diff mbox series

[v4,8/9] media: v4l: subdev: Return NULL from pad access functions on error

Message ID 20231026070329.948847-9-sakari.ailus@linux.intel.com
State New
Headers show
Series Unify sub-device state access functions | expand

Commit Message

Sakari Ailus Oct. 26, 2023, 7:03 a.m. UTC
Return NULL from sub-device pad state access functions
(v4l2_subdev_state_get_{format,crop,compose}) for non-existent pads. While
this behaviour differs from older set of pad state information access
functions, we've had a WARN_ON() there for a long time and callers also do
validate the pad index nowadays. Therefore problems are not expected.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
---
 drivers/media/v4l2-core/v4l2-subdev.c | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)

Comments

Hans Verkuil Oct. 26, 2023, 7:35 a.m. UTC | #1
On 26/10/2023 09:03, Sakari Ailus wrote:
> Return NULL from sub-device pad state access functions
> (v4l2_subdev_state_get_{format,crop,compose}) for non-existent pads. While
> this behaviour differs from older set of pad state information access
> functions, we've had a WARN_ON() there for a long time and callers also do
> validate the pad index nowadays. Therefore problems are not expected.

Huh? Patch 2 adds the WARN_ON, and it is removed in patch 8 again?

I'm really confused.

Regards,

	Hans

> 
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> ---
>  drivers/media/v4l2-core/v4l2-subdev.c | 18 +++---------------
>  1 file changed, 3 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/media/v4l2-core/v4l2-subdev.c b/drivers/media/v4l2-core/v4l2-subdev.c
> index 9d4ff9b4fcec..bd0d89c2996f 100644
> --- a/drivers/media/v4l2-core/v4l2-subdev.c
> +++ b/drivers/media/v4l2-core/v4l2-subdev.c
> @@ -1684,12 +1684,8 @@ __v4l2_subdev_state_get_format_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_fmt;
>  	}
> @@ -1722,12 +1718,8 @@ __v4l2_subdev_state_get_crop_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_crop;
>  	}
> @@ -1760,12 +1752,8 @@ __v4l2_subdev_state_get_compose_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_compose;
>  	}
Sakari Ailus Oct. 26, 2023, 7:43 a.m. UTC | #2
Hi Hans,

On Thu, Oct 26, 2023 at 09:35:55AM +0200, Hans Verkuil wrote:
> On 26/10/2023 09:03, Sakari Ailus wrote:
> > Return NULL from sub-device pad state access functions
> > (v4l2_subdev_state_get_{format,crop,compose}) for non-existent pads. While
> > this behaviour differs from older set of pad state information access
> > functions, we've had a WARN_ON() there for a long time and callers also do
> > validate the pad index nowadays. Therefore problems are not expected.
> 
> Huh? Patch 2 adds the WARN_ON, and it is removed in patch 8 again?
> 
> I'm really confused.

Patch two makes the generic subdev state information access functions do
exactly what the subdev state pad-specific information access functions
did. As we're merging two classes of users, doing one thing at a time is
prudent.

This patch aligns the behaviour of the single set of subdev state access
functions. Should there a problem there, we can just revert this one.
Laurent Pinchart Oct. 26, 2023, 1:55 p.m. UTC | #3
Hi Sakari,

Thank you for the patch.

On Thu, Oct 26, 2023 at 10:03:28AM +0300, Sakari Ailus wrote:
> Return NULL from sub-device pad state access functions
> (v4l2_subdev_state_get_{format,crop,compose}) for non-existent pads. While
> this behaviour differs from older set of pad state information access
> functions, we've had a WARN_ON() there for a long time and callers also do
> validate the pad index nowadays. Therefore problems are not expected.
> 
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

> ---
>  drivers/media/v4l2-core/v4l2-subdev.c | 18 +++---------------
>  1 file changed, 3 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/media/v4l2-core/v4l2-subdev.c b/drivers/media/v4l2-core/v4l2-subdev.c
> index 9d4ff9b4fcec..bd0d89c2996f 100644
> --- a/drivers/media/v4l2-core/v4l2-subdev.c
> +++ b/drivers/media/v4l2-core/v4l2-subdev.c
> @@ -1684,12 +1684,8 @@ __v4l2_subdev_state_get_format_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_fmt;
>  	}
> @@ -1722,12 +1718,8 @@ __v4l2_subdev_state_get_crop_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_crop;
>  	}
> @@ -1760,12 +1752,8 @@ __v4l2_subdev_state_get_compose_stream(struct v4l2_subdev_state *state,
>  		if (stream)
>  			return NULL;
>  
> -		/*
> -		 * Set the pad to 0 on error as this is aligned with the
> -		 * behaviour of the pad state information access functions.
> -		 */
>  		if (WARN_ON(pad >= state->sd->entity.num_pads))
> -			pad = 0;
> +			return NULL;
>  
>  		return &state->pads[pad].try_compose;
>  	}
diff mbox series

Patch

diff --git a/drivers/media/v4l2-core/v4l2-subdev.c b/drivers/media/v4l2-core/v4l2-subdev.c
index 9d4ff9b4fcec..bd0d89c2996f 100644
--- a/drivers/media/v4l2-core/v4l2-subdev.c
+++ b/drivers/media/v4l2-core/v4l2-subdev.c
@@ -1684,12 +1684,8 @@  __v4l2_subdev_state_get_format_stream(struct v4l2_subdev_state *state,
 		if (stream)
 			return NULL;
 
-		/*
-		 * Set the pad to 0 on error as this is aligned with the
-		 * behaviour of the pad state information access functions.
-		 */
 		if (WARN_ON(pad >= state->sd->entity.num_pads))
-			pad = 0;
+			return NULL;
 
 		return &state->pads[pad].try_fmt;
 	}
@@ -1722,12 +1718,8 @@  __v4l2_subdev_state_get_crop_stream(struct v4l2_subdev_state *state,
 		if (stream)
 			return NULL;
 
-		/*
-		 * Set the pad to 0 on error as this is aligned with the
-		 * behaviour of the pad state information access functions.
-		 */
 		if (WARN_ON(pad >= state->sd->entity.num_pads))
-			pad = 0;
+			return NULL;
 
 		return &state->pads[pad].try_crop;
 	}
@@ -1760,12 +1752,8 @@  __v4l2_subdev_state_get_compose_stream(struct v4l2_subdev_state *state,
 		if (stream)
 			return NULL;
 
-		/*
-		 * Set the pad to 0 on error as this is aligned with the
-		 * behaviour of the pad state information access functions.
-		 */
 		if (WARN_ON(pad >= state->sd->entity.num_pads))
-			pad = 0;
+			return NULL;
 
 		return &state->pads[pad].try_compose;
 	}