- wine-gitlab - winehq.org

Re: [PATCH v2 0/8] MR7597: wintypes: Implement IBufferByteAccess::Buffer().
by Mohamad Al-Jaf (＠maljaf) 19 Mar '25

19 Mar '25

On Wed Mar 19 05:07:15 2025 +0000, Rémi Bernon wrote: > Is it really useful to test such corner case here? I'm not sure what you mean by corner case, the test fails on i386 because 0xdeadbeef is too big of an allocation. In any case, letting malloc handle it makes the check unnecessary. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7597#note_98369

1 0

[PATCH v2 0/2] MR7606: user32/edit: Block key input when mouse input is captured.
by Nikolay Sivov (＠nsivov) 19 Mar '25

19 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56108 -- v2: comctl32/edit: Block key input when mouse input is captured. user32/edit: Block key input when mouse input is captured. https://gitlab.winehq.org/wine/wine/-/merge_requests/7606

3 4

[PATCH 0/1] MR7599: COMCTL32: LVN_ENDLABELEDIT with empty text must send "" and not NULL
by Anders Kjersem (＠anders) 19 Mar '25

19 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7599

4 5

[PATCH v11 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 19 Mar '25

19 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v11: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses. https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

3 3

Re: [PATCH v5 0/6] MR6866: ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code.
by Paul Gofman (＠gofman) 18 Mar '25

18 Mar '25

Paul Gofman (@gofman) commented about dlls/ntdll/unix/signal_x86_64.c: > "callq *(%r10,%rax,8)\n\t" > "leaq -0x98(%rbp),%rcx\n\t" > __ASM_LOCAL_LABEL("__wine_syscall_dispatcher_return") ":\n\t" > + /* push rbp-based kernel stack cfi */ > + __ASM_CFI(".cfi_remember_state\n\t") > + __ASM_CFI_CFA_IS_AT2(rcx, 0xa8, 0x01) /* frame->syscall_cfa */ > + "leaq 0x70(%rcx),%rsp\n\t" /* %rsp > frame means no longer inside syscall */ Looks like patch subject is a bit misleading? "leaq 0x70(%rcx),%rsp\n\t" doesn't leave kernel stack, maybe "ntdll: Move stack to machine frame before accessing %gs in x86_64 syscall dispatcher."? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/6866#note_98356

1 0

Re: [PATCH v5 0/6] MR6866: ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code.
by Brendan Shanks (＠bshanks) 18 Mar '25

18 Mar '25

On Tue Mar 18 23:36:40 2025 +0000, Jinoh Kang wrote: > Yeah, I think so. FWIW, even with two MOVs, should be faster than one > MOV from GS indirect. Thanks, I suspect the performance impact is negligible but the patch is certainly cleaner. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/6866#note_98354

1 0

[PATCH v4 0/6] MR6866: ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code.
by Brendan Shanks (＠bshanks) 18 Mar '25

18 Mar '25

x86_64 Windows and macOS both use `%gs` to access thread-specific data (Windows TEB, macOS TSD). To date, Wine has worked around this conflict by filling the most important TEB fields (`0x30`/`Self`, `0x58`/`ThreadLocalStorage`) in the macOS TSD structure (Apple reserved the fields for our use). This was sufficient for most Windows apps. CrossOver's Wine had an additional hack to handle `0x60`/`ProcessEnvironmentBlock`, and binary patches for certain CEF binaries which directly accessed `0x8`/`StackBase`. Additionally, Apple's libd3dshared could activate a special mode in Rosetta 2 where code executing in certain regions would use the Windows TEB when accessing `%gs`. Now that the PE separation is complete, GSBASE can be swapped when entering/exiting PE code. This is done in the syscall dispatcher, unix-call dispatcher, and for user-mode callbacks. GSBASE also needs to be set to the macOS TSD when entering signal handlers (in `init_handler()`), and then restored to the Windows TEB when exiting (in `leave_handler()`). Some changes to the syscall dispatcher were needed to ensure that the TEB is not accessed through `%gs` while on the kernel stack (since a SIGUSR1 while on the kernel stack will result in GSBASE being set to the TSD). --- I've tested this successfully on macOS 15 (Apple Silicon and Intel) and macOS 10.13 with several apps and games, including the `cefclient.exe` CEF sample. Encouragingly, in some simple tests I didn't see a noticeable performance regression from this MR. There are drawbacks though: - libraries which jump directly from PE code into Unix code (expecting that %gs is always pointing to the macOS TSD) will crash. Notable examples are D3DMetal and DXMT. These will need to be changed to use Unix calls. - If Windows code uses the `syscall` instruction directly, the stack pointer likely needs to be valid (which is probably not true on Windows). This is due to the syscall dispatcher saving registers onto the user stack and having to call `_thread_set_tsd_base`. I can't say I've ever seen direct syscalls done with an invalid `%rsp`, but it seems like something anticheat code might do. --- macOS does not have a public API for setting GSBASE, but the private `_thread_set_tsd_base()` works and was added in macOS 10.12. `_thread_set_tsd_base()` is a small thunk that sets `%esi`, `%eax`, and does the `syscall`: https://github.com/apple-oss-distributions/xnu/blob/8d741a5de7ff4191bf97d57…. The syscall instruction itself clobbers `%rcx` and `%r11`. I've tried to save as few registers as possible when calling `_thread_set_tsd_base()`, but there may be room for improvement there. --- I also tested an alternate implementation strategy for this which took advantage of the expanded "full" thread state which is passed to signal handlers when a process has set a user LDT. The full thread state includes GSBASE, so GSBASE is set back to whatever is in the sigcontext on return (like every other field in the context). This would avoid needing to explicitly reset GSBASE in `leave_handler()`. This strategy was simpler, but I'm not using it for 2 reasons: - the "full" thread state is only available starting with macOS 10.15, and we still support 10.13. - more crucially, Rosetta 2 doesn't seem to correctly implement the GS.base field of the full thread state. It's set to 0 on entry, and isn't read on exit. -- v4: ntdll: Remove x86_64 Mac-specific TEB access workarounds that are no longer needed. ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code. ntdll: Leave kernel stack before accessing %gs in x86_64 syscall dispatcher. ntdll: Don't access the TEB through %gs when using the kernel stack in x86_64 syscall dispatcher. ntdll: Ensure init_handler runs in signal handlers before any compiler-generated memset calls. ntdll: Remove ugly fallback method for getting a thread's GSBASE on macOS. https://gitlab.winehq.org/wine/wine/-/merge_requests/6866

2 6

[PATCH 0/1] MR7609: ntoskrnl.exe: Implement MmMapLockedPages.
by Alexander Morozov (＠amorozov) 18 Mar '25

18 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7609

2 1

[PATCH 0/5] MR7608: server: Avoid delegating console waits to other objects.
by Elizabeth Figura (＠zfigura) 18 Mar '25

18 Mar '25

Some console objects currently do several unique things: * Delegate waits onto the queue of another object. This is not really a problem for in-process waits, since we can just return the sync object for the delegate. However, it's also unnecessary, adds to the complexity of the server logic, and is one out of one places where this is done. * Make the wait state dependent on the process. This is difficult to emulate with ntsync and would require creating separate server objects for each process, hacking into duplicate_handle. * Fail a wait entirely in certain circumstances. This is pretty much impossible to emulate with in-process waits. Although ntsync has been in development for some time, I have regrettably failed to notice these problems until now. Fortunately, none of these behaviours happen on modern Windows. Although I/O on unbound handles delegates to the console of the current process, the signaled state does not. As the tests here show, the signaled state of a handle depends on the active console of the process in which the handle was created. If that console no longer exists, the signaled state is no longer updated [with one rather inexplicable exception]. Crucially, in current Windows waits never fail, and the state of an object is the same across all process which hold handles to it. Therefore this patch brings our behaviour to closer match current Windows. In theory these are fds and should use default_fd_signaled(). However, the points at which the handles are signaled are completely different, and I/O does not trigger console handles to become signaled when it normally would. Therefore for the time being I've kept the code using custom signaled ops. There is one other oddity related to consoles, which is the existence of console_add_queue(), which seeks to lazily create an input thread when a console is first waited on. This is one out of two places, after this patch, when the wait process is hijacked (the other being message queues). Fortunately this is easy to handle for in-process synchronization objects, by queueing the ioctl from the callback used to retrieve the in-process synchronization object itself. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7608

2 5

[PATCH v11 0/4] MR6855: server: Fix renaming file to a hardlink of itself.
by Gabriel Ivăncescu (＠insn) 18 Mar '25

18 Mar '25

-- v11: server: Handle hardlinks and casefolding when renaming the same file. server: Handle renames to destinations containing trailing slashes. kernel32/tests: Test renaming a file into a hardlink of itself. kernel32/tests: Use FindClose instead of CloseHandle when closing https://gitlab.winehq.org/wine/wine/-/merge_requests/6855

3 11

Re: [PATCH v5 0/1] MR7550: ntdll: Print name for TOKEN_INFORMATION_CLASS - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7550

1 0

Re: [PATCH v2 0/1] MR7398: winex11.drv: Don't add MWM_DECOR_BORDER to windows without a caption. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7398

1 0

Re: [PATCH v3 0/2] MR7601: bluetoothapis: Implement BluetoothGetDeviceInfo. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7601

1 0

Re: [PATCH v3 0/2] MR7595: comctl32/tests: Add a test for LVM_FINDITEM with LVS_OWNERDATA. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7595

1 0

Re: [PATCH v4 0/1] MR7594: msvcrt: Add _IONBF flag check in _filbuf to avoid dead loop in application which sets the flag. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7594

1 0

Re: [PATCH v2 0/6] MR7605: PDB rewrite #3 - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7605

1 0

Re: [PATCH v6 0/2] MR7290: kernelbase: Handle correctly paths with forward slashes in ReplaceFileW. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7290

1 0

Re: [PATCH v2 0/1] MR7603: kerberos: Fix imported target name leak. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7603

1 0

Re: [PATCH v2 0/1] MR7582: comctl32: Implement EVENT_OBJECT_VALUECHANGE for datetime control. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7582

1 0

Re: [PATCH v2 0/2] MR7602: comctl32/tests: Add a LVN_ENDLABELEDIT test with empty text. - approved
by Alexandre Julliard (＠julliard) 18 Mar '25

18 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7602

1 0

[PATCH v9 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v9: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses. https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

2 1

[PATCH v7 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v7: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses. https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

2 1

[PATCH v6 0/2] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v6: Fix format string try3 https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

2 2

[PATCH v5 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v5: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses. https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

4 6

[PATCH v3 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- v3: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses. https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

2 1

[PATCH 0/1] MR7607: win32u: Add stub for NtUserSetAdditionalForegroundBoostProcesses.
by Louis Lenders (＠xe) 18 Mar '25

18 Mar '25

Microsoft Edge worked if windows version is set to win8.1 (or win7) but now crashes in current wine due to this unimplemented function. This patch makes it work again in win8.1 mode; (It's still broken in win10 mode, see https://bugs.winehq.org/show_bug.cgi?id=56378 for more info) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7607

2 1

Re: [PATCH v7 0/15] MR7516: Server-side thread priorities implementation (Part 3)
by Rémi Bernon (＠rbernon) 18 Mar '25

18 Mar '25

Rémi Bernon (@rbernon) commented about include/winternl.h: > UINT64 InheritedFromUniqueProcessId; > } PROCESS_BASIC_INFORMATION64; > > -#define PROCESS_PRIOCLASS_IDLE 1 > -#define PROCESS_PRIOCLASS_NORMAL 2 > -#define PROCESS_PRIOCLASS_HIGH 3 > -#define PROCESS_PRIOCLASS_REALTIME 4 > -#define PROCESS_PRIOCLASS_BELOW_NORMAL 5 > -#define PROCESS_PRIOCLASS_ABOVE_NORMAL 6 > +#define PROCESS_PRIORITY_CLASS_UNKNOWN 0 > +#define PROCESS_PRIORITY_CLASS_IDLE 1 > +#define PROCESS_PRIORITY_CLASS_NORMAL 2 > +#define PROCESS_PRIORITY_CLASS_HIGH 3 > +#define PROCESS_PRIORITY_CLASS_REALTIME 4 > +#define PROCESS_PRIORITY_CLASS_BELOW_NORMAL 5 > +#define PROCESS_PRIORITY_CLASS_ABOVE_NORMAL 6 I haven't found neither the PROCESS_PRIOCLASS nor the PROCESS_PRIORITY_CLASS constants in the SDK headers, where are you seeing them? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7516#note_98242

1 0

Re: [PATCH v7 0/15] MR7516: Server-side thread priorities implementation (Part 3)
by Rémi Bernon (＠rbernon) 18 Mar '25

18 Mar '25

Rémi Bernon (@rbernon) commented about include/winnt.h: > #define THREAD_BASE_PRIORITY_MIN -2 > #define THREAD_BASE_PRIORITY_IDLE -15 > > +#define LOW_PRIORITY 0 /* Lowest thread priority level */ > +#define LOW_REALTIME_PRIORITY 16 /* Lowest realtime priority level */ > +#define HIGH_PRIORITY 31 /* Highest thread priority level */ > +#define MAXIMUM_PRIORITY 32 /* Number of thread priority levels */ > + This should be in `ddk/wdm.h`. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7516#note_98241

1 0

[PATCH 0/2] MR7606: user32/edit: Block key input when mouse input is captured.
by Nikolay Sivov (＠nsivov) 18 Mar '25

18 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56108 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7606

3 3

[PATCH v2 0/2] MR7595: comctl32/tests: Add a test for LVM_FINDITEM with LVS_OWNERDATA.
by Nikolay Sivov (＠nsivov) 18 Mar '25

18 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=57746 Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- v2: comctl32/listview: Use correct LVN_ODFINDITEM notification. comctl32/tests: Add a test for LVM_FINDITEM with LVS_OWNERDATA. https://gitlab.winehq.org/wine/wine/-/merge_requests/7595

3 3

[PATCH v2 0/5] MR7466: mf: Fix some issues with seeking.
by Conor McCarthy (＠cmccarthy) 18 Mar '25

18 Mar '25

This fixes failure to play the prologue video in Planet of the Apes: Last Frontier, which unpauses by calling Start() with a location. The exact state leading to this issue does not occur in the Start() tests, and it's not clear how to reproduce it reliably. -- v2: mf: Reset transform nodes when seeking. mf: Initialise the grabber sample count when setting state after a seek. mf: Reset transform node outputs when seeking. mf: Drop transform node input events when unpausing at a specific position. https://gitlab.winehq.org/wine/wine/-/merge_requests/7466

3 7

[PATCH v3 0/5] MR7559: cfgmgr32: Implement CM_Register_Notification and CM_Unregister_Notification.
by Vibhav Pant (＠vibhavp) 18 Mar '25

18 Mar '25

The methods are needed to implement Bluetooth device pairing in `bluetoothapis.dll` without having to create a dummy window for `RegisterDeviceNotification`, as that may break CLI applications. `CM_Register_Notification` is implemented as a thin wrapper over `I_ScRegisterDeviceNotification`, mapping `DBT_*` flags to their associated `CM_NOTIFY_*` constants. -- v3: user32: Remove incorrect FIXME warning while registering for DBT_DEVTYP_HANDLE notifications. cfgmgr32: Implement CM_Register_Notification and CM_Unregister_Notification. dinput/tests: Add tests for CM_Register_Notification. cfgmgr32/tests: Add basic tests for CM_(Un)Register_Notification. https://gitlab.winehq.org/wine/wine/-/merge_requests/7559

3 7

[PATCH 0/1] MR3862: user32/tests: Ignore other messages in test_broadcast.
by Esme Povirk (＠madewokherd) 18 Mar '25

18 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54194 If the SendMessageTimeout call takes a long time, we can get other messages which also set the observed wparam value. Apparently, this is especially likely on Windows 7. This also removes the (wParam == 0xbaadbeef) check which may have been intended to serve the same goal but doesn't work because the observed wParam value is still assigned. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3862

3 2

[PATCH v3 0/1] MR7344: winex11: Fix alpha blending in X11DRV_UpdateLayeredWindow.
by Dmitry Timoshkov (＠dmitry) 18 Mar '25

18 Mar '25

Original patch description https://www.winehq.org/pipermail/wine-patches/2017-February/157574.html winehq.org no longer provides the wine-patches archives, so it's necessary to dig into [web.archive.org](https://web.archive.org/web/20211011014926/https://www.win…: https://web.archive.org/web/20211011014926/https://www.winehq.org/pipermail… https://web.archive.org/web/20211011014930/https://www.winehq.org/pipermail… =================================================================== Based on a patch by Dmitry Timoshkov. ## Signed-off-by: Sebastian Lackner sebastian(a)fds-team.de Fixes https://bugs.winehq.org/show_bug.cgi?id=33943. Battle.Net initially creates completely transparent popup windows and afterwards uses UpdateLayeredWindow without hdcSrc to let them fade in. Testing confirms that this works fine on Windows - on Wine however the original content is lost and no longer available. To avoid that, this patch stores the original content in the surface and uses \_NET_WM_WINDOW_OPACITY to create a transparency effect. The same method is already used by SetLayeredWindowAttributes. If compositing is disabled, the window content will be visible immediately. Dmitrys original patch used a different approach, and manually implemented the alpha blending in x11drv_surface_flush using additional buffers, however this shouldn't really be necessary (yet). We can still use it later if there is any benefit compared to using \_NET_WM_WINDOW_OPACITY. -- v3: winex11: Fix alpha blending in X11DRV_UpdateLayeredWindow. https://gitlab.winehq.org/wine/wine/-/merge_requests/7344

3 9

[PATCH v3 0/1] MR7594: msvcrt: Add _IONBF flag check in _filbuf to avoid dead loop in application which sets the flag.
by Kun Yang (＠yangkun) 18 Mar '25

18 Mar '25

Signed-off-by: Kun Yang <yangkun(a)uniontech.com> Change-Id: Ia68e11ccb0aa563eeb73a2f5cb93afcd328e1937 [fgetws_test.7z](/uploads/7b706f2c4f639e4260c548d834658b32/fgetws_test.7z) If the application calls setvbuf(.., _IONBF, ..) before fgetws, wine sets the flag. In fgetws, this flag is not checked. read_i is called with paramenter count 1 which is not allowed in read_i implemantion. Function read_i returns -1 and the file stream is not actually read by program. The file pointer is not moved. If the application writes fgetws in a loop, it never reach EOF and falls in dead loop. By adding check for _IONBF in _filbuf, the application will run into another branch(file->_bufsiz=2). Thats's correct. Please checkout the example in my attachment. This application exits normally in Windows but falls in dead loop in wine. -- v3: msvcrt: Add MSVCRT__NOBUF flag check in _filbuf to avoid dead loop in application which sets the flag. https://gitlab.winehq.org/wine/wine/-/merge_requests/7594

3 2

Re: [PATCH v3 0/6] MR7530: win32u, server: Improve tracking mouse messages from pointer messages.
by Rémi Bernon (＠rbernon) 18 Mar '25

18 Mar '25

On Tue Mar 11 14:30:34 2025 +0000, Paul Gofman wrote: > I don't see how to avoid that in principle? > - the logic is triggered by calling DefWindowProc, so we can't avoid > some action from DefWindowProc; > - after WM_POINTERUPDATE with changing coordinates the messages should > be generated (and, separately, updating cursor position) regardless of > DefWindowProc calls; also generating that in DefWindowProc or > accept_hardware_message would be clumsy because > NtUserSendHardwareInput() can't generate the exact message we need, that > would need some custom interface (and won't avoid a server call to send > message anyway); > - we could in principle stop sending those `track_mouse_from_pointer` > calls once the tracking switched to server side (by returning the > corresponding flag from this call and tracking this part of state in > thread data), but we also want to update driver cursor position which > seems to cope nicely with this call, in that case it would need to be > done in some other more complicated ways. So reducing the amount of > those calls (by not sending them once server tracking started) is > possible but it will require these parts. I thought it doesn't add too > much of calls, do you think it worth it? > EDIT: updated. I don't know, this call feels inherently racy and I would like to give it more thought and time to check how windows behave. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7530#note_98225

1 0

[PATCH 0/6] MR7605: PDB rewrite #3
by eric pouech (＠epo) 18 Mar '25

18 Mar '25

This serie (PDB rewrite #3) mostly handles the debug information needed to support C++ enum underlying integral type. - it generalizes the use of VARIANT to gather enum values, - to make things simple, in PDB, the type of variant used has nothing to do with the size of the C++ integral type, but is the smallest size to hold the considered enum value, - adapt the dwarf backend to behave the same way as the PDB one, - adapt debugger to support up to 8-bytes variant types. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7605

2 6

[PATCH v7 0/1] MR6938: winewayland.drv: Implement support for xdg-toplevel-icon.
by Etaash Mathamsetty (＠etaash.mathamsetty) 18 Mar '25

18 Mar '25

-- v7: winewayland.drv: Implement support for xdg-toplevel-icon. https://gitlab.winehq.org/wine/wine/-/merge_requests/6938

3 2

[PATCH v5 0/2] MR7290: kernelbase: Handle correctly paths with forward slashes in ReplaceFileW.
by Roman Pišl (＠rpisl) 18 Mar '25

18 Mar '25

PathRemoveFileSpecW keeps only the drive if the path contains all forward slashes as shown in tests. But then the temporary file is created in the root folder which fails for drive Z:. This fixes a corner case when using CMake in Wine. Another usage of PathRemoveFileSpecW is in PathRelativePathToW which should be also checked. I'm leaving this as a draft until getting some feedback and adding tests. -- v5: kernelbase: Handle correctly paths with forward slashes in ReplaceFileW. kernel32: Test ReplaceFileW with forward slashes. https://gitlab.winehq.org/wine/wine/-/merge_requests/7290

2 2

[PATCH 0/1] MR7603: kerberos: Fix imported target name leak.
by Dmitry Timoshkov (＠dmitry) 18 Mar '25

18 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7603

3 2

[PATCH v2 0/4] MR7540: kernel32/tests: Add tests for maximum path length limitation.
by Yongjie Yao (＠yaoyongjie) 18 Mar '25

18 Mar '25

In CopyFileEx, and DeleteFile functions, by default, the file name and path are limited to MAX_PATH characters. To extend this limit to 32,767 wide characters, need prepend "\\\\?\\" to the path. -- v2: kernelbase: Limit the maximum path length for DeleteFile. kernelbase: Fix DeleteFileA doesn't support long path. kernelbase: Limit the maximum path length for filesystem. kernel32/tests: Add tests for maximum path length limitation. https://gitlab.winehq.org/wine/wine/-/merge_requests/7540

3 10

Re: [PATCH v3 0/1] MR5052: krnl386: Restore simulated real mode interrupts. - closed
by Alex Henrie (＠alexhenrie) 18 Mar '25

18 Mar '25

This merge request was closed by Alex Henrie. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5052

1 0

Re: [PATCH v3 0/1] MR5052: krnl386: Restore simulated real mode interrupts.
by Alex Henrie (＠alexhenrie) 18 Mar '25

18 Mar '25

On Mon Mar 17 22:03:12 2025 +0000, Alexandre Julliard wrote: > You should add a separate handler just for what the app needs, instead > of forcing all handlers to support real-mode pointers. > I added an example in 7e3bfd4e5e4f9ca689e1734da05a90b1b14351c9, it seems > to work here but it could be extended if necessary. I see what you mean now about having a separate handler. I still don't understand why it's better to have two implementations of the DOS "get volume serial number" ioctl, a more realistic implementation in [int21.c](https://gitlab.winehq.org/wine/wine/-/blob/6298b0cab2086ae61f46b28… and a less realistic implementation in [int31.c](https://gitlab.winehq.org/wine/wine/-/blob/6298b0cab2086ae61f46b28…, or why it's hard to support real-mode pointers (the original [CTX_SEG_OFF_TO_LIN](https://gitlab.winehq.org/wine/wine/-/commit/526b245237… macro seemed to work just fine). But the two-implementation solution works, so my unresolved questions are just curiosities at this point. Thank you very much for taking the time to fix Myst. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5052#note_98205

1 0

[PATCH v3 0/6] MR6866: Draft: ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code.
by Brendan Shanks (＠bshanks) 17 Mar '25

17 Mar '25

x86_64 Windows and macOS both use `%gs` to access thread-specific data (Windows TEB, macOS TSD). To date, Wine has worked around this conflict by filling the most important TEB fields (`0x30`/`Self`, `0x58`/`ThreadLocalStorage`) in the macOS TSD structure (Apple reserved the fields for our use). This was sufficient for most Windows apps. CrossOver's Wine had an additional hack to handle `0x60`/`ProcessEnvironmentBlock`, and binary patches for certain CEF binaries which directly accessed `0x8`/`StackBase`. Additionally, Apple's libd3dshared could activate a special mode in Rosetta 2 where code executing in certain regions would use the Windows TEB when accessing `%gs`. Now that the PE separation is complete, GSBASE can be swapped when entering/exiting PE code. This is done in the syscall dispatcher, unix-call dispatcher, and for user-mode callbacks. GSBASE also needs to be set to the macOS TSD when entering signal handlers (in `init_handler()`), and then restored to the Windows TEB when exiting (in `leave_handler()`). There is a special-case needed in `usr1_handler`: when inside a syscall (on the kernel stack), GSBASE may need to be reset to either the TEB or the TSD. The only way to tell is to determine what GSBASE was set to on entry to the signal handler. --- macOS does not have a public API for setting GSBASE, but the private `_thread_set_tsd_base()` works and was added in macOS 10.12. `_thread_set_tsd_base()` is a small thunk that sets `%esi`, `%eax`, and does the `syscall`: https://github.com/apple-oss-distributions/xnu/blob/8d741a5de7ff4191bf97d57…. The syscall instruction itself clobbers `%rcx` and `%r11`. I've tried to save as few registers as possible when calling `_thread_set_tsd_base()`, but there may be room for improvement there. --- I've tested this successfully on macOS 15 (Apple Silicon and Intel) with several apps and games, including the `cefclient.exe` CEF sample. I still need to test this patch on macOS 10.13, and I'd also like to do some performance testing. --- I also tested an alternate implementation strategy for this which took advantage of the expanded "full" thread state which is passed to signal handlers when a process has set a user LDT. The full thread state includes GSBASE, so GSBASE is set back to whatever is in the sigcontext on return (like every other field in the context). This would avoid needing to explicitly reset GSBASE in `leave_handler()`, and avoid the special-case in `usr1_handler()`. This strategy was simpler, but I'm not using it for 2 reasons: - the "full" thread state is only available starting with macOS 10.15, and we still support 10.13. - more crucially, Rosetta 2 doesn't seem to correctly implement the GS.base field of the full thread state. It's set to 0 on entry, and isn't read on exit. -- v3: ntdll: Remove x86_64 Mac-specific TEB access workarounds that are no longer needed. ntdll: On macOS x86_64, swap GSBASE between the TEB and macOS TSD when entering/leaving PE code. ntdll: Ensure init_handler runs in signal handlers before any compiler-generated memset calls. ntdll: Remove ugly fallback method for getting a thread's GSBASE on macOS. ntdll: Leave kernel stack before accessing %gs in x86_64 syscall dispatcher. ntdll: Do %gs accesses before switching to kernel stack in x86_64 syscall dispatcher. https://gitlab.winehq.org/wine/wine/-/merge_requests/6866

5 11

Re: [PATCH v2 0/5] MR7600: PDB rewrite #2 - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7600

1 0

Re: [PATCH v3 0/1] MR7510: msxml6/tests: Make test pass with Windows 7. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7510

1 0

Re: [PATCH v3 0/3] MR7583: ntdll: Stop using chdir() in nt_to_unix_file_name(). - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7583

1 0

Re: [PATCH v3 0/1] MR5052: krnl386: Restore simulated real mode interrupts.
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

On Mon Mar 17 04:39:02 2025 +0000, Alex Henrie wrote: > Now that simulated hardware interrupts for audio processing have been > implemented in commit ff0fbb1ff5f76d1d206aea452907ad2dd29206ea, > simulated real mode interrupts are the only thing left that's keeping > 16-bit Myst from running. When I wrote this patch, I tried to not > include anything more than what Myst requires. Could you please clarify > which parts of the patch are not strictly necessary? You should add a separate handler just for what the app needs, instead of forcing all handlers to support real-mode pointers. I added an example in 7e3bfd4e5e4f9ca689e1734da05a90b1b14351c9, it seems to work here but it could be extended if necessary. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5052#note_98176

1 0

Re: [PATCH v6 0/2] MR7579: ntdll: Return STATUS_ACCESS_VIOLATION if ret_len is not writable - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7579

1 0

Re: [PATCH v2 0/1] MR7593: ntoskrnl.exe/tests: Improve device properties test, avoid BSOD. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7593

1 0

Re: [PATCH v2 0/1] MR7598: bluetoothapis/tests: Use the correct file name while skipping tests when no radios are found. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7598

1 0

Re: [PATCH v3 0/8] MR7402: comctl32: Implement IAccessible methods. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7402

1 0

Re: [PATCH v4 0/7] MR7565: windows.system.profile.systemid: Implement ISystemIdentificationInfo::get_Source(). - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7565

1 0

Re: [PATCH v2 0/1] MR7596: wldap32: Also initialize idW.Flags field. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7596

1 0

Re: [PATCH v4 0/5] MR7584: gdiplus: Fix GdipFlattenPath and GdiAddPath*Curve drawing accuracy - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7584

1 0

Re: [PATCH v3 0/1] MR7589: gdiplus: Fix widening of LineCapArrowAnchor - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7589

1 0

Re: [PATCH v2 0/1] MR7592: d3dx9/tests: Skip some shader tests if we can't create a D3D object. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7592

1 0

Re: [PATCH v2 0/1] MR7585: d3dx9/tests: Enable tests for d3dx9_35.dll. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7585

1 0

Re: [PATCH v2 0/2] MR7587: msvcr120: Add imaxdiv implementation. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7587

1 0

Re: [PATCH v2 0/2] MR7588: kernelbase: Use ProcessImageFileNameWin32 in GetModuleFileNameExW. - approved
by Alexandre Julliard (＠julliard) 17 Mar '25

17 Mar '25

This merge request was approved by Alexandre Julliard. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7588

1 0

[PATCH 0/2] MR7602: comctl32/tests: Add a LVN_ENDLABELEDIT test with empty text.
by Nikolay Sivov (＠nsivov) 17 Mar '25

17 Mar '25

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7602

3 3

Re: [PATCH v6 0/1] MR6938: winewayland.drv: Implement support for xdg-toplevel-icon.
by Etaash Mathamsetty (＠etaash.mathamsetty) 17 Mar '25

17 Mar '25

I guess I can undraft it -- https://gitlab.winehq.org/wine/wine/-/merge_requests/6938#note_98128

1 0

[PATCH v2 0/1] MR7510: msxml6/tests: Make test pass with Windows 7.
by Bernhard Übelacker (＠bernhardu) 17 Mar '25

17 Mar '25

Currently test patterns show a systematic failure with Windows 7. https://test.winehq.org/data/patterns.html#msxml6:saxreader This is caused by the VisualBasic wrappers crashing when receiving some NULL parameters. Would this qualify for a version check like in a few other more central tests? If it would, is leaving the indentation that way ok? [This is a testbot run](https://testbot.winehq.org/JobDetails.pl?Key=151665) with this patch. -- v2: msxml6/tests: Make test pass with Windows 7. https://gitlab.winehq.org/wine/wine/-/merge_requests/7510

3 3

[PATCH v2 0/3] MR7583: ntdll: Stop using chdir() in nt_to_unix_file_name().
by Brendan Shanks (＠bshanks) 17 Mar '25

17 Mar '25

-- v2: ntdll: Stop using chdir() in nt_to_unix_file_name(). ntdll: Use *at() functions in get_dir_case_sensitivity(). https://gitlab.winehq.org/wine/wine/-/merge_requests/7583

2 4

Re: [PATCH v6 0/1] MR6938: Draft: winewayland.drv: Implement support for xdg-toplevel-icon.
by Alex Schwartz (＠alexschwartz01) 17 Mar '25

17 Mar '25

I just tested this on KDE Plasma 6.3 and it works perfectly This would be a great addition to the Wayland driver @etaash.mathamsetty is this still a draft? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/6938#note_98123

1 0

[PATCH v2 0/2] MR7601: bluetoothapis: Implement BluetoothGetDeviceInfo.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

-- v2: bluetoothapis/tests: Add tests for BluetoothGetDeviceInfo. https://gitlab.winehq.org/wine/wine/-/merge_requests/7601

2 2

[PATCH 0/2] MR7601: bluetoothapis: Implement BluetoothGetDeviceInfo.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7601

2 2

[PATCH 0/5] MR7600: PDB rewrite #2
by eric pouech (＠epo) 17 Mar '25

17 Mar '25

- fixes support for <mod>! syntax in source file enumeration - reduces memory usage for types (no longer having both a vector and a hash table as all types are now named) - some code refactoring -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7600

2 5

[PATCH 0/1] MR7592: d3dx9/tests: Skip some shader tests if we can't create a D3D object.
by Matteo Bruni (＠Mystral) 17 Mar '25

17 Mar '25

Hopefully it should help with some crashes on Rémi's test machines -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7592

2 2

[PATCH 0/1] MR7582: comctl32: Implement EVENT_OBJECT_VALUECHANGE for datetime control.
by Esme Povirk (＠madewokherd) 17 Mar '25

17 Mar '25

Windows implemented this event only for comctl32 v6. Making this behavior version-specific in Wine would be difficult, and I don't expect it to matter. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7582

3 2

[PATCH v10 0/4] MR6855: server: Fix renaming file to a hardlink of itself.
by Gabriel Ivăncescu (＠insn) 17 Mar '25

17 Mar '25

-- v10: server: Handle hardlinks and casefolding when renaming the same file. server: Handle renames to destinations containing trailing slashes. kernel32/tests: Test renaming a file into a hardlink of itself. kernel32/tests: Use FindClose instead of CloseHandle when closing https://gitlab.winehq.org/wine/wine/-/merge_requests/6855

3 8

[PATCH v2 0/8] MR7402: comctl32: Implement IAccessible methods.
by Esme Povirk (＠madewokherd) 17 Mar '25

17 Mar '25

-- v2: comctl32: Implement IOleWindow for SysLink. comctl32: Implement accChild for SysLink. comctl32: Implement get_accChildCount for SysLink. comctl32: Implement accLocation for SysLink. comctl32: Implement get_accDefaultAction for SysLink. comctl32: Implement acc_getName for SysLink. comctl32: Implement get_accState for SysLink controls. comctl32: Include only link items as IAccessible children. https://gitlab.winehq.org/wine/wine/-/merge_requests/7402

3 9

Re: [PATCH v2 0/1] MR7594: msvcrt: Add _IONBF flag check in _filbuf to avoid dead loop in application which sets the flag.
by Piotr Caban (＠piotr) 17 Mar '25

17 Mar '25

The patch generally looks good. It needs some changes for ucrtbase (see comment above). Thanks. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7594#note_98087

1 0

Re: [PATCH v2 0/1] MR7594: msvcrt: Add _IONBF flag check in _filbuf to avoid dead loop in application which sets the flag.
by Piotr Caban (＠piotr) 17 Mar '25

17 Mar '25

Piotr Caban (@piotr) commented about dlls/msvcrt/file.c: > return EOF; > } > > - if(!(file->_flag & (_IOMYBUF | MSVCRT__USERBUF))) { > + if(!(file->_flag & (_IONBF | _IOMYBUF | MSVCRT__USERBUF))) { ```suggestion:-0+0 if(!(file->_flag & (MSVCRT__NOBUF | _IOMYBUF | MSVCRT__USERBUF))) { ``` -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7594#note_98086

1 0

[PATCH v2 0/5] MR7559: cfgmgr32: Implement CM_Register_Notification and CM_Unregister_Notification.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

The methods are needed to implement Bluetooth device pairing in `bluetoothapis.dll` without having to create a dummy window for `RegisterDeviceNotification`, as that may break CLI applications. `CM_Register_Notification` is implemented as a thin wrapper over `I_ScRegisterDeviceNotification`, mapping `DBT_*` flags to their associated `CM_NOTIFY_*` constants. -- v2: user32: Remove incorrect FIXME warning while registering for DBT_DEVTYP_HANDLE notifications. cfgmgr32: Implement CM_Register_Notification and CM_Unregister_Notification. dinput/tests: Add tests for CM_Register_Notification. cfgmgr32/tests: Add basic tests for CM_(Un)Register_Notification. https://gitlab.winehq.org/wine/wine/-/merge_requests/7559

2 7

Re: [PATCH v10 0/7] MR7472: winebth.sys part 6: Add support for device inquiry/scanning.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

On Mon Mar 17 12:34:11 2025 +0000, Rémi Bernon wrote: > This broke some test validation, see > https://test.winehq.org/data/patterns.html#bluetoothapis:device. The > issue is about using skip from a different file than the test module > being executed. You need to either add a "subtest" call with the other > file name, or pass __FILE__ from the executing test module to the skip_ call. Sorry about that, I have created !7598 to fix this. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7472#note_98080

1 0

[PATCH 0/1] MR7598: bluetoothapis/tests: Use the correct file name while skipping tests when no radios are found.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7598

2 1

[PATCH v3 0/7] MR7565: windows.system.profile.systemid: Implement ISystemIdentificationInfo::get_Source().
by Mohamad Al-Jaf (＠maljaf) 17 Mar '25

17 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=55759 -- v3: windows.system.profile.systemid: Implement ISystemIdentificationInfo::get_Source(). windows.system.profile.systemid/tests: Add ISystemIdentificationInfo::get_Source() tests. windows.system.profile.systemid: Partially implement ISystemIdentificationStatics::GetSystemIdForPublisher(). https://gitlab.winehq.org/wine/wine/-/merge_requests/7565

3 8

[PATCH 0/4] MR7559: cfgmgr32: Implement CM_Register_Notification and CM_Unregister_Notification.
by Vibhav Pant (＠vibhavp) 17 Mar '25

17 Mar '25

The methods are needed to implement Bluetooth device pairing in `bluetoothapis.dll` without having to create a dummy window for `RegisterDeviceNotification`, as that may break CLI applications. `CM_Register_Notification` is implemented as a thin wrapper over `I_ScRegisterDeviceNotification`, mapping `DBT_*` flags to their associated `CM_NOTIFY_*` constants. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7559

3 8

[PATCH 0/8] MR7597: wintypes: Implement IBufferByteAccess::Buffer().
by Mohamad Al-Jaf (＠maljaf) 17 Mar '25

17 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=55759 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7597

3 10

Re: [PATCH v10 0/7] MR7472: winebth.sys part 6: Add support for device inquiry/scanning.
by Rémi Bernon (＠rbernon) 17 Mar '25

17 Mar '25

This broke some test validation, see https://test.winehq.org/data/patterns.html#bluetoothapis:device. The issue is about using skip from a different file than the test module being executed. You need to either add a "subtest" call with the other file name, or pass __FILE__ from the executing test module to the skip_ call. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7472#note_98062

1 0

[PATCH 0/1] MR7596: wldap32: Also initialize idW.Flags field.
by Dmitry Timoshkov (＠dmitry) 17 Mar '25

17 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7596

3 2

[PATCH 0/1] MR7533: gdi32: Avoid failing FillPath() for a EMF hdc.
by Dmitry Timoshkov (＠dmitry) 17 Mar '25

17 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7533

4 7

[PATCH 0/1] MR7595: comctl32/tests: Add a test for LVM_FINDITEM with LVS_OWNERDATA.
by Nikolay Sivov (＠nsivov) 17 Mar '25

17 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=57746 Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7595

2 1

Re: [PATCH v3 0/1] MR5052: krnl386: Restore simulated real mode interrupts.
by Alex Henrie (＠alexhenrie) 17 Mar '25

17 Mar '25

On Mon Mar 17 04:39:03 2025 +0000, Alexandre Julliard wrote: > Please let's not reintroduce support for vm86 contexts and pointers. > If necessary, you can add a separate real-mode handler for that specific > call. But I'd suggest to first make sure that the app works without any > actual vm86 support. Now that simulated hardware interrupts for audio processing have been implemented in commit ff0fbb1ff5f76d1d206aea452907ad2dd29206ea, simulated real mode interrupts are the only thing left that's keeping 16-bit Myst from running. When I wrote this patch, I tried to not include anything more than what Myst requires. Could you please clarify which parts of the patch are not strictly necessary? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5052#note_98044

1 0

[PATCH 0/1] MR7594: msvcrt: Add _IONBF flag check in _filbuf to avoid dead loop in application which sets the flag.
by Kun Yang (＠yangkun) 17 Mar '25

17 Mar '25

Signed-off-by: Kun Yang <yangkun(a)uniontech.com> Change-Id: Ia68e11ccb0aa563eeb73a2f5cb93afcd328e1937 [fgetws_test.7z](/uploads/7b706f2c4f639e4260c548d834658b32/fgetws_test.7z) If the application calls setvbuf(.., _IONBF, ..) before fgetws, wine sets the flag. In fgetws, this flag is not checked. read_i is called with paramenter count 1 which is not allowed in read_i implemantion. Function read_i returns -1 and the file stream is not actually read by program. The file pointer is not moved. If the application writes fgetws in a loop, it never reach EOF and falls in dead loop. By adding check for _IONBF in _filbuf, the application will run into another branch(file->_bufsiz=2). Thats's correct. Please checkout the example in my attachment. This application exits normally in Windows but falls in dead loop in wine. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7594

2 1

[PATCH 0/1] MR7593: ntoskrnl.exe/tests: Improve device properties test, avoid BSOD.
by Alexander Morozov (＠amorozov) 17 Mar '25

17 Mar '25

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=57940 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7593

2 1

[PATCH 0/1] MR3857: d3d11/tests: Properly mark the indexed sample position test as succeeding with the Vulkan backend.
by Henri Verbeet (＠hverbeet) 16 Mar '25

16 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/3857

4 4

[PATCH v2 0/1] MR7589: gdiplus: Fix widening of LineCapArrowAnchor
by Bartosz Kosiorek (＠gang65) 16 Mar '25

16 Mar '25

-- v2: gdiplus: Fix widening of LineCapArrowAnchor https://gitlab.winehq.org/wine/wine/-/merge_requests/7589

3 2

[PATCH v3 0/5] MR7584: gdiplus: Fix GdipFlattenPath and GdiAddPath*Curve drawing accuracy
by Bartosz Kosiorek (＠gang65) 16 Mar '25

16 Mar '25

With this PR, the GdipFlattenPath and GdiAddPath*Curve are drawing the same way as the native gdiplus -- v3: gdiplus: Fix GdipFlattenPath return path precision gdiplus/test: Add GdipFlattenPath tests with default flatness 0.25 https://gitlab.winehq.org/wine/wine/-/merge_requests/7584

3 8

[PATCH v9 0/4] MR6855: server: Fix renaming file to a hardlink of itself.
by Gabriel Ivăncescu (＠insn) 16 Mar '25

16 Mar '25

-- v9: server: Handle hardlinks and casefolding when renaming the same file. server: Handle renames to destinations containing trailing slashes. kernel32/tests: Test renaming a file into a hardlink of itself. kernel32/tests: Use FindClose instead of CloseHandle when closing https://gitlab.winehq.org/wine/wine/-/merge_requests/6855

3 8

[PATCH 0/1] MR7585: d3dx9/tests: Enable tests for d3dx9_35.dll.
by Nikolay Sivov (＠nsivov) 16 Mar '25

16 Mar '25

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7585

3 2

[PATCH 0/3] MR7577: d3dx9: Implement D3DXSaveVolumeToFile{A,W,InMemory}().
by Connor McAdams (＠cmcadams) 16 Mar '25

16 Mar '25

After this there is one more set of patches to finish up the file saving functions, I split it into two to hopefully make it easier to review. :) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7577

3 14

[PATCH v3 0/2] MR3833: iphlpapi/tests: Quadruple the buffer size in a GetAdaptersAddresses() test.
by Hans Leidekker (＠hans) 16 Mar '25

16 Mar '25

As shown by the testbot, doubling is not always sufficient. -- v3: iphlpapi/tests: Call GetExtendedTcp/UdpTable() in a loop. https://gitlab.winehq.org/wine/wine/-/merge_requests/3833

3 4

[PATCH v2 0/2] MR3833: iphlpapi/tests: Quadruple the buffer size in a GetAdaptersAddresses() test.
by Hans Leidekker (＠hans) 16 Mar '25

16 Mar '25

As shown by the testbot, doubling is not always sufficient. -- v2: iphlpapi/tests: Call GetExtendedTcp/UdpTable() in a loop. iphlpapi/tests: Call GetAdaptersAddresses() in a loop. https://gitlab.winehq.org/wine/wine/-/merge_requests/3833

4 6

[PATCH 0/1] MR3838: kernel32/tests: Allow a 0 removed entry count in the completion test.
by Francois Gouget (＠fgouget) 16 Mar '25

16 Mar '25

This matches the ntdll:file fix. Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=54554 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3838

3 2

[PATCH 0/1] MR7589: gdiplus: Fix widening of LineCapArrowAnchor
by Bartosz Kosiorek (＠gang65) 16 Mar '25

16 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/7589

2 1

[PATCH 0/2] MR7588: kernelbase: Use ProcessImageFileNameWin32 in GetModuleFileNameExW.
by Jacek Caban (＠jacek) 16 Mar '25

16 Mar '25

Allows retrieving the process image name even if the process has not yet completed loader initialization. Fixes LLDB versions newer than 15. LLVM-Issue: https://github.com/llvm/llvm-project/issues/131319 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7588

2 2

[PATCH 0/1] MR3837: quartz/tests: Fix segfault in changed3ddevice test.
by Alfred Agrell (＠Alcaro) 16 Mar '25

16 Mar '25

Fixes regression from https://gitlab.winehq.org/wine/wine/-/merge_requests/3792 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3837

4 4

[PATCH 0/1] MR3835: winscard/tests: Skip tests when SCardListReadersA() returns an empty list.
by Hans Leidekker (＠hans) 16 Mar '25

16 Mar '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/3835

3 2

2025

2024

2023

2022

wine-gitlab