February 2026 - Wine-gitlab

Re: [PATCH v2 0/1] MR9765: ws2_32 / Add stub for WSCInstallProvider64_32 - approved
by Alexandre Julliard (＠julliard) 09 Feb '26

09 Feb '26

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9765

1 0

[PATCH 0/1] MR10051: ws2_32: Allow using duplicated socket handles in closesocket().
by Bernhard Übelacker (＠bernhardu) 09 Feb '26

09 Feb '26

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56927 This is similar to a08d6d8b, but just for `closesocket()`. [Testbot run with this patch](https://testbot.winehq.org/JobDetails.pl?Key=161786) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10051

4 3

Re: [PATCH v3 0/2] MR10058: Draft: ntdll: Spawn main thread as a separate thread.
by Paul Gofman (＠gofman) 09 Feb '26

09 Feb '26

Why I also think it is unlikely possible on Windows because if it was some DRMs would already be calling NtCreateThreadEx with appropriate flags or otherwise try to do it, but I don't think we've seen such attempts. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10058#note_129260

1 0

Re: [PATCH v3 0/2] MR10058: Draft: ntdll: Spawn main thread as a separate thread.
by Paul Gofman (＠gofman) 09 Feb '26

09 Feb '26

There is also THREAD_CREATE_FLAGS_HIDE_FROM_DEBUGGER flag but that only makes thread not send debug events on Windows to the attached debugger but doesn't hide the thread per se. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10058#note_129259

1 0

Re: [PATCH v3 0/2] MR10058: Draft: ntdll: Spawn main thread as a separate thread.
by Paul Gofman (＠gofman) 09 Feb '26

09 Feb '26

Does THREAD_CREATE_FLAGS_LOADER_WORKER somehow hide thread completely on Windows? I doubt so, probably there is no way to hide user space thread completely. Loader is a different aspect. Threads may skip loader but they are still visible through NtGetNextThread or NtQuerySystemInformation( SystemProcessInformation ). So I am not 100% sure that Windows doesn't have such a way but doubt it does. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10058#note_129258

1 0

Re: [PATCH v3 0/2] MR10058: Draft: ntdll: Spawn main thread as a separate thread.
by Rémi Bernon (＠rbernon) 09 Feb '26

09 Feb '26

What about the RTL_USER_PROCESS_PARAMETERS LoaderThreads field and the THREAD_CREATE_FLAGS_LOADER_WORKER flags? This seems to indicate some sort of ntdll loader threading support? Are we sure that Windows has no such thing? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10058#note_129256

1 0

[PATCH v2 0/2] MR10053: mf: Fix crash when stream is null during sample request.
by Brendan McGrath (＠redmcg) 09 Feb '26

09 Feb '26

Windows waits until it receives the MEStreamStarted event(s) before subscribing to the sink event queues. This is the successor to MR !9716. -- v2: mf: Fix crash when stream is null during sample request. https://gitlab.winehq.org/wine/wine/-/merge_requests/10053

2 5

[PATCH v2 0/1] MR10062: Draft: win32u: Ignore EGL device if context cannot be created.
by Rémi Bernon (＠rbernon) 09 Feb '26

09 Feb '26

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59372 -- v2: win32u: Ignore EGL device if context cannot be created. https://gitlab.winehq.org/wine/wine/-/merge_requests/10062

2 1

[PATCH 0/1] MR10062: win32u: Ignore EGL device if context cannot be created.
by Rémi Bernon (＠rbernon) 09 Feb '26

09 Feb '26

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59372 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10062

2 1

[PATCH v7 0/1] MR10052: dinput: Return DIERR_UNSUPPORTED in dinput_device_Escape
by Viktor Balogh (＠viktorb) 09 Feb '26

09 Feb '26

At least Thrustmaster wheels are known for using the Escape method for communication between the driver and the game. This is not utilized by every game, but one that does is RaceRoom Racing Experience (RRRE). RRRE uses it to query/set the wheel range and set force strengths. The problematic part is the wheel range query, because the returned value (0) is not the real range and that throws off the internal logic of RRRE's engine and causes the wheel to snap in one direction. Returning DIERR_UNSUPPORTED clearly indicates that the call failed and RRRE no longer tries to use these methods for interaction with the wheel. Returning DIERR_UNSUPPORTED is also aligns with the behaviour of `hid_joystick_effect_Escape`. Captures of what RRRE used are available over here: https://github.com/Kimplul/hid-tmff2/issues/165 See this issue for more context and the resolution: https://github.com/Kimplul/hid-tmff2/issues/39#issuecomment-3104689106 -- v7: dinput: Return DIERR_UNSUPPORTED in dinput_device_Escape https://gitlab.winehq.org/wine/wine/-/merge_requests/10052

3 4