Hi Haogang,
On Tuesday 29 November 2011 22:32:25 Haogang Chen wrote:
> There is a potential integer overflow in uvc_ioctl_ctrl_map(). When a
> large xmap->menu_count is passed from the userspace, the subsequent call
> to kmalloc() will allocate a buffer smaller than expected.
> map->menu_count and map->menu_info would later be used in a loop (e.g.
> in uvc_query_v4l2_ctrl), which leads to out-of-bound access.
>
> The patch checks the ioctl argument and returns -EINVAL for zero or too
> large values in xmap->menu_count.
Thanks for the patch.
> Signed-off-by: Haogang Chen <haogangc...@gmail.com>
> ---
> drivers/media/video/uvc/uvc_v4l2.c | 6 ++++++
> 1 files changed, 6 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/media/video/uvc/uvc_v4l2.c
> b/drivers/media/video/uvc/uvc_v4l2.c index dadf11f..9a180d6 100644
> --- a/drivers/media/video/uvc/uvc_v4l2.c
> +++ b/drivers/media/video/uvc/uvc_v4l2.c
> @@ -58,6 +58,12 @@ static int uvc_ioctl_ctrl_map(struct uvc_video_chain
> *chain, break;
>
> case V4L2_CTRL_TYPE_MENU:
> + if (xmap->menu_count == 0 ||
> + xmap->menu_count > INT_MAX / sizeof(*map->menu_info)) {
I'd like to prevent excessive memory consumption by limiting the number of
menu entries, similarly to how the driver limits the number of mappings.
Defining UVC_MAX_CONTROL_MENU_ENTRIES to 32 in uvcvideo.h should be a
reasonable value.
> + kfree(map);
> + return -EINVAL;
I'd rather do
ret = -EINVAL;
goto done;
to centralize error handling.
If you're fine with both changes I can modify the patch, there's no need to
resubmit.
> + }
> +
> size = xmap->menu_count * sizeof(*map->menu_info);
> map->menu_info = kmalloc(size, GFP_KERNEL);
> if (map->menu_info == NULL) {
--
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
