--
v2: vkd3d-shader/dxil: No longer synthesize DCL instructions.
vkd3d-shader/spirv: Declare I/O registers from the signature.
vkd3d-shader/dxil: Map SEMANTIC_KIND_TARGET to VKD3D_SHADER_SV_TARGET.
vkd3d-shader/tpf: Do not uninvert used masks for domain shader patch constants.
vkd3d-shader/spirv: Do not use the output_info array for patch constants.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/464
Currently we are not properly handling register(cX) reservations for SM1, this is one of the things required for the SNK shaders (CW Bug Bug 18092).
register(cX) reservations also change the offset in the $Globals buffer in SM4, so support for this is also included.
---
Patch 1/4 is required to specify:
```
[require]
shader model < 4.0
```
so that the tests that follow do not get run with the vulkan backend on SM4. I think nobody disagreed with that patch.
--
v8: vkd3d-shader/hlsl: Turn register(cX) reservations into buffer offset for SM4.
vkd3d-shader/hlsl: Make register(cX) reservations work for SM1.
tests: Test register(cX) reservations.
tests: Rename register-reservations.shader_test to register-reservations-resources.shader_test.
tests/shader-runner: Run compilation tests with SM1 when SM1 models are selected.
tests/shader-runner: Allow passing (sm<4) and (sm>=4) to "fail" and "todo" qualifiers.
tests/shader-runner: Discern between the minimum_shader_model and the selected_shader_model.
tests/shader-runner: Add missing requirement checks for backends.
ci: Execute the shader runner on the correct test data on Windows.
ci: Deduplicate the CI configuration for Windows.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
--
v10: vkd3d-shader/ir: Store code block names in struct vkd3d_shader_desc.
vkd3d-shader/ir: Flatten structured control flow instructions.
vkd3d-shader: Rename shader_instruction_array_add_icb() to shader_instruction_array_add_opaque_param().
vkd3d-shader/spirv: Emit descriptor offset loads in the function entry block.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/450
`winecoreaudio` currently uses the `AudioDeviceID` as the device string (stored in the registry and used to correlate between Core Audio devices and generated GUIDs passed to Windows applications), but an `AudioDeviceID` may change between boots or even when a device is connected/disconnected.
Use the UID instead, which is persistent across boots and connects/disconnects.
These are strings like `BuiltInSpeakerDevice` or `AppleUSBAudioEngine:Apple Inc.:Studio Display:00008030-000128469718180E:6,7`.
This only seems to be documented in the header files, from `AudioHardwareBase.h`:
"`kAudioDevicePropertyDeviceUID`: A CFString that contains a persistent identifier for the AudioDevice. An
AudioDevice's UID is persistent across boots. The content of the UID string
is a black box and may contain information that is unique to a particular
instance of an AudioDevice's hardware or unique to the CPU. Therefore they
are not suitable for passing between CPUs or for identifying similar models
of hardware."
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/4632
This MR implements mouselook, which requires support for ClipCursor and relative motion events. The latter we can implement directly without using `SetCursorPos` for which no Wayland protocol exists at the moment.
1. Implement setting the foreground window, as this is required for clipping to work properly (i.e., for the ClipCursor driver callback to be called in the proper process/thread, see commit (2)). We can't unconditionally set the foreground window on keyboard focus, since compositors are eager to give the focus and some windows strongly dislike that (see https://gitlab.winehq.org/wine/wine/-/merge_requests/4102#note_51733). This MR borrows the "managed" window concept from WineX11 to detect windows that we shouldn't allow the compositor to manage, and avoids setting such windows as the foreground on keyboard focus.
2. Implement cursor clipping using the pointer-constraints protocol.
3. Switch to emitting relative motion events using the relative-pointer protocol when the cursor is not visible.
---
4. Finally, a potential fix for an issue I have been seeing where apps may transiently lose the active status (see commit message for more) as a result of the foreground state changes in this MR, which has undesirable side-effects. I have included this here for MR completeness/correctness, but it can be moved to a subsequent MR if preferred.
Note: the velocity of relative motion events when scaling is involved is somewhat inconsistent among Wayland compositors.
Thanks!
--
v5:
https://gitlab.winehq.org/wine/wine/-/merge_requests/4593