2.6.19-rc1: hda_intel: azx_get_response timeout, switching to polling mode...

2.6.19-rc1: hda_intel: azx_get_response timeout, switching to polling mode...

[Prakash Punnoor]

[-- Attachment #1: Type: text/plain, Size: 941 bytes --]

Hi,

I didn't get above message with 2.6.18. Usign irqpoll above message doesn't 
appear, but I think neither it optimal.

The kernel is patched with reiser4 and acpi_skip_timer_override quirk is 
deactivated (see last link why).

I tried different combinations (dmesg + .config). Differences are mostly pci 
mt init, irqpoll, nforce eth napi, pata/ide amd driver

http://www.prakash.gmxhome.de/linux/2.6.19-rc1-1.txt.bz2
http://www.prakash.gmxhome.de/linux/2.6.19-rc1-2.txt.bz2
http://www.prakash.gmxhome.de/linux/2.6.19-rc1-3.txt.bz2
http://www.prakash.gmxhome.de/linux/2.6.19-rc1-4.txt.bz2

cat /proc/interrupts for 2.6.18 resp .19-rc1
http://www.prakash.gmxhome.de/linux/irqs18.txt
http://www.prakash.gmxhome.de/linux/irqs19.txt

lspci can be found here:
http://marc.theaimsgroup.com/?l=linux-kernel&m=115545986619977&w=2

Cheers,
-- 
(°=                 =°)
//\ Prakash Punnoor /\\
V_/                 \_V

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.19-rc1: hda_intel: azx_get_response timeout, switching to polling mode...

[Takashi Iwai]

At Thu, 5 Oct 2006 09:54:57 +0200,
Prakash Punnoor wrote:
> 
> Hi,
> 
> I didn't get above message with 2.6.18. Usign irqpoll above message doesn't 
> appear, but I think neither it optimal.
> 
> The kernel is patched with reiser4 and acpi_skip_timer_override quirk is 
> deactivated (see last link why).
> 
> I tried different combinations (dmesg + .config). Differences are mostly pci 
> mt init, irqpoll, nforce eth napi, pata/ide amd driver
> 
> http://www.prakash.gmxhome.de/linux/2.6.19-rc1-1.txt.bz2
> http://www.prakash.gmxhome.de/linux/2.6.19-rc1-2.txt.bz2
> http://www.prakash.gmxhome.de/linux/2.6.19-rc1-3.txt.bz2
> http://www.prakash.gmxhome.de/linux/2.6.19-rc1-4.txt.bz2
> 
> cat /proc/interrupts for 2.6.18 resp .19-rc1
> http://www.prakash.gmxhome.de/linux/irqs18.txt
> http://www.prakash.gmxhome.de/linux/irqs19.txt
> 
> lspci can be found here:
> http://marc.theaimsgroup.com/?l=linux-kernel&m=115545986619977&w=2

The latest snd-hda-intel driver uses MSI as default.
Pass disable_msi=1 module option and see whether it works.

BTW, the switching to polling mode is relatively harmless (that's why
it's KERN_WARNING) in comparison with the single_cmd mode that the
driver uses when polling mode failed too.


Takashi

Re: [Alsa-devel] 2.6.19-rc1: hda_intel: azx_get_response timeout, switching to polling mode...

[Takashi Iwai]

At Thu, 5 Oct 2006 23:11:35 +0200,
Prakash Punnoor wrote:
> 
> Am Donnerstag 05 Oktober 2006 12:18 schrieb Takashi Iwai:
> > At Thu, 5 Oct 2006 09:54:57 +0200,
> > Prakash Punnoor wrote:
> > > I didn't get above message with 2.6.18. Usign irqpoll above message
> > > doesn't appear, but I think neither it optimal.
> >
> > The latest snd-hda-intel driver uses MSI as default.
> > Pass disable_msi=1 module option and see whether it works.
> 
> Yes, no more message, but what is even better: My nforce nic works again (w/o 
> using irqpoll)! So the alsa driver broke it. Now both are using same irq and 
> peaceful again.

It's more likely a problem of IRQ routing than the driver itself...

> (Though Alsa still reports that no soundcard was detetcted.)

It's a matter of PCI-multi probe, I guess, too.  The message comes
from sound/last.c, and it assumes that the probe has been called for
all sound devices while it isn't the case with PCI multi probe.


Takashi

Re: [Alsa-devel] 2.6.19-rc1: hda_intel: azx_get_response timeout, switching to polling mode...

[Prakash Punnoor]

[-- Attachment #1: Type: text/plain, Size: 914 bytes --]

Am Freitag 06 Oktober 2006 12:58 schrieb Takashi Iwai:
> At Thu, 5 Oct 2006 23:11:35 +0200,
>
> Prakash Punnoor wrote:
> > Am Donnerstag 05 Oktober 2006 12:18 schrieb Takashi Iwai:
> > > At Thu, 5 Oct 2006 09:54:57 +0200,
> > >
> > > Prakash Punnoor wrote:
> > > > I didn't get above message with 2.6.18. Usign irqpoll above message
> > > > doesn't appear, but I think neither it optimal.
> > >
> > > The latest snd-hda-intel driver uses MSI as default.
> > > Pass disable_msi=1 module option and see whether it works.
> >
> > Yes, no more message, but what is even better: My nforce nic works again
> > (w/o using irqpoll)! So the alsa driver broke it. Now both are using same
> > irq and peaceful again.
>
> It's more likely a problem of IRQ routing than the driver itself...

Well, whom should I contact then?
-- 
(°=                 =°)
//\ Prakash Punnoor /\\
V_/                 \_V

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]