September 2022 - Wine-gitlab

[PATCH v2 0/1] MR752: winecrt0: run C++ object destructors in DLL_PROCESS_DETACH.
by Kevin Puetz (＠puetzk) 01 Sep '22

01 Sep '22

Run C++ global/static destructors during DLL_PROCESS_DETACH while other other dllimport functions they may want to call are still viable. While windows imposes many restrictions on what may be done in such destructors (given that the run inside the loader lock), there are lots of legal and useful kernel32 functions like DestroyCriticalSection, DeleteAtom, TlsFree, etc that are both useful and legal. Currently this does not work for builtin modules because all the Win32 structures are discarded well before NtUnmapViewOfSection finally does the dlllose. Even for a winelib .dll.so module, it would be preferable for destructors to execute during process_detach (before wine tears down the MODREF and detaches dependant dlls), rather than be left until after the last NtUnmapViewOfSection (when we finally reach dlclose) Therefore, winegcc now always uses the DllMainCRTStartup entry point unless you specify your own --entry=func. Previously it did this only for PE modules using msvcrt. Making this default consistent matches cl.exe, which also always defaults to _DllMainCRTStartup unless overridden by /entry:foo https://docs.microsoft.com/en-us/cpp/build/reference/entry-entry-point-symb… The ELF version of winecrt0.a now provides a DllMainCRTStartup which, per the Itanium ABI that is in practice what is used by gcc and clang, performs this this destruction by calling __cxa_finalize(&__dso_handle).. This libc function is required to be idempotent, so it's OK that dlclose still calls it again later (there will just be no further work to do). Multiple calls to __cxa_finalize shall not result in calling termination function entries multiple times; the implementation may either remove entries or mark them finished. https://itanium-cxx-abi.github.io/cxx-abi/abi.html#dso-dtor-runtime-api This has two main effects; it moves ELF destructors earlier (before imports are unmapped), and it moves them inside the Nt loader lock. Being earlier was the intended goal, and moving them inside the lock seems fine. Any Win32 API calls in destructors are just being subjected to the same lock hierarchy rules as usual on windows (MSVC also runs destructors from DllMainCrtStartup) https://docs.microsoft.com/en-us/cpp/build/run-time-library-behavior?view=m… And any purely-ELF destructors that happen to also run earlier should never call functions exported from wine (and thus don't care about ntdll's locks). -- v2: winecrt0: run C++ object destructors in DLL_PROCESS_DETACH. https://gitlab.winehq.org/wine/wine/-/merge_requests/752

2 4

[PATCH v3 0/5] MR729: windows.media.speech: Initial recognition session work.
by Bernhard Kölbl 01 Sep '22

01 Sep '22

-- v3: windows.media.speech: Implement a worker thread for the recognition session. https://gitlab.winehq.org/wine/wine/-/merge_requests/729

2 7

[PATCH v7 0/8] MR717: Implement application hives in the registry
by Santino Mazza (＠tati1454) 01 Sep '22

01 Sep '22

Initial implementation for registry application hives, this is a step forward to get Visual Studio to run. :-) -- v7: server: save app hive into it's file when closing handle. server/registry: pass file name instead of file handle to server. kernelbase: Implement RegLoadAppKey. ntdll: Initial implementation for application hives. ntdll: Move NtLoadKey implementation to NtLoadKeyEx. https://gitlab.winehq.org/wine/wine/-/merge_requests/717

3 12

[PATCH 0/2] MR762: winevulkan: Update to VK spec version 1.3.226.
by Georg Lehmann (＠DadSchoorse) 01 Sep '22

01 Sep '22

Not trivial this time because VK_NV_mesh_shader now aliases 64bit enum values that are only later defined by VK_EXT_mesh_shader. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/762

2 2

[PATCH 0/3] MR758: d3d10: Test D3D10CreateEffectFromMemory.
by Ziqing Hui (＠zhui) 01 Sep '22

01 Sep '22

The test results show that data passed to D3D10CreateEffectFromMemory won't be modified. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/758

4 7

[PATCH v2 0/1] MR724: d3dx10/tests: Test block compressed texture data which has more than one block line.
by Ziqing Hui (＠zhui) 01 Sep '22

01 Sep '22

-- v2: d3dx10/tests: Test block compressed texture data which has more than one block line. https://gitlab.winehq.org/wine/wine/-/merge_requests/724

3 4

[PATCH 0/2] MR761: msvcrt: Use size_t for the size in wcsdup().
by Nikolay Sivov (＠nsivov) 01 Sep '22

01 Sep '22

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/761

3 3

[PATCH 0/1] MR756: win32u: Only cache font glyph metrics from GGO_METRICS.
by Zhiyi Zhang (＠zhiyi) 01 Sep '22

01 Sep '22

Fonts loaded with GGO_BITMAP may report different font metrics than that with GGO_METRICS. If the font metrics from GGO_BITMAP are used and later getting font metrics with GGO_METRICS or vice versa, the font metric difference may cause UI glitchs. Fix Steam installer next button text moving to the left for 1 pixel when hovered on Mac and Wine is built with FreeType > 2.8.0. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/756

3 2

[PATCH v10 0/6] MR576: win32u: Move display placement logic out of graphics drivers.
by Rémi Bernon 01 Sep '22

01 Sep '22

In preparation for nulldrv display modes support. -- v10: win32u: Move display placement logic out of graphics drivers. winemac.drv: Remove unnecessary display mode flags checks. win32u: Support interlaced and stretched display modes. winex11.drv: Remove unnecessary display mode flags checks. win32u: Move full display mode lookup out of graphics drivers. win32u: Sort adapter display modes after reading from the registry. https://gitlab.winehq.org/wine/wine/-/merge_requests/576

2 11

[PATCH 0/1] MR760: wbemdisp: Use wcsdup() instead of internal helper.
by Nikolay Sivov (＠nsivov) 01 Sep '22

01 Sep '22

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/760

3 2