2021-10-18 03:30:15

by Christophe JAILLET

Subject: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

'priv->workqueue' is destroyed in the error handling path of the probe but
not in the remove function.

Add the missing call to release some resources.

Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
Signed-off-by: Christophe JAILLET <[email protected]>
---
This patch is speculative. I'm not sure if the added function call is at
the right place in the remove function.
Review with care.
---
drivers/remoteproc/imx_rproc.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
index ff8170dbbc3c..0a45bc0d3f73 100644
--- a/drivers/remoteproc/imx_rproc.c
+++ b/drivers/remoteproc/imx_rproc.c
@@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device *pdev)
clk_disable_unprepare(priv->clk);
rproc_del(rproc);
imx_rproc_free_mbox(rproc);
+ destroy_workqueue(priv->workqueue);
rproc_free(rproc);

return 0;
--
2.30.2

2021-10-21 17:43:54

by Mathieu Poirier

Subject: Re: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

On Sat, Oct 16, 2021 at 08:44:28AM +0200, Christophe JAILLET wrote:
> 'priv->workqueue' is destroyed in the error handling path of the probe but
> not in the remove function.
>
> Add the missing call to release some resources.
>
> Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
> Signed-off-by: Christophe JAILLET <[email protected]>
> ---
> This patch is speculative. I'm not sure if the added function call is at
> the right place in the remove function.
> Review with care.
> ---
> drivers/remoteproc/imx_rproc.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
> index ff8170dbbc3c..0a45bc0d3f73 100644
> --- a/drivers/remoteproc/imx_rproc.c
> +++ b/drivers/remoteproc/imx_rproc.c
> @@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device *pdev)
> clk_disable_unprepare(priv->clk);
> rproc_del(rproc);
> imx_rproc_free_mbox(rproc);
> + destroy_workqueue(priv->workqueue);

This seems to be the right thing to do. Peng - please review and test on your
side.

Thanks,
Mathieu

> rproc_free(rproc);
>
> return 0;
> --
> 2.30.2
>

2021-11-10 08:49:49

by Peng Fan

Subject: RE: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

> Subject: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove
> function
>
> 'priv->workqueue' is destroyed in the error handling path of the probe but not
> in the remove function.
>
> Add the missing call to release some resources.
>
> Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
> Signed-off-by: Christophe JAILLET <[email protected]>


Tested-by: Peng Fan <[email protected]>

> ---
> This patch is speculative. I'm not sure if the added function call is at the right
> place in the remove function.
> Review with care.
> ---
> drivers/remoteproc/imx_rproc.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/remoteproc/imx_rproc.c
> b/drivers/remoteproc/imx_rproc.c index ff8170dbbc3c..0a45bc0d3f73
> 100644
> --- a/drivers/remoteproc/imx_rproc.c
> +++ b/drivers/remoteproc/imx_rproc.c
> @@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device
> *pdev)
> clk_disable_unprepare(priv->clk);
> rproc_del(rproc);
> imx_rproc_free_mbox(rproc);
> + destroy_workqueue(priv->workqueue);
> rproc_free(rproc);
>
> return 0;
> --
> 2.30.2

2021-11-10 08:54:58

by Peng Fan

Subject: RE: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

> Subject: Re: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the
> remove function
>
> On Sat, Oct 16, 2021 at 08:44:28AM +0200, Christophe JAILLET wrote:
> > 'priv->workqueue' is destroyed in the error handling path of the probe
> > but not in the remove function.
> >
> > Add the missing call to release some resources.
> >
> > Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
> > Signed-off-by: Christophe JAILLET <[email protected]>
> > ---
> > This patch is speculative. I'm not sure if the added function call is
> > at the right place in the remove function.
> > Review with care.
> > ---
> > drivers/remoteproc/imx_rproc.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/remoteproc/imx_rproc.c
> > b/drivers/remoteproc/imx_rproc.c index ff8170dbbc3c..0a45bc0d3f73
> > 100644
> > --- a/drivers/remoteproc/imx_rproc.c
> > +++ b/drivers/remoteproc/imx_rproc.c
> > @@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device
> *pdev)
> > clk_disable_unprepare(priv->clk);
> > rproc_del(rproc);
> > imx_rproc_free_mbox(rproc);
> > + destroy_workqueue(priv->workqueue);
>
> This seems to be the right thing to do. Peng - please review and test on your
> side.

Sorry for late reply, I built the driver as a module with this patch. And no issue found.

THnaks,
Peng.

>
> Thanks,
> Mathieu
>
> > rproc_free(rproc);
> >
> > return 0;
> > --
> > 2.30.2
> >

2021-11-10 08:55:23

by Peng Fan

Subject: RE: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

> Subject: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove
> function
>
> 'priv->workqueue' is destroyed in the error handling path of the probe but not
> in the remove function.
>
> Add the missing call to release some resources.
>
> Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
> Signed-off-by: Christophe JAILLET <[email protected]>

Reviewed-and-Tested-by: Peng Fan <[email protected]>

> ---
> This patch is speculative. I'm not sure if the added function call is at the right
> place in the remove function.
> Review with care.
> ---
> drivers/remoteproc/imx_rproc.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/remoteproc/imx_rproc.c
> b/drivers/remoteproc/imx_rproc.c index ff8170dbbc3c..0a45bc0d3f73
> 100644
> --- a/drivers/remoteproc/imx_rproc.c
> +++ b/drivers/remoteproc/imx_rproc.c
> @@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device
> *pdev)
> clk_disable_unprepare(priv->clk);
> rproc_del(rproc);
> imx_rproc_free_mbox(rproc);
> + destroy_workqueue(priv->workqueue);
> rproc_free(rproc);
>
> return 0;
> --
> 2.30.2

2021-11-11 18:20:37

by Mathieu Poirier

Subject: Re: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove function

On Wed, 10 Nov 2021 at 01:54, Peng Fan <[email protected]> wrote:
>
> > Subject: [PATCH] remoteproc: imx_rproc: Fix a resource leak in the remove
> > function
> >
> > 'priv->workqueue' is destroyed in the error handling path of the probe but not
> > in the remove function.
> >
> > Add the missing call to release some resources.
> >
> > Fixes: 2df7062002d0 ("remoteproc: imx_proc: enable virtio/mailbox")
> > Signed-off-by: Christophe JAILLET <[email protected]>
>
> Reviewed-and-Tested-by: Peng Fan <[email protected]>
>

Thank you for that - I will pick up this patch when -rc1 or -rc2 gets released.

Thanks,
Mathieu

> > ---
> > This patch is speculative. I'm not sure if the added function call is at the right
> > place in the remove function.
> > Review with care.
> > ---
> > drivers/remoteproc/imx_rproc.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/remoteproc/imx_rproc.c
> > b/drivers/remoteproc/imx_rproc.c index ff8170dbbc3c..0a45bc0d3f73
> > 100644
> > --- a/drivers/remoteproc/imx_rproc.c
> > +++ b/drivers/remoteproc/imx_rproc.c
> > @@ -804,6 +804,7 @@ static int imx_rproc_remove(struct platform_device
> > *pdev)
> > clk_disable_unprepare(priv->clk);
> > rproc_del(rproc);
> > imx_rproc_free_mbox(rproc);
> > + destroy_workqueue(priv->workqueue);
> > rproc_free(rproc);
> >
> > return 0;
> > --
> > 2.30.2
>