Re: Conexant 5051 lenovo-x200 fixes

[Takashi Iwai <tiwai@suse.de>]

At Tue, 23 Aug 2011 14:32:21 +0300,
Ulo Mets wrote:
> 
> On 8/22/2011 3:00 PM, Takashi Iwai wrote:
> > At Sat, 20 Aug 2011 18:14:20 +0300,
> > Ülo Mets wrote:
> >> On 08/19/2011 08:43 AM, Takashi Iwai wrote:
> >>
> >>> At Fri, 19 Aug 2011 03:00:58 +0300,
> >>> Ulo Mets wrote:
> >>>> On 08/18/2011 01:08 PM, Takashi Iwai wrote:
> >>>>> At Thu, 18 Aug 2011 03:45:33 +0300,
> >>>>> Ulo Mets wrote:
> >>>>>> Hi,
> >>>>>> I am not a kernel developer, but occasionally dig into it if something gets in my way.
> >>>>>> On Lenovo X200 Tablet none of the mixers showed Mic Boost adjustments, but HDA-Analyzer found them
> >>>>>> and adjusted properly. So I inserted the corresponding controls, and it worked.
> >>>>>> Then, during testing I found that the labels of External and Docking Mic are swapped,
> >>>>>> so I exchanged these too. Since I could not test it on other hardware, I put it all on lenovo-x200 branch.
> >>>>>> The patch applies to 3.1-rc2, I hope it is sufficiently recent.
> >>>>>> The Mic Switches don't have any effect on my unit (X200T without modem)
> >>>>> Could you try model=auto instead of patching?
> >>>>> The recent auto-parser should work better.
> >>>> Model=auto seems to give a very generic configuration, which indeed does
> >>>> include working
> >>>> Mic boost for Internal Mic. (Ext?) Mic Boost and gain are also present,
> >>>> but jack sensing does not work,
> >>>> so these cannot be tested.
> >>> The automatic mic-switch by jack-sensing works without model=auto,
> >>> right?  Then we need to fix it.  Could you give alsa-info.sh output?
> >>>
> >> Right. The alsa-info.sh output is attached.
> > Thanks.
> >
> >>>> It would be nice to get the mute switches to work as well.
> >>> Which mute switch do you mean exactly?
> >> I see the "Mic Switch" controls in the input nodes 0x14 and 0x15, and some Mute controls
> >>
> >> with hda-analyzer. But it may well be just wishful thinking. I wonder if anybody has seen the
> >> datasheet of CX20561?
> > Google is your friend :)
> Is that so? I'll give you $100 for a set of keywords that leads to the datasheet.

Then it has been removed.  Ask Conexant if you still need it.

> BTW, the chip revision in question is -15Z

It should be compatible no matter which revision is.

> > But you can figure out almost all things in the codec proc file.
> Yes, the Pin Defaults are correct in the proc file, and the control names in 0x14 and 0x15 contradict these.
> That's why it was so easy to fix.
> 
> > What you should test is to check which I/O the pins 0x16, 0x17, 0x18
> > and 0x19 correspond.  I suppose 0x16 = HP, 0x17 = mic-jack, 0x18 =
> > dock-mic, 0x19 = dock-HP.
> Almost. 0x16 = HP, 0x17 = dock-mic, 0x18 = ext-mic, 0x19 = dock-HP

OK.

> > An easy way would be to test the jack-detection via hda-verb.
> > 	# hda-verb /dev/snd/hwC0D0 0x16 GET_PIN_SENSE 0
> >
> > When any jack corresponds to the pin 0x16, it'll report 0x80000000
> > upon plugged.
> Nice tool. But ver. 0.3 reports 0xffffffff when plugged and 0x7fffffff when unplugged.

It doesn't matter.  The only meaningful bit is the bit 31.

For making auto-parser working, apply the patch below.
It fixes the multi-headphone assignment case.

Then, put lines below to /lib/firmware/x200t-pincfg file.

----------------------------------------------------------------
0x16 0x042140ff
0x17 0x23a11040
0x18 0x04a190f0
0x19 0x2121103f
0x1a 0x901701f0
0x1d 0x90a601f0
----------------------------------------------------------------

After creating the file, pass "patch=x200t-pincfg model=auto" module
options to snd-hda-intel driver.  Now the driver will be set up using
the auto-parser with the given pin-configuration table.

Let me know if this works.  Once after confirming it's working, we can
add a static configuration in the driver itself.


thanks,

Takashi

---
From: Takashi Iwai <tiwai@suse.de>
Subject: [PATCH] ALSA: hda - Fix double-headphone/speaker paths for Cxt
 auto-parser

When multiple headphones or speakers are assigned but no individual
DACs are available, the driver should take the first HP/SPK DAC instead
of another primary output.  The patch adds a bit-flag to dac field of
struct pin_dac_pair indicating that it's a slave DAC.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_conexant.c |   13 ++++++++++---
 1 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c
index 502fc94..4c462c3 100644
--- a/sound/pci/hda/patch_conexant.c
+++ b/sound/pci/hda/patch_conexant.c
@@ -3348,6 +3348,8 @@ static hda_nid_t get_unassigned_dac(struct hda_codec *codec, hda_nid_t pin,
 
 #define MAX_AUTO_DACS	5
 
+#define DAC_SLAVE_FLAG	0x8000	/* filled dac is a slave */
+
 /* fill analog DAC list from the widget tree */
 static int fill_cx_auto_dacs(struct hda_codec *codec, hda_nid_t *dacs)
 {
@@ -3379,6 +3381,8 @@ static int fill_dacs_for_pins(struct hda_codec *codec, hda_nid_t *pins,
 		filled[nums].pin = pins[i];
 		filled[nums].type = type;
 		filled[nums].dac = get_unassigned_dac(codec, pins[i], dacs, rest);
+		if (!filled[nums].dac && i > 0 && filled[0].dac)
+			filled[nums].dac = filled[0].dac | DAC_SLAVE_FLAG;
 		nums++;
 	}
 	return nums;
@@ -3407,7 +3411,7 @@ static void cx_auto_parse_output(struct hda_codec *codec)
 	/* fill multiout struct */
 	for (i = 0; i < nums; i++) {
 		hda_nid_t dac = spec->dac_info[i].dac;
-		if (!dac)
+		if (!dac || (dac & DAC_SLAVE_FLAG))
 			continue;
 		switch (spec->dac_info[i].type) {
 		case AUTO_PIN_LINE_OUT:
@@ -4035,6 +4039,8 @@ static void cx_auto_init_output(struct hda_codec *codec)
 		nid = spec->dac_info[i].dac;
 		if (!nid)
 			nid = spec->multiout.dac_nids[0];
+		else if (nid & DAC_SLAVE_FLAG)
+			nid &= ~DAC_SLAVE_FLAG;
 		select_connection(codec, spec->dac_info[i].pin, nid);
 	}
 	if (spec->auto_mute) {
@@ -4191,7 +4197,8 @@ static int cx_auto_build_output_controls(struct hda_codec *codec)
 	for (i = 0; i < spec->dac_info_filled; i++) {
 		const char *label;
 		int idx, type;
-		if (!spec->dac_info[i].dac)
+		hda_nid_t dac = spec->dac_info[i].dac;
+		if (!dac || (dac & DAC_SLAVE_FLAG))
 			continue;
 		type = spec->dac_info[i].type;
 		if (type == AUTO_PIN_LINE_OUT)
@@ -4211,7 +4218,7 @@ static int cx_auto_build_output_controls(struct hda_codec *codec)
 			idx = num_spk++;
 			break;
 		}
-		err = try_add_pb_volume(codec, spec->dac_info[i].dac,
+		err = try_add_pb_volume(codec, dac,
 					spec->dac_info[i].pin,
 					label, idx);
 		if (err < 0)
-- 
1.7.6

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