March 2026 - Wine-GitLab - list.winehq.org

Re: [PATCH v9 0/3] MR10306: mfmediaengine: Implement looping behaviour. - approved
by Alexandre Julliard (＠julliard) March 16, 2026

March 16, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10306

1 0

Re: [PATCH v5 0/1] MR10330: ntdll: Implement syscall emulation on Linux using Syscall User Dispatch. - approved
by Alexandre Julliard (＠julliard) March 16, 2026

March 16, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10330

1 0

Re: [PATCH v2 0/2] MR10331: wined3d: Two Vulkan fixes. - approved
by Alexandre Julliard (＠julliard) March 16, 2026

March 16, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10331

1 0

[PATCH v10 0/1] MR10339: quartz/tests: Add tests for IMediaSeeking time formats in avisplit.
by José Luis Jiménez López March 16, 2026

March 16, 2026

Added comprehensive tests for IMediaSeeking_ConvertTimeFormat. Replaced magic numbers with descriptive constants. Verified on Windows 10. Signed-off-by: José Luis Jiménez López josejl1987(a)gmail.com -- v10: quartz/tests/avisplit: Add tests for IMediaSeeking time formats. https://gitlab.winehq.org/wine/wine/-/merge_requests/10339

1 2

[PATCH v3 0/2] MR10075: wow64win: Redirect system32 in NtGdiMakeFontDir().
by Elizabeth Figura (＠zfigura) March 16, 2026

March 16, 2026

-- v3: gdi32: Redirect system32 in AddFontResourceEx(). https://gitlab.winehq.org/wine/wine/-/merge_requests/10075

2 2

[PATCH v2 0/1] MR10256: winegstreamer: Do not copy the incorrect buffer.
by Conor McCarthy (＠cmccarthy) March 16, 2026

March 16, 2026

Rewritten to implement Elizabeth's solution. -- v2: winegstreamer: Do not copy the incorrect buffer. https://gitlab.winehq.org/wine/wine/-/merge_requests/10256

3 2

[PATCH v9 0/1] MR10339: quartz/tests: Add tests for IMediaSeeking time formats in avisplit.
by José Luis Jiménez López March 16, 2026

March 16, 2026

Added comprehensive tests for IMediaSeeking_ConvertTimeFormat. Replaced magic numbers with descriptive constants. Verified on Windows 10. Signed-off-by: José Luis Jiménez López josejl1987(a)gmail.com -- v9: quartz/tests/avisplit: Add tests for IMediaSeeking time formats. https://gitlab.winehq.org/wine/wine/-/merge_requests/10339

3 2

[PATCH 0/2] MR10332: ddraw/tests: Fix vertex winding in test_ck_rgba.
by Stefan Dösinger March 16, 2026

March 16, 2026

On recent Windows 11 versions, probably starting with 26H1, RGB software devices accidentally enable culling by default for transformed geometry. Since this test is run on RGB in addition to HAL devices that causes failues. HAL devices are fine either way, and ddraw2+ looks OK too. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10332

1 3

[PATCH v3 0/1] MR10348: ntdll: Avoid expensive inverted DST check on get_timezone_info() cache hit path
by Zhou Qiankang (＠wszqkzqk) March 16, 2026

March 16, 2026

inverted DST detection (two mktime() calls) out of the cache hit fast path in get_timezone_info(). Keep the cache key based on the cheap raw bias value and only compute inverted_dst/adjusted_bias on cache miss. Preserve the inverse DST correctness fix by still applying adjusted_bias to tzi->Bias in the slow path. This restores fast-path behavior for frequent GetTimeZoneInformation callers without regressing inverse DST handling. Fixes regression: bfae6eb8b4f6 ("ntdll: Fix get_timezone_info() for timezones defined with inverse DST on Unix.") Fixes: https://bugs.winehq.org/show_bug.cgi?id=59514 ## Test results Using a simple test demo: [time_bench.c](/uploads/700a5b087b09a0d93388dbcf437ad42d/time_bench.c) On AMD Ryzen 5800H with power connected it's **24 times** faster. Wine 11.4 without this fix: ``` iterations: 10000 time() : 1.093640 s (109364 ns/call) localtime() : 1.109118 s (110912 ns/call) sink=17736685140379 ``` With this fix: ``` iterations: 10000 time() : 0.046216 s (4622 ns/call) localtime() : 0.046636 s (4664 ns/call) sink=17736685140000 ``` -- v3: ntdll: Avoid expensive inverted DST check on get_timezone_info() cache hit path https://gitlab.winehq.org/wine/wine/-/merge_requests/10348

3 2

[PATCH v2 0/1] MR10348: ntdll: Avoid expensive inverted DST check on get_timezone_info() cache hit path
by Zhou Qiankang (＠wszqkzqk) March 16, 2026

March 16, 2026

inverted DST detection (two mktime() calls) out of the cache hit fast path in get_timezone_info(). Keep the cache key based on the cheap raw bias value and only compute inverted_dst/adjusted_bias on cache miss. Preserve the inverse DST correctness fix by still applying adjusted_bias to tzi->Bias in the slow path. This restores fast-path behavior for frequent GetTimeZoneInformation callers without regressing inverse DST handling. Fixes regression: bfae6eb8b4f6 ("ntdll: Fix get_timezone_info() for timezones defined with inverse DST on Unix.") Fixes: https://bugs.winehq.org/show_bug.cgi?id=59514 ## Test results Using a simple test demo: [time_bench.c](/uploads/700a5b087b09a0d93388dbcf437ad42d/time_bench.c) On AMD Ryzen 5800H with power connected it's **24 times** faster. Wine 11.4 without this fix: ``` iterations: 10000 time() : 1.093640 s (109364 ns/call) localtime() : 1.109118 s (110912 ns/call) sink=17736685140379 ``` With this fix: ``` iterations: 10000 time() : 0.046216 s (4622 ns/call) localtime() : 0.046636 s (4664 ns/call) sink=17736685140000 ``` -- v2: ntdll: Avoid expensive inverted DST check on get_timezone_info() cache hit path https://gitlab.winehq.org/wine/wine/-/merge_requests/10348

3 2