Author: Andrew Eikum aeikum@codeweavers.com Date: Mon Dec 3 10:28:11 2012 -0600
wineoss.drv: Improve IsFormatSupported handling.
:040000 040000 5bd844a8be682e09519e17485f43bd425b35eacc abf616c99f804d1cf6acf771e4d344a0ba903498 M dlls
I have a look at the wikipage about regression testing now.
http://bugs.winehq.org/show_bug.cgi?id=32706
Marenz winehq@supradigital.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |aeikum@codeweavers.com
http://bugs.winehq.org/show_bug.cgi?id=32706
Marenz winehq@supradigital.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winehq@supradigital.org
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #4 from Marenz winehq@supradigital.org 2013-01-11 23:27:27 CST --- Okay, I guess I did already what the wiki page said :D
http://bugs.winehq.org/show_bug.cgi?id=32706
Marenz winehq@supradigital.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |dc0cef3d3b68ae10c80c1384626 | |5a9317eecdfd3
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #5 from Marenz winehq@supradigital.org 2013-01-12 06:35:54 CST --- I just applied the reverse patch of that commit on top of master and sure enough, sound works again now.
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #6 from Andrew Eikum aeikum@codeweavers.com 2013-01-14 09:19:25 CST --- Thanks for the nice bug report, Marenz.
It's working fine for me with Arch package oss-4.2_2007-1. Which OSS version do you have, and can you attach a log with the debug channels from http://wiki.winehq.org/Sound?
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #7 from Marenz winehq@supradigital.org 2013-01-14 16:15:14 CST --- Created attachment 43169 --> http://bugs.winehq.org/attachment.cgi?id=43169 Log of the problem
Version info: OSS 4.2 (b 2007/201209142246) (0x00040100) GPL Platform: Linux/x86_64 3.7-pf #1 SMP PREEMPT Thu Jan 10 09:10:26 EET 2013 (rebooc)
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #8 from Andrew Eikum aeikum@codeweavers.com 2013-01-15 08:27:16 CST --- This looks like a bug in your OSS driver:
0031:trace:oss:AUDDRV_GetAudioEndpoint OSS audioinfo: 0031:trace:oss:AUDDRV_GetAudioEndpoint devnode: /dev/oss/oss_hdaudio0/spdout0 0031:trace:oss:AUDDRV_GetAudioEndpoint name: HD Audio play spdifout-mix (vmix) 0031:trace:oss:AUDDRV_GetAudioEndpoint busy: 2 0031:trace:oss:AUDDRV_GetAudioEndpoint caps: 19077500 0031:trace:oss:AUDDRV_GetAudioEndpoint iformats: 5010 0031:trace:oss:AUDDRV_GetAudioEndpoint oformats: 5010 0031:trace:oss:AUDDRV_GetAudioEndpoint enabled: 1 0031:trace:oss:AUDDRV_GetAudioEndpoint min_rate: 0 0031:trace:oss:AUDDRV_GetAudioEndpoint max_rate: 0 0031:trace:oss:AUDDRV_GetAudioEndpoint min_channels: 2 0031:trace:oss:AUDDRV_GetAudioEndpoint max_channels: 0
Notice that the allowed rates are both 0, and that min_channels is 2 while max_channels is 0.
I guess we can add some sanity checks for this. Maybe check if it's 0 and use a reasonable default (44.1kHz stereo).
I'll send an email to the OSS mailing list, too.
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #9 from Andrew Eikum aeikum@codeweavers.com 2013-01-15 08:50:20 CST --- Created attachment 43173 --> http://bugs.winehq.org/attachment.cgi?id=43173 wineoss.drv: Set reasonable defaults if driver reports invalid values
Here's a patch that sets defaults on broken drivers. I think it should fix your problem.
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #10 from Marenz winehq@supradigital.org 2013-01-15 08:53:18 CST --- I will try it when I get home. Would it help to show the same log for the old wine version where it still worked? How did the old wine behave?
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #11 from Andrew Eikum aeikum@codeweavers.com 2013-01-15 09:02:38 CST --- No, I think I understand what's going on here.
We had a change to dsound recently (605cd62c7dc8) which stopped using a hard-coded dsound format and instead uses the preferred format from mmdevapi. As part of that, we made the format validation in mmdevapi more strict (dc0cef3d3b6), to help dsound do its detection correctly. Your OSS driver was causing an invalid format from mmdevapi, but the lax validation would let it slip through, so stricter validation ended up causing the breakage.
I've sent a mail to opensound-devel: http://sourceforge.net/mailarchive/forum.php?thread_name=20130115143710.GJ10...
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #12 from Marenz winehq@supradigital.org 2013-01-15 09:05:08 CST --- One thing makes me wonder though. The sound DID work work for the test in winecfg. Shouldn't that have the same problem?
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #13 from Andrew Eikum aeikum@codeweavers.com 2013-01-15 09:10:38 CST --- No, winecfg uses the winmm API, not dsound. winmm doesn't have to do any mixing, so it doesn't attempt to use the preferred mmdevapi format. It just feeds the WAV data directly(-ish) to mmdevapi, so it doesn't matter that mmdevapi was reporting a bogus preferred format.
http://bugs.winehq.org/show_bug.cgi?id=32706
Jörg Höhle hoehle@users.sourceforge.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |mmdevapi
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #14 from Andrew Eikum aeikum@codeweavers.com 2013-01-22 08:27:24 CST --- Hi Marenz,
When you get an opportunity, can you confirm that the patch fixes your problem? I'd like to get it into Wine if it does.
Thanks!
http://bugs.winehq.org/show_bug.cgi?id=32706
--- Comment #15 from Marenz winehq@supradigital.org 2013-01-25 16:02:38 CST --- Sorry for the long delay, lots of things happened in RL.
I can confirm that your patch fixes the problem!
Thanks! :)
http://bugs.winehq.org/show_bug.cgi?id=32706
Andrew Eikum aeikum@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |6da3ce276bd76a5242c8340c500 | |42c95ea74a787 Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #16 from Andrew Eikum aeikum@codeweavers.com 2013-01-28 13:48:23 CST --- The patch is in Wine as 6da3ce276bd76a5242c8340c50042c95ea74a787, and will be in the 1.5.23 release.
Thanks for reporting and retesting, Marenz!
http://bugs.winehq.org/show_bug.cgi?id=32706
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #17 from Alexandre Julliard julliard@winehq.org 2013-02-01 04:54:37 CST --- Closing bugs fixed in 1.5.23.
-
wine-bugs＠winehq.org