[alsa-devel] patch_realtek: support Vaio Type G

[Takashi Iwai]

At Tue, 17 Nov 2009 20:15:37 +0900,
Mattia Dongili wrote:
> 
> On Tue, Nov 17, 2009 at 07:40:12AM +0100, Takashi Iwai wrote:
> > At Tue, 17 Nov 2009 07:47:56 +0900,
> > Mattia Dongili wrote:
> > > 
> > > On Mon, Nov 16, 2009 at 03:42:03PM +0100, Takashi Iwai wrote:
> > > > At Mon, 16 Nov 2009 23:02:27 +0900,
> > > > Mattia Dongili wrote:
> > > > > 
> > > > > Hi Takashi,
> > > > > 
> > > > > my laptop (Vaio Type G) used to work (although the valume was pretty
> > > > > low) with model=hippo until 2.6.29 then from 2.6.30 it stopped being
> > > > > supported throwing this message:
> > > > > hda-intel: azx_get_response timeout, switching to polling mode: last cmd=0x01570900
> > > > 
> > > > Still doesn't work 2.6.32-rc7 without your patch?
> > > 
> > > No, now tested on .32-rc7 and model=hippo still doesn't work.
> > 
> > What about model=auto?
> 
> err hmmm... well it works! I thought that model=auto was the same as no
> parameter at all so haven't tried it. Sorry about the noise.

Good to hear.  You are partly right -- the model=auto isn't always the
default.  If a quirk is already defined in the driver, another model
is picked up.

I guess the problem is that current quirk for your device doesn't work.

> 
> > > I have to
> > > admit thought that I see
> > > 
> > > [  286.236022] HDA Intel 0000:00:1b.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21
> > > [  286.236114] HDA Intel 0000:00:1b.0: setting latency timer to 64
> > > [  287.392510] hda-intel: azx_get_response timeout, switching to polling mode: last cmd=0x015b8000
> > > 
> > > even with my patch. But at least I get sound.
> > 
> > The message above is usually pretty harmless, and unlikely relevant
> > with your problem.
> > 
> > Could you give alsa-info.sh outputs on both vanilla 2.6.32-rc7 kernel,
> > with model=auto, and with your patch?
> 
> in case you still want it, it's attached

Thanks.

The fix patch is below.  Could you test it without model option?

I think we can get rid of the fallback of model=sony-assamd since
now model=auto works almost well as long as the BIOS pins are correct.


Takashi

---
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index ef7d210..ba339d7 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -11470,6 +11470,7 @@ static struct snd_pci_quirk alc262_cfg_tbl[] = {
 	SND_PCI_QUIRK(0x104d, 0x9016, "Sony VAIO", ALC262_AUTO), /* dig-only */
 	SND_PCI_QUIRK(0x104d, 0x9025, "Sony VAIO Z21MN", ALC262_TOSHIBA_S06),
 	SND_PCI_QUIRK(0x104d, 0x9035, "Sony VAIO VGN-FW170J", ALC262_AUTO),
+	SND_PCI_QUIRK(0x104d, 0x9047, "Sony VAIO Type G", ALC262_AUTO),
 	SND_PCI_QUIRK_MASK(0x104d, 0xff00, 0x9000, "Sony VAIO",
 			   ALC262_SONY_ASSAMD),
 	SND_PCI_QUIRK(0x1179, 0x0001, "Toshiba dynabook SS RX1",