On 18/09/17 20:27, Jason Ekstrand wrote:
On Sun, Sep 17, 2017 at 3:31 AM, Lionel Landwerlin
<lionel.g.landwer...@intel.com <mailto:lionel.g.landwer...@intel.com>>
wrote:
On 15/09/17 23:36, Jason Ekstrand wrote:
struct anv_format_plane {
enum isl_format isl_format:16;
struct isl_swizzle swizzle;
+
+ /* Whether this plane contains chroma channels */
+ bool has_chroma;
+
+ /* For downscaling of YUV planes */
+ uint8_t denominator_scales[2];
+
+ /* How to map sampled ycbcr planes to a single 4
component element. */
+ struct isl_swizzle ycbcr_swizzle;
How is this different from "swizzle" above?
Sorry for the delay in responding.
The spec mandates to hand channels in the Cr-Y-Cb order when no
conversion is applied.
With multiple planes you need to know which one is which. And the order
might not match what the spec requires.
I'm getting a bit concerned about how bloated the format table is
getting. The old table had 184 entries each of which was 32
bits. Now we're looking at more like 32 bytes per entry and most
of it isn't used for your average format. It probably isn't a
huge deal but it does seem a bit excessive. I'm not quite sure
what to do about it though.
I've come to the conclusion that I don't actually care about this
today. It's at most 6k which isn't that bad.
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
