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

[Bug 59333] VNyan leaks memory (race in __wine_syscall_dispatcher $rsp switching breaks Unity GC)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=59333 --- Comment #32 from Madalee <madalee(a)ohlmeyers.com> --- It is, indeed, hammering the CPU and running around 886 threads. The memory usage, however, didn't go up after leaving it running all night on GE-Proton10-10 (original, no leak patch) We know Unity has several different engine modes and stuff, maybe our project is in the wrong one? If you're willing, feel free to reach out to us on discord at "madalee" (In reply to Adalyn from comment #31) > if the script is applied correctly, it should be using at least 800 threads, > and basically every available CPU cycle the kernel will provide > all that should be necessary to use it is adding the script as a component > to any object in the scene -- 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.

Feb. 23, 2026

[Bug 59329] C++ catch() use-after-free with std::rethrow_exception()

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=59329 Piotr Pawłowski <p(a)perkele.cc> changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED --- Comment #9 from Piotr Pawłowski <p(a)perkele.cc> --- Fixed in 11.3 -- 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.

Feb. 23, 2026

[Bug 57484] VKB Joystick not detected in Wine 9.22 but works in 9.21

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=57484 mactan(a)edgetransit.dev <mactan(a)edgetransit.dev> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mactan(a)edgetransit.dev --- Comment #14 from mactan(a)edgetransit.dev <mactan(a)edgetransit.dev> --- with VKB NXTs, the device ID changes depending on what modules are connected, is there a reason the device IDs are specified in addition to the vendor ID? We have been comparing to the Virpil vendor which prefers hidraw for the entire vendor without specifying pid -- 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.

Feb. 23, 2026

[Bug 59333] VNyan leaks memory (race in __wine_syscall_dispatcher $rsp switching breaks Unity GC)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=59333 --- Comment #31 from Adalyn <adibtw(a)tuta.io> --- if the script is applied correctly, it should be using at least 800 threads, and basically every available CPU cycle the kernel will provide all that should be necessary to use it is adding the script as a component to any object in the scene -- 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.

Feb. 23, 2026

[Bug 59333] VNyan leaks memory (race in __wine_syscall_dispatcher $rsp switching breaks Unity GC)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=59333 --- Comment #30 from Madalee <madalee(a)ohlmeyers.com> --- Warudo has been running under this modified Proton 10 for 18 hours now and has not experienced "THE" memory leak. We say "THE" because it does seem like it's experiencing "A" memory leak. It's been, very slowly, increasing in memory use since around hour 2 or so. It's gone from 1.5G to 5.8G after 18 hours. Along with the increase in memory usage, the CPU usage has also climbed, with a setup that's basically just idling(not receiving VRM data or anything) the model stands and moves around in an idle pose. CPU was idling around 12% initially and it's up to an average of 17% now. The slowness of this leak and CPU increase could be on the Warudo side at this point. On a side nite, we tried running the Unity script for several hours, with normal Proton 10, and haven't seen the leak. We likely haven't setup the script properly. We don't develop in Unity, just use it enough to work with our VRM model so maybe we set it up wrong? -- 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.

Feb. 23, 2026

[Bug 31899] No keyboard input in La-Mulana remake (GetKeyState should behave similar to GetAsyncKeyState for specific window messages / queue states)

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=31899 Giacomo Orlandi <gia_(a)inwind.it> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gia_(a)inwind.it --- Comment #15 from Giacomo Orlandi <gia_(a)inwind.it> --- This was probably fixed upstream by the following commit: https://gitlab.winehq.org/wine/wine/-/commit/68b6e876f03acf5c7933d9b7a2d502… -- 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.

Feb. 23, 2026

[Bug 35907] Caps Lock state gets confused with multiple processes/threads

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=35907 Giacomo Orlandi <gia_(a)inwind.it> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gia_(a)inwind.it --- Comment #9 from Giacomo Orlandi <gia_(a)inwind.it> --- This was probably fixed by the following commit: https://gitlab.winehq.org/wine/wine/-/commit/68b6e876f03acf5c7933d9b7a2d502… -- 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.

Feb. 23, 2026

[Bug 45385] Wrong state of virtual keys after cycling windows. Usually VK_MENU, VK_SHIFT and VK_CONTROL, but every key can be affected.

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=45385 Giacomo Orlandi <gia_(a)inwind.it> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gia_(a)inwind.it --- Comment #3 from Giacomo Orlandi <gia_(a)inwind.it> --- This probably fixed by the following commit: https://gitlab.winehq.org/wine/wine/-/commit/68b6e876f03acf5c7933d9b7a2d502… -- 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.

Feb. 23, 2026

[Bug 26269] BioShock 2: Loss of keyboard input on loading screen

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=26269 Giacomo Orlandi <gia_(a)inwind.it> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gia_(a)inwind.it --- Comment #24 from Giacomo Orlandi <gia_(a)inwind.it> --- It looks like the patch attached in comment 23 was upstreamed about one year later, in the following commit: https://gitlab.winehq.org/wine/wine/-/commit/68b6e876f03acf5c7933d9b7a2d502… Is this fixed? By the way, the REOPENED status was set many years earlier. -- 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.

Feb. 23, 2026

[Bug 35331] Ellipse drawing problem using GM_ADVANCED graphics mode

by WineHQ Bugzilla

http://bugs.winehq.org/show_bug.cgi?id=35331 Giacomo Orlandi <gia_(a)inwind.it> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gia_(a)inwind.it --- Comment #13 from Giacomo Orlandi <gia_(a)inwind.it> --- The linked staging patchset was removed in the following commit: https://gitlab.winehq.org/wine/wine-staging/-/commit/683813d151ecef69835c2b… However, it does not appear that this issue has been fixed upstream. As far as I can tell, several related commits were added and later reverted, possibly due to regressions. The other bug addressed by this patchset (bug 34579) is still open as well. -- 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.

Feb. 23, 2026