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

[Bug 59193] New: Wine and MT4 don't work together very well
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59193 Bug ID: 59193 Summary: Wine and MT4 don't work together very well Product: Wine Version: 9.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)list.winehq.org Reporter: j-nws23(a)planet.nl Distribution: --- ✅ WINE Bug Report Author: ChatGPT Title: MT4 floating “Market Watch” window forces Wine topmost/focus-stealing behavior for all Wine windows Wine version: 9.0~repack-4build3 (Ubuntu package) Distribution: Linux Mint 22.2 Cinnamon 6.4.8 Muffin (X11 session) Application affected: MetaTrader 4 (MT4), build 1443 Also affects unrelated Wine applications (IrfanView, winecfg, etc.) Overview When MetaTrader 4's Market Watch panel is set to floating, MT4 triggers a Wine-wide focus and stacking problem: MT4’s main window and its floating windows become always-on-top Hovering the mouse over MT4 steals focus without clicking Other Wine applications (e.g., IrfanView, Winecfg) become topmost or behave incorrectly Docking the Market Watch window immediately restores normal behavior The issue does not occur in native Windows The bug appears to be caused by how MT4 requests window styles for the floating tool window, and how Wine interprets them. Steps to Reproduce Install MetaTrader 4 build 1443 in Wine 9.0 on Linux Mint 22.2 Cinnamon (X11). Start MT4. Open View → Market Watch. Right-click the Market Watch panel and select Docked → uncheck (floating mode). Move the floating Market Watch panel anywhere. Move the mouse over MT4 or over other Wine windows such as Winecfg or IrfanView. Actual Behavior MT4 forces itself to the top of the stacking order. MT4 grabs focus on mouse hover, without any click. Any other Wine window: may be pushed behind MT4 may adopt the same "tool window" styling may become unresponsive to normal focus rules Behavior persists until Market Watch is docked again. Expected Behavior MT4 floating tool windows should not force global focus-stealing. Other Wine applications should retain normal stacking and focus behavior. Floating Market Watch should behave like the floating Data Window, which does not cause any of these issues. Additional Notes The issue occurs even with a fresh Wine prefix. Both “direct rendering” and Mesa/i915 driver stack are normal. No compositor issues exist (Cinnamon compositor fully active). This behavior did not happen on Windows. MT4’s Data Window uses a different style and does not trigger the bug, suggesting MT4 applies the WS_EX_TOOLWINDOW or WS_EX_TOPMOST flag to Market Watch but Wine may be misinterpreting or forcing additional behaviors. Suggested Component Likely related to: win32u user32 window style handling for tool windows Workaround Docking the Market Watch window restores normal behavior and stops focus stealing. Impact This behavior makes MT4 nearly unusable under Wine with floating panels and affects unrelated Wine applications by corrupting stacking and focus behavior across the session. -- 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.

1 3

[Bug 59346] New: WineMine: lacking fallback option for "difficulty", "Width", ... reading in Registry leading to wrong menu check mark
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59346 Bug ID: 59346 Summary: WineMine: lacking fallback option for "difficulty", "Width", ... reading in Registry leading to wrong menu check mark Product: Wine Version: 11.0-rc4 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)list.winehq.org Reporter: kyle.kcsoftwares(a)gmail.com Distribution: --- Created attachment 80282 --> http://bugs.winehq.org/attachment.cgi?id=80282 Incorrect width for beginner Open Regedit and change "Difficulty" to an invalid value (eg: 99) Start Winmine Opens with last used row/col but does not set any difficulty (does not recognize from row/col and does not set custom) Guilty Code: SetDifficulty function where CheckMenuItem( hMenu, IDM_BEGINNER + difficulty, MF_CHECKED ); while p_board->difficulty is not checked for being in acceptable bounds : programs/winemine/main.c -- 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.

1 5

[Bug 39005] New: "http://xml.org/sax/features/namespace-prefixes" default value
by wine-bugs＠winehq.org Feb. 26, 2026

Feb. 26, 2026

https://bugs.winehq.org/show_bug.cgi?id=39005 Bug ID: 39005 Summary: "http://xml.org/sax/features/namespace-prefixes" default value Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: msxml3 Assignee: wine-bugs(a)winehq.org Reporter: ott(a)mirix.org Distribution: --- The specification is ambigous about this feature: "the "http://xml.org/sax/features/namespace-prefixes" feature is set to True. (It is False by default.)" (https://msdn.microsoft.com/en-us/library/ms757829%28v=vs.85%29.aspx) "only if the "http://xml.org/sax/features/namespace-prefixes" feature is set to True. This feature is set to False by default." (https://msdn.microsoft.com/en-us/library/ms762302%28v=vs.85%29.aspx) "The default value is true." (https://msdn.microsoft.com/en-us/library/ms759146%28v=vs.85%29.aspx) "namespace-prefixes | Boolean | True" (https://msdn.microsoft.com/en-us/library/ms762317%28v=vs.85%29.aspx) It should clarified from Microsoft's implementation whether the feature is enabled by default or not. Maybe the "namespace-prefixes" and "http://xml.org/sax/features/namespace-prefixes" are two different features? Perhaps it also makes sense notify Microsoft about the inconsistency. -- 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.

2 2

[Bug 59309] New: Wine regression, FreeArc external algorithm (cls-lolz_x64.exe) stopped working since wine 10.7
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59309 Bug ID: 59309 Summary: Wine regression, FreeArc external algorithm (cls-lolz_x64.exe) stopped working since wine 10.7 Product: Wine Version: 11.1 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: critical Priority: P2 Component: -unknown Assignee: wine-bugs(a)list.winehq.org Reporter: obscurexwing(a)disroot.org Distribution: --- //the actual problem The cls-lolz module acts as an external algorithm for FreeArc to handle high-ratio compression FreeArc with CLS-LOLZ to unpack data now just does nothing since wine 10.7, every version of wine after version 10.6 are exposed to this bug, including new 11.1. Expected behaviour (like on wine 10.6) : if program uses FreeArc with CLS-LOLZ, data unpacking process will spawn "cls-lolz_x64.exe" in task manager, and unpacking process will start Actual behaviour (10.7-11.1) : if program uses FreeArc with CLS-LOLZ, data unpacking will not happen, process "cls-lolz_x64.exe" will not appear in task manager, and files unpacking process completely stuck at this point //where you can find CLS-LOLZ ? For example, actual program placed here : https://github.com/CarldricGaming/Mini-Compressor 1.Download archive from github releases 2.Install program 3.Go to \%INSTALLPATH%\_MiniCompressor\Releases\Compression\FreeArc\LZ\ 4.Program that need to catch your attention placed here, with name "lolz_x64.exe" 5.Run with cmd through wine, like this : wine cmd C: cd _MiniCompressor\Releases\Compression\FreeArc\LZ\ lolz_x64.exe and you will get something like lolz v22c4b [Dec 30 2018 20:15:58] Settings: *this will output various launch options like -mt -bc -bm etc* -- 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.

1 6

[Bug 59272] New: COM ports resolved relative to working directory instead of dosdevices
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59272 Bug ID: 59272 Summary: COM ports resolved relative to working directory instead of dosdevices Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: kernel32 Assignee: wine-bugs(a)list.winehq.org Reporter: hilgert(a)mail.ru Distribution: --- Wine incorrectly resolves COM port paths relative to the current working directory instead of using the symlinks in ~/.wine/dosdevices/, causing CreateFile() calls for COM ports to fail when the application is launched from subdirectories. STEPS TO REPRODUCE: 1. Create a symlink in dosdevices: ln -s /dev/ttyUSB0 ~/.wine/dosdevices/com1 2. Create a test application that calls: CreateFile("COM1", GENERIC_READ | GENERIC_WRITE, 0, NULL, OPEN_EXISTING, 0, NULL); 3. Run the application from C:\: cd ~/.wine/drive_c/ wine test.exe Result: COM1 opens successfully via /dev/ttyUSB0 4. Run the application from a subdirectory: cd ~/.wine/drive_c/subdir/ wine test.exe Result: CreateFile fails with ERROR_FILE_NOT_FOUND EXPECTED BEHAVIOR: COM ports should always resolve through dosdevices symlinks regardless of working directory, matching Windows behavior where COM1 is a global device namespace. ACTUAL BEHAVIOR: Wine attempts to open "C:\subdir\COM1" as a regular file instead of resolving COM1 through dosdevices. DEBUG TRACE: trace:file:CreateFileW L"COM1" GENERIC_READ GENERIC_WRITE creation 3 attributes 0x0 trace:file:NtCreateFile name=L"\\??\\C:\\subdir\\COM1" warn:file:NtCreateFile L"\\??\\C:\\subdir\\COM1" not found (c0000034) WORKAROUND: Creating a symlink directly in the working directory allows the application to open the port, but this is not portable and breaks when changing directories: ln -s /dev/ttyUSB0 ~/.wine/drive_c/subdir/COM1 ADDITIONAL NOTES: This affects serial communication applications that rely on standard COM port naming conventions. The issue appears to be in how Wine's file path resolution handles device names. On real Windows, device names like COM1, LPT1, etc. are resolved through the device namespace regardless of the current directory. -- 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.

1 7

[Bug 59390] New: BoyScout GBA Software synthesizer hangs on boot
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59390 Bug ID: 59390 Summary: BoyScout GBA Software synthesizer hangs on boot Product: Wine Version: 11.1 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)list.winehq.org Reporter: sterophonick(a)gmail.com Distribution: --- Created attachment 80349 --> http://bugs.winehq.org/attachment.cgi?id=80349 Log that hangs for several seconds before being manually killed. +timestamp,+pid,+tid,+seh,+unwind,+threadname,+debugstr,+loaddll,+mscoree BoyScout is a tracker for emulating the Game Boy Advance chiptune capabilities and exporting them for use in homebrew. It requires MFC42, but once this is installed, it hangs forever, never reaching the main interface. Download: https://gbadev.org/tools.php?showinfo=187 -- 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.

1 3

[Bug 57706] New: 0414:err:msxml:mxwriter_QueryInterface interface {e19c7100-9709-4db7-9373-e7b518b47086} not implemented
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

https://bugs.winehq.org/show_bug.cgi?id=57706 Bug ID: 57706 Summary: 0414:err:msxml:mxwriter_QueryInterface interface {e19c7100-9709-4db7-9373-e7b518b47086} not implemented Product: Wine Version: 10.0-rc5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: msxml3 Assignee: wine-bugs(a)winehq.org Reporter: winehq.tabloid372(a)passmail.net Distribution: --- Hi Nikolay, How can I work out the correct stub to add form the guid? -- 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.

2 4

[Bug 53843] New: coturn support
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

https://bugs.winehq.org/show_bug.cgi?id=53843 Bug ID: 53843 Summary: coturn support Product: Packaging Version: unspecified Hardware: x86-64 OS: other Status: UNCONFIRMED Severity: blocker Priority: P2 Component: wine-packages Assignee: wine-bugs(a)winehq.org Reporter: Neustradamus(a)hotmail.com CC: dimesio(a)earthlink.net Can you look to add the WINE support of coturn directly in software? There is a ticket here, answer is needed: - https://github.com/coturn/coturn/issues/1044 Thanks in advance. -- 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.

2 5

[Bug 59283] New: Berkeley DB version mismatch crashes 32-bit GTK+ applications in Wine 11
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

http://bugs.winehq.org/show_bug.cgi?id=59283 Bug ID: 59283 Summary: Berkeley DB version mismatch crashes 32-bit GTK+ applications in Wine 11 Product: Wine Version: 11.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)list.winehq.org Reporter: nen24t(a)gmail.com Distribution: --- Created attachment 80208 --> http://bugs.winehq.org/attachment.cgi?id=80208 Full backtrace reaper 32bit ## Bug Summary **Title:** Berkeley DB version mismatch causes crashes in 32-bit applications using GTK+ **Component:** wine (core) **Severity:** normal **Priority:** P2 **Version:** Wine 11.0 **Hardware:** x86_64 **OS:** Linux (Ubuntu 24.04) --- ## Description Wine 11.0 exhibits Berkeley DB (libdb) version mismatch errors when running certain 32-bit Windows applications that use GTK+ libraries. The error occurs before the application can fully initialize, resulting in a crash. ### Error Message ``` BDB1539 Build signature doesn't match environment Cannot open DB environment: BDB0091 DB_VERSION_MISMATCH: Database environment version mismatch wine: Unhandled page fault on execute access to 0x00000000 ``` ### Affected Applications - REAPER 32-bit (Digital Audio Workstation) - Garritan CFX Lite 32-bit (VST instrument host) - Other applications using GTK+ with Berkeley DB backend ### Impact Applications crash immediately on startup, before reaching their main functionality. This affects audio production software and other professional tools that rely on GTK+ libraries. --- ## Steps to Reproduce 1. Install Wine 11.0 (wine-stable or custom build) 2. Set up a 32-bit Wine prefix 3. Install REAPER 32-bit or similar GTK+-based application 4. Run the application: ```bash WINEPREFIX=~/.wine wine "C:\Program Files (x86)\REAPER\reaper.exe" ``` 5. Observe crash with BDB error ### Environment - **Wine version:** wine-11.0 - **Distribution:** Ubuntu 24.04 LTS - **Kernel:** 6.14.0-1019-oem - **System libdb:** libdb-5.3.so (from libdb5.3-dev package) - **Architecture:** x86_64 (running 32-bit Windows apps via WoW64) --- ## Expected Behavior Application should start normally without Berkeley DB errors. GTK+ libraries should either: 1. Use a compatible libdb version, or 2. Handle version mismatches gracefully, or 3. Not use persistent Berkeley DB storage that can become outdated --- ## Actual Behavior Application crashes with page fault at NULL pointer (0x00000000) after Berkeley DB version mismatch error. The crash occurs in system libraries before the application's main code executes. ### Backtrace ``` Unhandled exception: page fault on write access to 0x00000000 in wow64 32-bit code (0x003600a9). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:003600a9 ESP:0141a864 EBP:00688f36 EFLAGS:00010203 EAX:00000000 EBX:006885e0 ECX:018eb0b0 EDX:00000000 Modules: Module Address Debug info Name (146 modules) PE 400000- 1317000 Deferred reaper ELF f2271000-f2464000 Deferred libdb-5.3.so [... more modules ...] ``` --- ## Analysis ### Root Cause The issue appears to be related to: 1. **GTK+ uses Berkeley DB** for recently-used file tracking (`.local/share/recently-used.xbel`) 2. **System libdb version** (5.3.x) may be incompatible with Wine's expectations 3. **Database files persist** between Wine versions, causing signature mismatch 4. **32-bit applications more affected** - possibly due to WoW64 library loading ### Not Application-Specific This is NOT an application bug. We verified this by: - Testing with multiple applications (REAPER, CFX Lite) - Creating a minimal test program that does NOT use Berkeley DB - works perfectly - The crash occurs in Wine/GTK+ code, not application code --- ## Workarounds ### Temporary Solutions 1. **Remove stale database files:** ```bash rm -rf ~/.wine/.local/share/recently-used.xbel* find ~/.wine -name "*.db" -delete ``` 2. **Use 64-bit applications** (not always possible) 3. **Downgrade to Wine 10.x** (loses Wine 11 improvements) ### Limitations These workarounds are temporary - the issue recurs after GTK+ recreates database files. --- ## Additional Information ### Testing Details - **Wine 11 new WoW64:** Crashes - **Wine 11 old WoW64 (wine-stable):** Crashes - **Wine 10.x:** Also affected (issue predates Wine 11) - **64-bit applications:** Generally unaffected ### Relevant Files - System libdb: `/lib/x86_64-linux-gnu/libdb-5.3.so` (64-bit) - System libdb: `/lib/i386-linux-gnu/libdb-5.3.so` (32-bit) - Wine loads: `ELF f2271000-f2464000 Deferred libdb-5.3.so` ### Related Issues This may be related to GTK+'s use of Berkeley DB for: - Recently used files tracking - Icon cache - Other persistent data structures --- ## Suggested Fixes 1. **Pin libdb version:** Bundle compatible libdb version with Wine 2. **Handle version mismatch:** Gracefully recreate database on version mismatch 3. **Alternative backend:** Consider switching GTK+ to use SQLite or other database 4. **Clear on upgrade:** Automatically clear incompatible databases when Wine version changes --- ## Impact Assessment - **Severity:** Medium-High - Affects professional audio production tools - **Scope:** 32-bit GTK+ applications on Wine 11 - **Workaround:** Available but temporary - **User Impact:** Cannot use certain professional Windows software on Linux --- ## Test Case A minimal test case that demonstrates the issue: 1. Create Wine prefix: ```bash WINEPREFIX=~/.wine wineboot ``` 2. Install REAPER 32-bit (or any GTK+ app) 3. Run with debug output: ```bash WINEDEBUG=+all wine reaper.exe 2>&1 | grep -i "bdb\|berkeley" ``` 4. Observe BDB error followed by crash --- ## References - **WineASIO project:** https://github.com/giang17/wineasio - **Issue documented in:** TEST-RESULTS.md (comprehensive 32-bit testing analysis) - **Berkeley DB:** https://www.oracle.com/database/technologies/related/berkeleydb.html - **GTK+ documentation:** https://www.gtk.org/ --- ## Reporter Information - **Reporter:** WineASIO v1.4.1 testing team - **Date:** January 21, 2026 - **Tested configurations:** Multiple Wine 11 builds, Ubuntu 24.04 - **Verification:** Extensive testing with multiple applications --- ## Additional Notes This bug was discovered during Wine 11 porting of WineASIO (ASIO to JACK audio driver). While investigating what appeared to be a WineASIO bug, we determined the root cause is a Wine/GTK+/Berkeley DB compatibility issue. The WineASIO library itself works correctly - we confirmed this with a minimal test program that bypasses GTK+. The crash occurs in Wine's GTK+ implementation before WineASIO is even initialized. --- **Follow-up questions for WineHQ developers:** 1. Is Wine's libdb build compatible with system libdb 5.3? 2. Should Wine bundle its own libdb version? 3. Can GTK+ database initialization be made more robust? 4. Is there a way to force database recreation on version mismatch? -- 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.

1 6

[Bug 51128] New: Linux Mint 20.x: Cant install wine.
by WineHQ Bugzilla Feb. 26, 2026

Feb. 26, 2026

https://bugs.winehq.org/show_bug.cgi?id=51128 Bug ID: 51128 Summary: Linux Mint 20.x: Cant install wine. Product: Wine Version: 6.8 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: trivial Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: dushansecondary(a)gmail.com Distribution: --- Trying to install wine with "wine-installer" fails with the following error: ---------------------------------------------------------- The following information may help to resolve the situation: The following packages have unmet dependencies: wine-installer : Depends: wine Recommends: wine64 but it is not going to be installed Recommends: wine32 E: Unable to correct problems, you have held broken packages. ---------------------------------------------------------- Installing using the method on your website fails with: ---------------------------------------------------------- The following packages have unmet dependencies: winehq-stable : Depends: wine-stable (= 6.0.0~focal-1) E: Unable to correct problems, you have held broken packages. ---------------------------------------------------------- We discussed about this bug on a Linux mint forum, here is a short statement: "...and I found some bug (?) in apt relating to the packages wine-installer and wine, which might be related to your message: "wine-installer depends on wine but wine isn't going to be installed" (why not??), and the fact that I can install wine-installer without installing wine, even though the package software tells me that wine is required for wine-installer." -- 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.

2 3