Linux kernel -stable discussions
 help / color / mirror / Atom feed
* [PATCH] drm/tegra: rgb: Fix use-after-free and leak in tegra_dc_rgb_probe()
@ 2026-04-07  8:46 Wentao Liang
  2026-05-28 14:06 ` Thierry Reding
  0 siblings, 1 reply; 2+ messages in thread
From: Wentao Liang @ 2026-04-07  8:46 UTC (permalink / raw)
  To: Thierry Reding, Mikko Perttunen, David Airlie, Simona Vetter,
	Jonathan Hunter
  Cc: dri-devel, linux-tegra, linux-kernel, Wentao Liang, stable

Move the of_device_is_available() check before devm_add_action_or_reset()
to avoid using np after it may have been freed (if the action registration
fails). Also release np immediately when the device is not available to
prevent a reference leak.

Fixes: 3c3642335065 ("drm/tegra: rgb: Fix the unbound reference count")
Cc: stable@vger.kernel.org
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
---
 drivers/gpu/drm/tegra/rgb.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index ff5a749710db..d7586fc820ce 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -215,13 +215,15 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
 	if (!np)
 		return -ENODEV;
 
+	if (!of_device_is_available(np)) {
+		of_node_put(np);
+		return -ENODEV;
+	}
+
 	err = devm_add_action_or_reset(dc->dev, tegra_dc_of_node_put, np);
 	if (err < 0)
 		return err;
 
-	if (!of_device_is_available(np))
-		return -ENODEV;
-
 	rgb = devm_kzalloc(dc->dev, sizeof(*rgb), GFP_KERNEL);
 	if (!rgb)
 		return -ENOMEM;
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] drm/tegra: rgb: Fix use-after-free and leak in tegra_dc_rgb_probe()
  2026-04-07  8:46 [PATCH] drm/tegra: rgb: Fix use-after-free and leak in tegra_dc_rgb_probe() Wentao Liang
@ 2026-05-28 14:06 ` Thierry Reding
  0 siblings, 0 replies; 2+ messages in thread
From: Thierry Reding @ 2026-05-28 14:06 UTC (permalink / raw)
  To: Wentao Liang
  Cc: Thierry Reding, Mikko Perttunen, David Airlie, Simona Vetter,
	Jonathan Hunter, dri-devel, linux-tegra, linux-kernel, stable

[-- Attachment #1: Type: text/plain, Size: 1497 bytes --]

On Tue, Apr 07, 2026 at 08:46:29AM +0000, Wentao Liang wrote:
> Move the of_device_is_available() check before devm_add_action_or_reset()
> to avoid using np after it may have been freed (if the action registration
> fails). Also release np immediately when the device is not available to
> prevent a reference leak.
> 
> Fixes: 3c3642335065 ("drm/tegra: rgb: Fix the unbound reference count")
> Cc: stable@vger.kernel.org
> Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
> ---
>  drivers/gpu/drm/tegra/rgb.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
> index ff5a749710db..d7586fc820ce 100644
> --- a/drivers/gpu/drm/tegra/rgb.c
> +++ b/drivers/gpu/drm/tegra/rgb.c
> @@ -215,13 +215,15 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
>  	if (!np)
>  		return -ENODEV;
>  
> +	if (!of_device_is_available(np)) {
> +		of_node_put(np);
> +		return -ENODEV;
> +	}
> +
>  	err = devm_add_action_or_reset(dc->dev, tegra_dc_of_node_put, np);
>  	if (err < 0)
>  		return err;
>  
> -	if (!of_device_is_available(np))
> -		return -ENODEV;
> -

I don't see how this achieves anything. If the action registration
fails, devm_add_action_or_reset() executes the action (that's the
"or_reset" part), so that takes care of the leak. Similarly, when things
fail, np is not going to be used anymore since we immediately return, so
there's no use-after-free either.

Thierry

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2026-05-28 14:06 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-07  8:46 [PATCH] drm/tegra: rgb: Fix use-after-free and leak in tegra_dc_rgb_probe() Wentao Liang
2026-05-28 14:06 ` Thierry Reding

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox