From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20041111 Firefox/1.0 Description of problem: In /etc/sysconfig/apm-scripts/apmscript there is fault when finding sound module. Script looks for modules with '-' in the name while at least my sound module uses '_' in the name. Here's how I got it finding my module: $ diff apmscript apmscript-new 174c174 < SOUNDMODULES=`modprobe -c | awk '/^alias sound-slot/ { print $3 }'` --- > SOUNDMODULES=`modprobe -c | awk '/^alias snd-card/ { print $3 }'| tr - _` and here's what those commands output from commandline: 1. original prints nothing: $ /sbin/modprobe -c | awk '/^alias sound-slot/ { print $3 }' 2. changed finds the module: $ /sbin/modprobe -c | awk '/^alias snd-card/ { print $3 }' | tr - _ snd_cs46xx OK, thats the part finding the module. Then the second problem is apmd does not find sound programs. It tries to look for /dev/sndtat by using command: cat /dev/sndstat 2>/dev/null| LC_ALL=C grep Open\ by|awk {'print $3'}|uniq First of all, there is no such file (/dev/sndstat) after having udev. But even if linking /proc/asound/oss/sndstat to the /dev/sndstat does not help. There are no such lines in sndstat file as "Open by". I tried playing sound with xmms using alsa and using oss emulation. No difference. So unloading the driver module does not succeed because soundprocs are not killed first. This all comes to problem that sound does not work after suspend and resume cyckle. kernel-2.6.9-1.681_FC3, machine is IBM T20 laptop Version-Release number of selected component (if applicable): apmd-3.0.2-24 How reproducible: Always Steps to Reproduce: 1. listen music via xmms 2. press suspend button on keyboard 3. wake up the machine Actual Results: sound does not work after resume, the sound driver does not get unloaded & reloaded Expected Results: music (or any sound) should just continue Additional info:
Just wanted to note that modifying apmscript as the reporter describes fixes the problem on my laptop (FC3, ThinkPad T21, kernel 2.6.10-1.741_FC3, acpi=off). Sound comes back after suspend/resume just fine. I haven't tried suspending while sound programs are running, so it's possible that the missing /dev/sndstat is also still a problem on my system.
SOUNDMODULES grep fixed in 3.2.2-1. The sndstat thing will require more work, and is deferred for the moment.
I just upgraded to 3.2.2-1 on my Thinkpad T21, FC3, kernel 2.6.10-1.770_FC3, and this did not fix the problem. On suspend, sound still doesn't work. If I try to play a sound, I get messages of the following kind: ALSA lib pcm_hw.c:521:(snd_pcm_hw_prepare) SNDRV_PCM_IOCTL_PREPARE failed: Device or resource busy ALSA lib pcm_hw.c:521:(snd_pcm_hw_prepare) SNDRV_PCM_IOCTL_PREPARE failed: Device or resource busy ALSA lib pcm_hw.c:521:(snd_pcm_hw_prepare) SNDRV_PCM_IOCTL_PREPARE failed: Device or resource busy ALSA lib pcm_hw.c:521:(snd_pcm_hw_prepare) SNDRV_PCM_IOCTL_PREPARE failed: Device or resource busy ALSA lib pcm_hw.c:521:(snd_pcm_hw_prepare) SNDRV_PCM_IOCTL_PREPARE failed: Device or resource busy Any thoughts? Should this bug be re-opened?
It just occurred to me that I'm probably being an idiot. I use ACPI for suspend and resume, so I imagine a fix to apmd won't do me much good. My apologies.