Re: [PATCH] ASoC: amd: correct pm_runtime enable sequence

Re: [PATCH] ASoC: amd: correct pm_runtime enable sequence

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 324 bytes --]

On Wed, Jun 14, 2023 at 06:01:59AM +0530, Arun Gopal Kondaveeti wrote:

> Replace pm_runtime_allow() with pm_runtime_mark_last_busy()
> & pm_runtime_set_active() in pm_runtime enable sequence for
> dma drivers.

This says what the change does but it doesn't say why the old sequence
was a problem and the new one is better.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[PATCH] ASoC: amd: correct pm_runtime enable sequence

[Arun Gopal Kondaveeti]

Replace pm_runtime_allow() with pm_runtime_mark_last_busy()
& pm_runtime_set_active() in pm_runtime enable sequence for
dma drivers.

Signed-off-by: Arun Gopal <arungopal.kondaveeti@amd.com>
---
 sound/soc/amd/ps/ps-pdm-dma.c         | 3 ++-
 sound/soc/amd/raven/acp3x-pcm-dma.c   | 3 ++-
 sound/soc/amd/renoir/acp3x-pdm-dma.c  | 3 ++-
 sound/soc/amd/vangogh/acp5x-pcm-dma.c | 4 ++--
 sound/soc/amd/yc/acp6x-pdm-dma.c      | 3 ++-
 5 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/sound/soc/amd/ps/ps-pdm-dma.c b/sound/soc/amd/ps/ps-pdm-dma.c
index bdbbb797c74d..d48f7c5af289 100644
--- a/sound/soc/amd/ps/ps-pdm-dma.c
+++ b/sound/soc/amd/ps/ps-pdm-dma.c
@@ -391,8 +391,9 @@ static int acp63_pdm_audio_probe(struct platform_device *pdev)
 	}
 	pm_runtime_set_autosuspend_delay(&pdev->dev, ACP_SUSPEND_DELAY_MS);
 	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
 	return 0;
 }
 
diff --git a/sound/soc/amd/raven/acp3x-pcm-dma.c b/sound/soc/amd/raven/acp3x-pcm-dma.c
index 7362dd15ad30..9538f3ffc5d9 100644
--- a/sound/soc/amd/raven/acp3x-pcm-dma.c
+++ b/sound/soc/amd/raven/acp3x-pcm-dma.c
@@ -416,8 +416,9 @@ static int acp3x_audio_probe(struct platform_device *pdev)
 
 	pm_runtime_set_autosuspend_delay(&pdev->dev, 2000);
 	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
 	return 0;
 }
 
diff --git a/sound/soc/amd/renoir/acp3x-pdm-dma.c b/sound/soc/amd/renoir/acp3x-pdm-dma.c
index 4e299f96521f..c3b47e9bd239 100644
--- a/sound/soc/amd/renoir/acp3x-pdm-dma.c
+++ b/sound/soc/amd/renoir/acp3x-pdm-dma.c
@@ -430,8 +430,9 @@ static int acp_pdm_audio_probe(struct platform_device *pdev)
 	}
 	pm_runtime_set_autosuspend_delay(&pdev->dev, ACP_SUSPEND_DELAY_MS);
 	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
 	return 0;
 }
 
diff --git a/sound/soc/amd/vangogh/acp5x-pcm-dma.c b/sound/soc/amd/vangogh/acp5x-pcm-dma.c
index 29901ee4bfe3..587dec5bb33d 100644
--- a/sound/soc/amd/vangogh/acp5x-pcm-dma.c
+++ b/sound/soc/amd/vangogh/acp5x-pcm-dma.c
@@ -409,9 +409,9 @@ static int acp5x_audio_probe(struct platform_device *pdev)
 	}
 	pm_runtime_set_autosuspend_delay(&pdev->dev, 2000);
 	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
-
 	return 0;
 }
 
diff --git a/sound/soc/amd/yc/acp6x-pdm-dma.c b/sound/soc/amd/yc/acp6x-pdm-dma.c
index d818eba48546..72c4591e451b 100644
--- a/sound/soc/amd/yc/acp6x-pdm-dma.c
+++ b/sound/soc/amd/yc/acp6x-pdm-dma.c
@@ -383,8 +383,9 @@ static int acp6x_pdm_audio_probe(struct platform_device *pdev)
 	}
 	pm_runtime_set_autosuspend_delay(&pdev->dev, ACP_SUSPEND_DELAY_MS);
 	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
 	return 0;
 }
 
-- 
2.25.1

Re: [PATCH] ASoC: amd: correct pm_runtime enable sequence

[kondaveeti, Arungopal]

On 6/13/2023 7:39 PM, Mark Brown wrote:
> On Wed, Jun 14, 2023 at 06:01:59AM +0530, Arun Gopal Kondaveeti wrote:
>
>> Replace pm_runtime_allow() with pm_runtime_mark_last_busy()
>> & pm_runtime_set_active() in pm_runtime enable sequence for
>> dma drivers.
> This says what the change does but it doesn't say why the old sequence
> was a problem and the new one is better.

pm_runtime_allow is not needed for child platform drivers .

Previous sequence as listed below which is incorrect.

enable sequence:

     pm_runtime_set_autosuspend_delay(&pdev->dev, ACP_SUSPEND_DELAY_MS);
     pm_runtime_use_autosuspend(&pdev->dev);
     pm_runtime_enable(&pdev->dev);
     pm_runtime_allow(&pdev->dev);

disable sequence:

pm_runtime_disable(&pdev->dev)

Replaced pm runtime enable sequence.

Re: [PATCH] ASoC: amd: correct pm_runtime enable sequence

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 656 bytes --]

On Fri, Jun 16, 2023 at 06:28:29PM +0530, kondaveeti, Arungopal wrote:
> On 6/13/2023 7:39 PM, Mark Brown wrote:
> > On Wed, Jun 14, 2023 at 06:01:59AM +0530, Arun Gopal Kondaveeti wrote:

> > > Replace pm_runtime_allow() with pm_runtime_mark_last_busy()
> > > & pm_runtime_set_active() in pm_runtime enable sequence for
> > > dma drivers.

> > This says what the change does but it doesn't say why the old sequence
> > was a problem and the new one is better.

> pm_runtime_allow is not needed for child platform drivers .

If it's not needed that explains the removal but not the addition of
set_active()...  This information should be in the changelog.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]