April 2025 - wine-gitlab

[PATCH v2 0/1] MR6288: wintypes: Partially implement RoResolveNamespace and add 'Windows.winmd'
by Aaron Hill (＠Aaron1011) 19 Jun '25

19 Jun '25

The Visual Studio 2022 installer needs to be able to lookup a valid WinMD file for the 'Windows.Networking.Connectivity' namespace in order for the installation to start. A normal Windows installation contains lots of individual '.winmd' files (e.g. 'Windows.Networking.winmd'), which we cannot redistribute. Fortunately, Microsoft provides an open-source project called 'windows-rs', which contains a combined 'Windows.winmd' file that is explicitly MIT-licensed: https://github.com/microsoft/windows-rs/blob/master/crates/libs/bindgen/def… Make RoResolveNamespace resolve anything starting with 'Windows' by just pointing to this giant file. A full implementation will require us to actually walk the 'WinMetadata' directory (if the caller provides a custom path) and parse the matched '.winmd' file. However, this implementation is good enough for Visual Studio 2022 To actually run the vs2022 installer, a patch for the SHELL_execute trailing-space bug https://bugs.winehq.org/show_bug.cgi?id=56940 is needed: https://gitlab.winehq.org/wine/wine/-/merge_requests/6079 -- v2: wintypes: Partially implement RoResolveNamespace and add 'Windows.winmd' https://gitlab.winehq.org/wine/wine/-/merge_requests/6288

7 6

[PATCH v19 0/5] MR820: Implement application hives
by Santino Mazza (＠tati1454) 31 May '25

31 May '25

-- v19: kernelbase: Implement RegLoadAppKey. ntdll: Implement REG_APP_HIVE flag server: Create \REGISTRY\A hive. ntdll: Rename NtLoadKeyEx parameters. ntdll: Pass filename to server in NtLoadKeyEx. https://gitlab.winehq.org/wine/wine/-/merge_requests/820

5 15

[PATCH v6 0/2] MR4812: user32: Fix the number of characters processed by DrawTextExW.
by JiangYi Chen (＠meshine) 29 May '25

29 May '25

Description: When flags does not include DT_CALCRECT, since len is calculated in the middle, it will be reduced to zero. Resulting in the length of the processed string that is finally returned to zero and the non-processing string length is unchanged. But some application taking the non-processing string length to zero as the loop end condition. Log: Signed-off-by: chenjiangyi <chenjiangyi(a)uniontech.com> Change-Id: Icc0f250f5f4faba1bee8326fc911a4fc9cd7c012 -- v6: user32: Fix the number of characters processed by DrawTextExW. user32/tests: Add tests for DrawTextExW. https://gitlab.winehq.org/wine/wine/-/merge_requests/4812

3 5

[PATCH v6 0/29] MR7226: ntdll: Implement in-process synchronization via the Linux "ntsync" driver.
by Elizabeth Figura (＠zfigura) 28 May '25

28 May '25

This introduces a faster implementation of signal and wait operations on NT events, semaphores, and mutexes, which improves performance to native levels for a wide variety of games and other applications. The goal here is similar to the long-standing out-of-tree "esync" and "fsync" patch sets, but without the flaws that make those patch sets not upstreamable. The Linux "ntsync" driver is not currently released. It has been accepted into the trunk Linux tree for 6.14, so barring any extraordinary circumstances, the API is frozen and it will be released in its current form in about 2 months. Since it has passed all relevant reviewers on the kernel side, and the API is all but released, it seems there is no reason any more not to submit the Wine side to match. Some important notes: * This patch set does *not* include any way to disable ntsync support, since that kind of configuration often seems to be dispreferred where not necessary. In essence, ntsync should just work everywhere. Probably the easiest way to effectively disable ntsync, for the purposes of testing, is to chmod the /dev/ntsync device to prevent its being opened. Regardless, a Wine switch to disable ntsync can be added simply enough. Note that it should probably not take the form of a registry key, however, since it needs to be easily accessible from the server itself. * It is, generally speaking, not possible for only some objects, or some processes, to have backing ntsync objects, while others use the old server path. The esync/fsync patch sets explicitly protected against this by making sure every process had a consistent view of whether esync was enabled. This is not provided here, since no switch is provided to toggle ntsync, and it should not be possible to get into such an inconsistent state without gross misconfiguration. * Similarly, no diagnostic messages are provided to note that ntsync is in use, or not in use. These messages are part of esync/fsync, as well as part of ntsync "testing" trees unofficially distributed. However, if ntsync is working correctly, no message should be necessary. The basic structure is: * Each type of server object which can be waited on by the client (including events, semaphores, mutexes, but also other types such as processes, files) must store an "inproc_sync" object. This "inproc_sync" object is a full server object (note that this differs from esync/fsync). A vector and server request is introduced to retrieve an NT handle to this object from an arbitrary NT handle. Since the actual ntsync objects are simply distinct file descriptions, we then call get_handle_fd from the client to retrieve an fd to the object, and then perform ioctls on it. * Objects signaled by the server (processes, files, etc) perform ntsync ioctls on that object. The backing object in all such cases is simply an event. * Signal and wait operations on the client side attempt to defer to an "inproc_\*" function, falling back to the server implementation if it returns STATUS_NOT_IMPLEMENTED. This mirrors how in-process synchronization objects (critical sections, SRW locks, etc) used to be implemented—attempting to use an architecture-specific "fast_\*" function and falling back if it returned STATUS_NOT_IMPLEMENTED. * The inproc_sync handles, once retrieved, are cached per-process. This caching takes a similar form to the fd cache. It does not reuse the same infrastructure, however. The primary reason for this is that the fd cache is designed to fit within a 64-bit value and uses 64-bit atomic operations to ensure consistency. However, we need to store more than 64 bits of information. [We also need to modify them after caching, in order to correctly implement handle closing—see below.] The secondary reason is that retrieving the ntsync fd from the inproc_sync handle itself uses the fd cache. * In order to keep the Linux driver simple, it does not implement access flags (EVENT_MODIFY_STATE etc.) Instead, the flags are cached locally and validated there. This too mirrors the fd cache. Note that this means that a malicious process can now modify objects it should not be able modify—which is less true than it is with wineserver—but this is no different from the way other objects (notably fds) are handled, and would require manual syscalls. * In order to achieve correct behaviour related to closing objects while they are used, this patch set essentially relies on refcounting. This is broadly true of the server as well, but because we need to avoid server calls when performing object operations, significantly more care must be taken. In particular, because waits need to be interruptable by signals and then be restarted, we need the backing ntsync object to remain valid until all users are done with it. On a process level, this is achieved by letting multiple processes own handles to the underlying inproc_sync server object. On a thread level, multiple simultaneous calls need to refcount the process's local handle. This refcount is stored in the sync object cache. When it reaches zero, the cache is cleared. Punting this behaviour to the Linux driver would have introduced a great deal more complexity, which is best kept in userspace and out of the kernel. * The cache is, as such, treated as a cache. The penultimate commit, which introduces client support but does not yet cache the objects, effectively illustrates this by never actually caching anything, and retrieving a new NT handle and fd every time. * Certain waits, on internal handles (such as async, startup_info, completion), are delegated to the server even when ntsync is used. Those server objects do not create an underlying ntsync object. -- v6: ntdll: Cache in-process synchronization objects. ntdll: Use server_wait_for_object() when waiting on only the queue object. ntdll: Use in-process synchronization objects. ntdll: Introduce a helper to wait on an internal server handle. server: Allow creating an event object for client-side user APC signaling. server: Introduce select_inproc_queue and unselect_inproc_queue requests. server: Add a request to retrieve the in-process synchronization object from a handle. server: Create in-process synchronization objects for fd-based objects. server: Create in-process synchronization objects for timers. server: Create in-process synchronization objects for threads. server: Create in-process synchronization objects for message queues. server: Create in-process synchronization objects for jobs. server: Create in-process synchronization objects for processes. server: Create in-process synchronization objects for keyed events. server: Create in-process synchronization objects for device managers. server: Create in-process synchronization objects for debug objects. server: Create in-process synchronization objects for console outputs. server: Create in-process synchronization objects for console inputs. server: Create in-process synchronization objects for screen buffers. server: Create in-process synchronization objects for console servers. server: Create in-process synchronization objects for consoles. server: Create in-process synchronization objects for completion ports. server: Create in-process synchronization objects for mutexes. server: Create in-process synchronization objects for semaphores. server: Create in-process synchronization objects for events. server: Add an object operation to retrieve an in-process synchronization object. ntdll: Retrieve and cache an ntsync device in wait calls. 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/7226

7 10

[PATCH v2 0/1] MR7360: mshtml: Try to guess the charset on streams with no BOM.
by Gabriel Ivăncescu (＠insn) 28 May '25

28 May '25

-- v2: mshtml: Try to guess the charset on streams with no BOM. https://gitlab.winehq.org/wine/wine/-/merge_requests/7360

4 6

[PATCH v2 0/3] MR7748: ucrtbase: Add cexp() implementation.
by Daniel Lehman (＠dlehman25) 27 May '25

27 May '25

will have followup with cexpf and others -- v2: msvcr120: Add cexp() implementation. libs/musl: Pass math_error callback to exp. msvcr120/tests: Test exp. https://gitlab.winehq.org/wine/wine/-/merge_requests/7748

3 7

[PATCH 0/1] MR7803: kernel32: Add export for CreateFileMapping2.
by Etaash Mathamsetty (＠etaash.mathamsetty) 24 May '25

24 May '25

Rainbow Six Siege checks if the export exists but then doesn't actually use it -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7803

6 8

[PATCH 0/2] MR7938: Draft: winemac: Use IOSurface to update native window layers instead of CGImage.
by Marc-Aurel Zent (＠mzent) 22 May '25

22 May '25

This consolidates the two separate `CGImage`s (`colorImage` and `shapeImage`) previously used in `WineContentView` into a single `IOSurface` that also holds the window mask information in its alpha channel now. The alpha channel is now being updated in `macdrv_surface_flush`, whenever the shape changes (and stores that change for the next update, to also keep the front buffer in sync). This reduces the delay from setting a `CGImage` to a layer from 50ms to about 1.5ms in the `IOSurface` case (also accounting for `vImageSelectChannels_ARGB8888` operation). Based on a patch by @bshanks. I am already putting it out for feedback, but I will so some more testing to see how the performance impact of this is and do more exhaustive testing with shaped windows. Ideally the double buffer solution can be avoided as well... Also the `HBITMAP` for the windows is currently backed by a `CGDataProviderRef` and could also become an `IOSurface` I believe. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7938

3 11

[PATCH v6 0/8] MR7466: mf: Fix some issues with seeking.
by Conor McCarthy (＠cmccarthy) 22 May '25

22 May '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. -- v6: mf: Reset transform nodes when seeking. mf: Initialise the grabber sample count when setting state after a seek. mf: Set pending requests for transform outputs to zero on flush. mf: Reset transform node outputs when seeking. mf: Flush nodes before restarting a paused source at a specific position. mf: Drop transform node input events when unpausing at a specific position. mf: Reset transforms before restarting a paused source instead of after. mf: Move transform node reset to a helper function. https://gitlab.winehq.org/wine/wine/-/merge_requests/7466

5 13

[PATCH v11 0/1] MR4809: Draft: wip! winewayland.drv: Try to support smooth scroll events
by Ray Strode (＠halfline) 21 May '25

21 May '25

Wayland has 3 types of scrolling events: - axis. Used for e.g., touchpad 2 finger smooth scrolling - axis_discrete. Used for mouse scroll wheels (i.e., notches) - axis_value120. Used for high resolution input devices Wine currently only supports axis_discrete, meaning that 2 finger scroll events get ignored. This commit tries to add basic support for axis scrolling events, by translating the smooth motion in scroll increments using some primitive assumptions about line height and number of lines to scroll. -- v11: wip! winewayland.drv: Try to support smooth scroll events https://gitlab.winehq.org/wine/wine/-/merge_requests/4809

6 13

2025

2024

2023

2022

wine-gitlab April 2025