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

[PATCH v2 0/1] MR4932: sapi: Create a new engine only when needed in ISpVoice.
by Shaun Ren (＠shaunren) 30 Jan '24

30 Jan '24

Instead of creating the TTS voice engine directly in SetVoice, we save the specified voice token, and only create them in speak_proc when necessary. -- v2: sapi: Create a new engine only when needed in ISpVoice. https://gitlab.winehq.org/wine/wine/-/merge_requests/4932

3 4

[PATCH v2 0/1] MR4933: winegstreamer: Ask GStreamer to stop messing with signal handlers.
by Rémi Bernon 30 Jan '24

30 Jan '24

-- v2: winegstreamer: Ask GStreamer to stop messing with signal handlers. https://gitlab.winehq.org/wine/wine/-/merge_requests/4933

2 3

[PATCH 0/1] MR4966: winebuild: On macOS, search for tools un-prefixed as a last resort.
by Brendan Shanks (＠bshanks) 30 Jan '24

30 Jan '24

When cross-compiling on macOS, like building on ARM64 to target x86_64 [1], `make install` would fail with `winebuild: cannot find the 'ar' tool`. `winebuild` was looking for `x86_64-apple-darwin-ar`, but macOS does not install prefixed tools. As a last step before failing in `find_tool()`, search for the un-prefixed tool name. [1]: with a `configure` invocation like: `../configure --enable-archs=i386,x86_64 --host=x86_64-apple-darwin CC="clang -arch x86_64"` -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4966

3 2

[PATCH 0/1] MR4975: winedump: Sync demangling code with msvcrt.
by Piotr Caban (＠piotr) 30 Jan '24

30 Jan '24

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/4975

2 1

[PATCH 0/1] MR4974: winex11.drv: Translate whole_rect to x11 root coordinates in set_size_hints().
by Zhiyi Zhang (＠zhiyi) 30 Jan '24

30 Jan '24

I haven't found an application affected by this but better to get it fixed. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4974

2 1

Re: [PATCH v2 0/1] MR4965: cfgmgr32: Add CM_Get_Device_Interface_PropertyW stub. - closed
by Helix Graziani (＠TornadoCookie) 30 Jan '24

30 Jan '24

This merge request was closed by Helix Graziani. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4965

1 0

[PATCH v3 0/2] MR4964: winebus.sys: Fix Unity crashing with certain input devices connected.
by Arek Hiler (＠ivyl) 30 Jan '24

30 Jan '24

Affected games: Phasmophobia, Lethal Company. The games do not use joystick input but Unity Input subsystem is initialized and that's enough to cause stack overflow when it tries to parse 8 bit hat switches. Most joysticks and some mice have hats. I own 2 Thrustmaster devices and this fixes the problem with them plugged in. I've also seen [report descriptors shared by Proton users](https://github.com/ValveSoftware/Proton/issues/7284) that reported the crash. The real devices always use 4 bits only. This MR was tested using a few dinput test programs (upstream Wine) and Elite Dangerous (change was applied on top of the Proton tree). Singular hat works. I don't have anything with more than that though. -- v3: winebus.sys: Use 4 bits for hat switches. https://gitlab.winehq.org/wine/wine/-/merge_requests/4964

3 3

Re: [PATCH v2 0/2] MR4964: winebus.sys: Fix Unity crashing with certain input devices connected.
by Arek Hiler (＠ivyl) 30 Jan '24

30 Jan '24

On Tue Jan 30 10:13:17 2024 +0000, Rémi Bernon wrote: > And here? > ```suggestion:-11+0 > static void hatswitch_compose(LONG x, LONG y, UINT index, BYTE *value) > { > BYTE mask = (index % 2) ? 0x0f : 0xf0, shift = (index % 2) ? 4 : 0; > if (x == 0 && y == 0) *value = (*value & mask) | (0 << shift); > else if (x == 0 && y < 0) *value = (*value & mask) | (1 << shift); > else if (x > 0 && y < 0) *value = (*value & mask) | (2 << shift); > else if (x > 0 && y == 0) *value = (*value & mask) | (3 << shift); > else if (x > 0 && y > 0) *value = (*value & mask) | (4 << shift); > else if (x == 0 && y > 0) *value = (*value & mask) | (5 << shift); > else if (x < 0 && y > 0) *value = (*value & mask) | (6 << shift); > else if (x < 0 && y == 0) *value = (*value & mask) | (7 << shift); > else if (x < 0 && y < 0) *value = (*value & mask) | (8 << shift); > } > ``` Ah, I haven't seen that in the email preview. If the current solution is not okay let me know. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4964#note_59501

1 0

[PATCH 0/1] MR612: vkd3d-shader/spirv: Emit line numbers if trace is enabled.
by Conor McCarthy (＠cmccarthy) 30 Jan '24

30 Jan '24

-- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/612

4 4

[PATCH v4 0/1] MR4973: ntdll: Add stub for RtlGetDeviceFamilyInfoEnum.
by Louis Lenders (＠xe) 30 Jan '24

30 Jan '24

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56259 Note: the stub now returns for the last two parameters the values that the testbots gave me (DEVICEFAMILYINFOENUM_DESKTOP and DEVICEFAMILYDEVICEFORM_UNKNOWN). The first parameter returned some value that I don't get (like 0xa00004a63066d), for now i set it to 0, as the Webview installer didn't use this anyway (null pointer). -- v4: ntdll: Add stub for RtlGetDeviceFamilyInfoEnum. https://gitlab.winehq.org/wine/wine/-/merge_requests/4973

2 1