LinuxLists.cc - [ALSA PATCH] alsa-git merge request

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

At Thu, 17 Jul 2008 19:06:57 +0200,
Ingo Molnar wrote:
>
>
> hi Jaroslav, et al,
>
> * Jaroslav Kysela <[email protected]> wrote:
>
> > ALSA: opti93x: use cs4231 lib
>
> -tip testing found the following bootup crash on latest -git:
>
> [ 44.827459] calling alsa_card_opti9xx_init+0x0/0x20
> [ 44.830435] bus: 'isa': add driver opti93x
> [ 44.833503] device: 'opti93x.0': device_add
> [ 44.837804] bus: 'isa': add device opti93x.0
> [ 44.841820] bus: 'isa': driver_probe_device: matched device opti93x.0 with driver opti93x
> [ 44.845327] bus: 'isa': really_probe: probing driver opti93x with device opti93x.0
> [ 44.851601] BUG: unable to handle kernel NULL pointer dereference at 00000010
> [ 44.855329] IP: [<786c0782>] snd_card_opti9xx_free+0x12/0x40
> [ 44.859370] *pde = 00000000
> [ 44.862651] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
> [ 44.862651]
> [ 44.862651] Pid: 1, comm: swapper Not tainted (2.6.26-00085-g9f240a5-dirty #20182)
> [ 44.862651] EIP: 0060:[<786c0782>] EFLAGS: 00010286 CPU: 0
> [ 44.862651] EIP is at snd_card_opti9xx_free+0x12/0x40
> [ 44.862651] EAX: 96892000 EBX: 00000000 ECX: 96892140 EDX: 786c0770
> [ 44.862651] ESI: 9689229c EDI: 9689229c EBP: 9782fe10 ESP: 9782fe08
> [ 44.862651] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
> [ 44.862651] Process swapper (pid: 1, ti=9782f000 task=97848000 task.ti=9782f000)
> [ 44.862651] Stack: 96892000 00000286 9782fe1c 786a2b5b 96892000 9782fe48 786a349d 00000000
> [ 44.862651] 00000286 78b34028 9688a100 00000286 9689229c fffff000 00000286 9689229c
> [ 44.862651] 9782fe84 787f4b18 968922c0 9689229f 9688dc00 96892000 00000004 00000000
> [ 44.862651] Call Trace:
> [ 44.862651] [<786a2b5b>] ? snd_card_do_free+0x4b/0x120
> [ 44.862651] [<786a349d>] ? snd_card_free+0x7d/0x90
> [ 44.862651] [<787f4b18>] ? snd_opti9xx_isa_probe+0x158/0xa30
> [ 44.862651] [<787f49c0>] ? snd_opti9xx_isa_probe+0x0/0xa30
> [ 44.862651] [<7832580d>] ? isa_bus_probe+0x1d/0x30
> [ 44.862651] [<783229c0>] ? driver_probe_device+0xa0/0x1c0
> [ 44.862651] [<78322b58>] ? __device_attach+0x8/0x10
> [ 44.862651] [<78321d60>] ? bus_for_each_drv+0x40/0x80
> [ 44.862651] [<78322c04>] ? device_attach+0x84/0xa0
> [ 44.862651] [<78322b50>] ? __device_attach+0x0/0x10
> [ 44.862651] [<78321b97>] ? bus_attach_device+0x47/0x50
> [ 44.862651] [<78320877>] ? device_add+0x3a7/0x500
> [ 44.862651] [<783209e2>] ? device_register+0x12/0x20
> [ 44.862651] [<783259c2>] ? isa_register_driver+0xa2/0x120
> [ 44.862651] [<78153631>] ? ktime_get+0x11/0x30
> [ 44.862651] [<78b78852>] ? alsa_card_opti9xx_init+0x12/0x20
> [ 44.862651] [<78b3d679>] ? kernel_init+0x1d9/0x300
> [ 44.862651] [<78b78840>] ? alsa_card_opti9xx_init+0x0/0x20
> [ 44.862651] [<781192ce>] ? syscall_exit+0x8/0x28
> [ 44.862651] [<78b3d4a0>] ? kernel_init+0x0/0x300
> [ 44.862651] [<78b3d4a0>] ? kernel_init+0x0/0x300
> [ 44.862651] [<7811a4e7>] ? kernel_thread_helper+0x7/0x10
> [ 44.862651] =======================
> [ 44.862651] Code: e8 8b 03 e8 11 e5 a9 ff 8b 03 5b 5d c3 b8 ff ff ff ff 5b 5d c3 8d 74 26 00 55 89 e5 56 53 8b b0 38 01 00 00 85 f6 74 21 8b 5e 1c <8b> 43 10 85 c0 7e 0f e8 32 5a aa ff 8b 43 10 89 da e8 68 5a aa
> [ 44.862651] EIP: [<786c0782>] snd_card_opti9xx_free+0x12/0x40 SS:ESP 0068:9782fe08
> [ 44.866056] Kernel panic - not syncing: Fatal exception
>
> i've bisected it down to:
>
> | 9f240a558f94bea839e14c46c8fccf93bdd35ef6 is first bad commit
> | commit 9f240a558f94bea839e14c46c8fccf93bdd35ef6
> | Author: Krzysztof Helt <[email protected]>
> | Date: Wed Jun 11 12:26:32 2008 +0200
> |
> | ALSA: opti93x: use cs4231 lib
>
> The bisection log is:
>
> # bad : [a3cf859f] Merge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6
> # good: [bce7f795] Linux 2.6.26
> # bad : [d1794f2a] Merge branch 'bkl-removal' of git://git.lwn.net/linux-2.6
> # good: [e30a44f3] x86: merge put_user.
> # bad : [6c118e4c] Merge git://git.kernel.org/pub/scm/linux/kernel/git/hskinnemoen/avr32-2.6
> # bad : [fe0a3fed] ALSA: Release v1.0.17
> # bad : [dbac7cbe] ALSA: ASoC: Add missing includes
> # good: [f2690025] [ALSA] hda - Add support of Teradici controller
> # good: [f10485eb] [ALSA] ASoC: Add WM8990 driver
> # good: [0a1b42d5] sound: sound/oss/dmasound/: cleanups
> # good: [a6a950a4] ALSA: hda - Add a warning if pending IRQ is found
> # bad : [d751b231] ALSA: ASoC: Fix register cache sizes for Wolfson codecs
> # bad : [9f240a56] ALSA: opti93x: use cs4231 lib
> # good: [abf1f5af] ALSA: opti93x: add support for Opti93x codec in cs4231-lib
>
> the patch below is a temporary workaround to keep my tests running.
>
> Can send more info about the system in question - it has ISA (it's an
> old Pentium-3-Coppermine based laptop) but otherwise it doesnt have
> opti93x hw so the crash should be reproducible elsewhere as well.

Could you try the patch below?

thanks,

Takashi

---
diff --git a/sound/isa/opti9xx/opti92x-ad1848.c b/sound/isa/opti9xx/opti92x-ad1848.c
index 41c047e..d20abb2 100644
--- a/sound/isa/opti9xx/opti92x-ad1848.c
+++ b/sound/isa/opti9xx/opti92x-ad1848.c
@@ -688,7 +688,7 @@ static void snd_card_opti9xx_free(struct snd_card *card)
if (chip) {
#ifdef OPTi93X
struct snd_cs4231 *codec = chip->codec;
- if (codec->irq > 0) {
+ if (codec && codec->irq > 0) {
disable_irq(codec->irq);
free_irq(codec->irq, codec);
}

2008-07-17 19:19:32

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

On 17-07-08 20:40, Takashi Iwai wrote:

> Could you try the patch below?

No, please hang on, it's much simpler. Above commit isn't actually
itself guilty. Have a patch ready, but am testing a bit.

Rene.

2008-07-17 19:37:43

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

On 17-07-08 21:20, Rene Herman wrote:

> On 17-07-08 20:40, Takashi Iwai wrote:
>
>> Could you try the patch below?
>
> No, please hang on, it's much simpler. Above commit isn't actually
> itself guilty. Have a patch ready, but am testing a bit.

I retract that. Your patch should fix things.

I would like to place something on top to make the init isa/pnp choice
look more generic but I'll submit that tomorrow.

Rene.

2008-07-18 08:45:15

by Ingo Molnar

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

* Takashi Iwai <[email protected]> wrote:

> > Can send more info about the system in question - it has ISA (it's
> > an old Pentium-3-Coppermine based laptop) but otherwise it doesnt
> > have opti93x hw so the crash should be reproducible elsewhere as
> > well.
>
> Could you try the patch below?

yeah, looks like it should fix the issue. Please consider this bug:

Tested-by: Ingo Molnar <[email protected]>
Acked-by: Ingo Molnar <[email protected]>

unless i mail about a new crash in the next 24 hours. I've queued up
your fix for testing. (and have removed the revert from the test setup)

Ingo

2008-07-18 09:17:29

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

>From fc64af262b44d41cd45fbe96be301e4662ba1c1a Mon Sep 17 00:00:00 2001
From: Rene Herman <[email protected]>
Date: Fri, 18 Jul 2008 11:15:12 +0200
Subject: [PATCH] ALSA: opti9xx: no isapnp param for !CONFIG_PNP

"isapnp" needs CONFIG_PNP to be useful.

Signed-off-by: Rene Herman <[email protected]>
---
sound/isa/opti9xx/opti92x-ad1848.c | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/sound/isa/opti9xx/opti92x-ad1848.c b/sound/isa/opti9xx/opti92x-ad1848.c
index e5d9028..5dfead9 100644
--- a/sound/isa/opti9xx/opti92x-ad1848.c
+++ b/sound/isa/opti9xx/opti92x-ad1848.c
@@ -68,7 +68,9 @@ MODULE_SUPPORTED_DEVICE("{{OPTi,82C924 (AD1848)},"
static int index = SNDRV_DEFAULT_IDX1; /* Index 0-MAX */
static char *id = SNDRV_DEFAULT_STR1; /* ID for this card */
//static int enable = SNDRV_DEFAULT_ENABLE1; /* Enable this card */
+#ifdef CONFIG_PNP
static int isapnp = 1; /* Enable ISA PnP detection */
+#endif
static long port = SNDRV_DEFAULT_PORT1; /* 0x530,0xe80,0xf40,0x604 */
static long mpu_port = SNDRV_DEFAULT_PORT1; /* 0x300,0x310,0x320,0x330 */
static long fm_port = SNDRV_DEFAULT_PORT1; /* 0x388 */
@@ -85,8 +87,10 @@ module_param(id, charp, 0444);
MODULE_PARM_DESC(id, "ID string for opti9xx based soundcard.");
//module_param(enable, bool, 0444);
//MODULE_PARM_DESC(enable, "Enable opti9xx soundcard.");
+#ifdef CONFIG_PNP
module_param(isapnp, bool, 0444);
MODULE_PARM_DESC(isapnp, "Enable ISA PnP detection for specified soundcard.");
+#endif
module_param(port, long, 0444);
MODULE_PARM_DESC(port, "WSS port # for opti9xx driver.");
module_param(mpu_port, long, 0444);
--
1.5.5

Attachments:

0001-ALSA-opti9xx-no-isapnp-param-for-CONFIG_PNP.patch (1.61 kB)

2008-07-18 12:24:44

by Krzysztof Helt

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

On Fri, 18 Jul 2008 11:19:19 +0200
Rene Herman <[email protected]> wrote:

> On 17-07-08 21:38, Rene Herman wrote:
>
> > On 17-07-08 21:20, Rene Herman wrote:
> >
> >> On 17-07-08 20:40, Takashi Iwai wrote:
> >>
> >>> Could you try the patch below?
> >>
>
> This bug was uncovered by !CONFIG_PNP and we had trouble in this driver
> due to that before so was rewriting the init to be a generic multi-card
> ALSA init but, actually, not much point.
>

The bug was introduced by the patch as it used (tried to free) the codec pointer
before it was assigned any value. It happened only if the card was missing and
I didn't test the driver on such a configuration. My mistake. The CONFIG_PNP
setting makes no difference.

>
> In fact, in the meantime I'll look at splitting 92x and 93x again. It's
> much better now that Krzysztof made it use the generic library code but
> I still hate this driver. It's also poking at ports without being told
> something is there by either system or user.
>

Please hold your axe until Monday. The last patch (not posted yet) from the
unification of the WSS library series removes some ifdefs in this driver.
This may (or may not) change your view on splitting the driver.

Regards,
Krzysztof

----------------------------------------------------------------------
Zobacz cala prawde o Lukaszu Podolskim!
kliknij >>> http://link.interia.pl/f1e57

2008-07-18 12:34:19

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

On 18-07-08 14:28, Krzysztof Helt wrote:

> On Fri, 18 Jul 2008 11:19:19 +0200
> Rene Herman <[email protected]> wrote:
>
>> On 17-07-08 21:38, Rene Herman wrote:
>>
>>> On 17-07-08 21:20, Rene Herman wrote:
>>>
>>>> On 17-07-08 20:40, Takashi Iwai wrote:
>>>>
>>>>> Could you try the patch below?
>> This bug was uncovered by !CONFIG_PNP and we had trouble in this driver
>> due to that before so was rewriting the init to be a generic multi-card
>> ALSA init but, actually, not much point.
>>
>
> The bug was introduced by the patch as it used (tried to free) the codec pointer
> before it was assigned any value. It happened only if the card was missing and
> I didn't test the driver on such a configuration. My mistake. The CONFIG_PNP
> setting makes no difference.

Yes it does. Note how the ISA match method fails due to "if (isapnp)"
with CONFIG_PNP and isapnp being initialized to 1. I did test without
the hardware but didn't see it fail due to that.

>> In fact, in the meantime I'll look at splitting 92x and 93x again. It's
>> much better now that Krzysztof made it use the generic library code but
>> I still hate this driver. It's also poking at ports without being told
>> something is there by either system or user.
>>
>
> Please hold your axe until Monday. The last patch (not posted yet) from the
> unification of the WSS library series removes some ifdefs in this driver.
> This may (or may not) change your view on splitting the driver.

Will do.

Rene.

2008-07-18 12:37:25

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

On 18-07-08 14:36, Rene Herman wrote:

> On 18-07-08 14:28, Krzysztof Helt wrote:
>
>> On Fri, 18 Jul 2008 11:19:19 +0200
>> Rene Herman <[email protected]> wrote:
>>
>>> On 17-07-08 21:38, Rene Herman wrote:
>>>
>>>> On 17-07-08 21:20, Rene Herman wrote:
>>>>
>>>>> On 17-07-08 20:40, Takashi Iwai wrote:
>>>>>
>>>>>> Could you try the patch below?
>>> This bug was uncovered by !CONFIG_PNP and we had trouble in this
>>> driver due to that before so was rewriting the init to be a generic
>>> multi-card ALSA init but, actually, not much point.
>>>
>>
>> The bug was introduced by the patch as it used (tried to free) the
>> codec pointer
>> before it was assigned any value. It happened only if the card was
>> missing and
>> I didn't test the driver on such a configuration. My mistake. The
>> CONFIG_PNP
>> setting makes no difference.
>
> Yes it does. Note how the ISA match method fails due to "if (isapnp)"
> with CONFIG_PNP and isapnp being initialized to 1. I did test without
> the hardware but didn't see it fail due to that.

(but please note, yes, "uncovered", not "caused". the bug was indeed in
this commit and was fixed by Takashi).

Rene.

2008-07-18 20:16:24

by Ingo Molnar

[permalink] [raw]

Subject: Re: [boot crash, bisected] in 9f240a55 "ALSA: opti93x: use cs4231 lib"

* Rene Herman <[email protected]> wrote:

> On 17-07-08 21:38, Rene Herman wrote:
>
>> On 17-07-08 21:20, Rene Herman wrote:
>>
>>> On 17-07-08 20:40, Takashi Iwai wrote:
>>>
>>>> Could you try the patch below?
>>>
>>> No, please hang on, it's much simpler. Above commit isn't actually
>>> itself guilty. Have a patch ready, but am testing a bit.
>>
>> I retract that. Your patch should fix things.
>
> It does. Acked/Tested-by: Rene Herman <[email protected]>

box survived dozens of reboots so the fix from yesterday indeed did the
trick.

thanks,

Ingo

2008-08-25 05:50:57

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

At Sun, 24 Aug 2008 23:30:08 +0200,
Guillaume Chazarain wrote:
>
> On Mon, Jul 14, 2008 at 11:09 AM, Jaroslav Kysela <[email protected]> wrote:
> > ALSA: hda - Added model selection for iMac 24"
>
> This commit introduces a regression from 2.6.26 on my MacBookPro 3,1.
> I lose the "Front" mixer entry but this I don't mind.
> The actual regression is that when I plug headphones, the internal
> speakers are not muted anymore so I hear the sound from both outputs
> (headphones & internal). Reverting this commit fixes the problem for
> me.

OK, this must be a conflict of PCI SSID. Apple seems to assign the
same PCI SSID for completely different models.

I guess it's fixed as well by passing model=auto?

Anyway, please run alsa-info.sh with --no-upload option, and attach
the generated file. The script is found in
http://www.alsa-project.org/alsa-info.sh

thanks,

Takashi

2008-08-24 21:30:26

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

On Mon, Jul 14, 2008 at 11:09 AM, Jaroslav Kysela <[email protected]> wrote:
> ALSA: hda - Added model selection for iMac 24"

This commit introduces a regression from 2.6.26 on my MacBookPro 3,1.
I lose the "Front" mixer entry but this I don't mind.
The actual regression is that when I plug headphones, the internal
speakers are not muted anymore so I hear the sound from both outputs
(headphones & internal). Reverting this commit fixes the problem for
me.

Thanks.

commit 3e0e469fa216ec70c93b1593821b759d19ee2e6b
Author: Travis Place <[email protected]>
Date: Fri Jun 20 16:51:45 2008 +0200

ALSA: hda - Added model selection for iMac 24"

Added the SSID of a known iMac 24" to automatically use
ALC885_IMAC24 quirk.

Signed-off-by: Travis Place <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>
Signed-off-by: Jaroslav Kysela <[email protected]>

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 61f8c13..d96a876 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -6192,6 +6192,7 @@ static struct snd_pci_quirk alc882_cfg_tbl[] = {
SND_PCI_QUIRK(0x1043, 0x817f, "Asus P5LD2", ALC882_6ST_DIG),
SND_PCI_QUIRK(0x1043, 0x81d8, "Asus P5WD", ALC882_6ST_DIG),
SND_PCI_QUIRK(0x105b, 0x6668, "Foxconn", ALC882_6ST_DIG),
+ SND_PCI_QUIRK(0x106b, 0x00a0, "Apple iMac 24''", ALC885_IMAC24),
SND_PCI_QUIRK(0x1458, 0xa002, "Gigabyte P35 DS3R", ALC882_6ST_DIG),
SND_PCI_QUIRK(0x1462, 0x28fb, "Targa T8", ALC882_TARGA), /*
MSI-1049 T8 */
SND_PCI_QUIRK(0x1462, 0x6668, "MSI", ALC882_6ST_DIG),

--
Guillaume

2008-08-25 21:37:10

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

On Mon, Aug 25, 2008 at 7:50 AM, Takashi Iwai <[email protected]> wrote:
> I guess it's fixed as well by passing model=auto?

Indeed that works too.

> Anyway, please run alsa-info.sh with --no-upload option, and attach
> the generated file.

Here you go.

Thanks.

--
Guillaume

Attachments:

(No filename) (276.00 B)
alsa-info-bad.txt (22.16 kB)
alsa-info-good.txt (24.74 kB)
Download all attachments

2008-08-25 22:44:18

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

On Mon, Aug 25, 2008 at 11:36 PM, Guillaume Chazarain <[email protected]> wrote:
> On Mon, Aug 25, 2008 at 7:50 AM, Takashi Iwai <[email protected]> wrote:
>> I guess it's fixed as well by passing model=auto?
>
> Indeed that works too.

I spoke too soon, with model=auto I get back the "Front" mixer that I
used during the git bisection to find the culprit, but I get no sound
at all in the plugged in headphones or the internal speakers.

>> Anyway, please run alsa-info.sh with --no-upload option, and attach
>> the generated file.

alsa-info-good.txt actually contained the output of alsa-info.sh with
model=auto, so it's not good.
Here is the output with the commit reverted, so sound is working.
Looks like the revert brings back the "Speaker" switch which I need.

Sorry for the confusion.

--
Guillaume

Attachments:

(No filename) (807.00 B)
alsa-info.sounds_working.txt (25.94 kB)
Download all attachments

2008-08-26 05:59:39

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

At Tue, 26 Aug 2008 00:43:59 +0200,
Guillaume Chazarain wrote:
>
> On Mon, Aug 25, 2008 at 11:36 PM, Guillaume Chazarain <[email protected]> wrote:
> > On Mon, Aug 25, 2008 at 7:50 AM, Takashi Iwai <[email protected]> wrote:
> >> I guess it's fixed as well by passing model=auto?
> >
> > Indeed that works too.
>
> I spoke too soon, with model=auto I get back the "Front" mixer that I
> used during the git bisection to find the culprit, but I get no sound
> at all in the plugged in headphones or the internal speakers.

This is odd.
Are you sure both are the same kernel but just one commit reverted?

Takashi

2008-08-26 21:11:12

[permalink] [raw]

Subject: Re: [ALSA PATCH] alsa-git merge request

On Tue, Aug 26, 2008 at 7:59 AM, Takashi Iwai <[email protected]> wrote:
> Are you sure both are the same kernel but just one commit reverted?

I retested, yes.
With or without the revert, model=auto disables sound completely, and
I get this in dmesg:

HDA Intel 0000:00:1b.0: PCI INT A disabled
HDA Intel 0000:00:1b.0: PCI INT A -> GSI 20 (level, low) -> IRQ 20
HDA Intel 0000:00:1b.0: setting latency timer to 64
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:2325: hda_codec:
model 'auto' is selected
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:3021: autoconfig:
line_outs=1 (0x14/0x0/0x0/0x0/0x0)
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:3025:
speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:3029: hp_outs=1
(0x15/0x0/0x0/0x0/0x0)
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:3030: mono: mono_out=0x0
ALSA /home/g/linux-2.6/sound/pci/hda/hda_codec.c:3038: inputs:
mic=0x18, fmic=0x0, line=0x1a, fline=0x0, cd=0x0, aux=0x0

I played with model=intel-mac-auto and model=intel-mac-v3 but it had
no effect, i.e. same as without the model= parameter.

So, without model=auto the problem is that I get the output on both
the internal speakers and the headphones, and this is fixed by
reverting the commit.

Thanks.

--
Guillaume

2008-08-27 06:02:21