Hi Hans,
Thank you for the patch.
On Saturday 15 March 2014 14:08:03 Hans Verkuil wrote:
> From: Hans Verkuil <hans.verk...@cisco.com>
>
> Drop the use of __user in the user_ptr variable since the v4l2 structs are
> actually defined without __user, instead cast to a __user pointer only
> there where it is really needed: in the copy_to/from_user calls.
>
> Also remove unnecessary casts in check_array_args and replace a wrong
> cast (void *) with the correct one (void **).
>
> This fixes these sparse warnings:
>
> drivers/media/v4l2-core/v4l2-ioctl.c:2284:35: warning: incorrect type in
> assignment (different address spaces)
> drivers/media/v4l2-core/v4l2-ioctl.c:2301:35: warning: incorrect type in
> assignment (different address spaces)
> drivers/media/v4l2-core/v4l2-ioctl.c:2319:35: warning: incorrect type in
> assignment (different address spaces)
> drivers/media/v4l2-core/v4l2-ioctl.c:2386:57: warning: incorrect type in
> argument 4 (different address spaces)
> drivers/media/v4l2-core/v4l2-ioctl.c:2420:29: warning: incorrect type in
> assignment (different address spaces)
>
> Signed-off-by: Hans Verkuil <hans.verk...@cisco.com>
> ---
> drivers/media/v4l2-core/v4l2-ioctl.c | 20 ++++++++++----------
> 1 file changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c
> b/drivers/media/v4l2-core/v4l2-ioctl.c index d9113cc..3e0cf4f 100644
> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> @@ -2260,7 +2260,7 @@ done:
> }
>
> static int check_array_args(unsigned int cmd, void *parg, size_t
> *array_size,
> - void * __user *user_ptr, void ***kernel_ptr)
> + void **user_ptr, void ***kernel_ptr)
> {
> int ret = 0;
>
> @@ -2276,8 +2276,8 @@ static int check_array_args(unsigned int cmd, void
> *parg, size_t *array_size, ret = -EINVAL;
> break;
> }
> - *user_ptr = (void __user *)buf->m.planes;
> - *kernel_ptr = (void *)&buf->m.planes;
> + *user_ptr = buf->m.planes;
> + *kernel_ptr = (void **)&buf->m.planes;
> *array_size = sizeof(struct v4l2_plane) * buf->length;
> ret = 1;
> }
> @@ -2293,8 +2293,8 @@ static int check_array_args(unsigned int cmd, void
> *parg, size_t *array_size, ret = -EINVAL;
> break;
> }
> - *user_ptr = (void __user *)edid->edid;
> - *kernel_ptr = (void *)&edid->edid;
> + *user_ptr = edid->edid;
> + *kernel_ptr = (void **)&edid->edid;
> *array_size = edid->blocks * 128;
> ret = 1;
> }
> @@ -2311,8 +2311,8 @@ static int check_array_args(unsigned int cmd, void
> *parg, size_t *array_size, ret = -EINVAL;
> break;
> }
> - *user_ptr = (void __user *)ctrls->controls;
> - *kernel_ptr = (void *)&ctrls->controls;
> + *user_ptr = ctrls->controls;
> + *kernel_ptr = (void **)&ctrls->controls;
> *array_size = sizeof(struct v4l2_ext_control)
> * ctrls->count;
> ret = 1;
> @@ -2334,7 +2334,7 @@ video_usercopy(struct file *file, unsigned int cmd,
> unsigned long arg, long err = -EINVAL;
> bool has_array_args;
> size_t array_size = 0;
> - void __user *user_ptr = NULL;
> + void *user_ptr = NULL;
> void **kernel_ptr = NULL;
>
> /* Copy arguments into temp kernel buffer */
> @@ -2395,7 +2395,7 @@ video_usercopy(struct file *file, unsigned int cmd,
> unsigned long arg, if (NULL == mbuf)
> goto out_array_args;
> err = -EFAULT;
> - if (copy_from_user(mbuf, user_ptr, array_size))
> + if (copy_from_user(mbuf, (void __user *)user_ptr, array_size))
> goto out_array_args;
> *kernel_ptr = mbuf;
> }
> @@ -2413,7 +2413,7 @@ video_usercopy(struct file *file, unsigned int cmd,
> unsigned long arg,
>
> if (has_array_args) {
> *kernel_ptr = user_ptr;
> - if (copy_to_user(user_ptr, mbuf, array_size))
> + if (copy_to_user((void __user *)user_ptr, mbuf, array_size))
Moving the __user annotation to copy_from_user/copy_to_user defeats the whole
point of the annotation. user_ptr is really a user pointer here, and I believe
it should be treated as such. I'd rather fix the sparse warnings where they
really occur.
> err = -EFAULT;
> goto out_array_args;
> }
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
