[PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Dan Chen]

[-- Attachment #1: Type: text/plain, Size: 296 bytes --]

Hi,

In triaging bugs for Ubuntu 9.10, I've seen several Dell models that
are behaving strangely with the current driver. Attached please find
a patch against current sound-2.6 HEAD that allows these models
to have somewhat more sane behaviour when headphones are
inserted.

Thanks,
Dan



      

[-- Attachment #2: 0001-ALSA-hda-Use-3stack-quirk-for-some-broken-Dell-model.patch --]
[-- Type: application/octet-stream, Size: 2192 bytes --]

From 2ecc8de3cbd95f9a32a9f536c5f154f2fd42665c Mon Sep 17 00:00:00 2001
From: Daniel T Chen <crimsun@ubuntu.com>
Date: Wed, 4 Nov 2009 20:24:21 -0500
Subject: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

BugLink: https://bugs.launchpad.net/bugs/279478

The original (current) quirks rely on the bioses being correct, which
sadly is not the case empirically. Users are reporting one of the
following after inserting headphones:

* audio fades until inaudible
* internal speakers do not mute

Until we can pinpoint the real culprit, let's just use a working
quirk.

Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
---
 sound/pci/hda/patch_sigmatel.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index b513eba..3b346c4 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -2050,13 +2050,13 @@ static struct snd_pci_quirk stac927x_cfg_tbl[] = {
 			   "Intel D965", STAC_D965_3ST),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_INTEL, 0xff00, 0x2000,
 			   "Intel D965", STAC_D965_3ST),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01dd, "Dell Dimension E520", STAC_D965_3ST),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01ed, "Dell Dimension E521", STAC_D965_3ST),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01f3, "Dell Inspiron 1420", STAC_D965_3ST),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01f4, "Dell Dimension C521", STAC_D965_3ST),
 	/* Dell 3 stack systems */
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01f7, "Dell XPS M1730", STAC_DELL_3ST),
-	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01dd, "Dell Dimension E520", STAC_DELL_3ST),
-	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01ed, "Dell     ", STAC_DELL_3ST),
-	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01f4, "Dell     ", STAC_DELL_3ST),
 	/* Dell 3 stack systems with verb table in BIOS */
-	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x01f3, "Dell Inspiron 1420", STAC_DELL_BIOS),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x0227, "Dell Vostro 1400  ", STAC_DELL_BIOS),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x022e, "Dell     ", STAC_DELL_BIOS),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL,  0x022f, "Dell Inspiron 1525", STAC_DELL_BIOS),
-- 
1.6.3.3


[-- Attachment #3: Type: text/plain, Size: 160 bytes --]

_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Takashi Iwai]

At Wed, 4 Nov 2009 17:36:51 -0800 (PST),
Dan Chen wrote:
> 
> Hi,
> 
> In triaging bugs for Ubuntu 9.10, I've seen several Dell models that
> are behaving strangely with the current driver. Attached please find
> a patch against current sound-2.6 HEAD that allows these models
> to have somewhat more sane behaviour when headphones are
> inserted.

Try the very latest sound git tree.  Does the problem still happen?

The decreasing volume should come from the volume-knob setup, which
was fixed recently.  The headphone muting is another issue, but might
have been fixed :)

In anyway, please give alsa-info.sh output with the problematic
hardware (preferably with the latest driver).


thanks,

Takashi

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Daniel Chen]

On Thu, Nov 5, 2009 at 2:10 AM, Takashi Iwai <tiwai@suse.de> wrote:
> Try the very latest sound git tree.  Does the problem still happen?

The good news is that the symptom reportedly is gone with the 20091105
snapshot. Thanks for the suggestion! An example codec dump running
20091105 is found at
http://launchpadlibrarian.net/35172146/Card0.Codecs.codec.2.txt

> The decreasing volume should come from the volume-knob setup, which
> was fixed recently.  The headphone muting is another issue, but might
> have been fixed :)

This symptom remains reproducible. An example codec dump is at
http://launchpadlibrarian.net/35125538/codec%230
_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Takashi Iwai]

At Thu, 5 Nov 2009 11:45:51 -0500,
Daniel Chen wrote:
> 
> On Thu, Nov 5, 2009 at 2:10 AM, Takashi Iwai <tiwai@suse.de> wrote:
> > Try the very latest sound git tree.  Does the problem still happen?
> 
> The good news is that the symptom reportedly is gone with the 20091105
> snapshot. Thanks for the suggestion! An example codec dump running
> 20091105 is found at
> http://launchpadlibrarian.net/35172146/Card0.Codecs.codec.2.txt

Any chance to get alsa-info.sh output?
The codec proc file isn't enough, e.g. missing PCI IDs.


Takashi
_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Daniel Chen]

On Thu, Nov 5, 2009 at 12:09 PM, Takashi Iwai <tiwai@suse.de> wrote:
> Any chance to get alsa-info.sh output?
> The codec proc file isn't enough, e.g. missing PCI IDs.

Unfortunately there's no alsa-info.sh output, but the important bits
are as follows: SSIDs 0x102801dd and 0x102801ed are affected by the
volume-knob change.

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Takashi Iwai]

At Thu, 5 Nov 2009 12:56:43 -0500,
Daniel Chen wrote:
> 
> On Thu, Nov 5, 2009 at 12:09 PM, Takashi Iwai <tiwai@suse.de> wrote:
> > Any chance to get alsa-info.sh output?
> > The codec proc file isn't enough, e.g. missing PCI IDs.
> 
> Unfortunately there's no alsa-info.sh output, but the important bits
> are as follows: SSIDs 0x102801dd and 0x102801ed are affected by the
> volume-knob change.

So, do you mean that the decreasing volume still present?
Or the headphone auto-muting still doesn't work?


Takashi

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Daniel Chen]

On Fri, Nov 6, 2009 at 10:45 AM, Takashi Iwai <tiwai@suse.de> wrote:
> So, do you mean that the decreasing volume still present?
> Or the headphone auto-muting still doesn't work?

Sorry for being unclear. For SSIDs 0x102801dd and 0x102801ed, the
volume-knob issue is resolved, i.e., volume no longer fades to an
inaudible level once hp are inserted.

The hp automuting only affects 0x102801f3 (as tested thus far), and it
remains reproducible.

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Takashi Iwai]

At Fri, 6 Nov 2009 12:49:10 -0500,
Daniel Chen wrote:
> 
> On Fri, Nov 6, 2009 at 10:45 AM, Takashi Iwai <tiwai@suse.de> wrote:
> > So, do you mean that the decreasing volume still present?
> > Or the headphone auto-muting still doesn't work?
> 
> Sorry for being unclear. For SSIDs 0x102801dd and 0x102801ed, the
> volume-knob issue is resolved, i.e., volume no longer fades to an
> inaudible level once hp are inserted.
> 
> The hp automuting only affects 0x102801f3 (as tested thus far), and it
> remains reproducible.

OK, then for debugging I'd need either alsa-info.sh output, or the
codec#* contents after loaded with probe_only=1 option.
The previous codec file wasn't enough because it doesn't contain the
original pin-default values.

Could you get the file?


thanks,

Takashi

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Daniel Chen]

On Sat, Nov 7, 2009 at 7:26 AM, Takashi Iwai <tiwai@suse.de> wrote:
> OK, then for debugging I'd need either alsa-info.sh output, or the
> codec#* contents after loaded with probe_only=1 option.
> The previous codec file wasn't enough because it doesn't contain the
> original pin-default values.

The original reporter has confirmed that the 20091012 stable snapshot
fixes the jacksense issue. Sorry for the noise!

Re: [PATCH] ALSA: hda: Use 3stack quirk for some broken Dell models

[Takashi Iwai]

At Wed, 11 Nov 2009 01:33:46 -0500,
Daniel Chen wrote:
> 
> On Sat, Nov 7, 2009 at 7:26 AM, Takashi Iwai <tiwai@suse.de> wrote:
> > OK, then for debugging I'd need either alsa-info.sh output, or the
> > codec#* contents after loaded with probe_only=1 option.
> > The previous codec file wasn't enough because it doesn't contain the
> > original pin-default values.
> 
> The original reporter has confirmed that the 20091012 stable snapshot
> fixes the jacksense issue. Sorry for the noise!

Good to hear!  Thanks for confirmation.


Takashi