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

[PATCH 0/10] MR607: Draft: vkd3d-shader/ir: Introduce a Böhm-Jacopini structurizer.
by Giovanni Mascellani (＠giomasce) Jan. 29, 2024

Jan. 29, 2024

This goes atop !598 and !603. -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/607

3 14

[PATCH v3 0/12] MR4102: Draft: winewayland.drv: Implement keyboard input support.
by Rémi Bernon Jan. 29, 2024

Jan. 29, 2024

@afrantzis Can't seem to be possible to add you as a reviewer, probably you need to request "Access" to the wine/wine project (near the top of the project page). What do you think of something like that? It is very different from the current winex11 code, but I believe it will match keyboard layouts in a much more accurate way, and it's also IMO much simpler. If that works well with Wayland, I think it could be a good hint that it might work as well in X11 and make a case for my other MR to use that approach there. I was a bit annoyed that it doesn't seem possible to retrieve the Xkb "layout:variant" string here, but only the layout description, so I had to use xkbregistry to match it back to the known layouts. It is mostly only there to provide a more accurate HKL value (which should match the layout langid), and scan to vk mapping table, and custom layouts should still work. The lang would be neutral then, and the scan to vk table is QWERTY by default, which is the most common case, and doesn't enforce any vkey -> unicode mapping anyway. So, if the xkbregistry dependency is an issue we could probably make it optional and dynamically loaded, and skip the langid and scan to vk specialized mappings. -- v3: winewayland.drv: Implement Xkb composition using ImeProcessKey. winewayland.drv: Translate Xkb keyboard layouts to KBDTABLES. win32u: Allow KBDTABLES conversion from CTLR + ALT to WCHAR. win32u: Force US layout in ToUnicode when CTRL is pressed. win32u: Introduce KbdLayerDescriptor user driver entry. win32u: Avoid accessing NULL key name string pointer. winewayland.drv: Enumerate Xkb layouts and create matching HKL. winewayland.drv: Handle and parse Xkb keymap events. win32u: Implement opt-in auto-repeat for WM_(SYS)KEYDOWN messages. winewayland.drv: Configure win32u keyboard repeat delay and speed. winewayland.drv: Basic handling of Wayland keyboard events. gitlab: Install libxkbcommon and libxkbregistry dependencies. https://gitlab.winehq.org/wine/wine/-/merge_requests/4102

5 37

[PATCH v2 0/4] MR603: Draft: vkd3d-shader: Use 64 bit swizzles for 64 bit data types in VSIR.
by Giovanni Mascellani (＠giomasce) Jan. 29, 2024

Jan. 29, 2024

The goal here is to be able to convert between SSA and TEMP easily, which will be the second preliminary pass for the CFG structurizer. Ideally I would have added the check in commit 4/4 for TEMPs as well, but typing information from TPF sources is insufficient. This leaves the current validation inconsistent: we check that a register is always used either as a scalar or a vec4, but that "scalar" might be 32 bit in some cases and 64 bit in others. Maybe the right thing to do here is just give up on trying to put some typing on TEMPs and just say they're always 4x32 bit words. I don't know, I'll leave this for when the dust is more settled. -- v2: vkd3d-shader/ir: Check that SSA registers are used with compatible data types. vkd3d-shader: Use 64 bit swizzles for 64 bit data types in VSIR. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/603

3 7

[PATCH v3 0/4] MR601: Draft: vkd3d-shader/dxil: Implement DX intrinsic TextureLoad.
by Conor McCarthy (＠cmccarthy) Jan. 29, 2024

Jan. 29, 2024

Goes atop !583. The last two commits belong to this MR. -- v3: vkd3d-shader/spirv: Emit a vector bitcast if necessary in spirv_compiler_emit_load_ssa_reg(). vkd3d-shader/dxil: Implement DX intrinsic TextureLoad. vkd3d-shader/dxil: Implement DX intrinsic RawBufferLoad. vkd3d-shader/dxil: Load raw/structured buffer SRV/UAV descriptors. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/601

2 4

[PATCH 0/5] MR4950: Extending dmime tests.
by eric pouech (＠epo) Jan. 28, 2024

Jan. 28, 2024

This serie is the first result of investigation of loops not being played on dmusic for wave tracks. It basically: - extends tests so that we can check if loop work at some extend (they don't right now) - add and fix a couple of existing tests. And contain fixes for crufts in existing code: - fix an errouneous prototype - removed unused fields. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4950

3 8

[PATCH v2 0/1] MR4839: Draft: serialui:confdlg: Simple test failure.
by Francois Gouget (＠fgouget) Jan. 28, 2024

Jan. 28, 2024

-- v2: serialui:confdlg: Simple test failure. https://gitlab.winehq.org/wine/wine/-/merge_requests/4839

4 3

[PATCH 0/1] MR4840: Draft: Check if GitLab unnecessarily wakes up the Windows VM.
by Francois Gouget (＠fgouget) Jan. 28, 2024

Jan. 28, 2024

If an MR does not touch the tests, then there is no need to send winetest.exe and run a job on the Windows VM. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4840

4 3

[PATCH v2 0/1] MR4944: dlls/msvcrt: Free previous environment variable when clearing
by David Kahurani (＠kahurani) Jan. 28, 2024

Jan. 28, 2024

Shifting the rest of the variables by one towards indices of lesser value overwrites the variable being cleared and effectively leaks it. Signed-off-by: David Kahurani <k.kahurani(a)gmail.com> -- v2: dlls/msvcrt: Free previous environment variable when clearing https://gitlab.winehq.org/wine/wine/-/merge_requests/4944

3 2

[PATCH 0/1] MR4953: krnl386: Emulate the VGA status register.
by Alex Henrie (＠alexhenrie) Jan. 28, 2024

Jan. 28, 2024

Sega Bug uses it as a timer. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4953

2 1

[PATCH 0/1] MR4948: winewayland.drv: Track and apply latest window cursor on pointer enter.
by Alexandros Frantzis (＠afrantzis) Jan. 28, 2024

Jan. 28, 2024

Keep track of the window cursor set for each surface and apply it when the pointer enters the corresponding surface. This fixes a problem where due to transient disagreements between win32 pointer focus and Wayland pointer focus a cursor update may be lost. For example: 1. Assume existing cursor C1. 2. Window is created beneath cursor. 3. The SetCursor(hwnd, C2) driver callback is called, but because the wl_pointer hasn't entered the Wayland surface yet we don't update the wl_pointer cursor. 4. wl_pointer enters the surface, we apply the latest cursor we know about, which is C1. 5. Since no change in cursor window occurs, we don't get any further SetCursor(hwnd, C2) callbacks, so we get stuck with C1. This commit fixes the problem by tracking the per-surface HCURSOR in step 3, regardless of whether the cursor is actually applied. So, in step 4 we can use that HCURSOR for the surface. This change also fixes some cases of our mouselook heuristics not kicking in properly due to missing a SetCursor(hwnd, NULL) in the manner described above. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4948

3 2