[PATCH v2 1/2] dmaengine: ti: edma: fix pm_runtime usage counter leak on error

[PATCH v2 1/2] dmaengine: ti: edma: fix pm_runtime usage counter leak on error

[phucduc.bui]

From: bui duc phuc <phucduc.bui@gmail.com>

pm_runtime_get_sync() leaves the runtime PM usage counter incremented even
when it fails, but the error path in edma_probe() does not call
pm_runtime_put_noidle() to balance it, leaking a reference each time
resume fails.

Use pm_runtime_resume_and_get() instead, which automatically drops the
usage counter on failure, fixing the leak.

Fixes: 2a03c1314506 ("dmaengine: ti: edma: add missed operations")
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
---

Changes in v2:
 - Update the commit message: Change the function name from netcp_probe() 
   to edma_probe().

 drivers/dma/ti/edma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/ti/edma.c b/drivers/dma/ti/edma.c
index d97db5af3555..8be9e71dab76 100644
--- a/drivers/dma/ti/edma.c
+++ b/drivers/dma/ti/edma.c
@@ -2345,9 +2345,9 @@ static int edma_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, ecc);
 
 	pm_runtime_enable(dev);
-	ret = pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
 	if (ret < 0) {
-		dev_err(dev, "pm_runtime_get_sync() failed\n");
+		dev_err(dev, "pm_runtime_resume_and_get() failed\n");
 		pm_runtime_disable(dev);
 		return ret;
 	}
-- 
2.43.0

[PATCH v2 2/2] dmaengine: ti: edma: Fix missing pm_runtime cleanup on probe failure

[phucduc.bui]

From: bui duc phuc <phucduc.bui@gmail.com>

pm_runtime_get_sync() errors were not checked, so on failure the usage
count leaked and pm_runtime_enable() was never undone, leaving
runtime PM in a half-initialized state after a failed probe.

Switch to pm_runtime_resume_and_get(), which cleans up the usage count
internally on error, and call pm_runtime_disable() to balance the earlier
pm_runtime_enable() on that path.

Fixes: 23f49fd2ea9b ("dmaengine: edma: Remove dynamic TPTC power management feature")
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
---
 drivers/dma/ti/edma.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/dma/ti/edma.c b/drivers/dma/ti/edma.c
index 8be9e71dab76..5254ce7b8514 100644
--- a/drivers/dma/ti/edma.c
+++ b/drivers/dma/ti/edma.c
@@ -2650,8 +2650,16 @@ static struct platform_driver edma_driver = {
 
 static int edma_tptc_probe(struct platform_device *pdev)
 {
+	int ret;
+
 	pm_runtime_enable(&pdev->dev);
-	return pm_runtime_get_sync(&pdev->dev);
+	ret = pm_runtime_resume_and_get(&pdev->dev);
+	if (ret < 0) {
+		pm_runtime_disable(&pdev->dev);
+		return ret;
+	}
+
+	return 0;
 }
 
 static struct platform_driver edma_tptc_driver = {
-- 
2.43.0

Re: [PATCH v2 1/2] dmaengine: ti: edma: fix pm_runtime usage counter leak on error

[Frank Li]

On Wed, Sep 23, 2026 at 02:25:31PM +0700, phucduc.bui@gmail.com wrote:
> From: bui duc phuc <phucduc.bui@gmail.com>
>
> pm_runtime_get_sync() leaves the runtime PM usage counter incremented even
> when it fails, but the error path in edma_probe() does not call
> pm_runtime_put_noidle() to balance it, leaking a reference each time
> resume fails.
>
> Use pm_runtime_resume_and_get() instead, which automatically drops the
> usage counter on failure, fixing the leak.
>
> Fixes: 2a03c1314506 ("dmaengine: ti: edma: add missed operations")
> Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
> ---

Reviewed-by: Frank Li <Frank.Li@nxp.com>

>
> Changes in v2:
>  - Update the commit message: Change the function name from netcp_probe()
>    to edma_probe().
>
>  drivers/dma/ti/edma.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/dma/ti/edma.c b/drivers/dma/ti/edma.c
> index d97db5af3555..8be9e71dab76 100644
> --- a/drivers/dma/ti/edma.c
> +++ b/drivers/dma/ti/edma.c
> @@ -2345,9 +2345,9 @@ static int edma_probe(struct platform_device *pdev)
>  	platform_set_drvdata(pdev, ecc);
>
>  	pm_runtime_enable(dev);
> -	ret = pm_runtime_get_sync(dev);
> +	ret = pm_runtime_resume_and_get(dev);
>  	if (ret < 0) {
> -		dev_err(dev, "pm_runtime_get_sync() failed\n");
> +		dev_err(dev, "pm_runtime_resume_and_get() failed\n");
>  		pm_runtime_disable(dev);
>  		return ret;
>  	}
> --
> 2.43.0
>

Re: [PATCH v2 2/2] dmaengine: ti: edma: Fix missing pm_runtime cleanup on probe failure

[Frank Li]

On Wed, Sep 23, 2026 at 02:25:32PM +0700, phucduc.bui@gmail.com wrote:
> From: bui duc phuc <phucduc.bui@gmail.com>
>
> pm_runtime_get_sync() errors were not checked, so on failure the usage
> count leaked and pm_runtime_enable() was never undone, leaving
> runtime PM in a half-initialized state after a failed probe.
>
> Switch to pm_runtime_resume_and_get(), which cleans up the usage count
> internally on error, and call pm_runtime_disable() to balance the earlier
> pm_runtime_enable() on that path.
>
> Fixes: 23f49fd2ea9b ("dmaengine: edma: Remove dynamic TPTC power management feature")
> Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
> ---

Reviewed-by: Frank Li <Frank.Li@nxp.com>

>  drivers/dma/ti/edma.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/dma/ti/edma.c b/drivers/dma/ti/edma.c
> index 8be9e71dab76..5254ce7b8514 100644
> --- a/drivers/dma/ti/edma.c
> +++ b/drivers/dma/ti/edma.c
> @@ -2650,8 +2650,16 @@ static struct platform_driver edma_driver = {
>
>  static int edma_tptc_probe(struct platform_device *pdev)
>  {
> +	int ret;
> +
>  	pm_runtime_enable(&pdev->dev);
> -	return pm_runtime_get_sync(&pdev->dev);
> +	ret = pm_runtime_resume_and_get(&pdev->dev);
> +	if (ret < 0) {
> +		pm_runtime_disable(&pdev->dev);
> +		return ret;
> +	}
> +
> +	return 0;
>  }
>
>  static struct platform_driver edma_tptc_driver = {
> --
> 2.43.0
>