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

[PATCH v19 0/16] MR10611: LPC: Implement proper LPC system
by Rose Hellsing (＠axtlos) June 5, 2026

June 5, 2026

Until now, LPC was just a stub, any application that relied on LPC for communication would fail. This change implements parts of LPC, to allow for IPC between programs. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=30069 - fully fixed, sqlservr is able to run as expected Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59494 - partially fixed, starmoney doesn't flat out crash anymore, but it doesn't start fully, possibly because it uses the ALPC system which was introduced in Windows Vista, the version Starmoney 10 was advertised for -- v19: wow64: properly convert lpc messages wip: fix LPC tests server: Fix double list_remove in LPC port logic ntdll/tests: Add LPC port error handling tests. server: Handle LPC port close to break circular references. ntdll: Implement NtRegisterThreadTerminatePort. server: Implement LPC thread termination notification. ntdll: Implement NtReplyPort, NtReplyWaitReceivePort and NtReplyWaitReceivePortEx. ntdll: Implement NtRequestPort and NtRequestWaitReplyPort. server: Implement LPC message request and reply handlers. ntdll: Implement NtListenPort, NtAcceptConnectPort and NtCompleteConnectPort. ntdll: Implement NtConnectPort and NtSecureConnectPort. server: Implement LPC connection acceptance and completion. server: Implement LPC connection request and listen handlers. ntdll: Implement NtCreatePort and NtCreateWaitablePort. server: Add LPC port object infrastructure. This merge request has too many patches to be relayed via email. Please visit the URL below to see the contents of the merge request. https://gitlab.winehq.org/wine/wine/-/merge_requests/10611

1 0

Re: [PATCH v3 0/1] MR11067: hhctrl: return a nonzero cookie from HH_INITIALIZE - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11067

1 0

Re: [PATCH v2 0/2] MR11009: widl: Add support for enums to SLTG typelib generator. - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11009

1 0

Re: [PATCH v7 0/5] MR10716: dsound: Speed up resampling, part 6 - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10716

1 0

Re: [PATCH v5 0/1] MR10966: dsound: Fail IDirectSoundCaptureBufferImpl_Lock() for invalid buffer range. - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10966

1 0

[PATCH 0/3] MR11069: Fix VEH/misaligned stack related crashes with copy protection software.
by Soham Nandy (＠natimerry) June 4, 2026

June 4, 2026

These patches aim to fix crashes and hardlocks when running software that uses a exception handler as a control flow stub, which also happens to violate stack alignment. The software register a VEH which calls a procedure that destroys stack alignment and restores a pre-saved context before returning. This VEH also calls OpenThread with a misaligned stack, causing GCC/Clang's movaps to fault and recurse through the VEH until signal stack is exhausted. Additionally the software installs hw breakpoints on KSHARED_USER_DATA. Wine's signal handlers access the same addresses triggering recursive SIGTRAP dispatches which crashes via `virtual_setup_exception()`'s nested exception check. Although all 3 patches are required together for the software in question to function, the HWBKPT signal fix and the VEH clobber fix Wine issues that may affect other software. I am the least certain about the viability of `OpenThread`'s handrolled ASM version. If the assembly implementation is not acceptable applying `__attribute__((optimize("no-tree-vectorize")))` would also prevent movaps emission. OpenThread is a small function that does not meaningfully benefit from SIMD acceleration so the performance effect should be negligible. Discussed on #winehackers prior to submission. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11069

4 12

Re: [PATCH v4 0/5] MR11066: msxml3/writer: Normalize newlines in CDATA content. - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11066

1 0

Re: [PATCH v10 0/3] MR11001: ntdll, ntoskrnl.exe: account for read-only cookie in loader - approved
by Alexandre Julliard (＠julliard) June 4, 2026

June 4, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11001

1 0

[PATCH v3 0/1] MR11071: ntdll: Ignore hardware breakpoint traps inside the signal stack.
by Soham Nandy (＠natimerry) June 4, 2026

June 4, 2026

Ignore hardware-breakpoint traps raised to avoid recursively dispatching a trap through the Unix signal handler when user code installs a hardware breakpoint that is also reached from Wine's signal-handling path. This is split up from https://gitlab.winehq.org/wine/wine/-/merge_requests/11069 as asked. -- v3: ntdll: Ignore hardware breakpoint traps inside the signal stack. https://gitlab.winehq.org/wine/wine/-/merge_requests/11071

3 3

[PATCH v2 0/2] MR11071: ntdll: Ignore hardware breakpoint traps inside the signal stack.
by Soham Nandy (＠natimerry) June 4, 2026

June 4, 2026

Ignore hardware-breakpoint traps raised to avoid recursively dispatching a trap through the Unix signal handler when user code installs a hardware breakpoint that is also reached from Wine's signal-handling path. This is split up from https://gitlab.winehq.org/wine/wine/-/merge_requests/11069 as asked. -- v2: Apply 1 suggestion(s) to 1 file(s) https://gitlab.winehq.org/wine/wine/-/merge_requests/11071

3 3

[PATCH 0/1] MR11071: ntdll: Ignore hardware breakpoint traps inside the signal stack.
by Soham Nandy (＠natimerry) June 4, 2026

June 4, 2026

Ignore hardware-breakpoint traps raised to avoid recursively dispatching a trap through the Unix signal handler when user code installs a hardware breakpoint that is also reached from Wine's signal-handling path. This is split up from https://gitlab.winehq.org/wine/wine/-/merge_requests/11069 as asked. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11071

4 6

[PATCH 0/2] MR11070: d3dcompiler: Retrieve more information using vkd3d in d3dcompiler_shader_reflection_init().
by Henri Verbeet (＠hverbeet) June 4, 2026

June 4, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/11070

2 2

[PATCH 0/1] MR11062: wined3d: Add Qualcomm vendor and Adreno 640 GPU info
by Stian Low (＠stianlow) June 4, 2026

June 4, 2026

https://bugs.winehq.org/show_bug.cgi?id=59812 Patch adds Qualcomm as new GPU vendor and uses Adreno 640 GPU as default/fallback placeholder for future entries based on existing vendor/gpu entries. Driver qdcmlib.ARM64.dll as default is a best guess and may need to change to something more accurate in future. Unclear if wbemprox support is necessary but added it for consistency and to cover all cases. vidmem assigned to 5540 MB for gpu_description_table as reported by user's glxinfo: - https://bugs.winehq.org/attachment.cgi?id=81077 videmem also boosted to 32016 MB for CARD_INTEL_UHD620_2 (recently added) to match that user's glxinfo instead of vidmem copied from CARD_INTEL_UHD620_1. Unclear if CARD_INTEL_UHD620_1 should also be boosted to match CARD_INTEL_UHD620_2. Figured this change is related/recent enough to be rolled into this MR. I'll update that bug report about this change also: - https://bugs.winehq.org/show_bug.cgi?id=59186 - https://bugs.winehq.org/attachment.cgi?id=80058 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11062

2 4

[PATCH 0/2] MR11009: widl: Add support for enums to SLTG typelib generator.
by Dmitry Timoshkov (＠dmitry) June 4, 2026

June 4, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/11009

3 3

[PATCH v6 0/5] MR10716: dsound: Speed up resampling, part 6
by Anton Baskanov (＠baskanov) June 4, 2026

June 4, 2026

-- v6: dsound: Add an SSE version of downsample. dsound: Add an SSE version of upsample. https://gitlab.winehq.org/wine/wine/-/merge_requests/10716

4 9

[PATCH v3 0/2] MR11018: combase: Implement delayed marshaling for RoGetAgileReference
by Vibhav Pant (＠vibhavp) June 4, 2026

June 4, 2026

When RoGetAgileReference is used with AGILEREFERENCE_DELAYEDMARSHAL, the returned IAgileReference also contains a reference to the apartment/context the object belongs to. Thus, when IAgileReference::Resolve() gets called, the object will get marshaled inside the owning apartment/context, not the caller's apartment. -- v3: combase: Implement delayed marshaling for RoGetAgileReference. combase/tests: Add tests for delayed marshaling of objects with apartment affinity. https://gitlab.winehq.org/wine/wine/-/merge_requests/11018

4 5

[PATCH v4 0/1] MR10966: dsound: Fail IDirectSoundCaptureBufferImpl_Lock() for invalid buffer range.
by Paul Gofman (＠gofman) June 4, 2026

June 4, 2026

-- v4: dsound: Fail IDirectSoundCaptureBufferImpl_Lock() for invalid buffer range. https://gitlab.winehq.org/wine/wine/-/merge_requests/10966

4 5

[PATCH v3 0/5] MR11066: msxml3/writer: Normalize newlines in CDATA content.
by Nikolay Sivov (＠nsivov) June 4, 2026

June 4, 2026

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- v3: msxml3/writer: Normalize newlines in CDATA content. msxml3/tests: Avoid string conversion in a few places. msxml3/tests: Use writer output test helper throughout the tests. https://gitlab.winehq.org/wine/wine/-/merge_requests/11066

2 5

[PATCH v2 0/1] MR11067: hhctrl: return a nonzero cookie from HH_INITIALIZE
by Russell Greene (＠russelltg) June 4, 2026

June 4, 2026

Fixes launch of Simberian Simbeor PCB simulation software. It seems it must do a check to see if the cookie is 0. The software is not open source but a view-only license can be acquired for free by making an account on their website. This doesn't seem to be documented behavior (https://learn.microsoft.com/en-us/previous-versions/windows/desktop/htmlhel…) does not mention that 0 is an invalid cookie, but Windows must not return zero otherwise this app would not work on windows? -- v2: hhctrl: return a nonzero cookie from HH_INITIALIZE https://gitlab.winehq.org/wine/wine/-/merge_requests/11067

2 1

[PATCH 0/1] MR11067: hhctrl: return a nonzero cookie from HH_INITIALIZE
by Russell Greene (＠russelltg) June 4, 2026

June 4, 2026

Fixes launch of Simberian Simbeor PCB simulation software. It seems it must do a check to see if the cookie is 0. The software is not open source but a view-only license can be acquired for free by making an account on their website. This doesn't seem to be documented behavior (https://learn.microsoft.com/en-us/previous-versions/windows/desktop/htmlhel…) does not mention that 0 is an invalid cookie, but Windows must not return zero otherwise this app would not work on windows? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11067

1 0

[PATCH v2 0/5] MR11066: msxml3/writer: Normalize newlines in CDATA content.
by Nikolay Sivov (＠nsivov) June 4, 2026

June 4, 2026

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- v2: msxml3/writer: Normalize newlines in CDATA content. msxml3/tests: Avoid string conversion in a few places. msxml3/tests: Use writer output test helper throughout the tests. msxml3/tests: Add more tests for the writer output. https://gitlab.winehq.org/wine/wine/-/merge_requests/11066

2 5

[PATCH 0/2] MR11063: sane.ds: TWAIN improvements.
by Dmitry Timoshkov (＠dmitry) June 4, 2026

June 4, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/11063

3 3

[PATCH 0/5] MR11066: msxml3/writer: Normalize newlines in CDATA content.
by Nikolay Sivov (＠nsivov) June 4, 2026

June 4, 2026

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/11066

2 5

[PATCH v2 0/1] MR10848: winewayland.drv: Add alpha-modifier-v1 protocol to support change alpha by...
by JiangYi Chen (＠meshine) June 4, 2026

June 4, 2026

winewayland.drv: Add alpha-modifier-v1 protocol to support change alpha by WAYLAND_SetLayeredWindowAttributes. The attachment is the test source code on Windows and the corresponding generated test program. [test-layered-alpha.cpp](/uploads/dba6d38e8990978d4bde562011c85995/test-layered-alpha.cpp) [test-layered-alpha.exe](/uploads/bba6c165533d56ca85af9d461e488b03/test-layered-alpha.exe) The test program runs as shown in the figure below. ![test](/uploads/d862049eaeda3c57265b648a529f4af8/test.bmp){width=505 height=543} The purpose of the patch is that when adjusting the Alpha slider in the figure, the transparency of the window changes accordingly. -- v2: winewayland.drv: Add alpha-modifier-v1 protocol to support change alpha by WAYLAND_SetLayeredWindowAttributes. https://gitlab.winehq.org/wine/wine/-/merge_requests/10848

3 2

Re: [PATCH v2 0/1] MR10587: server: Update user_time when receive MSG_NOTIFY message.
by Rémi Bernon (＠rbernon) June 4, 2026

June 4, 2026

Well we may indeed tweak the foreground window changes rules a bit but whether the tray popup window is supposed to actually get foreground or not when displayed, would first need to be asserted with some tests. Using MSG_NOTIFY seem to be too wide, and we should probably try to make a more targeted change. It's also possible that explorer.exe should be somehow unrestricted wrt foreground window changes, but again it's difficult to say what kind of effect this would have. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10587#note_142187

1 0