Hi Geert,
On Friday 18 December 2015 18:37:51 Geert Uytterhoeven wrote:
> On Fri, Dec 18, 2015 at 6:16 PM, Laurent Pinchart wrote:
> >> > @@ -1987,7 +1988,8 @@ struct v4l2_pix_format_mplane {
> >> > __u8 ycbcr_enc;
> >> > __u8 quantization;
> >> > __u8 xfer_func;
> >> > - __u8 reserved[7];
> >> > + __u8 reserved[3];
> >> > + __u32 request;
> >>
> >> I think I mentioned this before: I feel uncomfortable using 4 bytes of
> >> the reserved fields when the request ID is limited to 16 bits anyway.
> >
> > I'm still unsure whether request IDs should be 16 or 32 bits long. If we
> > go for 16 bits then I'll of course make this field a __u16.
> >
> >> I would prefer a __u16 here. Also put the request field *before* the
> >> reserved array, not after.
> >
> > The reserved array isn't aligned to a 32 bit (or even 16 bit) boundary. I
> > can put the request field before it, with a 8 bit hole before the field.
>
> There's no alignment at all due to:
>
> >> > } __attribute__ ((packed));
Oops, indeed. Still, isn't it better to keep 16-bit or 32-bit values aligned
to 16-bit or 32-bit boundaries ?
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
