April 2024 - Wine-GitLab - list.winehq.org

[PATCH v7 0/5] MR767: vkd3d-shader/dxil: Tessellation part 1.
by Conor McCarthy (＠cmccarthy) April 15, 2024

April 15, 2024

Doesn't support tessellation yet, but adds some requirements. -- v7: vkd3d-shader/ir: Validate tessellation declarations. vkd3d-shader/dxil: Implement DX intrinsic PrimitiveID. vkd3d-shader/dxil: Load hull shader properties. vkd3d-shader/dxil: Load domain shader properties. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/767

4 7

[PATCH v19 0/3] MR734: vkd3d-shader/dxil: SamplePosition and SampleCount.
by Conor McCarthy (＠cmccarthy) April 15, 2024

April 15, 2024

-- v19: vkd3d-shader/dxil: Implement DX intrinsic RenderTargetGetSampleCount. vkd3d-shader/dxil: Implement DX intrinsics Texture2DMSGetSamplePosition and RenderTargetGetSamplePosition. tests/hlsl: Add tests for GetSamplePosition() and GetSampleCount(). https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/734

3 4

[PATCH v10 0/4] MR777: vkd3d-shader/dxil: Support SV_Depth, SV_DepthGreaterEqual and SV_DepthLessEqual.
by Conor McCarthy (＠cmccarthy) April 15, 2024

April 15, 2024

Includes !782 to prevent the 32-bit Windows CI crashing. -- v10: vkd3d-shader/dxil: Support SV_Depth, SV_DepthGreaterEqual and SV_DepthLessEqual. tests/hlsl: Add tests for SV_DepthLessEqual and SV_DepthGreaterEqual. tests/hlsl: Add tests for SV_Depth. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/777

3 7

[PATCH v2 0/3] MR5500: d3d9: Add a few swapchain tests
by Robin Kertels (＠K0bin) April 14, 2024

April 14, 2024

Creates test for the following things: - Whether a combination of BackBufferCount + SwapEffect results in another buffer being added that acts as a front buffer. We used to have bugs caused by this in DXVK ages ago. - GetFrontBufferData actually copying data from the front buffer to the target surface and not from one of the back buffers. This is motivated by a DXVK bug with a modded version of Silent Hill 2 we had recently. - GetFrontBufferData produces a specific fixed alpha value depending on whether the swapchain is windowed or fullscreen. That's a TODO for now because neither Wine nor DXVK implements this and it's a bit annoying to implement this while not a single application is known to rely on this. - GetFrontBufferData is supposed to capture a screenshot of the entire monitor. So this MR adds a test to make sure that the application window is positioned correctly in the target surface after calling GetFrontBufferData. Total War Medieval 2 actually hit this issue with DXVK on its loading screen in windowed mode. All the new tests pass on Windows 11 on the Nvidia D3D9 driver. -- v2: d3d9: Add test for GetFrontBufferData window placement. https://gitlab.winehq.org/wine/wine/-/merge_requests/5500

3 4

[PATCH 0/3] MR5500: d3d9: Add a few swapchain tests
by Robin Kertels (＠K0bin) April 14, 2024

April 14, 2024

Creates test for the following things: - Whether a combination of BackBufferCount + SwapEffect results in another buffer being added that acts as a front buffer. We used to have bugs caused by this in DXVK ages ago. - GetFrontBufferData actually copying data from the front buffer to the target surface and not from one of the back buffers. This is motivated by a DXVK bug with a modded version of Silent Hill 2 we had recently. - GetFrontBufferData produces a specific fixed alpha value depending on whether the swapchain is windowed or fullscreen. That's a TODO for now because neither Wine nor DXVK implements this and it's a bit annoying to implement this while not a single application is known to rely on this. - GetFrontBufferData is supposed to capture a screenshot of the entire monitor. So this MR adds a test to make sure that the application window is positioned correctly in the target surface after calling GetFrontBufferData. Total War Medieval 2 actually hit this issue with DXVK on its loading screen in windowed mode. All the new tests pass on Windows 11 on the Nvidia D3D9 driver. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5500

3 4

[PATCH 0/1] MR5499: mscms: Add stub for WcsGetDefaultColorProfile.
by Vijay Kiran Kamuju (＠infyquest) April 14, 2024

April 14, 2024

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56554 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5499

2 1

[PATCH v4 0/1] MR5497: ntdll: Implement NtQueueApcThreadEx()
by Isaac Marovitz (＠IsaacMarovitz) April 13, 2024

April 13, 2024

Should resolve a crash in `libogg` in the game "Cabal Online" -- v4: ntdll: Implement NtQueueApcThreadEx() https://gitlab.winehq.org/wine/wine/-/merge_requests/5497

3 2

[PATCH v2 0/1] MR5497: ntdll: Implement NtQueueApcThreadEx()
by Isaac Marovitz (＠IsaacMarovitz) April 13, 2024

April 13, 2024

Should resolve a crash in `libogg` in the game "Cabal Online" -- v2: ntdll: Implement NtQueueApcThreadEx() https://gitlab.winehq.org/wine/wine/-/merge_requests/5497

3 2

[PATCH v2 0/1] MR5498: krnl386.exe16: Start DOSVM timer on GET_SYSTEM_TIME
by Liam Middlebrook (＠liam-middlebrook) April 13, 2024

April 13, 2024

Start the DOSVM tick timer when the GET_SYSTEM_TIME interrupt (int 1a ah 00) is fired. This fixes a bug when running Sid Meier's CIVILIZATION for Windows 3.1 where the game would appear to hang at the launch screen (although sound would play as expected). This was due to the BIOSDATA.Ticks field never being updated. I originally had written a version of this patch which moved away from using `BIOSDATA.Ticks` until I saw that the timer thread existed and seemed like a more sensible option. -- v2: krnl386.exe16: Start DOSVM timer on GET_SYSTEM_TIME https://gitlab.winehq.org/wine/wine/-/merge_requests/5498

2 1

[PATCH 0/1] MR5497: ntdll: Implement NtQueueApcThreadEx()
by Isaac Marovitz (＠IsaacMarovitz) April 13, 2024

April 13, 2024

Should resolve a crash in `libogg` in the game "Cabal Online" -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5497

2 1