Bug 113709 - Sound not restored after APM event on IBM T21
Summary: Sound not restored after APM event on IBM T21
Status: RESOLVED WONTFIX
Alias: None
Product: SUSE LINUX 10.0
Classification: openSUSE
Component: Basesystem (show other bugs)
Version: Beta 2
Hardware: i686 All
: P5 - None : Normal
Target Milestone: ---
Assignee: Takashi Iwai
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-08-29 00:08 UTC by Phil Stopford
Modified: 2007-02-16 14:04 UTC (History)
0 users

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Phil Stopford 2005-08-29 00:08:25 UTC 
As summary. This affects 9.3 (not had time to test 10.0 just yet). You can
restart the sound system using 'rcalsasound restart', but this kills all apps
using sound. It would be better to have the sound recovered without this problem.

Trying to use the in-built recovery mechanisms in the powermanagement related
config files doesn't help.
Comment 1 Takashi Iwai 2005-08-29 10:03:33 UTC 
Which driver is used on this machine?  Show /proc/asound/cards.

If it's CS46xx, the PM won't work always, especially on laptops -> WONTFIX
If it's an Intel chip, it should be fixed...
Comment 2 Takashi Iwai 2005-08-29 10:19:08 UTC 
Also, try beta3.  The PM stuff has been fixed in each release.
Comment 3 Phil Stopford 2005-08-30 21:08:53 UTC 
Beta 3 fails to respond to APM beyond locking the machine up. It is a CS46xx
chip, but Windows naturally manages to revive this hardware without issue so
there must be something that can be done.

Note that rcalsasound restart does fix the problem, but the nasty effect is to
kill all applications that have some interaction with the sound system. If the
same effect could be achieved without killing apps off, things would be much better.

*shrug*
Comment 4 Takashi Iwai 2005-09-01 17:21:40 UTC 
If you unload snd-cs46xx before suspend, does lock up still occur?
Do you have any kernel error messages from cs46xx driver when resume?

The problem is that the behavior of controller doesn't follow to the datasheet
at resume time.  (Also we have not this hardware for debugging until now.)
Comment 5 Phil Stopford 2005-09-04 22:12:56 UTC 
The APM lock-up appears related to the framebuffer drivers being loaded. The
rcalsasound restart approach fixes sound on resume, but I haven't had the time
yet to look into the logs for kernel error messages. I only get a couple of
evenings a week to look at non-work related stuff.
Comment 6 Takashi Iwai 2005-09-06 10:27:51 UTC 
For testing just only the sound stuff, you can use "alsactl".  Run like:

   # alsactl power 0 D3hot

to suspend the sound function, and

   # alsactl power 0 D0

to resume.  Check the behavior of sound devices with the sequence above.
Also, please show the output of "lspci -nv", and attach any relevant kernel
messages before/after suspend/resume.
Comment 7 Matej Horvath 2007-02-16 14:04:27 UTC 
No reaction on this bug for more than 17 months. Closing as CANTFIX.