[Mesa-dev] [PATCH 0/4] i965: Properly handle CCS in glxBindTexImageEXT

[Daniel Stone]

Hi,

On 13 September 2017 at 00:23, Jason Ekstrand <jason at jlekstrand.net> wrote:
> This little series fixes (I think!) a bug in glxBindTexImageEXT when using
> modifiers.  Before, we were using make_shareable which resolves everything
> and then permanently throws away any aux information.  In the world of
> modifiers, that aux information is suddenly important.  This was causing
> rendering corruptions when running with Daniel's X11 modifiers work and
> using a compositor (mutter in this case).
>
> I've Cc'd a pile of people on this patch most of whom no longer work for
> Intel.  While I'm probably the most qualified person we have to work on
> this, I'm nowhere near as qualified as some other people out there are.
> Please review at least the commit messages and give me a sanity check.  I
> have very limited knowledge of how glx and compositors interact.

So, I've gone through the code and it all makes sense to me. It does
also fix corruption with GLX/BindTexImage compositing managers when
CCS is being used. But without any insight into why the original
codepath is the way it is, the best I can offer is a weak:
Reviewed-by: Daniel Stone <daniels at collabora.com>

Cheers,
Daniel