[Bug 60046] New: Multiple 32-bit applications including wine builtins (winecfg / regedit / etc.) crash under wine-11.13.
http://bugs.winehq.org/show_bug.cgi?id=60046 Bug ID: 60046 Summary: Multiple 32-bit applications including wine builtins (winecfg / regedit / etc.) crash under wine-11.13. Product: Wine Version: 11.13 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@list.winehq.org Reporter: hibbsncc1701@gmail.com Target Milestone: --- Distribution: --- Created attachment 81505 --> http://bugs.winehq.org/attachment.cgi?id=81505 Console log of winecfg in a newly created prefix crashing after moving the mouse cursor out of the window As reported in bug #56565, 32-bit applications crash under wine-11.13 when processing certain window messages. Sometimes certain programs (winecfg, explorer) can be made to work without crashing by setting HKCU\Software\Wine\X11 Driver\UseEGL to "N". (Either by manually adding it to drive_c/user.reg, or by very carefully controlling input to avoid triggering window messages that cause the crash.) For other programs (CNC3 installer, regsrv32, etc.) the registry setting has no effect, and the apps crash instantly or whenever the bug decides to show itself. In a worse case scenario, winedbg can get stuck in a crash loop. (I.e. winedbg itself crashing, and thus spawning another copy of itself. Which will also crash.) Sometimes it appears on the screen and can be closed, other times it crashes before it has a chance to draw a window, but it always remains in the process list, and needs to be sigkill'd after wineserver itself is sigkill'd to break the loop. (Which then requires all other wine processes to be sigkill'd to finish cleaning up.) I believe the EGL issue is a separate one, and probably caused by my own system's NVIDIA card. (It's a Pascal card so it's on borrowed time.) As I've mentioned in bug #56565, wine's EGL support has never worked on my system regardless of NVIDIA driver source (distro, NVIDIA's cuda packages, or the dot run blobs), regardless of which window manager was used (MATE, XFCE, KDE, etc.), and regardless of wine prefix / app bitness. (64-bit or 32-bit.) Despite EGL working for other host applications on the system in question. Wine's EGL rendering backend always errors out in the console, but games and other wine apps seemed to ignore the failures and worked regardless until wine-11.13. As wine's window messages are clearly impacted by what rendering backend is being used, I'm including that info here for context. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Patrick Hibbs <hibbsncc1701@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Distribution|--- |Debian -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Jonas <winehq@jonass.user.lysator.liu.se> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |winehq@jonass.user.lysator. | |liu.se --- Comment #1 from Jonas <winehq@jonass.user.lysator.liu.se> --- Can you do bisecting / regression test as described on https://gitlab.winehq.org/wine/wine/-/wikis/Regression-Testing ? -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Patrick Hibbs <hibbsncc1701@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |f50b263562c002c61ab049baa12 | |12c3bd36994bb --- Comment #2 from Patrick Hibbs <hibbsncc1701@gmail.com> --- OK, on the first revision that git bisect had me test, the err messages for the msg debug channel disappeared. Yet, the segfaults remained. Which indicates that the commit that created those msg debug channel errors is in the second half of the commits between 11.12 and 11.13, while the commit that caused the segfaults are in the first half. I'm not sure if I should spin off the msg errors as a new bug or not. As for the segfaults, git bisect revealed: f50b263562c002c61ab049baa1212c3bd36994bb is the first bad commit commit f50b263562c002c61ab049baa1212c3bd36994bb Author: Rémi Bernon <rbernon@codeweavers.com> Date: Fri Jun 5 09:03:34 2026 +0200 server: Move window monitor DPI to the shared memory. dlls/win32u/sysparams.c | 40 ++++------------------------------------ dlls/win32u/win32u_private.h | 5 +++++ dlls/win32u/window.c | 28 ++++++++++++++++++++++++---- include/wine/server_protocol.h | 7 ++++--- server/protocol.def | 6 ++++-- server/request_handlers.h | 11 +++++------ server/request_trace.h | 1 - server/window.c | 37 ++++++++++++++++++++++++++----------- 8 files changed, 72 insertions(+), 63 deletions(-) -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |wineserver CC| |rbernon@codeweavers.com -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 --- Comment #3 from Rémi Bernon <rbernon@codeweavers.com> --- Created attachment 81636 --> http://bugs.winehq.org/attachment.cgi?id=81636 Possible fix I'm not able to reproduce any crash but I believe I have found some possible cause. Does this patch help? Please test the patch on top of current git HEAD, as there are other fixes that have been merged recently that could solve some other crashes. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 --- Comment #4 from Patrick Hibbs <hibbsncc1701@gmail.com> --- (In reply to Rémi Bernon from comment #3)
Created attachment 81636 [details] Possible fix
I'm not able to reproduce any crash but I believe I have found some possible cause. Does this patch help?
Please test the patch on top of current git HEAD, as there are other fixes that have been merged recently that could solve some other crashes.
Yes, that patch applied on top of a011ce57245c958445cddb938721333e2890fd55 seems to have fixed it. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Rémi Bernon <rbernon@codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Fixed by SHA1| |3ec14378590503ffbadfa936b99 | |4e5bf849fe967 Resolution|--- |FIXED --- Comment #5 from Rémi Bernon <rbernon@codeweavers.com> --- This should be fixed with 3ec14378590503ffbadfa936b994e5bf849fe967. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=60046 Alexandre Julliard <julliard@winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #6 from Alexandre Julliard <julliard@winehq.org> --- Closing bugs fixed in 11.15. -- 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.
participants (1)
-
WineHQ Bugzilla