August 2026 - Wine-Bugs - list.winehq.org

[Bug 60241] Claude Core review report(a lot of copy paste and logic bugs uncovered)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60241 --- Comment #22 from Stian Low <wineryyyyy(a)gmail.com> --- (In reply to Olivier F. R. Dierick from comment #17) > I have an atypical personality. I'm not like you, any of you. Decades of career across industries has proved such atypical is quite typical especially among tech/dev workers. Your alternative perspectives/opinions are appreciated. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60241] Claude Core review report(a lot of copy paste and logic bugs uncovered)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60241 Alexandre Julliard <julliard(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|gitlab-unknown |-unknown Severity|enhancement |normal Product|WineHQ Gitlab |Wine --- Comment #21 from Alexandre Julliard <julliard(a)winehq.org> --- Olivier, please stop messing with this bug, it's fine as it is. Please just move on. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60241] Claude Core review report(a lot of copy paste and logic bugs uncovered)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60241 --- Comment #20 from Stian Low <wineryyyyy(a)gmail.com> --- (In reply to Rafał Mużyło from comment #19) > This bug is a *wine* bug. The fact that a discussion about wine bug > wrangling policies has arisen while discussing bug's merits is irrelevant. > Perhaps making a copy of this bug to discuss those would be reasonable, but > changing bug component to an unrelated topic borders on vandalism. Whatever allows this to remain open and appear in folks search results when seeking worthy bugs to work is most relevant. Any other lesser relevant topics raised were in service of that most relevant goal. I was unaware any such code analysis tools like Coverty were commonplace part of workflows so the discussion has been informative. Maybe Coverty is used in a less official way because it seems to be missing from mention by the wiki. Resistance to this bug seems primarily caused by being a bug about bugs vs just one of them but auto creating one bug per report seems impractical vs addressing each more manually and vetting on case-by-case basis. If a tool like Coverty were made more official and part of Gitlab workflow then the report of bugs/findings may be available officially from Gitlab. Then it may be reasonable for someone to create a bug to resolve all Gitlab/Coverty or alt-tools findings. It may also be reasonable for folks to create separate bug reports per items as individually addressed and link them to that wider report/effort to resolve all findings. > Of course, I'm not a part of the team, so that's just my personal, twisted > opinion on the current state of things. Nor do I have trust/privileges to change Bugzilla despite always being driven in service of whatever optimizes folks Bugzilla search results. Luckily Bugzilla supports much of that effort without needing elevated privileges. Thanks to reporter for the help raising awareness and igniting a relevant conversation about how to manage such a report going forward. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60219] Ankh - Anniversary Edition: Some textures are rendered black

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60219 --- Comment #4 from Béla Gyebrószki <gyebro69(a)gmail.com> --- FlatOut: Ultimate Carnage appears to be another affected application of the same regression. The game starts normally and the intro videos and main menu work, but starting any race fails with: 'Fatal error: Couldn't load texture.' I reproduced this with both the Steam version and the Zoom Platform version, and with both Low and High graphics settings. The regression is caused by the same Wine commit as the Ankh issue. I also tested the proposed fix/merge request for this issue and it fixes FlatOut as well. https://store.steampowered.com/app/12360/FlatOut_Ultimate_Carnage_Collector… wine-11.16-156-g2ecc2f84b45 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60256] New: Wine application prevents X11 root window from receiving mouse button events

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60256 Bug ID: 60256 Summary: Wine application prevents X11 root window from receiving mouse button events Product: Wine Version: 11.16 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winex11.drv Assignee: wine-bugs(a)list.winehq.org Reporter: achurch+wine(a)achurch.org Target Milestone: --- Distribution: --- Since wine-11.13, when a Wine application window has the input focus, the X11 root window does not receive mouse button events, preventing the user from (for example) opening a root window menu. This appears to be the same issue as was reported for the Dolphin emulator (https://bugs.dolphin-emu.org/issues/10668) caused by XInput2 enforcing an implicit input grab when an application asks for XI_ButtonPress inputs, and was fixed there by switching to raw button events (https://github.com/dolphin-emu/dolphin/commit/323d09f47e902cbf2b504a7ec11b4…) I can trivially fix this in my case by changing XI_ButtonPress to XI_RawButtonPress in x11drv_xinput2_enable(), but I don't see any change directly relevant to button press handling in 11.12 -> 11.13 diffs, so I presume this is a preexisting issue exposed by changes to use XInput2 in more conditions (the behavior change for my environment occurred in commit a2ceea3e "winex11: Enable XI2 raw events when window is focused."). -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60241] Claude Core review report(a lot of copy paste and logic bugs uncovered)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60241 --- Comment #19 from Rafał Mużyło <galtgendo(a)o2.pl> --- (In reply to Olivier F. R. Dierick from comment #18) > If this is an enhancement request, it relates to the development workflow on > Gitlab. > Changing the product to reflect that. > Someone is trying to be passive-aggressive here and in a pretty silly way too. First part is a whatever, second on the other hand... This bug is a *wine* bug. The fact that a discussion about wine bug wrangling policies has arisen while discussing bug's merits is irrelevant. Perhaps making a copy of this bug to discuss those would be reasonable, but changing bug component to an unrelated topic borders on vandalism. Of course, I'm not a part of the team, so that's just my personal, twisted opinion on the current state of things. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60255] New: ntdll: NtQuerySystemInformation(SystemPerformanceInformation) causes unhandled page fault (execute access to NULL) on native Win64 app

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60255 Bug ID: 60255 Summary: ntdll: NtQuerySystemInformation(SystemPerformanceInformation) causes unhandled page fault (execute access to NULL) on native Win64 app Product: Wine Version: 11.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: blocker Priority: P2 Component: ntdll Assignee: wine-bugs(a)list.winehq.org Reporter: atayksll(a)gmail.com Target Milestone: --- Distribution: --- # Summary Unhandled page fault (jump to NULL instruction pointer) immediately after `fixme:ntdll:NtQuerySystemInformation info_class SYSTEM_PERFORMANCE_INFORMATION`, reproducible with SpaceEngine (beta 0.990.48.2034) on startup. Same crash occurs across three different Wine-derived runtimes. # Environment - Distro: Fedora Linux (KDE Plasma 6, Wayland) - Kernel: 7.1.10-200.fc44.x86_64 - GPU: AMD Radeon RX 5600 XT (Navi 10), Mesa 26.1.6 (git-ffa422e53d), RADV/radeonsi, ACO - CPU: Intel Core i7-10700F (16 threads) - RAM: 32 GB - Wine builds tested (all produce the identical crash): - wine-experimental.bleeding.edge.11.0.411656.20260810 (TkG Plain) - Wine 11.0 stable - Proton (added SpaceEngine.exe as a non-Steam game) - Application: SpaceEngine beta 0.990.48.2034 (native Win64 build, not DXVK/vkd3d — game uses Wine's built-in OpenGL translation directly) # Reproduction steps 1. Install SpaceEngine (Win64 beta build) into a 64-bit Wine prefix. 2. Launch `SpaceEngine.exe` from its `system` folder. 3. The launcher/splash appears for under a second, then the process crashes with no dialog. # Observed behavior With `WINEDEBUG=+loaddll` the process log shows normal startup: all core DLLs load, `SpaceEngine.exe`, `SDL2.dll`, `openvr_api.dll`, etc. all load successfully as `native`, OpenGL context creation succeeds, and the game's own log (`se.log`, included below) shows OpenGL init, SDL init, and SDL_mixer init all completing normally before entering its hardware-detection ("INITIALIZING LOADER") step. The crash consistently happens at this exact point, right after: ``` 00ec:trace:loaddll:build_module Loaded L"C:\windows\system32\dinput8.dll" at 00006FFFFDB70000: builtin 015c:fixme:avrt:AvSetMmThreadCharacteristicsW (L"Pro Audio",...): stub 00ec:fixme:ntdll:NtQuerySystemInformation info_class SYSTEM_PERFORMANCE_INFORMATION wine: Unhandled page fault on execute access to 0000000000000000 at address 0000000000000000 (thread 00ec), starting debugger... ``` No further trace lines appear between the `fixme` line and the fault, which suggests the fault happens inside ntdll's own handling of this info class rather than after control returns to the application. Register dump (identical register values and stack contents reproduced across separate runs and separate Wine builds): ``` rip:0000000000000000 rsp:000000000010f8e8 rbp:0000000000000000 rax:0000000000000000 rbx:0000000000000000 rcx:0000000000000000 rdx:0000000000000000 rsi:ffffffffffffffff rdi:0000000000000004 r8:00000000ffffffff r9:0000000000000000 r11:0000000000000000 r12:0000000000001000 r13:0000000140d8f2a8 r14:0000000000000003 r15:0000000000000001 Backtrace: =>0 0000000000000000 (0000000000000000) 0000000000000000: -- no code accessible -- ``` `r13` (0000000140d8f2a8) consistently points to a fixed static offset inside `spaceengine.exe`'s own address space (module base 0000000140000000), the same value across every crash, on every Wine build tested — behavior is fully deterministic. # se.log excerpt (SpaceEngine's own log, up to the last line printed before crash) ``` INITIALIZING LOADER [MT] Number of CPUs: 16 [MT] Worker threads: 14 [MT] Total RAM: 31997 Mb [MT] Available RAM: 26908 Mb [MT] Max SE RAM: 8192 Mb ``` No further lines are ever written to se.log after this point — the crash happens before the next planned log line. # What has been ruled out - GPU/Vulkan driver stack: `vulkaninfo` runs cleanly and correctly identifies the RX 5600 XT via RADV. - `GL_ARB_bindless_texture`: hiding it via `MESA_EXTENSION_OVERRIDE=-GL_ARB_bindless_texture` had no effect on the crash (ruling out the known AMD bindless-texture driver bug class as the cause here). - `openvr_api.dll`: this is a hard load-time import of the executable; removing it just fails earlier with `STATUS_DLL_NOT_FOUND` rather than avoiding the real crash, so VR init isn't the trigger. - Wine build/runner choice: identical crash (same register values, same crash address) on wine-experimental TkG bleeding-edge, wine-11.0 stable, and Proton. # Question / request Is `SystemPerformanceInformation` handling in `ntdll`'s `NtQuerySystemInformation` known to be incomplete or buggy on x86-64 Linux hosts in a way that could produce this kind of null-pointer-execute fault rather than a clean error return? Happy to provide the full debug logs, `se.log`, or test further patches/builds if that would help narrow this down — this crash is 100% reproducible on this hardware. generated via claude. it did all the work finding this bug while trying to fix space engine for me. sorry. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60252] fix wow64 type cast bug + use-after-free of the demuxer context in demuxer_destroy/demuxer_create

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60252 --- Comment #4 from Austin English <austinenglish(a)gmail.com> --- (In reply to Rafał Mużyło from comment #2) > (In reply to Ken Sharp from comment #1) > > Which Wine version? > > > > What app is causing this crash? > > Could we just call it 'current git head' and move on ? $ git describe works fine. Keep in mind that bugs are often read months/years later. It takes way less effort for the commenter to write the actual wine version than it does for someone to try to figure it out via git history after the fact. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 60252] fix wow64 type cast bug + use-after-free of the demuxer context in demuxer_destroy/demuxer_create

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=60252 Nikolay Sivov <bunglehead(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|winegstreamer |-unknown Version|unspecified |11.16 Status|UNCONFIRMED |NEW Ever confirmed|0 |1 --- Comment #3 from Nikolay Sivov <bunglehead(a)gmail.com> --- I've submitted the fix for that a day ago https://gitlab.winehq.org/wine/wine/-/merge_requests/11781 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026

[Bug 56681] 3d games continue receiving mouse input while not focused

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=56681 --- Comment #14 from Mel <pmargeti34(a)gmail.com> --- (In reply to Zhai Can from comment #13) > (In reply to Mel from comment #12) > > I've just posted a patch here: > > https://bugs.winehq.org/show_bug.cgi?id=59740#c2 > > > > It fixes this bug by fixing the root issue. I've tested it on Warframe > > successfully, camera stops spinning and cursor is no longer invisible over > > Warframe's surface when it doesn't have the focus. Please test with other > > affected software. > > Tested with 4D Miner and my winit demo. It fixes the issue. But on Hyprland > it doesn't work (app still not receiving focus loss on focus change to a > non-wine app) for unknown reason. (works for kwin_wayland, mutter, niri, > sway compositor but not for hyprland). might be worth opening a hyprland issue for this then, maybe they're doing something differently -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Aug. 29, 2026