[PATCH] sound working on the macbook pro

[PATCH] sound working on the macbook pro

[Rafael Espíndola]

[-- Attachment #1: Type: text/plain, Size: 658 bytes --]

The problem with the pin config of the macbook pro is that it has no
line outs. Because of this, snd_hda_parse_pin_def_config tries to hack
the speaker into being a line out. This hack makes the subsequent
setup fail.

The attached patch does the followig:
1) install the correct pin config (extracted from macos)
2) removes the hack from snd_hda_parse_pin_def_config
3) stac92xx_parse_auto_config no longuer aborts if line_outs == 0

item 2 might cause some problems for parts of the system that expect
this hack. IMHO, we should try to fix these problems and remove the
hack. It is very confusing to see an Speaker appearing as a line out!

Cheers,
Rafael

[-- Attachment #2: sound.patch --]
[-- Type: text/x-patch, Size: 1824 bytes --]

diff -r c7540f430d45 pci/hda/hda_codec.c
--- a/pci/hda/hda_codec.c	Fri Feb 16 13:27:18 2007 +0100
+++ b/pci/hda/hda_codec.c	Sun Feb 18 16:19:20 2007 +0000
@@ -2179,26 +2179,6 @@ int snd_hda_parse_pin_def_config(struct 
 		   cfg->input_pins[AUTO_PIN_CD],
 		   cfg->input_pins[AUTO_PIN_AUX]);
 
-	/*
-	 * FIX-UP: if no line-outs are detected, try to use speaker or HP pin
-	 * as a primary output
-	 */
-	if (! cfg->line_outs) {
-		if (cfg->speaker_outs) {
-			cfg->line_outs = cfg->speaker_outs;
-			memcpy(cfg->line_out_pins, cfg->speaker_pins,
-			       sizeof(cfg->speaker_pins));
-			cfg->speaker_outs = 0;
-			memset(cfg->speaker_pins, 0, sizeof(cfg->speaker_pins));
-		} else if (cfg->hp_outs) {
-			cfg->line_outs = cfg->hp_outs;
-			memcpy(cfg->line_out_pins, cfg->hp_pins,
-			       sizeof(cfg->hp_pins));
-			cfg->hp_outs = 0;
-			memset(cfg->hp_pins, 0, sizeof(cfg->hp_pins));
-		}
-	}
-
 	return 0;
 }
 
diff -r c7540f430d45 pci/hda/patch_sigmatel.c
--- a/pci/hda/patch_sigmatel.c	Fri Feb 16 13:27:18 2007 +0100
+++ b/pci/hda/patch_sigmatel.c	Sun Feb 18 16:26:17 2007 +0000
@@ -530,9 +530,9 @@ static unsigned int macbook_pin_configs[
 };
 
 static unsigned int macbook_pro_pin_configs[10] = {
-	0x0221401f, 0x90a70120, 0x01813024, 0x01014010,
-	0x400000fd, 0x01016011, 0x1345e240, 0x13c5e22e,
-	0x400000fc, 0x400000fb,
+	0x012b4050, 0x90a00110, 0x90100140, 0x400000f0,
+	0x400000f0, 0x018b3020, 0x014be060, 0x01cbe030,
+	0x400000f0, 0x400000f0,
 };
 
 static unsigned int *stac922x_brd_tbl[STAC_922X_MODELS] = {
@@ -1401,8 +1401,6 @@ static int stac92xx_parse_auto_config(st
 						&spec->autocfg,
 						spec->dmic_nids)) < 0)
 		return err;
-	if (! spec->autocfg.line_outs)
-		return 0; /* can't find valid pin config */
 
 	if ((err = stac92xx_add_dyn_out_pins(codec, &spec->autocfg)) < 0)
 		return err;

[-- Attachment #3: Type: text/plain, Size: 345 bytes --]

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

[-- Attachment #4: Type: text/plain, Size: 161 bytes --]

_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Sun, 18 Feb 2007 17:24:37 +0000,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> ---------- Forwarded message ----------
> From: Rafael Espíndola <rafael.espindola@gmail.com>
> Date: Feb 18, 2007 5:06 PM
> Subject: [PATCH] sound working on the macbook pro
> To: alsa-devel@alsa-project.org
> 
> 
> The problem with the pin config of the macbook pro is that it has no
> line outs. Because of this, snd_hda_parse_pin_def_config tries to hack
> the speaker into being a line out. This hack makes the subsequent
> setup fail.

How (and what) does it fail exactly?


> The attached patch does the followig:
> 1) install the correct pin config (extracted from macos)
> 2) removes the hack from snd_hda_parse_pin_def_config
> 3) stac92xx_parse_auto_config no longuer aborts if line_outs == 0
> 
> item 2 might cause some problems for parts of the system that expect
> this hack. IMHO, we should try to fix these problems and remove the
> hack. It is very confusing to see an Speaker appearing as a line out!

I'm sure that this would break many other devices.  So, the patch is
unacceptable as it is, unfortunately.

The line_outs are actually primary outputs, not exactly "line-out".
You can find pretty many codes that refer to this as the primary
output.


Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Rafael Espíndola]

> How (and what) does it fail exactly?
a "diff" of the trace without and with the fix-up is:

1) snd_hda_parse_pin_def_config moves the speaker to the line_out
2) stac92xx_add_dyn_out_pins adds two line_outs
3) stac92xx_auto_create_multi_out_ctls creates the controls "Front",
"Surround", Center, LFE
4) stac92xx_auto_create_hp_ctls doesn't run add_spec_dacs or
create_controls for the speaker. It then creates a control name Master
instead of Headphone.

> I'm sure that this would break many other devices.  So, the patch is
> unacceptable as it is, unfortunately.
>
> The line_outs are actually primary outputs, not exactly "line-out".
> You can find pretty many codes that refer to this as the primary
> output.
Do you consider changing this a long term goal? I agree that we should
first try to get the sound working on the mac book, but the current
implementation is strange IMHO.

>
> Takashi
>

Thanks,
Rafael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Tue, 20 Feb 2007 09:40:35 +0000,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> > How (and what) does it fail exactly?
> a "diff" of the trace without and with the fix-up is:
> 
> 1) snd_hda_parse_pin_def_config moves the speaker to the line_out
> 2) stac92xx_add_dyn_out_pins adds two line_outs
> 3) stac92xx_auto_create_multi_out_ctls creates the controls "Front",
> "Surround", Center, LFE

OK, this is basically the problem of stac92xx_add_dyn_out_pins() that
it adds surrounds unconditionally.
(BTW, just a curisotiy - would these surround and/or CLFE work as
 outputs?  Or does it have no [mic/line] input jack?)

> 4) stac92xx_auto_create_hp_ctls doesn't run add_spec_dacs or
> create_controls for the speaker. It then creates a control name Master
> instead of Headphone.

Hm, I don't get this yet.  The current auto-config code in
patch_sigmatel isn't 100% suitable for the laptop-style configurations
but rather for the desktop machines.  So, we need a slight redesign of
the configurator at first.  Let's get things straight now.

Please write down the pins you have on your device and what controls
should appear.  I suppose you have

- a HP pin
- a speaker pin
- a mic-in pin

What else?  And, for these controls, we'll have

- Speaker volume/switch
- Headphone volume/switch
- Captuer volume/switch
- Input Mux

but no Master (as STAC codec controls these outputs individually).


> > I'm sure that this would break many other devices.  So, the patch is
> > unacceptable as it is, unfortunately.
> >
> > The line_outs are actually primary outputs, not exactly "line-out".
> > You can find pretty many codes that refer to this as the primary
> > output.
> Do you consider changing this a long term goal? I agree that we should
> first try to get the sound working on the mac book, but the current
> implementation is strange IMHO.

Yes, there is certainly a room for improvements.
For example, we can add another field to indicate what is the primary
output type (line-out, speaker, HP) and stop moving the pin
information itself, so that the check of primary output is kept as
simple as now, and the original information won't be lost.


Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Rafael Espíndola]

> OK, this is basically the problem of stac92xx_add_dyn_out_pins() that
> it adds surrounds unconditionally.
> (BTW, just a curisotiy - would these surround and/or CLFE work as
>  outputs?  Or does it have no [mic/line] input jack?)

The control status with the FIX-UP (and the new pin config) is:
Master: controls the hp volume
Front: dummy
Surround: dummy
Center: dummy
LFE: dummy

> > 4) stac92xx_auto_create_hp_ctls doesn't run add_spec_dacs or
> > create_controls for the speaker. It then creates a control name Master
> > instead of Headphone.
>
> Hm, I don't get this yet.
The speaker was moved to line_out, so cfg->speaker_outs == 0.

>  The current auto-config code in
> patch_sigmatel isn't 100% suitable for the laptop-style configurations
> but rather for the desktop machines.  So, we need a slight redesign of
> the configurator at first.  Let's get things straight now.
>
> Please write down the pins you have on your device and what controls
> should appear.  I suppose you have
>
> - a HP pin
> - a speaker pin
> - a mic-in pin
> What else?  And, for these controls, we'll have
>
> - Speaker volume/switch
> - Headphone volume/switch
> - Captuer volume/switch
> - Input Mux
>

the mac book pro has a built in speaker, a built in microphone and two
jacks. One is a spdif output and headphone. The other one is a spdif
in and a line in.

The controls should be:
Speaker, Microphone, select builtin mic/line-in . I am not sure if
there should be two different controls or one control and one select
switch for the mic/line-in. Currently I also have IEC958 to turn on
the spdif output.

I don't have any spdif equipment. I also don't have an external
microphone, but I am sure I can get one for testing.

> but no Master (as STAC codec controls these outputs individually).
Right. I love that :-)

> Yes, there is certainly a room for improvements.
> For example, we can add another field to indicate what is the primary
> output type (line-out, speaker, HP) and stop moving the pin
> information itself, so that the check of primary output is kept as
> simple as now, and the original information won't be lost.
I will give it a try latter on.

> Takashi
>

Rafael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Tue, 20 Feb 2007 20:05:49 +0000,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> > OK, this is basically the problem of stac92xx_add_dyn_out_pins() that
> > it adds surrounds unconditionally.
> > (BTW, just a curisotiy - would these surround and/or CLFE work as
> >  outputs?  Or does it have no [mic/line] input jack?)
> 
> The control status with the FIX-UP (and the new pin config) is:
> Master: controls the hp volume
> Front: dummy
> Surround: dummy
> Center: dummy
> LFE: dummy

Interesting.  So, the front (that is hooked to speaker-pin) has no
influence, too?  And did you get speaker working if you avoid moving
from speaker_pin to line_outs?

> > > 4) stac92xx_auto_create_hp_ctls doesn't run add_spec_dacs or
> > > create_controls for the speaker. It then creates a control name Master
> > > instead of Headphone.
> >
> > Hm, I don't get this yet.
> The speaker was moved to line_out, so cfg->speaker_outs == 0.

But, its corresponding dac_nid is mapped, so basically 'Front' should
control the speaker output as it is now.


Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Rafael Espíndola]

> Interesting.  So, the front (that is hooked to speaker-pin) has no
> influence, too?  And did you get speaker working if you avoid moving
> from speaker_pin to line_outs?
No influence. For some reason the speaker doesn't work. It works when
I remove the fix-up.

> But, its corresponding dac_nid is mapped, so basically 'Front' should
> control the speaker output as it is now.
I don't know why it doesn't. Do you have any suggestion on how to find why?

>
> Takashi
>

Rafael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Thu, 22 Feb 2007 20:01:06 +0000,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> > Interesting.  So, the front (that is hooked to speaker-pin) has no
> > influence, too?  And did you get speaker working if you avoid moving
> > from speaker_pin to line_outs?
> No influence. For some reason the speaker doesn't work. It works when
> I remove the fix-up.

Even if you use the new right pin_config?

> > But, its corresponding dac_nid is mapped, so basically 'Front' should
> > control the speaker output as it is now.
> I don't know why it doesn't. Do you have any suggestion on how to find why?

Try to compare the codec proc files between with and without
speaker->line_out hacks.


Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Mon, 26 Feb 2007 21:34:17 -0800,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> > Even if you use the new right pin_config?
> Yes
> 
> > Try to compare the codec proc files between with and without
> > speaker->line_out hacks.
> I believe that the problem is this part:
> 
>    Pin Default 0x90100140: [Fixed] Speaker at Int N/A
>      Conn = Unknown, Color = Unknown
> -  Pin-ctls: 0x40: OUT
> +  Pin-ctls: 0x20: IN
>    Connection: 1

Then it means that stac92xx_hp_detect() isn't called properly, or the
pin-detection doesn't work with your hardware.  Otherwise
stac92xx_auto_init_multiout() should enable this pin.  Could you check
whether stac92xx_hp_detect() is called and how is it processed
(e.g. by adding printk's)?  Also, add a printk in
stac92xx_auto_init_multiout() to make sure that this isn't called.


Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Rafael Espíndola]

 > Then it means that stac92xx_hp_detect() isn't called properly, or the
> pin-detection doesn't work with your hardware.  Otherwise
> stac92xx_auto_init_multiout() should enable this pin.  Could you check
> whether stac92xx_hp_detect() is called and how is it processed
> (e.g. by adding printk's)?  Also, add a printk in
> stac92xx_auto_init_multiout() to make sure that this isn't called.

all tests performed with the new pin config.
 with the fix-up enabled:
*) stac92xx_auto_init_multi_out is not called
*) stac92xx_hp_detect is called:
 cfg->hp_outs = 1
 presence = 0 after the first loop
  cfg->line_outs = 3
  cfg->speaker_outs = 0
*) speakers don't work

 with the fix-up disabled:
*) stac92xx_auto_init_multi_out in not called
*) stac92xx_hp_detect is called:
 cfg->hp_outs = 1
 presense = 0 after the first loop
 cfg->line_outs = 0
 cfg->speaker_outs = 1
*) speakers work

> Takashi
>
Thanks,
Rafael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Fri, 9 Mar 2007 21:57:45 -0800,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
>  > Then it means that stac92xx_hp_detect() isn't called properly, or the
> > pin-detection doesn't work with your hardware.  Otherwise
> > stac92xx_auto_init_multiout() should enable this pin.  Could you check
> > whether stac92xx_hp_detect() is called and how is it processed
> > (e.g. by adding printk's)?  Also, add a printk in
> > stac92xx_auto_init_multiout() to make sure that this isn't called.
> 
> all tests performed with the new pin config.
>  with the fix-up enabled:
> *) stac92xx_auto_init_multi_out is not called
> *) stac92xx_hp_detect is called:
>  cfg->hp_outs = 1
>  presence = 0 after the first loop
>   cfg->line_outs = 3
>   cfg->speaker_outs = 0
> *) speakers don't work
> 
>  with the fix-up disabled:
> *) stac92xx_auto_init_multi_out in not called
> *) stac92xx_hp_detect is called:
>  cfg->hp_outs = 1
>  presense = 0 after the first loop
>  cfg->line_outs = 0
>  cfg->speaker_outs = 1
> *) speakers work

I guess it's a missing initialization for the first line-out jack.
It works on ohter devices because the default PIN setup has
AC_PIN_OUT_EN.  But, on Mac, it has AC_PIN_IN_EN by some reason.

The patch below should fix the problm.  Give it a try.


Takashi

diff -r e87b2114ceae pci/hda/patch_sigmatel.c
--- a/pci/hda/patch_sigmatel.c	Mon Mar 12 12:54:23 2007 +0100
+++ b/pci/hda/patch_sigmatel.c	Mon Mar 12 15:56:58 2007 +0100
@@ -1600,6 +1600,11 @@ static int stac92xx_init(struct hda_code
 		for (i = 0; i < cfg->hp_outs; i++)
 			enable_pin_detect(codec, cfg->hp_pins[i],
 					  STAC_HP_EVENT);
+		/* force to enable the first line-out; the others are set up
+		 * in unsol_event
+		 */
+		stac92xx_auto_set_pinctl(codec, spec->autocfg.line_out_pins[0],
+					 AC_PINCTL_OUT_EN);
 		stac92xx_auto_init_hp_out(codec);
 		/* fake event to set up pins */
 		codec->patch_ops.unsol_event(codec, STAC_HP_EVENT << 26);

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Rafael Espíndola]

> I guess it's a missing initialization for the first line-out jack.
> It works on ohter devices because the default PIN setup has
> AC_PIN_OUT_EN.  But, on Mac, it has AC_PIN_IN_EN by some reason.
>
> The patch below should fix the problm.  Give it a try.

It does :-)
The "Front" control now sets the speaker volume.

> Takashi
Rafael

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: Fwd: [PATCH] sound working on the macbook pro

[Takashi Iwai]

At Tue, 13 Mar 2007 07:14:19 +0000,
=?UTF-8?Q?Rafael_Esp=C3=ADndola?= wrote:
> 
> > I guess it's a missing initialization for the first line-out jack.
> > It works on ohter devices because the default PIN setup has
> > AC_PIN_OUT_EN.  But, on Mac, it has AC_PIN_IN_EN by some reason.
> >
> > The patch below should fix the problm.  Give it a try.
> 
> It does :-)
> The "Front" control now sets the speaker volume.

OK, the patch is not on HG tree.


thanks,

Takashi

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV