[Bug 60005] New: Pegasus just dies. Downgrading to 11.12 makes it work fine again.
http://bugs.winehq.org/show_bug.cgi?id=60005 Bug ID: 60005 Summary: Pegasus just dies. Downgrading to 11.12 makes it work fine again. 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: mikes@guam.net Target Milestone: --- Distribution: --- Just upgraded to 11.13, and noticed a few errors that cuase Pegasus Mail to completely die. Sometimes the screen is just gone, and no task or error message I can see. Sometimes screen just stops responding completely fozen. Had also noticed at one point typing ( resulded in cursor moving but no character on screen. Did find if I changed to 20pt font the ( would display, but nothing at regular size. Crash would allow restarting, and it would come up again. Did the downgrade to 11.12 and it once again, works fine, so can only assume something in 11.13 is causing issue. Not sure where I could get details of what cuases issue. dmesg, journalctl some wine option. No debug. Option. -- 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=60005 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> --- I think this should have Keywords: download, regression (But I am not allowed to set them) -- 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=60005 --- Comment #2 from Jonas <winehq@jonass.user.lysator.liu.se> --- Created attachment 81419 --> http://bugs.winehq.org/attachment.cgi?id=81419 Various logs I may have the same problem, after updating from wine-11.12 to wine-11.13. Running from a shell window I get various error messages, see attached file. First impression is that it works fine if Pegasus is run, left with open window and nothing else. Whenever I work with other windows like minimize Pegasus or some other window, change focus and thus the depth order and send Pegasus back from top or such seems to trigger some problem. -- 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=60005 --- Comment #3 from Jonas <winehq@jonass.user.lysator.liu.se> --- (In reply to Jonas from comment #2) This is on XUbuntu 24.04.4 LTS -- 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=60005 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Pegasus just dies. |Pegasus hangs and crashes |Downgrading to 11.12 makes | |it work fine again. | --- Comment #4 from Ken Sharp <imwellcushtymelike@gmail.com> --- Please run a regression test. https://gitlab.winehq.org/wine/wine/-/wikis/Regression-Testing Though this could be a clue: 0308:err:msg:process_hardware_message unknown message type fffffffa 0308:err:msg:process_hardware_message unknown message type 0 0308:err:msg:process_hardware_message unknown message type 3 0308:err:system:user_check_not_lock BUG: holding USER lock dlls/win32u/sysparams.c:351: user_check_not_lock: Assertion `0' failed. wine: Assertion failed at address F34C9579 (thread 0308), starting debugger... Can this application be downloaded? -- 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=60005 --- Comment #5 from Jonas <winehq@jonass.user.lysator.liu.se> --- (In reply to Ken Sharp from comment #4)
Can this application be downloaded?
Yes, look for "Pegasus Mail v4.91 for Windows [Current offical version]" at https://www.pmail.com/downloads.htm -- 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=60005 --- Comment #6 from Jonas <winehq@jonass.user.lysator.liu.se> --- git bisect says this may be related to: 08f7b746b00c07312a70d2f978d86bc80e6df88a is the first bad commit commit 08f7b746b00c07312a70d2f978d86bc80e6df88a Author: Rémi Bernon <rbernon@codeweavers.com> Date: Thu Jun 18 14:43:00 2026 +0200 winex11: Send raw mouse motion frames from XI2 RawEvents. dlls/winex11.drv/event.c | 2 +- dlls/winex11.drv/mouse.c | 30 ++++++++++++++++++------------ 2 files changed, 19 insertions(+), 13 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=60005 --- Comment #7 from Jonas <winehq@jonass.user.lysator.liu.se> --- Could this be a duplicate of Bug 59999 ? Bisecting suggests the same commit. -- 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=60005 --- Comment #8 from Michael Setzer II <mikes@guam.net> --- Don't know will look at that one in a minute. Using Fedora 43 with XFCE using X11. Worked fine with 11.2, but issues with 11.3. Downgrading to 11.2 resolves. Went to check again, and now 11.3 downloads, but fails checksum. Was looking around at different places, and saw one screen mention a read-only file system. Also, get checksum message on a Fedora 44 machine to try reinstalling 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.
http://bugs.winehq.org/show_bug.cgi?id=60005 Austin English <austinenglish@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, 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=60005 Austin English <austinenglish@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |rbernon@codeweavers.com Regression SHA1| |08f7b746b00c07312a70d2f978d | |86bc80e6df88a -- 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=60005 --- Comment #9 from Michael Setzer II <mikes@guam.net> --- Was getting checksum error. Downloaded fine, but then checksum failed with Fedora 43. But later it downloaded and installed the 11.13, and not seeing issue? But don't know if something was fixed. So Fedora 43 downloads and installs, but now on Fedora 44 machine that had the 11.13 installed get this. [1/1] winehq-devel-1:11.13-1.1.x86_64 100% | 379.1 KiB/s | 75.1 KiB | 00m00s
Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256) - https://dl.winehq.org/wine-builds/fedora/44/x86_64/winehq-devel-11.13-1.1.x8... Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256) - https://dl.winehq.org/wine-builds/fedora/44/x86_64/winehq-devel-11.13-1.1.x8... Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256) - https://dl.winehq.org/wine-builds/fedora/44/x86_64/winehq-devel-11.13-1.1.x8... Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256) - https://dl.winehq.org/wine-builds/fedora/44/x86_64/winehq-devel-11.13-1.1.x8... Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256)
[1/1] Total 100% | 189.6 KiB/s | 75.1 KiB | 00m00s Failed to download packages Librepo error: Downloading successful, but checksum doesn't match. Calculated: a7b5ded5428237fc3f80717b9910d873f0f2dc1f6bfdf7e94873a6e876daa9e4(sha256) Expected: 6a0c250ad2bbfe186e3f8267e79296bc95927fd0fed95fa98c5a3abfe952d409(sha256) So, not sure if issue is fixed, or if some issue is going on with repos?? -- 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=60005 --- Comment #10 from Jonas <winehq@jonass.user.lysator.liu.se> --- (In reply to Michael Setzer II from comment #9)
So, not sure if issue is fixed, or if some issue is going on with repos??
Please do not mix up issues. Repos were probably fixed by Alexandre Julliard, see comment on issue 60006. -- 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=60005 --- Comment #11 from Michael Setzer II <mikes@guam.net> --- Update gone back to 11.12. Had the 11.13 installed afther the checksum error disappeared, and thought something had been fixed. Then used Pegasus right after last post, and once again it just blanked closed with no warning. Had it running via a terminal start windows, and both the Pegasus WIndows, and Terminal window disappeared, so no message. So downgraded to 11.12, and it worked fine. Currently seeing checksum errors with Fedora 44 machines. Downloads fine, but then says checksum is wrong. Note sure what I could do fine info. Past crashes would give debug info, or leave terminal window with something. Hopefully it is something minor. Was movng mouse at time. Fedora 43 using XFCE and X11. -- 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=60005 Ken Sharp <imwellcushtymelike@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- URL| |https://www.pmail.com/downl | |oads.htm -- 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=60005 --- Comment #12 from Jonas <winehq@jonass.user.lysator.liu.se> --- (In reply to Michael Setzer II from comment #11)
Note sure what I could do fine info.
Did you try to compile and bisect as described on https://gitlab.winehq.org/wine/wine/-/wikis/Regression-Testing ? To verify if my conclusion about relevant commit is correct for you too. -- 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=60005 --- Comment #13 from Michael Setzer II <mikes@guam.net> --- Looked at link? But seems the 11.13 has 22 bug fixes and not sure if that would be what the bisects would be one, or something totally different. Have cloned git projects and built the projects, but doesn't look like I see how the bisect process would be done. It I new if one of the 22 might be causing the issue, but then sometimes it seemed to be working, and then popped up again. -- 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=60005 --- Comment #14 from Jonas <winehq@jonass.user.lysator.liu.se> --- Today I run with 11.13 by mistake. After a while it crashed with "wine client error:7dc: write: Bad file descriptor" -- 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=60005 Stian Low <wineryyyyy@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |wineryyyyy@gmail.com --- Comment #15 from Stian Low <wineryyyyy@gmail.com> --- All bugs recently reporting this log and similar behavior may be duplicates: 0494:err:msg:process_hardware_message unknown message type 1 https://bugs.winehq.org/buglist.cgi?quicksearch=process_hardware_message -- 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=60005 --- Comment #16 from Jonas <winehq@jonass.user.lysator.liu.se> --- Problem seems to be in wine-11.14 also. -- 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=60005 --- Comment #17 from Michael Setzer II <mikes@guam.net> --- Also seeing after install 1.14 In terminal window behind wine run see lots of these lines. 0184:err:msg:process_hardware_message unknown message type (seeing number 1 2 3 or 5) pattern of numbers is 1 2 5 3? Don't know how to find what device. No usb devices plugged in. Dell Latitude 5580HQ does have a touchpad? -- 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=60005 --- Comment #18 from Michael Setzer II <mikes@guam.net> --- Did little more test, and note 1 difference. With 1.13 it would do the stop responding, but sometimes just shot windows completely. with 1.14 have noticed the Pegasus window stops responding to any clicks. Only clicking x in upper right has a window come up that windows has stopped responding. THen closes. Restarting Pegasus then comes up with message that pegasus was not shut down correctly since lock file still exist. Have to pass that, and then turn preview mode back on. Have downgraded twice to get back to 1.12 that works fine. -- 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=60005 --- Comment #19 from Michael Setzer II <mikes@guam.net> --- Update 11.14 seems to no longer show issue? But don't know if it fixed something or if some other update solved problem. 11.12 worked 11.13 would have the Pegasus lock and only click on x in top right to close with wine not responding, or sometimes the Pegasus Mail window and the terminual wendo would both just close. 11.14 would to the screen lock, but never had windows both close. Went back to 11.12 aand all was good. Using Fedora 43 with xfce. Tried to do a bisect with help of google ai, but it kept giving errors about freetype and downloaded git, and had to build without freetype even thou the packages that both google ai an claude ai later showed as needed were install. bisect wine build, but running it resulted with pegasus opening with a messed up window, and it had background windos that would open like terminal windows in top panel, but would no actual show anythig only option was to close them. Gave up on google ai. claude ai had same issue, but then it came up with an additional option that did have the build work with no error. Tran thru link 7 builds/bisects and all came up as good no locks at all. Moved mouse all over sccreen, and mouse over worked, and clicking on menu options work.. Then installed 11.14, and did tests with it, and again, no issues? Have had a number of packages updated with dnf, so I'm thinking it was something in 11.13 and 11.14 that had issues with something, but the 11.12 didn't. Wish I could point to something exactly and say this fixed it?, but at moment running 11.14 for about a day, and don't see issue? Problem with both 11.13 and 11.14 would generally happen within about 20 seconds of just moving the mouse around. So, not clear. What caused problem or what has now seemed to fix 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=60005 --- Comment #20 from Michael Setzer II <mikes@guam.net> --- Created attachment 81589 --> http://bugs.winehq.org/attachment.cgi?id=81589 dmesg lines. ? had phone plugged in to charge. Don't know if these might have any issue. Sometime when it starts shows messages about usb devices. And also some like the other log file attached. right now wine terminal window shows. 0128:err:virtual:allocate_virtual_memory out of memory for allocation, base (nil) size fad50000 0128:err:virtual:allocate_virtual_memory out of memory for allocation, base (nil) size fad50000 0128:err:virtual:allocate_virtual_memory out of memory for allocation, base (nil) size fad50000 Just closed again completely. These are new dmesg lines. Going back to 11.12 [64797.409041] usb 3-1: USB disconnect, device number 10 [64797.552572] xhci_hcd 0000:3c:00.0: xHCI host controller not responding, assume dead [64797.552579] xhci_hcd 0000:3c:00.0: HC died; cleaning up [64797.570275] xhci_hcd 0000:3c:00.0: remove, state 4 [64797.570283] usb usb4: USB disconnect, device number 1 [64797.571872] xhci_hcd 0000:3c:00.0: USB bus 4 deregistered [64797.571982] xhci_hcd 0000:3c:00.0: remove, state 1 [64797.571987] usb usb3: USB disconnect, device number 1 [64797.572659] xhci_hcd 0000:3c:00.0: USB bus 3 deregistered [64797.589188] pci_bus 0000:05: Allocating resources [64797.589195] pcieport 0000:04:00.0: bridge window [io size 0x2000]: can't claim; no address assigned [64797.589199] pcieport 0000:04:00.0: [io 0x1000-0x2fff] clipped to [io 0x2000-0x2fff] [64797.589205] pcieport 0000:04:00.0: bridge window [io 0x2000-0x2fff] [64797.589219] pcieport 0000:05:01.0: bridge window [io size 0x1000]: can't claim; no address assigned [64797.589232] pcieport 0000:05:02.0: bridge window [io size 0x1000]: can't claim; no address assigned [64797.589244] pcieport 0000:05:02.0: bridge window [mem size 0x00200000 64bit pref]: can't claim; no address assigned [64797.589272] pcieport 0000:05:01.0: bridge window [io 0x1000-0x0fff] to [bus 07-3b] add_size 1000 [64797.589276] pcieport 0000:05:02.0: bridge window [io 0x1000-0x0fff] to [bus 3c] add_size 1000 [64797.589280] pcieport 0000:05:02.0: bridge window [mem 0x00100000-0x000fffff 64bit pref] to [bus 3c] add_size 200000 add_align 100000 [64797.589291] pcieport 0000:05:02.0: bridge window [mem size 0x00200000 64bit pref]: can't assign; no space [64797.589294] pcieport 0000:05:02.0: bridge window [mem size 0x00200000 64bit pref]: failed to assign [64797.589298] pcieport 0000:05:01.0: bridge window [io 0x2000-0x2fff]: assigned [64797.589301] pcieport 0000:05:02.0: bridge window [io size 0x1000]: can't assign; no space [64797.589304] pcieport 0000:05:02.0: bridge window [io size 0x1000]: failed to assign [64797.589307] pcieport 0000:05:01.0: bridge window [io 0x2000-0x2fff]: releasing [64797.589311] pcieport 0000:05:02.0: bridge window [mem size 0x00200000 64bit pref]: can't assign; no space [64797.589314] pcieport 0000:05:02.0: bridge window [mem size 0x00200000 64bit pref]: failed to assign [64797.589317] pcieport 0000:05:02.0: bridge window [io 0x2000-0x2fff]: assigned [64797.589321] pcieport 0000:05:01.0: bridge window [io size 0x1000]: can't assign; no space [64797.589323] pcieport 0000:05:01.0: bridge window [io size 0x1000]: failed to assign [64802.608691] pcieport 0000:05:00.0: Unable to change power state from D3hot to D0, device inaccessible [64802.615769] pci_bus 0000:06: busn_res: [bus 06] is released [64802.619142] pci_bus 0000:07: busn_res: [bus 07-3b] is released [64802.619816] pci_bus 0000:3c: busn_res: [bus 3c] is released [64802.620513] pci_bus 0000:05: busn_res: [bus 05-3c] is released [64823.809719] perf: interrupt took too long (4196 > 4065), lowering kernel.perf_event_max_sample_rate to 47000 -- 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=60005 --- Comment #21 from Rémi Bernon <rbernon@codeweavers.com> --- I haven't been able to reproduce this yet, although the process_hardware_message message flood reports are making me think that it might be some memory corruption on wineserver side. I don't see how these unsupported messages could be generated otherwise. Starting the prefix by running wineserver under valgrind with: WINEPREFIX=<prefix path> valgrind wineserver -f Prior to starting any other application might tell a bit more. -- 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=60005 --- Comment #22 from Michael Setzer II <mikes@guam.net> --- Not clear what message is. env WINE_LARGE_ADDRESS_AWARE=1 WINEARCH=win64 WINEDEBUG=fixme-all,err+all wine /home/msetzerii/.wine/drive_c/PMAIL/Programs/WINPM-32.EXE that is the command. Nothing else is being run under wine. With 11.12 there is no problem. With both 11.13 and later withe 11.14 both have lock ups or freezes as I see it. Moving the mouse across them menus and various buttons. Under normal opperations, the buttons and menu items adjust or menus pull down. This suddenly just stops, the mouse continues to move fine, buth mouse over and clicking on menus results in no changes. With the exception of clicking on the x in upper right, that brings up the windows note responding window and closing it. The other condition, is some times both the Pegasus and the terminal window both disappear. Going back to 11.12 resolves issue. Last time test with 11.14 Pegasus was running, and went down and clicked on mail folder and both Pegasus and terminal window closed. Don't know how to get more info, since no dbg screen and error messages. -- 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=60005 --- Comment #23 from Rémi Bernon <rbernon@codeweavers.com> --- Could you attach a log file with WINEDEBUG=+pid,+timestamp,+server,+cursor,+win,+event,+x11drv ? Thanks -- 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=60005 --- Comment #24 from Michael Setzer II <mikes@guam.net> --- Created attachment 81619 --> http://bugs.winehq.org/attachment.cgi?id=81619 windows-log env WINE_LARGE_ADDRESS_AWARE=1 WINEARCH=win64 WINEDEBUG=+pid,+timestamp,+server,+cursor,+win,+event,+x11drv wine /home/msetzerii/.wine/drive_c/PMAIL/Programs/WINPM-32.EXE >/home/m Ran env WINE_LARGE_ADDRESS_AWARE=1 WINEARCH=win64 WINEDEBUG=+pid,+timestamp,+server,+cursor,+win,+event,+x11drv wine /home/msetzerii/.wine/drive_c/PMAIL/Programs/WINPM-32.EXE
/home/msetzerii/windows-log 2>&1
Took a while, but then wine window with pegasus just closed completely. Assuming the lines near end might say something. Original file before compressing. 45660770 Jul 30 01:30 windows-log -- 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=60005 --- Comment #25 from Rémi Bernon <rbernon@codeweavers.com> --- Yeah looks like a buffer overflow. Could you check whether https://gitlab.winehq.org/wine/wine/-/merge_requests/11468/diffs?commit_id=a... has 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=60005 --- Comment #26 from Michael Setzer II <mikes@guam.net> --- Took a litte look at windows-log with cat -n windows-log | grep x11drv | tail -n 20; cat -n windows-log |tail -n 10 ; cat -n windows-log | grep x11drv | grep warn last lines with x11drv last lines before crash lines with x11drv and warn Perhaps someting in those lines will point to the issue. small sample 1st and 2nd. 432388 35241.734:0074:0078:trace:x11drv:destroy_whole_window win 0x1004a xwin 3c00007/0 432415 35241.736:0074:0078:trace:x11drv:destroy_whole_window win 0x10048 xwin 3c00006/0 432442 35241.738:0074:0078:trace:x11drv:destroy_whole_window win 0x10046 xwin 3c00005/0 432465 35241.740:0074:0078:trace:x11drv:destroy_whole_window win 0x10044 xwin 3c00004/0 433148 003c: terminate_process( handle=ffffffff, exit_code=0 ) 433149 003c: terminate_process() = 0 { self=1 } 433150 003c: *killed* exit_code=0 433151 0044: *process killed* 433152 006c: *process killed* 433153 008c: *process killed* 433154 00a4: *process killed* 433155 00fc: *process killed* 433156 0038: *process killed* 433157 wineserver: exiting (pid=86835) -- 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=60005 --- Comment #27 from Michael Setzer II <mikes@guam.net> --- Looking at log file with claude.ai and it reports this. Mentions audio, but its not doing any audio?? Found it. Line ~432225: thread 012c calls terminate_process(handle=0000, exit_code=2147319808). That's not a normal clean exit — 2147319808 in hex is 0x80010000, which is an HRESULT with FACILITY_RPC set (the high bit + facility code 0001 = RPC/COM facility). In other words, some COM/OLE component call returned an RPC-facility error, and something (either that component or Pegasus itself) responded by calling ExitProcess with that HRESULT as the exit code — not a system-level crash, but the app tearing itself down after an internal RPC/COM failure. Notably, right before that termination call, 012c was doing registry writes to Software\Microsoft\AudioCompressionManager\DriverCache\... — ACM (Audio Compression Manager) codec cache entries (imaadpcm, msadpcm, msg711, l3acm, msgsm610). So thread 012c was in the middle of ACM/multimedia driver registration when it apparently hit that RPC-facility error and self-terminated. That's very plausibly your actual "freeze point" — the app didn't truly hang, it silently died mid-ACM-registration, and because it was in a modal/parent role, the rest of the process tree (0020, 0120, 0128, etc.) got torn down after it, which is what looks like a frozen/blank window until you force-close. To confirm and pin the specific call that failed, it'd help to re-run with +ole,+rpc added to WINEDEBUG (in addition to what you had) to see exactly which COM interface call returned that error right before the terminate_process: -- 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=60005 --- Comment #28 from Rémi Bernon <rbernon@codeweavers.com> --- The buffer overflow happens after: 012c: send_hardware_message( win=00000000, input={type=MOUSE,x=0,y=2,data=00000000,flags=00000002,time=0,info=00000000,raw_count=67}, flags=00000002, extra={...} The number of rawinput frames is 67, which is larger than what is allowed to keep in the static raw mouse frames buffer. This comes from a bogus flush that has then been fixed with a1bae27f21b53fa3ac6be6b08f4cbfb441360ade. It doesn't immediately cause issues, but it likely causes some memory corruption on wineserver side that later trigger various random errors, including sometimes some invalid hardware messages being sent (and process_hardware_message spam). -- 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=60005 --- Comment #29 from Michael Setzer II <mikes@guam.net> --- Great. Glad the issue has been found. Hope it is fixed in next release. Not sure if there is a way to implement fix now. Right now sticking with 11.12 works. Thanks for info. -- 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=60005 --- Comment #30 from Jonas <winehq@jonass.user.lysator.liu.se> --- (In reply to Rémi Bernon from comment #25)
Yeah looks like a buffer overflow. Could you check whether
Not original author of this issue but I had the same problem. Being back I notice that the correction has been merged so I did git checkout master, git pull and recompiled. I would say that this issue has been resolved and can be closed with the next release of 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=60005 --- Comment #31 from Michael Setzer II <mikes@guam.net> --- I did the git clone, and it showed the patch was already included in the clone. built the wine ./wine --version wine-11.14-98-g4dbf7bef7a0 Ran test with lots of mouse movement, clicking on things, and mouse over buttons, and no issues. For now will stick with 11.12, but sure will be resolved with 11.15. Interesting the build directory crreated was 6.4G while the 11.12 in /opt/wine-devel is 1.8G. But good to confirm the patch found, and fixes the issue. THanks. -- 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=60005 Rémi Bernon <rbernon@codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED Fixed by SHA1| |a1bae27f21b53fa3ac6be6b08f4 | |cbfb441360ade --- Comment #32 from Rémi Bernon <rbernon@codeweavers.com> --- Marking as fixed with a1bae27f21b53fa3ac6be6b08f4cbfb441360ade, please reopen if not. -- 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=60005 Alexandre Julliard <julliard@winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #33 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