June 2025 - Wine-gitlab - list.winehq.org

[PATCH 0/1] MR8176: d3d10/tests: Check d3d buffer sizes in effects tests.
by Nikolay Sivov (＠nsivov) 04 Jun '25

04 Jun '25

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8176

3 2

[PATCH 0/2] MR8198: ucrtbase: Fix case mapping and ctype1 info for UTF8 locale.
by Paul Gofman (＠gofman) 04 Jun '25

04 Jun '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/8198

3 8

Re: [PATCH v4 0/3] MR7748: ucrtbase: Add cexp() implementation.
by Bernhard Übelacker 04 Jun '25

04 Jun '25

Bernhard Übelacker (@bernhardu) commented about libs/musl/Makefile.in: > EXTLIB = libmusl.a > EXTRAINCL = -I$(srcdir)/src/internal -I$(srcdir)/arch/generic > -EXTRADEFS = -D_ACRTIMP= -D_NO_CRT_MATH_INLINE > +EXTRADEFS = -D_ACRTIMP= -D_NO_CRT_MATH_INLINE \ > + -fno-builtin Hello @dlehman25, I noticed the test pattern page shows following tests starting to fail since 2025-05-28: - [d3dcompiler_42:hlsl_d3d9](https://test.winehq.org/data/patterns.html#d3dcom… - [d3dcompiler_43:hlsl_d3d9](https://test.winehq.org/data/patterns.html#d3dcom… - [d3dcompiler_46:hlsl_d3d9](https://test.winehq.org/data/patterns.html#d3dcom… - [d3dcompiler_47:hlsl_d3d9](https://test.winehq.org/data/patterns.html#d3dcom… A git bisect led me to this merge request containing this commit: ``` 15485e6dd9be152286f08b3a3f6cf42895f3b377 is the first broken commit commit 15485e6dd9be152286f08b3a3f6cf42895f3b377 Author: Daniel Lehman <dlehman25(a)gmail.com> Date: Fri Apr 25 21:45:19 2025 -0700 msvcr120: Add cexp() implementation. ``` (git bisect needed a clean build each time, so ccache helps a lot. Only visible with a GCC-mingw build.) I tried to further reduce it and think it is related with the addition of `-fno-builtin` to `libs/musl`, can you confirm this finding? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7748#note_105396

1 0

[PATCH 0/1] MR8207: gitlab: Switch to a Sequoia-based VM for the macOS CI build.
by Tim Clem (＠tclem) 04 Jun '25

04 Jun '25

For the record this is a VM running 15.5 with Xcode 16.4. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8207

2 1

[PATCH 0/5] MR8206: windows.devices.bluetooth: Add stubs for (I)BluetoothDevice.
by Vibhav Pant (＠vibhavp) 04 Jun '25

04 Jun '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/8206

2 5

[PATCH v4 0/2] MR8200: cmd: COPY should output file names as they are copied.
by Joe Souza (＠JoeS209) 04 Jun '25

04 Jun '25

In native Windows, the COPY command will display the names of the files as they are copied. Wine should do the same. This change enables that. -- v4: cmd/tests: Add tests for unexpected COPY filename output. https://gitlab.winehq.org/wine/wine/-/merge_requests/8200

2 3

[PATCH v2 0/4] MR8191: ntdll: Duplicate handles for thread pool waits.
by Conor McCarthy (＠cmccarthy) 04 Jun '25

04 Jun '25

Windows allows closure of a waitable timer handle while a work item is waiting on it. Also, the current Wine ntdll implementation calls NtWaitForMultipleObjects() on multiple handles if multiple items are pending, and if one handle is not valid, no items will execute. Btw there are occurrences of `INVALID_HANDLE_VALUE` elsewhere, e.g. in `RtlDeleteTimer()`, which I think are incorrect, unless Windows internals are inconsistent with its use. -- v2: ntdll: Duplicate handles for thread pool waits. ntdll: Initialise waitable handles with NULL. ntdll/tests: Test early closure of handles used for threadpool waits. https://gitlab.winehq.org/wine/wine/-/merge_requests/8191

3 6

[PATCH v3 0/1] MR8200: cmd: COPY should output file names as they are copied.
by Joe Souza (＠JoeS209) 04 Jun '25

04 Jun '25

In native Windows, the COPY command will display the names of the files as they are copied. Wine should do the same. This change enables that. -- v3: cmd: COPY should output file names as they are copied. https://gitlab.winehq.org/wine/wine/-/merge_requests/8200

2 2

[PATCH 0/1] MR8204: comctl32: Fix use-after-free when page creation fails.
by Yuxuan Shui (＠yshui) 04 Jun '25

04 Jun '25

When page creation fails in PROPSHEET_SetCurSel, the dialog window is destroyed, which means the attached PropSheetInfo (psInfo) is freed. Because there is a call to PROPSHEET_SetCurSel in the handling of WM_INITDIAG, any use of psInfo thereafter could be invalid, which wasn't taken into account. * * * this feels a bit iffy but i think this is the right fix? previously we got 0 return value from `PropertySheet` because `do_loop` calls `IsWindow(hwnd)`, and when page creation failed the `hwnd` is destroyed, which resulted in `do_loop` returning 0. i just duplicated this check in `PROPSHEET_PropertySheet` too. as for the change in `DiaglogProc`, if `PROPSHEET_SetCurSel` destroyed the window then the SendMessageW isn't necessary, i believe? also, this bit is redundant i think? ```c /* wizards set their focus during init */ if (psInfo->ppshheader.dwFlags & INTRNL_ANY_WIZARD) return FALSE; return FALSE; ``` -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8204

3 2

[PATCH 0/3] MR8205: comctl32: Make sure line_widths is big enough.
by Yuxuan Shui (＠yshui) 04 Jun '25

04 Jun '25

When button_count is 0, we would allocate 0 bytes for line_widths, but later line_count would be 1 so we would still try to access line_widths[0] which is out-of-bound. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8205

2 3