May 2024 - wine-gitlab

[PATCH v2 0/10] MR5573: win32u: Implement vulkan child / other process surface rendering for winex11.
by Rémi Bernon (＠rbernon) 28 May '24

28 May '24

I'm not completely sure about the mechanism, but I think it's simple enough to consider having that upstream now. This shows at least how we can leverage win32u surface changes to decide to switch surfaces on/off-screen and fallback to manual blitting. Having the surfaces on-screen makes sure they are presented as efficiently as possible, having them off-screen we use GDI blit to indirectly call XCopyArea and this will be suboptimal, probably with visible tearing, but hopefully not too common. -- v2: win32u: Use GDI blit to implement partial or other process presentation. winex11: Return an offscreen HDC from vulkan_surface_detach. win32u: Pass a HDC parameter to vulkan_surface_detach. winex11: Create a window surface even when there is client window. winex11: Also attach child client windows to their toplevel window. win32u: Make sure vulkan windows have a pixel format selected. win32u: Detach vulkan surfaces that aren't fully visible. win32u: Detach vulkan surfaces that are offscreen or for another process. win32u: Move vulkan surfaces to their new parent when reparenting. win32u: Introduce a new vulkan offscreen surfaces list. https://gitlab.winehq.org/wine/wine/-/merge_requests/5573

5 19

[PATCH v2 0/5] MR5570: win32u: Some more refactor in preparation for vulkan-only GPU enumeration.
by Rémi Bernon (＠rbernon) 28 May '24

28 May '24

In order to implement the patch discussed in https://gitlab.winehq.org/wine/wine/-/merge_requests/5491#note_67907 -- v2: win32u: Return STATUS_ALREADY_COMPLETE from UpdateDisplayDevices. win32u: Pass gdi_gpu structure members as add_gpu parameters. win32u: Remove driver-specific id from struct gdi_gpu. win32u: Use struct pci_id in struct gdi_gpu. win32u: Remove now unnecessary reset_display_manager_ctx. https://gitlab.winehq.org/wine/wine/-/merge_requests/5570

3 8

[PATCH v2 0/2] MR5579: win32u: Write / read available display modes as a single binary blob.
by Rémi Bernon (＠rbernon) 28 May '24

28 May '24

Reduces the time needed to write them from 20ms to 2ms and to read them from 10ms to 1ms, as well as reducing the number of wineserver requests by a lot. -- v2: win32u: Read / write source modes as a single registry blob. winex11: Only request display modes driver data when needed. https://gitlab.winehq.org/wine/wine/-/merge_requests/5579

3 7

[PATCH 0/1] MR5586: configure: Test PE compilers after setting their target argument.
by Billy Laws (＠bylaws) 28 May '24

28 May '24

clang will default to the host if no target is specified, which causes compile test failures if arch-specific CFLAGS are used for a non-host arch. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5586

3 2

[PATCH v2 0/1] MR5553: wined3d: Fix GLSL version comparison.
by Zebediah Figura (＠zfigura) 28 May '24

28 May '24

-- v2: wined3d: Fix GLSL version comparison. https://gitlab.winehq.org/wine/wine/-/merge_requests/5553

5 4

[PATCH v3 0/3] MR2404: ddraw: Two improvements for Prince of Persia 3D.
by Zebediah Figura (＠zfigura) 28 May '24

28 May '24

This fixes a regression introduced by 15d53761a5fbfc12fc5f9974c029dace00eab33d. -- v3: ddraw/tests: Test GetVertexBufferDesc(). https://gitlab.winehq.org/wine/wine/-/merge_requests/2404

5 7

[PATCH 0/2] MR5579: win32u: Write / read available display modes as a single binary blob.
by Rémi Bernon (＠rbernon) 28 May '24

28 May '24

Reduces the time needed to write them from 20ms to 2ms and to read them from 10ms to 1ms, as well as reducing the number of wineserver requests by a lot. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5579

3 5

[PATCH 0/5] MR5570: win32u: Some more refactor in preparation for vulkan-only GPU enumeration.
by Rémi Bernon (＠rbernon) 28 May '24

28 May '24

In order to implement the patch discussed in https://gitlab.winehq.org/wine/wine/-/merge_requests/5491#note_67907 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5570

4 8

[PATCH v3 0/2] MR2724: wined3d: Submit command buffers that are built for longer than 2 ms.
by Zebediah Figura (＠zfigura) 28 May '24

28 May '24

This improves performance for the game "Grounded", on a AMD Radeon RX 6700 XT, with radv from Mesa 22.3.6. Testing was done with the "cb_access_map_w" option enabled, which also improves performance with the game by itself. From my testing, it's possible to raise the threshold from 2 ms up to 5 ms or so, before the driver or GPU seems to reclock back to the lower power level. However, this measurement is questionable for several reasons. It seems to vary depending on the scene being rendered, and of course this will be specific to the game and driver and GPU in question anyway. The game also has a weird approach to vsync that seems to involve it presenting stale frames (and hence artificially inflating the FPS), which I'm not fully sure I accounted for while measuring. And of course, it's hard to be sure that 5 ms is actually the threshold for how long the driver will go before powering down the GPU. In any case, it seems better to err on the side of submitting more often, to make sure the fix affects more drivers. While submission isn't cheap, it seems to me that submitting every 2 ms is unlikely to cause a bottleneck [consider that this is at most 8 (more) submissions per frame]. The maximum of 4 concurrent periodically submitted buffers was chosen arbitrarily. Removing the maximum altogether does not measurably affect performance for this game either way. Credit goes to Philip Rebohle and his work on DXVK for helping me to notice that periodic submission might make a difference. -- v3: wined3d: Submit command buffers after 512 draw or dispatch commands. wined3d: Retrieve the VkCommandBuffer from wined3d_context_vk after executing RTV barriers. https://gitlab.winehq.org/wine/wine/-/merge_requests/2724

4 4

[PATCH v3 0/1] MR5589: ntdll: Don't restore PC from LR after unwinding through frame switches.
by Billy Laws (＠bylaws) 28 May '24

28 May '24

Fixes unwinds from within a consolidation callback on ARM64EC, as the context frame could be x64 with UNWOUND_TO_CALL set, in which case LR contains garbage. See: https://github.com/dotnet/runtime/blob/71abc26cf8f14f5d15665d7f580365ddfa29… -- v3: ntdll: Don't restore PC from LR after unwinding through frame switches. https://gitlab.winehq.org/wine/wine/-/merge_requests/5589

2 2

2025

2024

2023

2022

wine-gitlab May 2024