[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#531090: linux-image-2.6.29-2-486: opl3sa2 sound fails after suspend/resume or rmmod/insmod on toshiba libretto 110 ct



On Fri, 29 May 2009, Friedrich Delgado Friedrichs wrote:
> 
> When I use hardware suspend (via apm -s) on my toshiba libretto 110
> ct, the soundcard is unusable after resume.
> 
> This used to work in kernel 2.6.22, however I upgraded a lot of
> packages in one go (I didn't use the laptop for some time), so it
> might be another package that's causing this.
> 
> alsamixer starts and displays levels which I can change, but any
> soundplayer will hang for a few seconds and then terminate with an
> error (as if the sound device is blocked, but there's no process
> blocking any sound device).
> 
> Sound works again only after reboot.
> 
> Unloading alsa (with alsa unload) and reloading opl3sa2 doesn't help,
> but produces the following error after unloading:
> 
> May 28 22:57:40 gimli kernel: [13098.270456] Trying to free already-free IRQ 5
> May 28 22:57:40 gimli kernel: [13098.278892] snd-opl3sa2-pnpbios 00:16: disabled
> 
> And when I try to modprobe snd-opl3sa2, I get the following error:
> 
> May 28 22:57:51 gimli kernel: [13109.846896] snd-opl3sa2-pnpbios 00:16: activated
> May 28 22:57:51 gimli kernel: [13109.847126] opl3sa2: can't grab IRQ 5
> May 28 22:57:51 gimli kernel: [13109.858891] snd-opl3sa2-pnpbios 00:16: disabled
> 
> After failed modprobe, there are no sound devices at all any more,
> only /dev/snd/timer and /dev/snd/seq are present.
> 
> /etc/apm/suspend.d/80alsa and /etc/apm/suspend.d/80alsa contain:
> 
> #!/bin/sh
> #
> # apmd proxy script for ALSA
> 
> [ -x /usr/sbin/alsactl ] || exit 0
> 
> case "$1,$2" in
>         suspend,*) /usr/sbin/alsactl store && /usr/sbin/alsa suspend ;;
>         resume,suspend) /usr/sbin/alsa resume && /usr/sbin/alsactl restore ;;
> esac
> 
> The notebook is *not* acpi compatible, so using acpi is not an
> alternative. It may work with s2disk or s2ram (I never tried), but
> that would probably be very slow compared to hardware suspend.
> 
> On a personal note: This is a bit of a bummer since I intended to
> revive the libretto as networked music player in my living room ;-}
> 
> Please don't hesitate to ask for clarification, if needed.
> Kind regards
>      FDF
> 

if you can still reproduce with Squeeze 2.6.32 linux image
please notify upstream at bugzilla.kernel.org and let us know
the bug nr so that we can track it.

squeeze is scheduled to be released with 2.6.32 and this kernel
will have long stable support, so it is a worthwile target to test.

thanks for the report.



Reply to: