2.6.33.1 regression: sound splutters

2.6.33.1 regression: sound splutters

[Harald Dunkel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi folks,

With 2.6.33.1 and 2.6.34-rc2 sound splutters for my PC. Moving
back to 2.6.29.8 makes the problem go away.

lspci:
00:1b.0 Audio device: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) High Definition Audio Controller (rev 04)
        Subsystem: AOPEN Inc. Device 0575
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 32 bytes
        Interrupt: pin A routed to IRQ 18
        Region 0: Memory at d4340000 (64-bit, non-prefetchable) [size=16K]
        Capabilities: <access denied>
        Kernel driver in use: HDA Intel

Can anybody reproduce this? Please mail if I can help to track
this down.


Regards

Harri
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkux40sACgkQUTlbRTxpHjcXZwCeMyvXvah8HDeRd0h6luGSP5Yy
b5kAniUhFJehUzHZGIashlGPGzw6kIW9
=tSbA
-----END PGP SIGNATURE-----

Re: 2.6.33.1 regression: sound splutters

[Takashi Iwai]

At Tue, 30 Mar 2010 13:40:59 +0200,
Harald Dunkel wrote:
> 
> Hi folks,
> 
> With 2.6.33.1 and 2.6.34-rc2 sound splutters for my PC. Moving
> back to 2.6.29.8 makes the problem go away.

Hm, it's a bit old to bisect.  What about 2.6.32.*?

> lspci:
> 00:1b.0 Audio device: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) High Definition Audio Controller (rev 04)
>         Subsystem: AOPEN Inc. Device 0575
>         Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
>         Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
>         Latency: 0, Cache Line Size: 32 bytes
>         Interrupt: pin A routed to IRQ 18
>         Region 0: Memory at d4340000 (64-bit, non-prefetchable) [size=16K]
>         Capabilities: <access denied>
>         Kernel driver in use: HDA Intel
> 
> Can anybody reproduce this? Please mail if I can help to track
> this down.

First try the freshly released 2.6.34-rc3.  If the problem persists,
run alsa-info.sh with --no-upload option, and attach the generated file
for further analysis.
The script is found in www.alsa-project.org.


thanks,

Takashi

Re: 2.6.33.1 regression: sound splutters

[Harald Dunkel]

[-- Attachment #1.1: Type: text/plain, Size: 584 bytes --]

On 03/30/10 20:23, Takashi Iwai wrote:
> At Tue, 30 Mar 2010 13:40:59 +0200,
> Harald Dunkel wrote:
>>
>> With 2.6.33.1 and 2.6.34-rc2 sound splutters for my PC. Moving
>> back to 2.6.29.8 makes the problem go away.
> 
> Hm, it's a bit old to bisect.  What about 2.6.32.*?
> 

Sorry, that was a typo. The good version is 2.6.32.8.

> 
> First try the freshly released 2.6.34-rc3.  If the problem persists,
> run alsa-info.sh with --no-upload option, and attach the generated file
> for further analysis.

Same problem with rc3. See attachment.


Regards

Harri

[-- Attachment #1.2: alsa-info.txt.gz --]
[-- Type: application/x-tar, Size: 5367 bytes --]

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 262 bytes --]

Re: 2.6.33.1 regression: sound splutters

[Takashi Iwai]

At Thu, 01 Apr 2010 11:30:21 +0200,
Harald Dunkel wrote:
> 
> On 03/30/10 20:23, Takashi Iwai wrote:
> > At Tue, 30 Mar 2010 13:40:59 +0200,
> > Harald Dunkel wrote:
> >>
> >> With 2.6.33.1 and 2.6.34-rc2 sound splutters for my PC. Moving
> >> back to 2.6.29.8 makes the problem go away.
> > 
> > Hm, it's a bit old to bisect.  What about 2.6.32.*?
> > 
> 
> Sorry, that was a typo. The good version is 2.6.32.8.

OK, what about enable_msi=0 option for snd-hda-intel module?


Takashi

Re: 2.6.33.1 regression: sound splutters

[Harald Dunkel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 04/01/10 11:39, Takashi Iwai wrote:
> 
> OK, what about enable_msi=0 option for snd-hda-intel module?
> 

Now it works.

I have no idea what this enable_msi thing is about, but according
to Documentation/sound/alsa/ALSA-Configuration.txt it should be off
by default. Would you suggest to make this change permanent?


Regards

Harri
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAku0sywACgkQUTlbRTxpHjcevACfWThvjFATNwcA1q5vtz4LOVFz
JF8An3/+2d3OGmrJoEOJWEOY7n7If/GX
=sJ1q
-----END PGP SIGNATURE-----

Re: 2.6.33.1 regression: sound splutters

[Takashi Iwai]

At Thu, 01 Apr 2010 16:52:28 +0200,
Harald Dunkel wrote:
> 
> On 04/01/10 11:39, Takashi Iwai wrote:
> > 
> > OK, what about enable_msi=0 option for snd-hda-intel module?
> > 
> 
> Now it works.
> 
> I have no idea what this enable_msi thing is about, but according
> to Documentation/sound/alsa/ALSA-Configuration.txt it should be off
> by default.

This was enabled as default but for Nvidia controller since 2.6.33.
I'll need to update the document, too...

> Would you suggest to make this change permanent?

Apply the patch like below.  Replace "Unknown PC" with something more
useful.

Could you tell more details about your machine, which vendor, product
name etc to fill there?  Then I'll commit the fix to the upstream.


thanks,

Takashi

---
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index b185afd..f45ab94 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2366,6 +2366,7 @@ static struct snd_pci_quirk msi_black_list[] __devinitdata = {
 	SND_PCI_QUIRK(0x1043, 0x81f6, "ASUS", 0), /* nvidia */
 	SND_PCI_QUIRK(0x1043, 0x822d, "ASUS", 0), /* Athlon64 X2 + nvidia MCP55 */
 	SND_PCI_QUIRK(0x1849, 0x0888, "ASRock", 0), /* Athlon64 X2 + nvidia */
+	SND_PCI_QUIRK(0xa0a0, 0x0575, "Unknown PC", 0), /* Unknown PC */
 	{}
 };
 

Re: 2.6.33.1 regression: sound splutters

[Harald Dunkel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 04/01/10 17:01, Takashi Iwai wrote:
> 
> Could you tell more details about your machine, which vendor, product
> name etc to fill there?  Then I'll commit the fix to the upstream.
> 

The PC is a (pretty old) Aopen MZ915-M, made in 2006, AFAIR. See
the specs on http://global.aopen.com/products_detail.aspx?Auno=2170 .


Many thanx for your help

Harri
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAku0zjYACgkQUTlbRTxpHjdy/wCdG8PhFqQzVyGmiiDxDR4usitE
wNsAn1nMYByvdkUFaV/KnLiscQZHm1sK
=x78t
-----END PGP SIGNATURE-----

Re: 2.6.33.1 regression: sound splutters

[Takashi Iwai]

At Thu, 01 Apr 2010 18:47:50 +0200,
Harald Dunkel wrote:
> 
> On 04/01/10 17:01, Takashi Iwai wrote:
> > 
> > Could you tell more details about your machine, which vendor, product
> > name etc to fill there?  Then I'll commit the fix to the upstream.
> > 
> 
> The PC is a (pretty old) Aopen MZ915-M, made in 2006, AFAIR. See
> the specs on http://global.aopen.com/products_detail.aspx?Auno=2170 .

Thanks.  I added the device to the quirk list now.


Takashi