I/O normalisation is already a rather complicated pass that does a
lot of things, so it's better to keep anything not strictly
related to it somewhere else.
ERR is used to indicate internal inconsistencies in vkd3d. Here that's
not the case, we simply have to forward the error condition to the
caller.
This fixes failures on the CI with llvmpipe, because the build we use is
compiled without support for VK_KHR_surface and related extensions.
Similarly to RADV, this is a kind of perverted situation: in
principle Vulkan doesn't allow vkCmdResolveImage() to be
executed conditionally (i.e., it is unaffected by conditional
execution), which means that vkd3d cannot implement conditional
rendering for ResolveSubresource(), hence the todo. However,
like RADV, llvmpipe apparently violates the specification and
still executes the image resolution command conditionally. So
that's a llvmpipe bug, even if one that helps us doing the right
thing.
Fixes output signature normalisation failures in Silent Hill 2 caused by
failure of shader_signature_find_element_for_reg().
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=57359
The number of components needed for texel offsets is the number of
components needed to address a single image/layer of an array. I.e.,
"coordinate_component_count - 1" for array textures, and
"coordinate_component_count" for non-array textures.
This change will also fix sampling of cube textures with explicit
gradients. Because texel offsets are unsupported for cube textures,
"offset_component_count" is currently 0 for cube textures. However, the
SAMPLE_GRAD handler also uses "offset_component_count" to determine the
number of components needed for the explicit gradients, and SAMPLE_GRAD
is supposed to work with cube textures.
This commit fixes the compilation of shaders in Star Wars Jedi:
Survivor.
I have to admit I'm not even sure of why most of those got marked
as todo in the first place. Running again now tests on commit
dff7c0e7b8 doesn't show all those
failures.
Using ok() may result in todo's succeeding when create_shader_stage()
succeeds, but vkCreateGraphicsPipelines() fails. There's not much point
in using ok() here in the first place though, because ultimately the
draw operation is going to fail when shader compilation failed.
The Metal bug was reported to Apple as FB15617433. Since we don't
care specifically about that bug, I'm just tweaking the test so that
it doesn't hit that bad path any more.
