Skip to contentSkip to conversation

ASoC: codec: hdac_hdmi add device_link to card device

PATCH Superseded [email protected] 3 participants 9 messages Apr 12, 2019 – Apr 13, 2019

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ASoC: codec: hdac_hdmi add device_link to card device

From: Libin Yang <[email protected]>

In resume from S3, HDAC HDMI codec driver dapm event callback may be operated before HDMI codec driver turns on the display audio power domain because of the contest between display driver and hdmi codec driver.

This patch adds the device_link between soc card device (consumer) and hdmi codec device (supplier) to make sure the sequence is always correct.

---
 sound/soc/codecs/hdac_hdmi.c | 9 +++++++++
 1 file changed, 9 insertions(+)
sound/soc/codecs/hdac_hdmi.c +9 -0
index 5eeb0fe..8bb883e 100644
@@ -1855,6 +1855,15 @@ static int hdmi_codec_probe(struct snd_soc_component *component)
 	hdmi->card = dapm->card->snd_card;
 
 	/*
+	 * Setup a device_link between card device and HDMI codec device.
+	 * The card device is the consumer and the HDMI codec device is
+	 * the supplier. With this setting, we can make sure that the audio
+	 * domain in display power will be always turned on before operating
+	 * on the HDMI audio codec registers.
+	 */
+	device_link_add(component->card->dev, &hdev->dev, DL_FLAG_RPM_ACTIVE |
+			DL_FLAG_AUTOREMOVE_CONSUMER);
+	/*
 	 * hdac_device core already sets the state to active and calls
 	 * get_noresume. So enable runtime and set the device to suspend.
 	 */
Signature · 1 line
2.7.4
#1 lore raw
Quoted 20 lines · On 4/11/19 9:40 PM, [email protected] wrote:

From: Libin Yang <[email protected]>

In resume from S3, HDAC HDMI codec driver dapm event callback may be operated before HDMI codec driver turns on the display audio power domain because of the contest between display driver and hdmi codec driver.

This patch adds the device_link between soc card device (consumer) and hdmi codec device (supplier) to make sure the sequence is always correct.

---
  sound/soc/codecs/hdac_hdmi.c | 9 +++++++++
  1 file changed, 9 insertions(+)
sound/soc/codecs/hdac_hdmi.c +0 -0
index 5eeb0fe..8bb883e 100644
@@ -1855,6 +1855,15 @@ static int hdmi_codec_probe(struct snd_soc_component *component)
  	hdmi->card = dapm->card->snd_card;

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 10 lines
sound/soc/codecs/hdac_hdmi.c +8 -0
 
  	/*
+	 * Setup a device_link between card device and HDMI codec device.
+	 * The card device is the consumer and the HDMI codec device is
+	 * the supplier. With this setting, we can make sure that the audio
+	 * domain in display power will be always turned on before operating
+	 * on the HDMI audio codec registers.
+	 */
+	device_link_add(component->card->dev, &hdev->dev, DL_FLAG_RPM_ACTIVE |
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then? Thanks!

#2 lore raw

Hi Pierre,

Quoted 41 lines

-----Original Message-----
From: Pierre-Louis Bossart [mailto:[email protected]]
Sent: Friday, April 12, 2019 11:00 AM
To: Yang, Libin <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: Re: [alsa-devel] [PATCH] ASoC: codec: hdac_hdmi add device_link to card device

Quoted 5 lines · On 4/11/19 9:40 PM, [email protected] wrote:

From: Libin Yang <[email protected]>

In resume from S3, HDAC HDMI codec driver dapm event callback may be operated before HDMI codec driver turns on the display audio power domain because of the contest between display driver and hdmi codec

driver.

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct

Quoted 14 lines

This patch adds the device_link between soc card device (consumer) and hdmi codec device (supplier) to make sure the sequence is always correct.

---
  sound/soc/codecs/hdac_hdmi.c | 9 +++++++++
  1 file changed, 9 insertions(+)
(quoted diff) +0 -0

b/sound/soc/codecs/hdac_hdmi.c index 5eeb0fe..8bb883e 100644

sound/soc/codecs/hdac_hdmi.c +0 -0
@@ -1855,6 +1855,15 @@ static int hdmi_codec_probe(struct

snd_soc_component *component)

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component)

Quoted 10 lines
sound/soc/codecs/hdac_hdmi.c +7 -0
  	hdmi->card = dapm->card->snd_card;
 
  	/*
+	 * Setup a device_link between card device and HDMI codec device.
+	 * The card device is the consumer and the HDMI codec device is
+	 * the supplier. With this setting, we can make sure that the audio
+	 * domain in display power will be always turned on before operating
+	 * on the HDMI audio codec registers.
+	 */
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

Quoted 3 lines

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER flag. This will make sure the link will be freed when machine driver are removed. And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

Regards,
Libin

Quoted 1 line

Thanks!

#3 lore raw
Quoted 11 lines
Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER flag. This will make sure the link will be freed when machine driver are removed. And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

#4 lore raw

Hi Pierre,

Quoted 15 lines

-----Original Message-----
From: Pierre-Louis Bossart [mailto:[email protected]]
Sent: Friday, April 12, 2019 9:52 PM
To: Yang, Libin <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: Re: [alsa-devel] [PATCH] ASoC: codec: hdac_hdmi add device_link to card device

Quoted 6 lines
Quoted 5 lines

Review

Quoted 1 line
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

Quoted 7 lines
Quoted 4 lines

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER flag. This will make sure the link will be freed when machine driver are removed. And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

Thanks for suggestion. I will add the comment.

Regards,
Libin

#5 lore raw
In reply to Pierre-Louis Bossart
Quoted 12 lines
Quoted 12 lines
Quoted 11 lines
Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER flag. This will make sure the link will be freed when machine driver are removed. And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

Quoted 4 lines
Quoted 2 lines

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

Thanks for suggestion. I will add the comment.

After a second thought, is there any possibility that the machine driver does not depend on the hdac_hdmi module? I checked our current driver, there is the dependency between machine driver and hdac_hdmi module.

Regards,
Libin

#6 lore raw
Quoted 12 lines · Libin
Quoted 12 lines
Quoted 12 lines
Quoted 11 lines
Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER flag. This will make sure the link will be freed when machine driver are removed. And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

Quoted 8 lines
Quoted 4 lines
Quoted 2 lines

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

Thanks for suggestion. I will add the comment.

After a second thought, is there any possibility that the machine driver does not depend on the hdac_hdmi module? I checked our current driver, there is the dependency between machine driver and hdac_hdmi module.

I don't get your question.
All machine drivers supporting the iDISP link explicitly use a select HDAC_HDMI in the Kconfig and refer to DAIs created/exposed by the hdac_hdmi module.
It's possible that the hdac_hdmi codec is probed as a result of the platform driver configuration but the machine driver does not support HDMI. In that case, is there any issue with your solution?

#7 lore raw

Hi Pierre,

Quoted 33 lines

-----Original Message-----
From: Pierre-Louis Bossart [mailto:[email protected]]
Sent: Friday, April 12, 2019 11:03 PM
To: Yang, Libin <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: Re: [alsa-devel] [PATCH] ASoC: codec: hdac_hdmi add device_link to card device

Quoted 8 lines · Libin
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

sound/soc/codecs/hdac_hdmi.c static int hdmi_codec_probe(struct snd_soc_component *component) · line 1865

Quoted 1 line
sound/soc/codecs/hdac_hdmi.c +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER

flag.

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

This will make sure the link will be freed when machine driver are

removed.

Quoted 11 lines
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

Thanks for suggestion. I will add the comment.

After a second thought, is there any possibility that the machine driver does not depend on the hdac_hdmi module? I checked our current driver, there is the dependency between machine driver and hdac_hdmi

Quoted 6 lines

module.

I don't get your question.
All machine drivers supporting the iDISP link explicitly use a select HDAC_HDMI in the Kconfig and refer to DAIs created/exposed by the hdac_hdmi module.

So the dependency should always exists. I have checked all the machine drivers and they all depends on the hdac_hdmi if they are using it.

Quoted 3 lines

It's possible that the hdac_hdmi codec is probed as a result of the platform driver configuration but the machine driver does not support HDMI. In that case, is there any issue with your solution?

If machine driver doesn't support HDMI, component will never be called. It is safe.

Hi Takashi,

My patch doesn't use component->init() as you suggested. But they should be similar. Are you OK with this patch?

Regards,
Libin

#8 lore raw
Quoted 50 lines · Libin

Hi Pierre,

Quoted 33 lines

-----Original Message-----
From: Pierre-Louis Bossart [mailto:[email protected]]
Sent: Friday, April 12, 2019 11:03 PM
To: Yang, Libin <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: Re: [alsa-devel] [PATCH] ASoC: codec: hdac_hdmi add device_link to card device

Quoted 8 lines · Libin
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -0
+	device_link_add(component->card->dev, &hdev->dev,

DL_FLAG_RPM_ACTIVE |

Review

Quoted 1 line
(quoted diff) +1 -0
+			DL_FLAG_AUTOREMOVE_CONSUMER);

Should device_link_free() be added to hdmi_codec_remove then?

As Takashi suggested, I add the DL_FLAG_AUTOREMOVE_CONSUMER

flag.

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

This will make sure the link will be freed when machine driver are

removed.

Quoted 11 lines
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

And as machine driver depends on the hdac_hdmi module, when hdmi_codec_remove() is called, the link is freed already.

ok, maybe adding a comment would help dummies like me who didn't know about this flag? Thanks!

Thanks for suggestion. I will add the comment.

After a second thought, is there any possibility that the machine driver does not depend on the hdac_hdmi module? I checked our current driver, there is the dependency between machine driver and hdac_hdmi

Quoted 6 lines

module.

I don't get your question.
All machine drivers supporting the iDISP link explicitly use a select HDAC_HDMI in the Kconfig and refer to DAIs created/exposed by the hdac_hdmi module.

So the dependency should always exists. I have checked all the machine drivers and they all depends on the hdac_hdmi if they are using it.

Quoted 3 lines

It's possible that the hdac_hdmi codec is probed as a result of the platform driver configuration but the machine driver does not support HDMI. In that case, is there any issue with your solution?

Quoted 7 lines

If machine driver doesn't support HDMI, component will never be called. It is safe.

Hi Takashi,

My patch doesn't use component->init() as you suggested. But they should be similar. Are you OK with this patch?

Yes.

Takashi