[alsa-devel] Suspend/Resume failure of Alsa but OSS emulation still works

Shane Volpe shanevolpe at gmail.com
Wed Jul 13 21:47:14 CEST 2011

I also saved the alsa registers before and after suspend/resume:
"cat /proc/asound/card0/codec97#0/ac97#0-0+regs > file"
and did a diff on the 2 files, there was no difference.

On Wed, Jul 13, 2011 at 3:01 PM, Shane Volpe <shanevolpe at gmail.com> wrote:
> My embedded system's audio is not resuming after the system resumes
> from suspend to RAM, below is a detailed discription of the problem
> and what I have done, to date, to identify the source of the problem.
> I have tested the following scenario with several audio applications
> such as Alsa aplay and ogg123:
> If I have an audio player (aplay or ogg123) using Alsa and I suspend
> my system to RAM when the system wakes up (resumes) there is no sound.
>  If I kill the audio player and start it back up the sound returns so
> it seems the act of opening and closing Alsa fixes the failure. During
> the "no sound state" I also opened alsamixer and adjusted
> (successfully) several alsa settings including volume but the sound
> still did not resume until I restarted the player.
> I also configured one player to use the OSS (emulated by Alsa).  When
> I used the emulated OSS audio driver (/dev/dsp) the sound did resume
> after the system woke up (resumed) from memory so everything WORKED! I
> then configured the player back to use the native Alsa driver and
> again there was no sound after resume.
> Has anyone seen this issue before or does anyone have additional tests
> or debugging ideas they can recommend to further narrow down the
> issue.  I'm very comfortable with modifying and compiling the Kernel
> so please do not limit your suggestions to user-space tweaks.
> Embedded System Information:
> Processor: ARM/Xscale (PXA270)
> Audio Codec: AC97 (UCB1400)
> alsa lib version:
> Linux Kernel Version:
> Regards,
> Shane

More information about the Alsa-devel mailing list