[v2] rpmsg: char: Avoid double destroy of default endpoint

[PATCH v2] rpmsg: char: Avoid double destroy of default endpoint

Posted by Shengjiu Wang 3 years, 6 months ago

The rpmsg_dev_remove() in rpmsg_core is the place for releasing
this default endpoint.

So need to avoid destroying the default endpoint in
rpmsg_chrdev_eptdev_destroy(), this should be the same as
rpmsg_eptdev_release(). Otherwise there will be double destroy
issue that ept->refcount report warning:

refcount_t: underflow; use-after-free.

Call trace:
 refcount_warn_saturate+0xf8/0x150
 virtio_rpmsg_destroy_ept+0xd4/0xec
 rpmsg_dev_remove+0x60/0x70

The issue can be reproduced by stopping remoteproc before
closing the /dev/rpmsgX.

Fixes: bea9b79c2d10 ("rpmsg: char: Add possibility to use default endpoint of the rpmsg device")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Arnaud Pouliquen <arnaud.pouliquen@foss.st.com>
---
changes in v2:
- Add comments

 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 0850ae34fb88..3e0b8f3496ed 100644
--- a/drivers/rpmsg/rpmsg_char.c
+++ b/drivers/rpmsg/rpmsg_char.c
@@ -76,7 +76,9 @@ int rpmsg_chrdev_eptdev_destroy(struct device *dev, void *data)
 
 	mutex_lock(&eptdev->ept_lock);
 	if (eptdev->ept) {
-		rpmsg_destroy_ept(eptdev->ept);
+		/* The default endpoint is released by the rpmsg core */
+		if (!eptdev->default_ept)
+			rpmsg_destroy_ept(eptdev->ept);
 		eptdev->ept = NULL;
 	}
 	mutex_unlock(&eptdev->ept_lock);
-- 
2.34.1

Re: [PATCH v2] rpmsg: char: Avoid double destroy of default endpoint

Posted by Mathieu Poirier 3 years, 6 months ago

On Wed, Sep 21, 2022 at 09:58:43AM +0800, Shengjiu Wang wrote:
> The rpmsg_dev_remove() in rpmsg_core is the place for releasing
> this default endpoint.
> 
> So need to avoid destroying the default endpoint in
> rpmsg_chrdev_eptdev_destroy(), this should be the same as
> rpmsg_eptdev_release(). Otherwise there will be double destroy
> issue that ept->refcount report warning:
> 
> refcount_t: underflow; use-after-free.
> 
> Call trace:
>  refcount_warn_saturate+0xf8/0x150
>  virtio_rpmsg_destroy_ept+0xd4/0xec
>  rpmsg_dev_remove+0x60/0x70
> 
> The issue can be reproduced by stopping remoteproc before
> closing the /dev/rpmsgX.
> 
> Fixes: bea9b79c2d10 ("rpmsg: char: Add possibility to use default endpoint of the rpmsg device")
> Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> Reviewed-by: Arnaud Pouliquen <arnaud.pouliquen@foss.st.com>
> ---
> changes in v2:
> - Add comments
> 
>  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 0850ae34fb88..3e0b8f3496ed 100644
> --- a/drivers/rpmsg/rpmsg_char.c
> +++ b/drivers/rpmsg/rpmsg_char.c
> @@ -76,7 +76,9 @@ int rpmsg_chrdev_eptdev_destroy(struct device *dev, void *data)
>  
>  	mutex_lock(&eptdev->ept_lock);
>  	if (eptdev->ept) {
> -		rpmsg_destroy_ept(eptdev->ept);
> +		/* The default endpoint is released by the rpmsg core */
> +		if (!eptdev->default_ept)
> +			rpmsg_destroy_ept(eptdev->ept);

Applied.

Thanks,
Mathieu

>  		eptdev->ept = NULL;
>  	}
>  	mutex_unlock(&eptdev->ept_lock);
> -- 
> 2.34.1
>

RE: [PATCH v2] rpmsg: char: Avoid double destroy of default endpoint

Posted by Peng Fan 3 years, 6 months ago

> Subject: [PATCH v2] rpmsg: char: Avoid double destroy of default endpoint
> 
> The rpmsg_dev_remove() in rpmsg_core is the place for releasing this
> default endpoint.
> 
> So need to avoid destroying the default endpoint in
> rpmsg_chrdev_eptdev_destroy(), this should be the same as
> rpmsg_eptdev_release(). Otherwise there will be double destroy issue that
> ept->refcount report warning:
> 
> refcount_t: underflow; use-after-free.
> 
> Call trace:
>  refcount_warn_saturate+0xf8/0x150
>  virtio_rpmsg_destroy_ept+0xd4/0xec
>  rpmsg_dev_remove+0x60/0x70
> 
> The issue can be reproduced by stopping remoteproc before closing the
> /dev/rpmsgX.
> 
> Fixes: bea9b79c2d10 ("rpmsg: char: Add possibility to use default endpoint
> of the rpmsg device")
> Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> Reviewed-by: Arnaud Pouliquen <arnaud.pouliquen@foss.st.com>
> ---
> changes in v2:
> - Add comments
> 
>  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 0850ae34fb88..3e0b8f3496ed 100644
> --- a/drivers/rpmsg/rpmsg_char.c
> +++ b/drivers/rpmsg/rpmsg_char.c
> @@ -76,7 +76,9 @@ int rpmsg_chrdev_eptdev_destroy(struct device *dev,
> void *data)
> 
>  	mutex_lock(&eptdev->ept_lock);
>  	if (eptdev->ept) {
> -		rpmsg_destroy_ept(eptdev->ept);
> +		/* The default endpoint is released by the rpmsg core */
> +		if (!eptdev->default_ept)
> +			rpmsg_destroy_ept(eptdev->ept);
>  		eptdev->ept = NULL;
>  	}
>  	mutex_unlock(&eptdev->ept_lock);
> --

Reviewed-by: Peng Fan <peng.fan@nxp.com>