[Bug 59970] New: after ea5211a86752f0ac988425a7a941229824b27449, gamescope will disappear for some reason
http://bugs.winehq.org/show_bug.cgi?id=59970 Bug ID: 59970 Summary: after ea5211a86752f0ac988425a7a941229824b27449, gamescope will disappear for some reason Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winewayland Assignee: wine-bugs@list.winehq.org Reporter: l12436.tw@gmail.com Target Milestone: --- Distribution: --- reproduce step 1. open windows steam big picture mode. 2. open cheatengine with wine desktop. gamescope will soon disappear. but before ea5211a86752f0ac988425a7a941229824b27449, it working as expected. LOG just shown seh 0074:err:seh:call_seh_handlers invalid frame 00007FFFFE10F210 (00007FFFFE112000-00007FFFFE310000) 0074:err:seh:NtRaiseException Exception frame is not in stack limits => unable to dispatch exception. 02b8:err:seh:call_seh_handlers invalid frame 00007FFFFE10F310 (00007FFFFE112000-00007FFFFE210000) 02b8:err:seh:NtRaiseException Exception frame is not in stack limits => unable to dispatch exception. 0440:fixme:kernelbase:AppPolicyGetProcessTerminationMethod FFFFFFFFFFFFFFFA, 00007FFFFE20FEE0 02e0:fixme:kernelbase:AppPolicyGetProcessTerminationMethod FFFFFFFFFFFFFFFA, 00007FFFFE20FEE0 0748:fixme:thread:NtSetInformationThread ThreadPowerThrottling stub! 0748:fixme:thread:NtSetInformationThread ThreadPowerThrottling stub! 0748:fixme:heap:RtlSetHeapInformation HEAP_INFORMATION_CLASS 1 not implemented! 0748:fixme:process:GetProcessMitigationPolicy (FFFFFFFFFFFFFFFF, 4, 00007FFFFE20F74C, 4): stub 0748:fixme:thread:NtSetInformationThread ThreadPowerThrottling stub! 0748:err:seh:call_seh_handlers invalid frame 00007FFFFE10EB80 (00007FFFFE112000-00007FFFFE210000) 0748:err:seh:NtRaiseException Exception frame is not in stack limits => unable to dispatch exception. -- 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=59970 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|after |Gamescope fails |ea5211a86752f0ac988425a7a94 | |1229824b27449, gamescope | |will disappear for some | |reason | --- Comment #1 from Ken Sharp <imwellcushtymelike@gmail.com> ---
winewayland: Use a flag instead of checking zwp_relative_pointer_v1.
This? https://github.com/ValveSoftware/gamescope It says it's a Linux app. How is Wine involved? Please provide the exact process to reproduce this issue, as well as the Wine version and the console log. https://gitlab.winehq.org/wine/wine/-/wikis/Bugs -- 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=59970 --- Comment #2 from TOM <l12436.tw@gmail.com> --- because it working before that commit so .... I run wine under gamescope -- 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=59970 Zeb Figura <z.figura12@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12@gmail.com Keywords| |regression Regression SHA1| |ea5211a86752f0ac988425a7a94 | |1229824b27449 --- Comment #3 from Zeb Figura <z.figura12@gmail.com> --- Gamescope is a nested window manager. It's similar to running inside of e.g. Xephyr, but it has some additional tweaks meant to improve the experience for games. This bug report is about running Wine inside of Gamescope. -- 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=59970 --- Comment #4 from TOM <l12436.tw@gmail.com> --- everything is working correctly before 685c5b6f6312d55c948ee15315d858777af72408. about yesterday. I build wine daily -- 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=59970 --- Comment #5 from Zeb Figura <z.figura12@gmail.com> --- (In reply to TOM from comment #4)
everything is working correctly before 685c5b6f6312d55c948ee15315d858777af72408. about yesterday. I build wine daily
That's a different SHA. Why have you reported two different SHAs? -- 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=59970 --- Comment #6 from TOM <l12436.tw@gmail.com> --- I just said yesterday daily build is all working correctly. sorry for confused -- 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=59970 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- URL| |https://github.com/ValveSof | |tware/gamescope Keywords| |download, source Version|unspecified |11.12 --- Comment #7 from Ken Sharp <imwellcushtymelike@gmail.com> --- So Gamescope is closing when you run Wine? -- 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=59970 --- Comment #8 from TOM <l12436.tw@gmail.com> --- yes, it close after I run another desktop under gamescope. before that commit, it normal -- 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=59970 --- Comment #9 from TOM <l12436.tw@gmail.com> --- Sorry about update regression commit Test procedure: After multiple test, regression seems not on ea5211a86752f0ac988425a7a941229824b27449 I initially thought. It seems just accidentally working due to memory or something. If I switch to master and revert ea5211a86752f0ac988425a7a941229824b27449 it still happened. I believe regression is not locate at that. Now I found is locate at f50b263562c002c61ab049baa1212c3bd36994bb. This time, Not only I test on checkout before that commit. I also check revert commit. but due to master also has a commit seems use dpi code, it will show conflict on server/window.c. I actually revert cba7652eac8fda005c53851e59fcc6f194168df9 and f50b263562c002c61ab049baa1212c3bd36994bb. All the above is test on the wine prefix that did not often used. If I test on the wineprefix that is use daily, I need to revert one more 2587d2822c986f525ce96e2176bd41d01dcb5593. In order to shown dialog like winecfg. or it will just stuck with log 00dc:err:sync:RtlpWaitForCriticalSection section 00006FFFFFC3D460 "../dlls/ntdll/loader.c: loader_section" wait timed out in thread 00dc, blocked by 00d8, retrying (60 sec) Even wineboot.exe -u dialog did not shown. So I can said the regression is start with 2587d2822c986f525ce96e2176bd41d01dcb5593 ? Summary: revert 2587d2822c986f525ce96e2176bd41d01dcb5593, and f50b263562c002c61ab049baa1212c3bd36994bb will fix stuck and seh issue. but It will show extra conflict on server/window. so need to revert cba7652eac8fda005c53851e59fcc6f194168df9 to prevent this conflict. All of the test is tested with wine winecfg. only when winecfg shown. than I will test my window steam setup five times to check if issue still there. -- 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=59970 TOM <l12436.tw@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1|ea5211a86752f0ac988425a7a94 |2587d2822c986f525ce96e2176b |1229824b27449 |d41d01dcb5593 -- 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=59970 --- Comment #10 from TOM <l12436.tw@gmail.com> --- So my current understanding is: f50b263... seems related to the SEH issue, while 2587d282... seems related to stuck dialog issue on my daily wineprefix. By the way, the 2587d282... test with winecfg dialog was not under gamescope, because it just stuck there before the dialog showed. I tested it directly on KDE. -- 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=59970 --- Comment #11 from TOM <l12436.tw@gmail.com> --- My environment: wine built from master ubuntu 22.04 monitor with HDMI/DP, and both connect to same graphic card with clone screen. That mean no matter DP or HDMI will show both same screen. -- 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=59970 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- 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=59970 --- Comment #12 from TOM <l12436.tw@gmail.com> --- I can reproduce this issue by just run WINEDEBUG= wine explorer /desktop=shell,1920x1080 winecfg This is the log libEGL warning: pci id for fd 32: 10de:2544, driver (null) pci id for fd 34: 10de:2544, driver (null) pci id for fd 35: 10de:2544, driver (null) libEGL warning: egl: failed to create dri2 screen libEGL warning: pci id for fd 32: 10de:2544, driver (null) pci id for fd 34: 10de:2544, driver (null) pci id for fd 35: 10de:2544, driver (null) libEGL warning: egl: failed to create dri2 screen libEGL warning: pci id for fd 32: 10de:2544, driver (null) 00cc:fixme:ver:GetCurrentPackageId (017CFF10 00000000): stub 00ec:fixme:netprofm:init_networks no support for detecting network changes 00ec:fixme:netprofm:connection_point_Advise 00F37A78, 00F179EC, 00F17A54 - semi-stub 00ec:fixme:netprofm:init_networks no support for detecting network changes 002c:fixme:ver:GetCurrentPackageId (00007FFFFE30FEF0 0000000000000000): stub 0168:fixme:toolhelp:CreateToolhelp32Snapshot Unimplemented: heap list snapshot 0168:fixme:toolhelp:Heap32ListFirst : stub 0168:fixme:ntdll:NtQuerySystemInformation info_class SYSTEM_PERFORMANCE_INFORMATION 0034:fixme:ver:GetCurrentPackageId (00007FFFFE30FEF0 0000000000000000): stub libEGL warning: pci id for fd 33: 10de:2544, driver (null) pci id for fd 35: 10de:2544, driver (null) pci id for fd 36: 10de:2544, driver (null) libEGL warning: egl: failed to create dri2 screen libEGL warning: pci id for fd 33: 10de:2544, driver (null) pci id for fd 35: 10de:2544, driver (null) pci id for fd 36: 10de:2544, driver (null) libEGL warning: egl: failed to create dri2 screen libEGL warning: pci id for fd 33: 10de:2544, driver (null) 017c:err:seh:call_seh_handlers invalid frame 00007FFFFE10F3D0 (00007FFFFE112000-00007FFFFE310000) 017c:err:seh:NtRaiseException Exception frame is not in stack limits => unable to dispatch exception. 015c:fixme:ver:GetCurrentPackageId (00007FFFFE30FEF0 0000000000000000): stub -- 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=59970 --- Comment #13 from TOM <l12436.tw@gmail.com> --- Issue is locate the following code which is in monitor_get_dpi if (source && type != MDT_EFFECTIVE_DPI) { scale_x.num = source->physical.dmPelsWidth; scale_x.den = source->current.dmPelsWidth; scale_y.num = source->physical.dmPelsHeight; scale_y.den = source->current.dmPelsHeight; } somehow source->physical.dmPelsWidth/source->current.dmPelsWidth/source->physical.dmPelsHeight/source->current.dmPelsHeight will became 0 for some reason. -- 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=59970 TOM <l12436.tw@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Gamescope fails |monitor_get_dpi will get | |zero | |dmPelsWidth/dmPelsHeight in | |both physical/current cause | |divide by zero in | |make_ratio due to gcd | |return 0 -- 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=59970 --- Comment #14 from TOM <l12436.tw@gmail.com> --- The following code make_ratio will pass the dpi * scale_x.num, scale_x.den/dpi * scale_y.num, scale_y.den which will be both 0, 0 causing the gcd also pass 0,0 then return 0 the divide zero is maybe happened in make_ratio struct ratio r = { num / d, den / d }; bacause the d is return 0 by gcd. -- 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=59970 --- Comment #15 from TOM <l12436.tw@gmail.com> --- Created attachment 81527 --> http://bugs.winehq.org/attachment.cgi?id=81527 Workaround patch This patch only just for workaround divide zero happened in monitor_get_dpi. I did not know if there are better way. -- 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=59970 TOM <l12436.tw@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|monitor_get_dpi will get |monitor_get_dpi will get |zero |zero |dmPelsWidth/dmPelsHeight in |dmPelsWidth/dmPelsHeight in |both physical/current cause |both physical/current cause |divide by zero in |divide by zero |make_ratio due to gcd | |return 0 | -- 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=59970 --- Comment #16 from Rémi Bernon <rbernon@codeweavers.com> --- I don't think dmPelsWidth/dmPelsHeight should ever be 0. Could you attach a log file with WINEDEBUG=+timestamp,+pid,+system,+x11drv,+event ? -- 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=59970 --- Comment #17 from Rémi Bernon <rbernon@codeweavers.com> --- Created attachment 81635 --> http://bugs.winehq.org/attachment.cgi?id=81635 Possible fix Or perhaps, does this help? -- 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=59970 --- Comment #18 from Chunhao Hung <l12436.tw@gmail.com> --- Created attachment 81637 --> http://bugs.winehq.org/attachment.cgi?id=81637 Log from +timestamp,+pid,+system,+x11drv,+event -- 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=59970 --- Comment #19 from Chunhao Hung <l12436.tw@gmail.com> --- (In reply to Rémi Bernon from comment #17)
Created attachment 81635 [details] Possible fix
Or perhaps, does this help?
Yes, it seems help. thank for the proper fix, at least it did not crash on winecfg with desktop (In reply to Rémi Bernon from comment #16)
I don't think dmPelsWidth/dmPelsHeight should ever be 0. Could you attach a log file with WINEDEBUG=+timestamp,+pid,+system,+x11drv,+event ?
That is also mine question, I also think that should never be zero, but on my PC, it actually became zero for some reason. I did not know how it came from. -- 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=59970 --- Comment #20 from Chunhao Hung <l12436.tw@gmail.com> --- seems like 446809.077:0178:017c:trace:system:set_winstation_monitors increment 1 446809.077:0178:017c:trace:system:set_winstation_monitors flags 0x4 virt (0,0)-(0,0) dpi 96:1 raw (0,0)-(0,0) raw_dpi 1:0 446809.077:0178:017c:trace:system:set_winstation_monitors flags 0x1 virt (0,0)-(1920,1080) dpi 96:1 raw (0,0)-(1920,1080) raw_dpi 96:1 this part ? (0,0)-(0,0) !? -- 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=59970 --- Comment #21 from Rémi Bernon <rbernon@codeweavers.com> --- Yeah it turns out that they actually can be 0, for detached (ie: inactive) monitors, this happens briefly when using virtual desktop mode, but for some reason it doesn't cause a crash with me. -- 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=59970 --- Comment #22 from Chunhao Hung <l12436.tw@gmail.com> --- so that is actually shown on your side? that weird, according to the code. it should cause crash. but I also sometimes did not crash as well. it only frequently crash when wine desktop is opened. -- 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=59970 --- Comment #23 from Chunhao Hung <l12436.tw@gmail.com> --- I submit your patch to my MR. Is that OK ? -- 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=59970 Chunhao Hung <l12436.tw@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED --- Comment #24 from Chunhao Hung <l12436.tw@gmail.com> --- fix by 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=59970 Chunhao Hung <l12436.tw@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |3ec14378590503ffbadfa936b99 | |4e5bf849fe967 -- 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=59970 Alexandre Julliard <julliard@winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #25 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