[alsa-devel] HDA regression on Fujitsu S7020 laptop (ALC260 codec)

[Takashi Iwai]

At Sun, 10 Mar 2013 15:28:54 +1030,
Jonathan Woithe wrote:
> 
> Hi Takashi
> 
> > > On Wed, Feb 20, 2013 at 02:14:20PM +0100, Takashi Iwai wrote:
> > > > > Sounds good - I will test in the next day or so and report back.  What's the
> > > > > most efficient way to get hold of this tree / branch given that I have the
> > > > > for-next branch of sound.git?  Can I somehow leverage this, or do I need to
> > > > > do a completely separate checkout?  Apologies if this is a basic question -
> > > > > I'm still coming to terms with git.
> > > > 
> > > > sound-unstable git tree always contains the sound git tree master
> > > > branch, so everything is there basically. ...
> > > 
> > > What's the next step in the process?  I've reported the outcome of the tests
> > > of sound-unstable as per that report progress is being made.  Is it a case
> > > of waiting for things to stabilise, or do you need me to do something to
> > > move things along?  If so, what?
> > 
> > I merged the patches to the main sound git tree for-next branch now.
> > 
> > Also a few updates are merged there, too: now the hp mic jack mode
> > should be able to change the pin control no matter whether which
> > capture source is.  Also, the capture source won't change the hp/mic
> > pinctl when the jack mode enum is created.  This reduces the
> > automation but also reduces the confusion, too.  After all, when jack
> > mode is present, user wants to control the pin by himself.
> > 
> > Let me know if anything else is missing or broken.
> 
> I've given this new revision a quick test; sorry for the delay.  With this
> revision, when using "model=fujitsu-jwse" things seem to be working pretty
> well.  In particular, the updates you mention greatly improve usability; the
> lack of automatic interaction between mixer controls makes it much easier to
> use and understand.
> 
> I think everything is present and accounted for in this latest revision. 
> Thanks for your work on this.
> 
> The only slightly odd thing I noticed was that whenever the driver is loaded
> the "headphone mic jack mode" defaults to "mic bias 80pc", rather than "line
> out" or "headphone" which I would have expected.  Obviously this can be
> manually or automatically overridden after module load, but in many ways it
> would be good if it defaulted to one of the output modes (given that the
> jack's primary purpose is for output).

Indeed, it's an overlooked bug.  I merged the fix patch below.

> Where does this leave us now?  Is the "fujitsu-jwse" a permanent thing, or
> will the respective changes be rolled into something more generic?

I'm inclined to keep as is.


thanks,

Takashi

---
Subject: [PATCH] ALSA: hda - Don't re-initialize shared hp/mic pinctl

When a headphone pin is set up as a shared hp/mic pin, we rather want
to keep it as a headphone primarily as default, but the driver
overrides it always as a mic pin, just because the input controls are
created after outputs.  Add a check of pin NID and skip the
re-initialization of pinctl for such a shared hp/mic pin.

Reported-by: Jonathan Woithe <jwoithe at just42.net>
Signed-off-by: Takashi Iwai <tiwai at suse.de>
---
 sound/pci/hda/hda_generic.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c
index fb232c1..aae6b10 100644
--- a/sound/pci/hda/hda_generic.c
+++ b/sound/pci/hda/hda_generic.c
@@ -2967,7 +2967,8 @@ static int create_input_ctls(struct hda_codec *codec)
 		val = PIN_IN;
 		if (cfg->inputs[i].type == AUTO_PIN_MIC)
 			val |= snd_hda_get_default_vref(codec, pin);
-		set_pin_target(codec, pin, val, false);
+		if (pin != spec->hp_mic_pin)
+			set_pin_target(codec, pin, val, false);
 
 		if (mixer) {
 			if (is_reachable_path(codec, pin, mixer)) {
-- 
1.8.1.4