On Tue, 20 Apr 2021 at 15:35, Jan Sikorski jsikorski@codeweavers.com wrote:
On 20 Apr 2021, at 15:20, Marvin testbot@winehq.org wrote:
=== w2008s64 (32 bit report) ===
d3d11: d3d11.c:17243: Test failed: Got 0xffffffff, expected 0xff000000 at (0, 0, 0), sub-resource 0.
Perhaps it doesn’t really makes sense to test that behavior after all, or only check it on wine, since it would be supported despite being invalid. I don’t know about any software that relies on it though.
I think the options are essentially: - Don't test that case at all, only include the original test. - Accept skipping the draw completely as a valid result here. - Accept skipping the draw completely as a broken() result here.
Those each have their advantages and disadvantages, and I don't have a clear preference, so just pick whichever one makes sense to you.