hda-intel (ALC889) - sound loops/repeats continuously, and general brokenness

hda-intel (ALC889) - sound loops/repeats continuously, and general brokenness

[Juliano F. Ravasi]

 Hello ALSA devs,

(Cc'ing alsa-devel since looks like it can be a driver problem/bug, but
still unsure. Need some expert advice.)

I'm experiencing a lot of problems reproducing audio with an HD Audio
"card" integrated on an Intel DX58SO mainboard. It happens only on this
card, everything is Ok with another (USB) card.

The most visible problem is that `aplay` plays only a small fragment of
audio (less than one second), then hangs and the card keeps reproducing
that fragment continuously in a loop, until killed with Ctrl-C. `aplay
-N` progresses, but each fragment repeats 3 times.

Doing an strace on `aplay`, it hangs here while repeating the audio
fragment:

...
semop(294912, {{0, 0, 0}, {0, 1, SEM_UNDO}}, 2) = 0
semop(294912, {{0, -1, SEM_UNDO|IPC_NOWAIT}}, 1) = 0
ioctl(4, 0x4122, 0x3)                   = 0
ioctl(4, 0x4122, 0)                     = 0
ioctl(4, 0x4122, 0xe4)                  = 0
poll([{fd=3, events=POLLIN|POLLERR|POLLNVAL}], 1, -1) = 1 ([{fd=3,
revents=POLLIN}])
ioctl(4, 0x4122, 0x3)                   = 0
ioctl(4, 0x4122, 0xe4)                  = 0
ioctl(4, 0x4122, 0x1)                   = 0
read(3,
"\6\0\0\0\1\210\377\377\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\1\210\377\377"...,
128) = 64
read(3, 0x7fff890f6810, 128)            = -1 EAGAIN (Resource
temporarily unavailable)
poll([{fd=3, events=POLLIN|POLLERR|POLLNVAL}], 1, -1^C <unfinished ...>

Descriptor 3 is /dev/snd/timer ... Perhaps time stops progressing for
the driver? `aplay -N` continues a little here because this poll() call
timeouts after 1000ms, but this still repeats each fragment 3 times.

Other audio software (mplayer, amarok) can play audio almost normally,
except when you pause. When paused, the same behavior happens: the last
fragment of sound before pausing keeps repeating continuously until
unpaused.

When the song finishes, the application (tested with mplayer) also
hangs, the last fragment keeps repeating while the process is blocked here:

ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {B38400 opost isig -icanon -echo
...}) = 0
ioctl(0, SNDCTL_TMR_START or TCSETS, {B38400 opost isig icanon echo
...}) = 0
ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {B38400 opost isig icanon echo
...}) = 0
ioctl(6, 0x4144^C <unfinished ...>

Descriptor 6 is /dev/snd/pcmC0D0p , and 0x4144 is SNDRV_PCM_IOCTL_DRAIN
ioctl in sound/asound.h. Sometimes mplayer timeouts and reports:

[AO_ALSA] alsa-lib: pcm_hw.c:616:(snd_pcm_hw_drain)
SNDRV_PCM_IOCTL_DRAIN failed (-5): Input/output error

I'm at a loss here. I have already read the sound/alsa/HD-Audio.txt and
tried many of the tweaks described there, but with no luck.

Any help is greatly appreciated.


System information:
http://www.alsa-project.org/db/?f=4ab249b0f301ff20260d7388432e624aa65092d6


Best regards,
Juliano.

-- 
Juliano F. Ravasi ·· http://juliano.info/
5105 46CC B2B7 F0CD 5F47 E740 72CA 54F4 DF37 9E96


------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in  U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store 
http://p.sf.net/sfu/nokia-dev2dev

Re: hda-intel (ALC889) - sound loops/repeats continuously, and general brokenness

[Soeren D. Schulze]

On 20.10.2010 21:40, Juliano F. Ravasi wrote:
>  Hello ALSA devs,
> 
> (Cc'ing alsa-devel since looks like it can be a driver problem/bug, but
> still unsure. Need some expert advice.)
> 
> I'm experiencing a lot of problems reproducing audio with an HD Audio
> "card" integrated on an Intel DX58SO mainboard. It happens only on this
> card, everything is Ok with another (USB) card.
> 
> The most visible problem is that `aplay` plays only a small fragment of
> audio (less than one second), then hangs and the card keeps reproducing
> that fragment continuously in a loop, until killed with Ctrl-C. `aplay
> -N` progresses, but each fragment repeats 3 times.

[...]

As a matter of accident, I reported a similar bug to the Ubuntu
bugtracker just 30 minutes afterwards:

http://bugs.launchpad.net/ubuntu/+source/linux/+bug/664138

The error is not as bad for me, but it might be related.


Thanks,

Sören

Re: hda-intel (ALC889) - sound loops/repeats continuously, and general brokenness

[Clemens Ladisch]

Juliano F. Ravasi wrote:
> I'm experiencing a lot of problems reproducing audio with an HD Audio
> "card" integrated on an Intel DX58SO mainboard. It happens only on this
> card, everything is Ok with another (USB) card.
> 
> The most visible problem is that `aplay` plays only a small fragment of
> audio (less than one second), then hangs and the card keeps reproducing
> that fragment continuously in a loop, until killed with Ctrl-C. `aplay
> -N` progresses, but each fragment repeats 3 times.

Please check in /proc/interrupts whether that interrupt actually
works.


Regards,
Clemens

Re: [alsa-devel] hda-intel (ALC889) - sound loops/repeats continuously, and general brokenness

[Juliano F. Ravasi]

On 10/21/2010 01:41 PM, Clemens Ladisch wrote:
> Please check in /proc/interrupts whether that interrupt actually
> works.

There is no interrupt at all being assigned to hda_intel.

Yesterday, all of a sudden, the problem fixed itself and everything
about this driver started working as expected. I restarted the driver to
be sure and all problems returned. Then I noticed these two lines in the
log at the time when the problem fixed itself:

hda-intel: No response from codec, disabling MSI: last cmd=0x20cb2001
hda_intel: azx_get_response timeout, switching to single_cmd mode: last
cmd=0x20cb2001

So, I added to my modprobe.conf:

	options snd-hda-intel enable_msi=0 single_cmd=1

And now everything works fine and there is an hda_intel line in
/proc/interrupts. I don't know what are the implications of these
parameters. Particularly single_cmd seems to flag an incorrect operation
of the driver. Perhaps the driver should detect and blacklist this card
from MSI?

Thank you very much for your help!

Regards,
Juliano.


-- 
Juliano F. Ravasi ·· http://juliano.info/
5105 46CC B2B7 F0CD 5F47 E740 72CA 54F4 DF37 9E96

------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in  U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store 
http://p.sf.net/sfu/nokia-dev2dev