[alsa-devel] [PATCH 1/1] ALSA: hda/realtek - Correction of fixup codes for PB V7900 laptop

[Takashi Iwai]

At Mon, 02 Jun 2014 14:01:10 +0200,
Takashi Iwai wrote:
> 
> At Mon, 2 Jun 2014 13:59:25 +0200 (CEST),
> Ronan wrote:
> > 
> > 
> > 
> > 
> > 
> > 
> > > Message du 02/06/14 08:46
> > > De : "Takashi Iwai" 
> > > A : ronan.marquet at orange.fr
> > > Copie à : alsa-devel at alsa-project.org
> > > Objet : Re: [PATCH 1/1] ALSA: hda/realtek - Correction of fixup codes for PB V7900 laptop
> > > 
> > > At Sun, 1 Jun 2014 18:38:53 +0200,
> > > ronan.marquet at orange.fr wrote:
> > > > 
> > > > From: Ronan Marquet 
> > > > 
> > > > Correcion of wrong fixup entries add in commit ca8f0424 to replace
> > > > static model quirk for PB V7900 laptop (will model).
> > > > 
> > > > Signed-off-by: Ronan Marquet 
> > > > 
> > > > diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
> > > > index 49e884f..0e3a7f6 100644
> > > > --- a/sound/pci/hda/patch_realtek.c
> > > > +++ b/sound/pci/hda/patch_realtek.c
> > > > @@ -1647,12 +1647,10 @@ static const struct hda_fixup alc260_fixups[] = {
> > > > [ALC260_FIXUP_COEF] = {
> > > > .type = HDA_FIXUP_VERBS,
> > > > .v.verbs = (const struct hda_verb[]) {
> > > > - { 0x20, AC_VERB_SET_COEF_INDEX, 0x07 },
> > > > - { 0x20, AC_VERB_SET_PROC_COEF, 0x3040 },
> > > > + { 0x1a, AC_VERB_SET_COEF_INDEX, 0x07 },
> > > > + { 0x1a, AC_VERB_SET_PROC_COEF, 0x3040 },
> > > 
> > > That part looks correct, however:
> > > 
> > > > { }
> > > > },
> > > > - .chained = true,
> > > > - .chain_id = ALC260_FIXUP_HP_PIN_0F,
> > > 
> > > Is this removal mandatory? In other words, is keeping this harmful?
> > > The original quirk had this, so it may be safer to keep it for other
> > > machines.
> > > Yes, if i keep this two lines, i have a crackling sound and the internal output is blinking into Gnome's sound setting (Jack detection problem ?).Whitout this lines, all is working find on my laptop but, perhpas, it could be fixe differently ? thanks, Ronan 
> 
> Interesting.  Could you attach alsa-info.sh outputs both with and
> without this chain?  If the removal of this chain is needed, we should
> split the patch, one for correcting the wrong widget NID and another
> for the removal of the chain.

Never mind, I checked the alsa-info.sh output of PB V7900 and also HP
B1900, and the chain for NID 0x0f looks invalid for these machines
that have really only a single output.  This may confuse PulseAudio.

I'll take your patch as is now.


thanks,

Takashi