Re: [RFC PATCH v1 5/7] media: v4l2: introduce two IOCTLs for face detection

From: Ming Lei
Date: Tue Dec 06 2011 - 01:30:38 EST


Hi,

On Mon, Dec 5, 2011 at 10:37 PM, Arnd Bergmann <arnd@xxxxxxxx> wrote:
> On Sunday 04 December 2011, Ming Lei wrote:
>> >
>> > This data structure is not 32/64 bit safe: running a 64 bit kernel with 32 bit
>> > user space will see an incompatible layout.
>>
>> I agree that this is not 32/64 bit safe, but I understand lib32 can handle
>> this correctly, otherwise many 32bit applications can't run on current
>> 64bit kernel
>> since many kernel structures used by user space contained pointer,
>> such as struct v4l2_buffer, struct v4l2_ext_controls in v4l2 ABI.
>
> The other ones you mentioned are handled in the kernel in
> drivers/media/video/v4l2-compat-ioctl32.c. For new ioctl commands,
> it's better to define the data structure in a compatible way
> so you do not need a wrapper like that.

OK, I opt to use __u64 to pass the user pointer, so how about the blow
changes on this patch?

diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 073eb4d..8aeaa1e 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -2214,7 +2214,12 @@ struct v4l2_fd_result {
__u32 buf_index;
__u32 face_cnt;
__u32 reserved[6];
- struct v4l2_fd_detection *fd;
+
+ /*make 64/32 compatible*/
+ union {
+ struct v4l2_fd_detection *fd;
+ __u64 dummy;
+ };
};

/**

thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/