Potential UAF and memory leak exsit in exception handling paths for
rpmsg_anonymous_eptdev_create(), fix them.

- If rpmsg_add_eptdev() failes, eptdev is freed in it.
  Subsequent call of dev_err(&eptdev->device) triggers a UAF.
- If __rpmsg_eptdev_open() fails, eptdev is supposed to be freed by
  put_device().

Fixes: 2410558f5f11 ("rpmsg: char: Implement eptdev based on anonymous inode")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/all/aPi6gPZE2_ztOjIW@stanley.mountain/
Signed-off-by: Dawei Li <dawei.li@linux.dev>
---
 drivers/rpmsg/rpmsg_char.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/rpmsg/rpmsg_char.c b/drivers/rpmsg/rpmsg_char.c
index 1b8297b373f0..0919ad0a19df 100644
--- a/drivers/rpmsg/rpmsg_char.c
+++ b/drivers/rpmsg/rpmsg_char.c
@@ -494,6 +494,7 @@ static int rpmsg_eptdev_add(struct rpmsg_eptdev *eptdev,
 	if (cdev)
 		ida_free(&rpmsg_minor_ida, MINOR(dev->devt));
 free_eptdev:
+	dev_err(&eptdev->dev, "failed to add %s\n", eptdev->chinfo.name);
 	kfree(eptdev);
 
 	return ret;
@@ -544,7 +545,6 @@ int rpmsg_anonymous_eptdev_create(struct rpmsg_device *rpdev, struct device *par
 
 	ret =  rpmsg_eptdev_add(eptdev, chinfo, false);
 	if (ret) {
-		dev_err(&eptdev->dev, "failed to add %s\n", eptdev->chinfo.name);
 		return ret;
 	}
 
@@ -560,6 +560,8 @@ int rpmsg_anonymous_eptdev_create(struct rpmsg_device *rpdev, struct device *par
 
 	if (!ret)
 		*pfd = fd;
+	else
+		put_device(&eptdev->dev);
 
 	return ret;
 }
-- 
2.25.1

On Thu, Nov 13, 2025 at 11:39:08PM +0800, Dawei Li wrote:
> Potential UAF and memory leak exsit in exception handling paths for
> rpmsg_anonymous_eptdev_create(), fix them.
> 
> - If rpmsg_add_eptdev() failes, eptdev is freed in it.
>   Subsequent call of dev_err(&eptdev->device) triggers a UAF.
> - If __rpmsg_eptdev_open() fails, eptdev is supposed to be freed by
>   put_device().
> 
> Fixes: 2410558f5f11 ("rpmsg: char: Implement eptdev based on anonymous inode")
> Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> Closes: https://lore.kernel.org/all/aPi6gPZE2_ztOjIW@stanley.mountain/
> Signed-off-by: Dawei Li <dawei.li@linux.dev>
> ---
>  drivers/rpmsg/rpmsg_char.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/rpmsg/rpmsg_char.c b/drivers/rpmsg/rpmsg_char.c
> index 1b8297b373f0..0919ad0a19df 100644
> --- a/drivers/rpmsg/rpmsg_char.c
> +++ b/drivers/rpmsg/rpmsg_char.c
> @@ -494,6 +494,7 @@ static int rpmsg_eptdev_add(struct rpmsg_eptdev *eptdev,
>  	if (cdev)
>  		ida_free(&rpmsg_minor_ida, MINOR(dev->devt));
>  free_eptdev:
> +	dev_err(&eptdev->dev, "failed to add %s\n", eptdev->chinfo.name);
>  	kfree(eptdev);
>  
>  	return ret;
> @@ -544,7 +545,6 @@ int rpmsg_anonymous_eptdev_create(struct rpmsg_device *rpdev, struct device *par
>  
>  	ret =  rpmsg_eptdev_add(eptdev, chinfo, false);
>  	if (ret) {
> -		dev_err(&eptdev->dev, "failed to add %s\n", eptdev->chinfo.name);

This doesn't neet to be done if kfree(eptdev) is moved out of
rpmsg_eptdev_add().

>  		return ret;
>  	}
>  
> @@ -560,6 +560,8 @@ int rpmsg_anonymous_eptdev_create(struct rpmsg_device *rpdev, struct device *par
>  
>  	if (!ret)
>  		*pfd = fd;
> +	else
> +		put_device(&eptdev->dev);

This goes in the previous patch.

>  
>  	return ret;
>  }
> -- 
> 2.25.1
>