2006-05-28 16:40:19

by Dave Jones

[permalink] [raw]
Subject: 2.6.17rc emu10k1 regression.

$ lspci | grep audio
00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02)
06:0d.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 04)

(That's 06:0d.1 0980: 1102:7002 (rev 01))

This worked fine in 2.6.16, however when I try a .17rc kernel I get this..

cannot find the slot for index 0 (range 0-1)
EMU10K1_Audigy: probe of 0000:06:0d.0 failed with error -12

(Unremarkable, considering it *isn't* an Audigy)


modprobe.conf has ..
alias snd-card-0 snd-emu10k1
options snd-card-0 index=0
options snd-emu10k1 index=0
remove snd-emu10k1 { /usr/sbin/alsactl store 0 >/dev/null 2>&1 || : ; }; /sbin/modprobe -r --ignore-remove snd-emu10k1

What changed ?

Dave

--
http://www.codemonkey.org.uk


2006-05-28 16:51:17

by James Courtier-Dutton

[permalink] [raw]
Subject: Re: [Alsa-devel] 2.6.17rc emu10k1 regression.

Dave Jones wrote:
> $ lspci | grep audio
> 00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02)
> 06:0d.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 04)
>
> (That's 06:0d.1 0980: 1102:7002 (rev 01))
>
> This worked fine in 2.6.16, however when I try a .17rc kernel I get this..
>
> cannot find the slot for index 0 (range 0-1)
> EMU10K1_Audigy: probe of 0000:06:0d.0 failed with error -12
>
> (Unremarkable, considering it *isn't* an Audigy)
>
>

The same driver is used for the EMU10K1 and the Prodigy.
the -12 is an "Out of memory", so something outside the alas driver tree
must have changed to cause this problem.

James

2006-05-28 16:53:14

by James Courtier-Dutton

[permalink] [raw]
Subject: Re: [Alsa-devel] 2.6.17rc emu10k1 regression.

James Courtier-Dutton wrote:
> The same driver is used for the EMU10K1 and the Prodigy.
Dam spell checker
s/Prodigy/Audigy/g


2006-05-28 16:53:48

by Dave Jones

[permalink] [raw]
Subject: Re: [Alsa-devel] 2.6.17rc emu10k1 regression.

On Sun, May 28, 2006 at 05:51:12PM +0100, James Courtier-Dutton wrote:

> > cannot find the slot for index 0 (range 0-1)
> > EMU10K1_Audigy: probe of 0000:06:0d.0 failed with error -12
> >
> > (Unremarkable, considering it *isn't* an Audigy)
> The same driver is used for the EMU10K1 and the Prodigy.
> the -12 is an "Out of memory", so something outside the alas driver tree
> must have changed to cause this problem.

I rebooted into exactly the same kernel, and everything magically
started working again. Puzzling. There were no other signs of memory
exhaustion.

Hmm. Will keep an eye on things to see if it happens again.

Dave

--
http://www.codemonkey.org.uk

2006-05-29 09:44:03

by Takashi Iwai

[permalink] [raw]
Subject: Re: 2.6.17rc emu10k1 regression.

At Sun, 28 May 2006 12:40:15 -0400,
Dave Jones wrote:
>
> $ lspci | grep audio
> 00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02)
> 06:0d.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 04)
>
> (That's 06:0d.1 0980: 1102:7002 (rev 01))
>
> This worked fine in 2.6.16, however when I try a .17rc kernel I get this..
>
> cannot find the slot for index 0 (range 0-1)
>
> EMU10K1_Audigy: probe of 0000:06:0d.0 failed with error -12
>
> (Unremarkable, considering it *isn't* an Audigy)
>
>
> modprobe.conf has ..
> alias snd-card-0 snd-emu10k1
> options snd-card-0 index=0
> options snd-emu10k1 index=0
> remove snd-emu10k1 { /usr/sbin/alsactl store 0 >/dev/null 2>&1 || : ; }; /sbin/modprobe -r --ignore-remove snd-emu10k1
>
> What changed ?

Nothing from ALSA side. Maybe the module loading mechanism (or slot
order) was changed on your system.

The scenario is:

intel8x0 driver is first loaded on the first empty slot (index=0)
before snd-emu10k1. Since you set index=0 module option for
snd-emu10k1, it tried to load on the same slot but failed.

A workaround is to set index=1 option for snd-intel8x0 in addition,
for using emu10k1 as the primary card.


Takashi

2006-05-29 14:50:43

by Dave Jones

[permalink] [raw]
Subject: Re: 2.6.17rc emu10k1 regression.

On Mon, May 29, 2006 at 11:43:56AM +0200, Takashi Iwai wrote:

> > modprobe.conf has ..
> > alias snd-card-0 snd-emu10k1
> > options snd-card-0 index=0
> > options snd-emu10k1 index=0
> > remove snd-emu10k1 { /usr/sbin/alsactl store 0 >/dev/null 2>&1 || : ; }; /sbin/modprobe -r --ignore-remove snd-emu10k1
> >
> > What changed ?
>
> Nothing from ALSA side. Maybe the module loading mechanism (or slot
> order) was changed on your system.
>
> The scenario is:
>
> intel8x0 driver is first loaded on the first empty slot (index=0)
> before snd-emu10k1. Since you set index=0 module option for
> snd-emu10k1, it tried to load on the same slot but failed.
>
> A workaround is to set index=1 option for snd-intel8x0 in addition,
> for using emu10k1 as the primary card.

The modprobe.conf also had an index=1 entry already for the intel driver.

Dave
--
http://www.codemonkey.org.uk

2006-05-29 14:53:21

by Takashi Iwai

[permalink] [raw]
Subject: Re: 2.6.17rc emu10k1 regression.

At Mon, 29 May 2006 10:50:31 -0400,
Dave Jones wrote:
>
> On Mon, May 29, 2006 at 11:43:56AM +0200, Takashi Iwai wrote:
>
> > > modprobe.conf has ..
> > > alias snd-card-0 snd-emu10k1
> > > options snd-card-0 index=0
> > > options snd-emu10k1 index=0
> > > remove snd-emu10k1 { /usr/sbin/alsactl store 0 >/dev/null 2>&1 || : ; }; /sbin/modprobe -r --ignore-remove snd-emu10k1
> > >
> > > What changed ?
> >
> > Nothing from ALSA side. Maybe the module loading mechanism (or slot
> > order) was changed on your system.
> >
> > The scenario is:
> >
> > intel8x0 driver is first loaded on the first empty slot (index=0)
> > before snd-emu10k1. Since you set index=0 module option for
> > snd-emu10k1, it tried to load on the same slot but failed.
> >
> > A workaround is to set index=1 option for snd-intel8x0 in addition,
> > for using emu10k1 as the primary card.
>
> The modprobe.conf also had an index=1 entry already for the intel driver.

Hmm, then something like a webcam with an audio device?


Takashi