[PATCH 1/8] ASoC: dapm: Fix marking widgets dirty when a route is added

[PATCH 1/8] ASoC: dapm: Fix marking widgets dirty when a route is added

[Lars-Peter Clausen]

The current calls to dapm_mark_dirty() in snd_soc_dapm_add_path() are on a path
that is only reached if the sink widget is either a mixer or a mux. Move the
calls further up so they are called for all widget types.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/soc-dapm.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c
index d84bd0f..7e9afbc4 100644
--- a/sound/soc/soc-dapm.c
+++ b/sound/soc/soc-dapm.c
@@ -2374,6 +2374,9 @@ static int snd_soc_dapm_add_path(struct snd_soc_dapm_context *dapm,
 			wsource->ext = 1;
 	}
 
+	dapm_mark_dirty(wsource, "Route added");
+	dapm_mark_dirty(wsink, "Route added");
+
 	/* connect static paths */
 	if (control == NULL) {
 		list_add(&path->list, &dapm->card->paths);
@@ -2436,9 +2439,6 @@ static int snd_soc_dapm_add_path(struct snd_soc_dapm_context *dapm,
 		return 0;
 	}
 
-	dapm_mark_dirty(wsource, "Route added");
-	dapm_mark_dirty(wsink, "Route added");
-
 	return 0;
 err:
 	kfree(path);
-- 
1.8.0

[PATCH 2/8] ASoC: dmic: Convert table based DAPM setup

[Lars-Peter Clausen]

Let the core take care of instantiating the DAPM widgets and routes, this makes
the code a bit shorter.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/codecs/dmic.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/sound/soc/codecs/dmic.c b/sound/soc/codecs/dmic.c
index 66967ba..b2090b2 100644
--- a/sound/soc/codecs/dmic.c
+++ b/sound/soc/codecs/dmic.c
@@ -50,20 +50,11 @@ static const struct snd_soc_dapm_route intercon[] = {
 	{"DMIC AIF", NULL, "DMic"},
 };
 
-static int dmic_probe(struct snd_soc_codec *codec)
-{
-	struct snd_soc_dapm_context *dapm = &codec->dapm;
-
-	snd_soc_dapm_new_controls(dapm, dmic_dapm_widgets,
-				  ARRAY_SIZE(dmic_dapm_widgets));
-        snd_soc_dapm_add_routes(dapm, intercon, ARRAY_SIZE(intercon));
-	snd_soc_dapm_new_widgets(dapm);
-
-	return 0;
-}
-
 static struct snd_soc_codec_driver soc_dmic = {
-	.probe	= dmic_probe,
+	.dapm_widgets = dmic_dapm_widgets,
+	.num_dapm_widgets = ARRAY_SIZE(dmic_dapm_widgets),
+	.dapm_routes = intercon,
+	.num_dapm_routes = ARRAY_SIZE(intercon),
 };
 
 static int dmic_dev_probe(struct platform_device *pdev)
-- 
1.8.0

Re: [PATCH 2/8] ASoC: dmic: Convert table based DAPM setup

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 194 bytes --]

On Tue, Aug 27, 2013 at 03:50:55PM +0200, Lars-Peter Clausen wrote:
> Let the core take care of instantiating the DAPM widgets and routes, this makes
> the code a bit shorter.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 3/8] ASoC: tlv320aic32x4: Convert table based control and DAPM setup

[Lars-Peter Clausen]

Let the core take care of instantiating the controls and DAPM widgets and
routes, this makes the code a bit shorter.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Cc: Javier Martin <javier.martin@vista-silicon.com>
---
 sound/soc/codecs/tlv320aic32x4.c | 22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)

diff --git a/sound/soc/codecs/tlv320aic32x4.c b/sound/soc/codecs/tlv320aic32x4.c
index 17df4e3..2ed57d4 100644
--- a/sound/soc/codecs/tlv320aic32x4.c
+++ b/sound/soc/codecs/tlv320aic32x4.c
@@ -338,18 +338,6 @@ static inline int aic32x4_get_divs(int mclk, int rate)
 	return -EINVAL;
 }
 
-static int aic32x4_add_widgets(struct snd_soc_codec *codec)
-{
-	snd_soc_dapm_new_controls(&codec->dapm, aic32x4_dapm_widgets,
-				  ARRAY_SIZE(aic32x4_dapm_widgets));
-
-	snd_soc_dapm_add_routes(&codec->dapm, aic32x4_dapm_routes,
-				ARRAY_SIZE(aic32x4_dapm_routes));
-
-	snd_soc_dapm_new_widgets(&codec->dapm);
-	return 0;
-}
-
 static int aic32x4_set_dai_sysclk(struct snd_soc_dai *codec_dai,
 				  int clk_id, unsigned int freq, int dir)
 {
@@ -683,9 +671,6 @@ static int aic32x4_probe(struct snd_soc_codec *codec)
 	}
 
 	aic32x4_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
-	snd_soc_add_codec_controls(codec, aic32x4_snd_controls,
-			     ARRAY_SIZE(aic32x4_snd_controls));
-	aic32x4_add_widgets(codec);
 
 	/*
 	 * Workaround: for an unknown reason, the ADC needs to be powered up
@@ -714,6 +699,13 @@ static struct snd_soc_codec_driver soc_codec_dev_aic32x4 = {
 	.suspend = aic32x4_suspend,
 	.resume = aic32x4_resume,
 	.set_bias_level = aic32x4_set_bias_level,
+
+	.controls = aic32x4_snd_controls,
+	.num_controls = ARRAY_SIZE(aic32x4_snd_controls),
+	.dapm_widgets = aic32x4_dapm_widgets,
+	.num_dapm_widgets = ARRAY_SIZE(aic32x4_dapm_widgets),
+	.dapm_routes = aic32x4_dapm_routes,
+	.num_dapm_routes = ARRAY_SIZE(aic32x4_dapm_routes),
 };
 
 static int aic32x4_i2c_probe(struct i2c_client *i2c,
-- 
1.8.0

Re: [PATCH 3/8] ASoC: tlv320aic32x4: Convert table based control and DAPM setup

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 399 bytes --]

On Tue, Aug 27, 2013 at 03:50:56PM +0200, Lars-Peter Clausen wrote:
> Let the core take care of instantiating the controls and DAPM widgets and
> routes, this makes the code a bit shorter.

Applied, thanks.

Please don't submit a bunch of unrelated patches as a series, submit
them separately.  It's a bit easier for review and will avoid later
patches getting ignored if there's problems early on.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: [PATCH 3/8] ASoC: tlv320aic32x4: Convert table based control and DAPM setup

[Lars-Peter Clausen]

On 08/27/2013 04:35 PM, Mark Brown wrote:
> On Tue, Aug 27, 2013 at 03:50:56PM +0200, Lars-Peter Clausen wrote:
>> Let the core take care of instantiating the controls and DAPM widgets and
>> routes, this makes the code a bit shorter.
> 
> Applied, thanks.
> 
> Please don't submit a bunch of unrelated patches as a series, submit
> them separately.

The last patch in the series depends on all the other patches in this series
(maybe except for the mark dirty fix).

- Lars

Re: [PATCH 3/8] ASoC: tlv320aic32x4: Convert table based control and DAPM setup

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 404 bytes --]

On Tue, Aug 27, 2013 at 04:41:21PM +0200, Lars-Peter Clausen wrote:
> On 08/27/2013 04:35 PM, Mark Brown wrote:

> > Please don't submit a bunch of unrelated patches as a series, submit
> > them separately.

> The last patch in the series depends on all the other patches in this series
> (maybe except for the mark dirty fix).

Yes, it was mainly the mark_dirty change which shouldn't have any
overlap.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 4/8] ASoC; wm8904: Remove unnecessary call to snd_soc_dapm_new_widgets()

[Lars-Peter Clausen]

The core will call snd_soc_dapm_new_widgets() once all components of the card
have been initialized, so there is no need to do this manually in the driver.
Calling it earlier also might result in a partially instantiated system being
powered up which cause undesired side effects.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/codecs/wm8904.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/sound/soc/codecs/wm8904.c b/sound/soc/codecs/wm8904.c
index 91dfbfe..4dfa8dc 100644
--- a/sound/soc/codecs/wm8904.c
+++ b/sound/soc/codecs/wm8904.c
@@ -1202,7 +1202,6 @@ static int wm8904_add_widgets(struct snd_soc_codec *codec)
 		break;
 	}
 
-	snd_soc_dapm_new_widgets(dapm);
 	return 0;
 }
 
-- 
1.8.0

Re: [PATCH 4/8] ASoC; wm8904: Remove unnecessary call to snd_soc_dapm_new_widgets()

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 375 bytes --]

On Tue, Aug 27, 2013 at 03:50:57PM +0200, Lars-Peter Clausen wrote:
> The core will call snd_soc_dapm_new_widgets() once all components of the card
> have been initialized, so there is no need to do this manually in the driver.
> Calling it earlier also might result in a partially instantiated system being
> powered up which cause undesired side effects.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 5/8] ASoC: jack: Remove unnecessary call to snd_soc_dapm_new_widgets()

[Lars-Peter Clausen]

snd_soc_jack_add_pins() does not create any new DAPM widgets, so there is no
need to call snd_soc_dapm_new_widgets().

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/soc-jack.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/sound/soc/soc-jack.c b/sound/soc/soc-jack.c
index 7aa26b5..71358e3 100644
--- a/sound/soc/soc-jack.c
+++ b/sound/soc/soc-jack.c
@@ -183,8 +183,6 @@ int snd_soc_jack_add_pins(struct snd_soc_jack *jack, int count,
 		list_add(&(pins[i].list), &jack->pins);
 	}
 
-	snd_soc_dapm_new_widgets(&jack->codec->card->dapm);
-
 	/* Update to reflect the last reported status; canned jack
 	 * implementations are likely to set their state before the
 	 * card has an opportunity to associate pins.
-- 
1.8.0

Re: [PATCH 5/8] ASoC: jack: Remove unnecessary call to snd_soc_dapm_new_widgets()

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 208 bytes --]

On Tue, Aug 27, 2013 at 03:50:58PM +0200, Lars-Peter Clausen wrote:
> snd_soc_jack_add_pins() does not create any new DAPM widgets, so there is no
> need to call snd_soc_dapm_new_widgets().

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 6/8] ASoC: Call snd_soc_dapm_new_widgets() only once during card initialization

[Lars-Peter Clausen]

Each time snd_soc_dapm_new_widgets() is called it will instantiate all the
widgets and routes that have been added so far and then power them. Doing this
multiple times before the card is fully initialized and all widgets have been
added can cause unnecessary and even invalid power state transitions which can
result in extra register writes and and also might cause clicks and pops.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/soc-core.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index 528f870..aef5e8c 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -1243,9 +1243,6 @@ static int soc_post_component_init(struct snd_soc_card *card,
 	}
 	rtd->card = card;
 
-	/* Make sure all DAPM widgets are instantiated */
-	snd_soc_dapm_new_widgets(&codec->dapm);
-
 	/* machine controls, routes and widgets are not prefixed */
 	temp = codec->name_prefix;
 	codec->name_prefix = NULL;
@@ -1741,8 +1738,6 @@ static int snd_soc_instantiate_card(struct snd_soc_card *card)
 		snd_soc_dapm_add_routes(&card->dapm, card->dapm_routes,
 					card->num_dapm_routes);
 
-	snd_soc_dapm_new_widgets(&card->dapm);
-
 	for (i = 0; i < card->num_links; i++) {
 		dai_link = &card->dai_link[i];
 		dai_fmt = dai_link->dai_fmt;
-- 
1.8.0

Re: [PATCH 6/8] ASoC: Call snd_soc_dapm_new_widgets() only once during card initialization

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 481 bytes --]

On Tue, Aug 27, 2013 at 03:50:59PM +0200, Lars-Peter Clausen wrote:
> Each time snd_soc_dapm_new_widgets() is called it will instantiate all the
> widgets and routes that have been added so far and then power them. Doing this
> multiple times before the card is fully initialized and all widgets have been
> added can cause unnecessary and even invalid power state transitions which can
> result in extra register writes and and also might cause clicks and pops.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 7/8] ASoC: Move call to snd_soc_dapm_new_widgets() after snd_soc_dapm_auto_nc_codec_pins()

[Lars-Peter Clausen]

Call snd_soc_dapm_new_widgets() before the auto non-connected pins have been
marked as not connected will power the system under the assumption that those
pins are connected. Once the pins have been marked as disconnected the system
there will be an additional power run. This can cause unnecessary power
transitions. Calling snd_soc_dapm_new_widgets() only after the pins have been
marked as non-connected avoids this.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 sound/soc/soc-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index aef5e8c..c3c4906 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -1816,12 +1816,12 @@ static int snd_soc_instantiate_card(struct snd_soc_card *card)
 		}
 	}
 
-	snd_soc_dapm_new_widgets(&card->dapm);
-
 	if (card->fully_routed)
 		list_for_each_entry(codec, &card->codec_dev_list, card_list)
 			snd_soc_dapm_auto_nc_codec_pins(codec);
 
+	snd_soc_dapm_new_widgets(&card->dapm);
+
 	ret = snd_card_register(card->snd_card);
 	if (ret < 0) {
 		dev_err(card->dev, "ASoC: failed to register soundcard %d\n",
-- 
1.8.0

Re: [PATCH 7/8] ASoC: Move call to snd_soc_dapm_new_widgets() after snd_soc_dapm_auto_nc_codec_pins()

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 518 bytes --]

On Tue, Aug 27, 2013 at 03:51:00PM +0200, Lars-Peter Clausen wrote:
> Call snd_soc_dapm_new_widgets() before the auto non-connected pins have been
> marked as not connected will power the system under the assumption that those
> pins are connected. Once the pins have been marked as disconnected the system
> there will be an additional power run. This can cause unnecessary power
> transitions. Calling snd_soc_dapm_new_widgets() only after the pins have been
> marked as non-connected avoids this.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

[PATCH 8/8] ASoC: Pass card instead of dapm context to snd_soc_dapm_new_widgets()

[Lars-Peter Clausen]

snd_soc_dapm_new_widgets() works on the ASoC card as a whole not on a specific
DAPM context. The DAPM context that is passed as the parameter is only used to
look up the pointer to the card. This patch updates the signature of
snd_soc_dapm_new_widgets() to take the card directly.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
---
 include/sound/soc-dapm.h | 2 +-
 sound/soc/soc-core.c     | 2 +-
 sound/soc/soc-dapm.c     | 3 +--
 3 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h
index c728d28..27a72d5 100644
--- a/include/sound/soc-dapm.h
+++ b/include/sound/soc-dapm.h
@@ -413,7 +413,7 @@ int snd_soc_dapm_new_pcm(struct snd_soc_card *card,
 			 struct snd_soc_dapm_widget *sink);
 
 /* dapm path setup */
-int snd_soc_dapm_new_widgets(struct snd_soc_dapm_context *dapm);
+int snd_soc_dapm_new_widgets(struct snd_soc_card *card);
 void snd_soc_dapm_free(struct snd_soc_dapm_context *dapm);
 int snd_soc_dapm_add_routes(struct snd_soc_dapm_context *dapm,
 			    const struct snd_soc_dapm_route *route, int num);
diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index c3c4906..7d13c4f 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -1820,7 +1820,7 @@ static int snd_soc_instantiate_card(struct snd_soc_card *card)
 		list_for_each_entry(codec, &card->codec_dev_list, card_list)
 			snd_soc_dapm_auto_nc_codec_pins(codec);
 
-	snd_soc_dapm_new_widgets(&card->dapm);
+	snd_soc_dapm_new_widgets(card);
 
 	ret = snd_card_register(card->snd_card);
 	if (ret < 0) {
diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c
index 7e9afbc4..548b1c9 100644
--- a/sound/soc/soc-dapm.c
+++ b/sound/soc/soc-dapm.c
@@ -2712,9 +2712,8 @@ EXPORT_SYMBOL_GPL(snd_soc_dapm_weak_routes);
  *
  * Returns 0 for success.
  */
-int snd_soc_dapm_new_widgets(struct snd_soc_dapm_context *dapm)
+int snd_soc_dapm_new_widgets(struct snd_soc_card *card)
 {
-	struct snd_soc_card *card = dapm->card;
 	struct snd_soc_dapm_widget *w;
 	unsigned int val;
 
-- 
1.8.0

Re: [PATCH 8/8] ASoC: Pass card instead of dapm context to snd_soc_dapm_new_widgets()

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 375 bytes --]

On Tue, Aug 27, 2013 at 03:51:01PM +0200, Lars-Peter Clausen wrote:
> snd_soc_dapm_new_widgets() works on the ASoC card as a whole not on a specific
> DAPM context. The DAPM context that is passed as the parameter is only used to
> look up the pointer to the card. This patch updates the signature of
> snd_soc_dapm_new_widgets() to take the card directly.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: [PATCH 1/8] ASoC: dapm: Fix marking widgets dirty when a route is added

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 307 bytes --]

On Tue, Aug 27, 2013 at 03:50:54PM +0200, Lars-Peter Clausen wrote:
> The current calls to dapm_mark_dirty() in snd_soc_dapm_add_path() are on a path
> that is only reached if the sink widget is either a mixer or a mux. Move the
> calls further up so they are called for all widget types.

Applied, thanks.

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]