In malidp_tegra_crtc_reset, new memory is allocated with kzalloc, but 
no check is performed. Before calling __drm_atomic_helper_crtc_reset, 
mw_state should be checked to prevent possible null pointer dereferene.

Fixes: b7e0b04ae450 ("drm/tegra: Convert to using __drm_atomic_helper_crtc_reset() for reset.")
Signed-off-by: Huai-Yuan Liu <qq810974084@gmail.com>
---
 drivers/gpu/drm/tegra/dc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index be61c9d1a4f0..7648e129c212 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -1388,7 +1388,8 @@ static void tegra_crtc_reset(struct drm_crtc *crtc)
 	if (crtc->state)
 		tegra_crtc_atomic_destroy_state(crtc, crtc->state);
 
-	__drm_atomic_helper_crtc_reset(crtc, &state->base);
+	if (state)
+		__drm_atomic_helper_crtc_reset(crtc, &state->base);
 }
 
 static struct drm_crtc_state *
-- 
2.34.1

On Fri May 31, 2024 at 3:56 PM CEST, Huai-Yuan Liu wrote:
> In malidp_tegra_crtc_reset, new memory is allocated with kzalloc, but 
> no check is performed. Before calling __drm_atomic_helper_crtc_reset, 
> mw_state should be checked to prevent possible null pointer dereferene.

Please check that all these variable name references actually are valid.

Also, "dereference".

> Fixes: b7e0b04ae450 ("drm/tegra: Convert to using __drm_atomic_helper_crtc_reset() for reset.")
> Signed-off-by: Huai-Yuan Liu <qq810974084@gmail.com>
> ---
>  drivers/gpu/drm/tegra/dc.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
> index be61c9d1a4f0..7648e129c212 100644
> --- a/drivers/gpu/drm/tegra/dc.c
> +++ b/drivers/gpu/drm/tegra/dc.c
> @@ -1388,7 +1388,8 @@ static void tegra_crtc_reset(struct drm_crtc *crtc)
>  	if (crtc->state)
>  		tegra_crtc_atomic_destroy_state(crtc, crtc->state);
>  
> -	__drm_atomic_helper_crtc_reset(crtc, &state->base);
> +	if (state)
> +		__drm_atomic_helper_crtc_reset(crtc, &state->base);

Looking at how other drivers do this, they call

	__drm_atomic_helper_crtc_reset(crtc, NULL);

if they fail to allocate a new state. Any reason why we wouldn't want to
do the same thing here?

Thierry