- Wine-GitLab - list.winehq.org

[PATCH 0/1] MR10488: win32u: Optimize the speed of halftone primitives by using 32.32 fixed point arithmetic.
by Zhiyi Zhang (＠zhiyi) April 21, 2026

April 21, 2026

Floating-point operations are expensive, especially when running on x86 CPU emulators. Tests show that speed after this commit is improved by 18% on AMD 9950x with no visual difference. [halftone-test.zip](/uploads/d9cede791109b4d32d767ad0b7cac79e/halftone-test.zip) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10488

3 2

[PATCH 0/2] MR10697: win32u: Reject DDBs that are not 1-bit or 32-bit for GetDIBits().
by Zhiyi Zhang (＠zhiyi) April 21, 2026

April 21, 2026

SDL3 applications depend on this to not overwrite the depth of its internal display mode. Fix SDL3 applications using the native display mode instead of the specified display mode when restoring to fullscreen from focus loss for the first time, and the specific display mode is 16-bit. Please see `WIN_UpdateDisplayMode()` in the SDL3 source code. ```c static void WIN_UpdateDisplayMode(SDL_VideoDevice *_this, LPCWSTR deviceName, DWORD index, SDL_DisplayMode *mode) { ... if (index == ENUM_CURRENT_SETTINGS && (hdc = CreateDCW(deviceName, NULL, NULL, NULL)) != NULL) { char bmi_data[sizeof(BITMAPINFOHEADER) + 256 * sizeof(RGBQUAD)]; LPBITMAPINFO bmi; HBITMAP hbm; SDL_zeroa(bmi_data); bmi = (LPBITMAPINFO)bmi_data; bmi->bmiHeader.biSize = sizeof(BITMAPINFOHEADER); hbm = CreateCompatibleBitmap(hdc, 1, 1); GetDIBits(hdc, hbm, 0, 1, NULL, bmi, DIB_RGB_COLORS); GetDIBits(hdc, hbm, 0, 1, NULL, bmi, DIB_RGB_COLORS); DeleteObject(hbm); DeleteDC(hdc); if (bmi->bmiHeader.biCompression == BI_BITFIELDS) { switch (*(Uint32 *)bmi->bmiColors) { ... case 0x7C00: mode->format = SDL_PIXELFORMAT_XRGB1555; break; } ... ``` When entering a 16-bit display mode, `WIN_UpdateDisplayMode(_this, displaydata->DeviceName, ENUM_CURRENT_SETTINGS, mode);` is called in `WIN_SetDisplayMode()`. The second `GetDIBits()` call in `WIN_UpdateDisplayMode()` succeeds and ends up overwriting the format to SDL_PIXELFORMAT_XRGB1555, which is a 15-bit display mode. So when SDL3 tries to enter the 16-bit display mode again, it won't be found because it's now 15-bit. Tests show that the second `GetDIBits()` should fail and thus leave the display mode format unchanged. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10697

3 4

[PATCH v4 0/2] MR10504: Draft: vbscript: Implement GetLocale and SetLocale functions.
by Francis De Brabandere (＠francisdb) April 21, 2026

April 21, 2026

-- v4: vbscript: Use script LCID in Format* and call SetThreadLocale. vbscript: Implement GetLocale and SetLocale functions. https://gitlab.winehq.org/wine/wine/-/merge_requests/10504

2 3

[PATCH v2 0/7] MR10706: msxml3: Respect document encoding in Save().
by Nikolay Sivov (＠nsivov) April 21, 2026

April 21, 2026

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- v2: msxml3/tests: Add a few missing return value checks (Coverity). msxml3/tests: Fix some use-after-free (Coverity). msxml3/tests: Fix a string double free (Coverity). msxml3/tests: Fix copy-paste issue when testing node value (Coverity). msxml3: Improve cleanup on error paths when creating DOM nodes (Coverity). msxml3: Fix a leak when parsing xml declaration body (Coverity). msxml3: Respect document encoding in Save(). https://gitlab.winehq.org/wine/wine/-/merge_requests/10706

2 7

[PATCH 0/3] MR10708: mfplat: Strip leading slashes from the URL in source_resolver_CreateObjectFromURL().
by Conor McCarthy (＠cmccarthy) April 21, 2026

April 21, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10708

3 4

Re: [PATCH v2 0/7] MR10688: msxml3: Improvements to setAttribute().
by Nikolay Sivov (＠nsivov) April 21, 2026

April 21, 2026

On Tue Apr 21 07:42:53 2026 +0000, Bernhard Übelacker wrote: > And also silence from the [gitlab run](https://gitlab.winehq.org/bernhardu/wine/-/pipelines/64016). That's very good, thank you for reporting. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10688#note_137190

1 0

Re: [PATCH v2 0/7] MR10688: msxml3: Improvements to setAttribute().
by Bernhard Übelacker (＠bernhardu) April 21, 2026

April 21, 2026

On Mon Apr 20 21:28:41 2026 +0000, Bernhard Übelacker wrote: > With 7cb26d6936 applied ASan stays silent on my local system, with > x86_64 and i386. And also silence from the [gitlab run](https://gitlab.winehq.org/bernhardu/wine/-/pipelines/64016). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10688#note_137186

1 0

[PATCH v13 0/12] MR10569: wineandroid: rework Android integration and switch to split process model
by Twaik Yont (＠twaik) April 21, 2026

April 21, 2026

This series reworks the wineandroid driver and moves it away from the current “Wine inside JVM process” model to a split architecture where the Android activity (UI) and Wine run in separate processes. The existing model relies on running Wine inside the JVM process and interacting with Android through JNI callbacks and APC/unixlib-based dispatch. This approach no longer works reliably on modern Android: - Android 10+ applies seccomp restrictions to app processes, making a number of required syscalls unavailable inside the activity process, while processes started via fork/exec are not subject to the same limitations. - The implementation depends on Wine thread context (TEB), signal handling, and register state assumptions that conflict with the JVM runtime (e.g. segment register usage on x86), requiring fragile workarounds. - Due to sandboxing, it is difficult to reliably debug or even observe all failure modes, and future Android changes are likely to tighten these restrictions further. At the same time, downstream projects such as MiceWine, Winlator, Termux’s Wine environment, GameHub and GameNative already use a model where the UI (activity) and Wine run in separate processes and communicate via IPC, which has proven to work reliably on current Android systems. This series implements that model in wineandroid: - Wine is now started in a separate process via fork/exec instead of running inside the JVM process. - Android-facing code no longer depends on the Wine APC/unixlib execution path and runs on the UI thread where appropriate (e.g. native window registration). - ioctl handling is reworked to remove reliance on Wine thread context, with explicit JNIEnv usage and fd-based reply plumbing. - Environment setup and initialization are moved to the Java side, simplifying the native startup path. - Logging for ioctl paths is routed through Android logcat without depending on a Wine TEB. The changes are split into multiple bisect-safe commits to keep intermediate states functional while migrating the code step by step. With Wine running in a separate process, it is no longer subject to JVM seccomp restrictions, and there are no longer conflicts between JVM and Wine execution contexts (signals, thread state, register usage, etc.). <details> <summary>Old message</summary> wineandroid: move ioctl handling to dedicated Java-only thread Move ioctl handling out of the Wine APC/unixlib execution path into a dedicated Java-only thread without Wine context. On Android 10+ a number of syscalls are restricted inside the activity process via seccomp, while processes started via fork/exec are not subject to the same limitations. Decoupling ioctl handling from the Wine context is required to eventually run this code outside of the activity process. This change introduces a new transport and execution model where ioctl dispatch no longer depends on Wine thread state, preparing for moving the Android backend into a separate process. The refactor is intentionally split into multiple bisect-safe commits to keep intermediate states functional while gradually migrating the dispatch path. Since this is a fairly large chunk of work, I’d like to make sure we’re aligned before continuing. </details> -- v13: wineandroid: split Android driver and Wine into separate processes wineandroid: terminate activity when desktop client disconnects wineandroid: make JNI initialization work with direct JVM loading wineandroid: create desktop view via ioctl and pass event pipe wineandroid: return ioctl errors directly wineandroid: remove unixlib/ntoskrnl ioctl path wineandroid: switch dequeueBuffer to socket transport wineandroid: switch ioctl dispatch to socket transport on main thread looper wineandroid: pass JNIEnv explicitly through ioctl handlers wineandroid: add reply_fd plumbing to ioctl path wineandroid: move native window registration to surface_changed wineandroid: use Android logging for ioctl dispatch paths https://gitlab.winehq.org/wine/wine/-/merge_requests/10569

2 12

[PATCH v3 0/1] MR10541: vbscript: Fast-path assign_value for simple non-refcounted types.
by Francis De Brabandere (＠francisdb) April 21, 2026

April 21, 2026

Skip VariantCopyInd and VariantClear when both source and destination are simple scalar types (VT_I2, VT_I4, VT_R8, VT_BOOL, VT_EMPTY, etc.) that require no memory management. A direct struct assignment suffices for these types. Benchmarks (10M iterations, measured in isolation on top of master): - Do-While loop: 824 -> 472 ms (1.75×) - For + assignment: 324 -> 265 ms (1.22×) - Dict.Items iteration: 683 -> 324 ms (2.1×) - If-Else branching: 726 -> 644 ms (1.13×) Every variable assignment in the interpreter benefits. The fast path eliminates two cross-DLL calls to oleaut32 (VariantCopyInd + VariantClear) for each assignment of simple scalar types (VT_I2, VT_I4, VT_R8, VT_BOOL, etc.), replacing them with a single struct copy. -- v3: vbscript: Fast-path assign_value for simple non-refcounted types. https://gitlab.winehq.org/wine/wine/-/merge_requests/10541

2 1

[PATCH v3 0/1] MR10485: vbscript: Return proper errors for orphan Loop and Next keywords.
by Francis De Brabandere (＠francisdb) April 21, 2026

April 21, 2026

When Loop or Next appear outside their enclosing Do/For blocks, Windows returns specific error codes: 1038 ("loop without do") and 1055 ("unexpected Next"). Detect these tokens in parser_error() and set the proper HRESULT instead of the generic E_FAIL. -- v3: vbscript: Return proper errors for orphan Loop and Next keywords. https://gitlab.winehq.org/wine/wine/-/merge_requests/10485

2 1

Re: [PATCH v2 0/4] MR10654: winegstreamer: Fix wg_parser/native video stride mismatch.
by Brendan McGrath (＠redmcg) April 20, 2026

April 20, 2026

> I check the `MF_MT_DEFAULT_STRIDE` and the bitmap info stride, both are 162 re: the `MF_MT_DEFAULT_STRIDE`, I've found with the source reader, you can get the `MF_SOURCE_READERF_CURRENTMEDIATYPECHANGED` flag once you start reading samples. I wonder if the `MF_MT_DEFAULT_STRIDE` value is updated at that point. Otherwise it feels like an MS bug to report one thing and do another. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10654#note_137178

1 0

Re: [PATCH v2 0/4] MR10654: winegstreamer: Fix wg_parser/native video stride mismatch.
by Brendan McGrath (＠redmcg) April 20, 2026

April 20, 2026

On Mon Apr 20 21:30:04 2026 +0000, Yuxuan Shui wrote: > > check divisibility? > And that was how I was figuring out the stride. Brief explanation of my observations: > Video I used: yuv420p pixel format, size 162x120. > I create source reader then set output type to YV12. I check the > `MF_MT_DEFAULT_STRIDE` and the bitmap info stride, both are 162. I then > `IMFMediaBuffer::Lock` it. The buffer size reported by `Lock` is `33792`. > I calculated the Y plan size: `33792 / 3 * 2 = 22528`, and realized that > it's not divisible by 162: `22528 = 162*139+10`. So I tried rounding up > to 8-bytes `22528 / 168`, doesn't work. Then rounding up to 16-bytes > `22528 / 176 = 128`. Oh interesting. I was curious how the `IMF2DBuffer::ContiguousCopyFrom` and `IMF2DBuffer::ContiguousCopyTo` methods would work in this case and that led me to this: https://learn.microsoft.com/en-us/windows/win32/api/mfobjects/nn-mfobjects-… That states: > Every video format defines a contiguous or packed representation. This representation is compatible with the standard layout of a DirectX surface in system memory, with no additional padding. So maybe that's where the 16-byte alignment comes from. But that would break compatibility with older MFTs that aren't aware to check for the `IMF2DBuffer` interface (in order to determine which rule is being used). Edit: Although I guess if the media type for the output has `MF_MT_DEFAULT_STRIDE` and that is honoured by the MFT, then it would still work. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10654#note_137177

1 0

[PATCH 0/1] MR10707: po: update Polish translation
by Adam Markowski (＠AMark) April 20, 2026

April 20, 2026

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10707

2 1

[PATCH 0/5] MR10706: msxml3: Respect document encoding in Save().
by Nikolay Sivov (＠nsivov) April 20, 2026

April 20, 2026

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10706

2 5

[PATCH v12 0/1] MR10484: vbscript: Return "Invalid for loop control variable" error for member expressions.
by Francis De Brabandere (＠francisdb) April 20, 2026

April 20, 2026

Windows VBScript rejects member expressions (e.g. x.y) as For/For Each loop control variables with error 1040. Accept MemberExpression in the parser rules and check for obj_expr to report the proper error instead of a generic syntax error. -- v12: vbscript: Return "Invalid for loop control variable" error for member expressions. https://gitlab.winehq.org/wine/wine/-/merge_requests/10484

2 1

[PATCH v22 0/3] MR10464: vbscript: Return "Name redefined" error for duplicate declarations.
by Francis De Brabandere (＠francisdb) April 20, 2026

April 20, 2026

When a name is defined twice (Const/Const, Dim/Const, Const/Dim, Dim/Dim, Sub vs Const/Dim, Class vs Const/Dim/Function, duplicate class members), Windows returns VBScript error 1041 ("Name redefined"). Wine was returning E_FAIL. Store source locations in const_decl_t, dim_decl_t, function_decl_t, and class_decl_t so errors point to the duplicate name. Also fix duplicate Sub/Function/Property/Dim detection inside classes, including the case where a Property collides with an existing Sub/Function. -- v22: vbscript: Return "Name redefined" error for duplicate class members. vbscript: Return "Name redefined" error for duplicate global declarations. vbscript: Track source location in const, class, and function declarations. https://gitlab.winehq.org/wine/wine/-/merge_requests/10464

2 3

Re: [PATCH v2 0/4] MR10654: winegstreamer: Fix wg_parser/native video stride mismatch.
by Yuxuan Shui (＠yshui) April 20, 2026

April 20, 2026

On Thu Apr 16 19:09:11 2026 +0000, Elizabeth Figura wrote: > > Yes, it's just that the test videos I created are encoded with h264, > and are decoded so the test is testing raw video (because this is the > most straightforward way I thought). > That's not going to test what you want; that goes through an > IMFTransform on Windows. > > And since you mentioned it I was thinking I shouldn't have started > with h264 at all. I am thinking a custom media source that output raw > RGB and use video processing in source reader to convert to yuv. > That's not going to test what you want either. You want to encode raw > video in an existing container and have the builtin demuxer deal with > that. Most containers support raw video. Oh right, I am dumb. I was only thinking about wine and forgot h264 would also be going through IMFTransform on Windows. I'll do what you suggested. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10654#note_137172

1 0

Re: [PATCH v2 0/4] MR10654: winegstreamer: Fix wg_parser/native video stride mismatch.
by Yuxuan Shui (＠yshui) April 20, 2026

April 20, 2026

> check divisibility? And that was how I was figuring out the stride. Brief explanation of my observations: Video I used: yuv420p pixel format, size 162x120. I create source reader then set output type to YV12. I check the `MF_MT_DEFAULT_STRIDE` and the bitmap info stride, both are 162. I then `IMFMediaBuffer::Lock` it. The buffer size reported by `Lock` is `33792`. I calculated the Y plan size: `33792 / 3 * 2 = 22528`, and realized that it's not divisible by 162: `22528 = 162*139+10`. So I tried rounding up to 8-bytes `22528 / 168`, doesn't work. Then rounding up to 16-bytes `22528 / 176 = 128`. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10654#note_137171

1 0

Re: [PATCH v2 0/7] MR10688: msxml3: Improvements to setAttribute().
by Bernhard Übelacker (＠bernhardu) April 20, 2026

April 20, 2026

On Mon Apr 20 21:06:09 2026 +0000, Nikolay Sivov wrote: > Could you try if !10705 helps? With 7cb26d6936 applied ASan stays silent on my local system, with x86_64 and i386. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10688#note_137169

1 0

Re: [PATCH v2 0/4] MR10654: winegstreamer: Fix wg_parser/native video stride mismatch.
by Yuxuan Shui (＠yshui) April 20, 2026

April 20, 2026

> the stride is suppose to come from the `MF_MT_DEFAULT_STRIDE` attribute within the media type. Otherwise you can fallback to using `MFGetStrideForBitmapInfoHeader`. It seems to me that this stride is more like a minimum stride. For example, for the YV12, width 162 video I used, this stride is 162, but the actual samples I got have stride 176 when locked with `IMFMediaBuffer::Lock`. I don't know how a game will figure this number out. Maybe calculate the size of the Y plane (= buffer_size / 3 * 2), then check divisibility? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10654#note_137168

1 0

Re: [PATCH v3 0/5] MR10677: amstream: Test and fix media type enum on ddraw stream. - approved
by Alexandre Julliard (＠julliard) April 20, 2026

April 20, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10677

1 0

Re: [PATCH v3 0/2] MR10695: ntdll: Improvements for RtlVirtualUnwind2 with NULL output handler data. - approved
by Alexandre Julliard (＠julliard) April 20, 2026

April 20, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10695

1 0

Re: [PATCH v2 0/1] MR10698: kernelbase: Add message string for error 302-316 - approved
by Alexandre Julliard (＠julliard) April 20, 2026

April 20, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10698

1 0

Re: [PATCH v7 0/1] MR10600: vbscript: Convert VT_DISPATCH arguments to string in Eval/Execute/ExecuteGlobal. - approved
by Alexandre Julliard (＠julliard) April 20, 2026

April 20, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10600

1 0

Re: [PATCH v3 0/1] MR10496: vbscript: Return error 1045 for non-literal constant expressions. - approved
by Alexandre Julliard (＠julliard) April 20, 2026

April 20, 2026

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10496

1 0