- Wine-GitLab - list.winehq.org

[PATCH v2 0/2] MR10120: win32u: Implement NtUserEnableMouseInPointer semi-stub.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

-- v2: win32u: Implement NtUserEnableMouseInPointer semi-stub. user32/tests: Test EnableMouseInPointer message translation. https://gitlab.winehq.org/wine/wine/-/merge_requests/10120

1 2

[PATCH 0/2] MR10120: win32u: Implement NtUserEnableMouseInPointer semi-stub.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10120

1 2

[PATCH 0/3] MR10119: winegcc: Use compiler-rt instead of -lgcc on mingw targets.
by Jacek Caban (＠jacek) Feb. 17, 2026

Feb. 17, 2026

This treats mingw targets the same way as msvc. We already have the necessary infrastructure; we are just missing a few functions to be able to use it for mingw as well. In my case, I was looking at adding emutls support for GCC. Since it is typically provided by libgcc or compiler-rt, this should be straightforward to implement on top of this MR. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10119

2 3

[PATCH v2 0/1] MR10056: dsound: Use a better FIR filter generated with the Parks-McClellan algorithm.
by Anton Baskanov (＠baskanov) Feb. 17, 2026

Feb. 17, 2026

The new filter provides a 98.5 dB stopband rejection and doesn't attenuate frequencies below 20 kHz (assuming 44.1 kHz sampling rate). ![magnitude](/uploads/e0947573942210102f85bc11082c1380/magnitude.png){width=900 height=403} ![sweep.wine.upstream](/uploads/411cb840ed5f7117a669e1a729101d0f/sweep.wine.upstream.png){width=675 height=356} ![sweep.wine.pm-fir](/uploads/b2f6d669c1bb62c70d6a4a725cb06aa8/sweep.wine.pm-fir.png){width=675 height=356} P.S.: Thanks to Matteo Bruni for providing the test tools. -- v2: dsound: Use a better FIR filter generated with the Parks-McClellan algorithm. https://gitlab.winehq.org/wine/wine/-/merge_requests/10056

4 8

[PATCH 0/1] MR10118: win32u: Or the GL_FLUSH_INTERVAL flag when updating the swap interval.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10118

1 1

[PATCH v8 0/8] MR10019: opengl32: Move wglGetProcAddress implementation to the PE side.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

This better splits client / host extension lists, allowing to more easily check for available host-side extensions in win32u later on, then moves the function registry to the PE side to get rid of the remaining extension logic in opengl32 unix side. Doing that, it also drops the driver function check, and always expose extension functions to the client if their extension dependency is satisfied. I'm not completely sure this is right, but I think the unix side tends to expose functions from eglGetProcAddress more often than the opposite. We could also keep a mechanism to check the driver availability, with a bit of additional complexity. On the unix side, all OpenGL functions are queried on initialization instead of just core functions, simplifying their use later on. -- v8: opengl32: Implement the function registry on the PE side. win32u: Query every OpenGL functions on initialization. opengl32: Implement client extension enumeration on the PE side. opengl32: Move major / minor version to the PE side context wrapper. opengl32: Parse host extension list to the opengl_extensions struct. opengl32: Remove now unused wgl pixel format unix thunks. opengl32: Implement wglGetExtensionsString(ARB|EXT) on the PE side. opengl32: Initialize enabled / disabled OpenGL extensions struct. https://gitlab.winehq.org/wine/wine/-/merge_requests/10019

1 9

[PATCH v7 0/8] MR10019: opengl32: Move wglGetProcAddress implementation to the PE side.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

This better splits client / host extension lists, allowing to more easily check for available host-side extensions in win32u later on, then moves the function registry to the PE side to get rid of the remaining extension logic in opengl32 unix side. Doing that, it also drops the driver function check, and always expose extension functions to the client if their extension dependency is satisfied. I'm not completely sure this is right, but I think the unix side tends to expose functions from eglGetProcAddress more often than the opposite. We could also keep a mechanism to check the driver availability, with a bit of additional complexity. On the unix side, all OpenGL functions are queried on initialization instead of just core functions, simplifying their use later on. -- v7: opengl32: Implement the function registry on the PE side. win32u: Query every OpenGL functions on initialization. opengl32: Implement client extension enumeration on the PE side. opengl32: Move major / minor version to the PE side context wrapper. opengl32: Parse host extension list to the opengl_extensions struct. opengl32: Remove now unused wgl pixel format unix thunks. opengl32: Implement wglGetExtensionsString(ARB|EXT) on the PE side. opengl32: Initialize enabled / disabled OpenGL extensions struct. https://gitlab.winehq.org/wine/wine/-/merge_requests/10019

2 10

[PATCH 0/1] MR10117: win32u: Don't offset client rect when using D3D present rect.
by Rémi Bernon Feb. 17, 2026

Feb. 17, 2026

The window rects are all in absolute screen position and the client rect shouldn't be made relative to the window position. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10117

1 1

[PATCH v2 0/4] MR10114: msi: Handle unusual installer files (padding, malformed string refs)
by Theodoros Chatzigiannakis (＠TChatzigiannakis) Feb. 17, 2026

Feb. 17, 2026

This PR addresses two issues that I observed preventing the Native Instruments Kontakt Player MSI (and other MSI files from the same vendor) from successfully finishing: 1. Some installers contain padding for 4-byte alignment of table streams. We relax Wine's parsing to allow for such padding. 2. Some installers declare that they have 3-byte string refs, but the data actually contains 2-byte string refs. This looks like a malformed MSI to me (a result of a misapplied transform perhaps). However, Windows presumably(\*) accepts this inconsistency, so we try to be forgiving here. (\*) It has been years since I last tried to install NI software on Windows. But Windows is a main platform for NI, so if it didn't accept these malformed files, then NI (and other vendors) would presumably have caught this issue in their installers and addressed it already. --- The above are technically separate issues, but they affect the same software in similar ways, so I opted for one PR. Please me know if I should split them. --- Quick repro hint: Native Instruments offers a program called Native Access (which acts as an installation hub on the user's machine). On user request, NA downloads and launches the affected MSI files in the background. If you try to repro, a couple of things to watch out for: 1. NA seems to always report a successful installation, even when the underlying MSI fails to complete (and therefore nothing is installed). This PR does nothing to address this false positive, as it is most likely a NA logic bug (perhaps ignoring the installer's status code). 2. To observe/examine the underlying MSI files (including the fix presented here), you'll need to grab them after NA downloads and extracts them on your machine (or find manual/direct downloads on the NI website, where applicable) --- _Disclaimers for reviewers:_ * _AI was used while tracking down the issues and coming up with fixes._ * _I am not affiliated with Native Instruments, I'm just a user._ -- v2: msi/tests: Use broken() for strref mismatch test on Windows. https://gitlab.winehq.org/wine/wine/-/merge_requests/10114

3 5

[PATCH v15 0/3] MR9369: wininet: Return correct errors and put handle into invalid state when seeking beyond end of file.
by Torge Matthies (＠tmatthies) Feb. 17, 2026

Feb. 17, 2026

When `InternetSetFilePointer` is used to seek and then read beyond the size of the remote file, the initial read will fail with `ERROR_NOACCESS` and put the handle into an invalid state such that `InternetSetFilePointer`, `InternetReadFile`, and `InternetQueryDataAvailable` will return errors from then on. Based on !9644. -- v15: wininet: Return correct errors and put handle into invalid state when seeking beyond end of file. wininet/tests: Add tests for seeking beyond end of file. wininet/tests: Skip GetUrlCacheConfigInfo and CommitUrlCacheEntry tests if the functions return ERROR_CALL_NOT_IMPLEMENTED. https://gitlab.winehq.org/wine/wine/-/merge_requests/9369

2 3

[PATCH v14 0/3] MR9369: wininet: Return correct errors and put handle into invalid state when seeking beyond end of file.
by Torge Matthies (＠tmatthies) Feb. 17, 2026

Feb. 17, 2026

When `InternetSetFilePointer` is used to seek and then read beyond the size of the remote file, the initial read will fail with `ERROR_NOACCESS` and put the handle into an invalid state such that `InternetSetFilePointer`, `InternetReadFile`, and `InternetQueryDataAvailable` will return errors from then on. Based on !9644. -- v14: wininet: Return correct errors and put handle into invalid state when seeking beyond end of file. wininet/tests: Add tests for seeking beyond end of file. wininet/tests: Skip GetUrlCacheConfigInfo and CommitUrlCacheEntry tests if the functions return ERROR_CALL_NOT_IMPLEMENTED. https://gitlab.winehq.org/wine/wine/-/merge_requests/9369

2 3

[PATCH 0/1] MR10116: include: iphlpapi.h avoid C++ keyword
by Alistair Leslie-Hughes (＠alesliehughes) Feb. 17, 2026

Feb. 17, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10116

2 1

[PATCH 0/1] MR10115: wine.inf: Add a default association for .reg files.
by Esme Povirk (＠madewokherd) Feb. 16, 2026

Feb. 16, 2026

The launcher for the Steam release of Dino Crisis relies on this. https://github.com/ValveSoftware/Proton/issues/9491 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10115

2 1

[PATCH 0/3] MR10114: msi: handle unusual installer files (padding, malformed string refs)
by Theodoros Chatzigiannakis (＠TChatzigiannakis) Feb. 16, 2026

Feb. 16, 2026

This PR addresses two issues that I observed preventing the Native Instruments Kontakt Player MSI (and presumably other MSI files from the same vendor): 1. Some installers contain padding for 4-byte alignment of table streams. We relax Wine's parsing to allow for such padding. 2. Some installers declare that they have 3-byte string refs, but the data actually contains 2-byte string refs. This looks like a malformed MSI to me (a result of a misapplied transform when creating the installer perhaps). However, Windows presumably(\*) accepts this inconsistency, so we try to be forgiving here -- even though the current behavior is not an actual bug in Wine (as far as I can tell). (\*) I have not tried to install NI software in Windows for years. But Windows is a main platform for NI, so if Windows didn't accept these malformed files, then NI (and other vendors) would presumably have caught this issue in their installers and addressed it already. --- The above are technically separate issues, but they affect the same software in similar ways, so I opted for one PR. Please me know if I should split them. --- Quick repro hint: Native Instruments offers a program called Native Access (which acts as an installation hub on the user's machine). On the user's request, NA downloads and launches the affected MSI files in the background. If you try to repro, a couple of things to watch out for: 1. NA seems to always report a successful installation, even when the underlying MSI fails to complete (and therefore nothing is installed). This PR does nothing to address this false positive, as it is most likely a NA logic bug (i.e. not checking the installer's status or something similar). 2. To observe/examine the underlying MSI files (including the fix presented here), you'll need to grab them after NA downloads them on your machine (or find manual/direct downloads on the NI website, if any) --- _Disclaimers for reviewers:_ * _AI was used while tracking down the issues and coming up with fixes._ * _I am not affiliated with Native Instruments, I'm just a user._ -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10114

2 3

[PATCH 0/2] MR10113: services: Allow ".\LocalSystem" account name for an interactive service.
by Andrew Nguyen (＠arethusa) Feb. 16, 2026

Feb. 16, 2026

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59426 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10113

2 2

[PATCH 0/1] MR10112: avicap32: Fail gracefully if V4L is unavailable.
by Tim Clem (＠tclem) Feb. 16, 2026

Feb. 16, 2026

This regressed in 414304658e5, causing a syscall fault from the missing Unix call on non-Linux platforms. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10112

2 1

[PATCH v2 0/5] MR9992: Winemac IME rework part 1
by Marc-Aurel Zent (＠mzent) Feb. 16, 2026

Feb. 16, 2026

-- v2: winemac: Implement and use macdrv_ImeToAsciiEx(). win32u: Mark rejected IME messages as unprocessed. imm32: Generate a proper WM_KEYDOWN message for unconsumed keys. imm32: Pass keyboard state into ImeToAsciiEx(). win32u: Introduce a new ImeToAsciiEx call through NtUserMessageCall. https://gitlab.winehq.org/wine/wine/-/merge_requests/9992

4 11

[PATCH 0/1] MR10111: ucrtbase: Fix scanset character range handling in scanf.
by Piotr Caban (＠piotr) Feb. 16, 2026

Feb. 16, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10111

2 1

[PATCH v4 0/11] MR10058: Draft: ntdll: Spawn main thread as a separate thread.
by Rémi Bernon Feb. 16, 2026

Feb. 16, 2026

This makes sure macOS main thread is a Wine thread as in https://gitlab.winehq.org/wine/wine/-/merge_requests/9579, while keeping its event loop usable for GUI, and makes every other platform work the same. Their main thread is parked for now, but the idea is to use the poll loop for various unix side internal purposes, such as winewayland pipe polling, compositing, etc. On macOS the main loop would be used similarly. -- v4: ntdll: Use CFRunLoopRun to implement sched_run on macOS. ntdll: Spawn the application main thread as a separate thread. server: Initialize process tracing in get_startup_info. ntdll: Send process PEB / initial TEB in get_startup_info. ntdll: Call get_startup_info request even when it is empty. ntdll: Pass the startup info size to init_startup_info. server: Create the new process first thread in new_process request. server: Return new process thread handle from get_new_process_info. ntdll: Pass thread security descriptor to new_process request. ntdll: Pass the main thread flags in the new_process request. server: Pass a thread flags parameter to create_thread. https://gitlab.winehq.org/wine/wine/-/merge_requests/10058

3 20

[PATCH v5 0/1] MR9362: opengl32: Simplify extension tokenization.
by Rémi Bernon Feb. 16, 2026

Feb. 16, 2026

This was requested in !9263 review, but ignored for some reason. -- v5: opengl32: Simplify extension tokenization. https://gitlab.winehq.org/wine/wine/-/merge_requests/9362

3 6

[PATCH 0/1] MR9635: makedep: Allow building modules with C++ sources.
by Rémi Bernon Feb. 16, 2026

Feb. 16, 2026

Used in Proton, could be useful in the future for fluidsynth and others. The flags used for compilation are taken from CXXFLAGS + extern libs EXTRACFLAGS (which is CFLAGS without the warnings I think). Some warnings are C-only and we would need to duplicate a lot of WINE_TRY_*FLAGS to make sure whether they work or not with the C++ compiler. The other EXTRACFLAGS are assumed to work with it and it seems that they do. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9635

2 2

[PATCH v3 0/5] MR10109: mmdevapi/tests: Test a curated list of wave formats.
by Giovanni Mascellani (＠giomasce) Feb. 16, 2026

Feb. 16, 2026

The main commit is quite large. I figured that shouldn't be a problem, since most of it is long lists of formats rather than code, but if that's better I can split it more granularly. -- v3: mmdevapi/tests: Test capturing with RATEADJUST in exclusive mode more specifically. mmdevapi/tests: Tweak test logging a little bit. mmdevapi/tests: Test a curated list of wave formats. mmdevapi/tests: Introduce a helper to decide whether a wave format is valid. https://gitlab.winehq.org/wine/wine/-/merge_requests/10109

2 5

[PATCH v2 0/5] MR10109: mmdevapi/tests: Test a curated list of wave formats.
by Giovanni Mascellani (＠giomasce) Feb. 16, 2026

Feb. 16, 2026

The main commit is quite large. I figured that shouldn't be a problem, since most of it is long lists of formats rather than code, but if that's better I can split it more granularly. -- v2: mmdevapi/tests: Test capturing with RATEADJUST in exclusive mode more specifically. mmdevapi/tests: Tweak test logging a little bit. mmdevapi/tests: Test a curated list of wave formats. mmdevapi/tests: Introduce a helper to decide whether a wave format is valid. https://gitlab.winehq.org/wine/wine/-/merge_requests/10109

2 5

[PATCH v3 0/1] MR10068: msvcrt: Create copy of C++ exception object in std::rethrow_exception()
by Piotr Pawłowski Feb. 16, 2026

Feb. 16, 2026

Fixes: https://bugs.winehq.org/show_bug.cgi?id=59329 Resolves whole lot of foobar2000 crashes after media indexing or playback errors (verified). Passes various C++ exception torture tests I quickly wrote. Notes about alloca() use- Size of C++ object is used, one that already existed elsewhere on stack, so it should not be an issue. MS runtime appears to do the same - copy the exception object on stack before rethrowing. -- v3: msvcrt: Create copy of C++ exception object in std::rethrow_exception(). https://gitlab.winehq.org/wine/wine/-/merge_requests/10068

1 2

[PATCH v2 0/5] MR10103: win32u, opengl: Misc opengl fixes.
by Paul Gofman (＠gofman) Feb. 16, 2026

Feb. 16, 2026

-- v2: opengl32: Also bind default framebuffer in pop_default_fbo(). win32u: Don't present new drawable in context_sync_drawables(). opengl32: Create compatibility context in wrap_wglCreateContext(). https://gitlab.winehq.org/wine/wine/-/merge_requests/10103

3 11