https://bugs.winehq.org/show_bug.cgi?id=55736
Bug ID: 55736 Summary: Solid Edge crashes after a couple of minutes Product: Wine Version: 8.17 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: luca.finizio@protonmail.com Distribution: ---
Created attachment 75230 --> https://bugs.winehq.org/attachment.cgi?id=75230 Terminal output - Wine 8.17
I tried to launch Solid Edge 2023 Community Edition (https://resources.sw.siemens.com/en-US/download-solid-edge-community-edition) and it does, but crashes after a couple of minutes; the crash seems to be independent from what I do and click. In order to reproduce, just download and install the software, then start it, try to create a new part file, and then simply wait for the application to crash. A attach both backtrace and terminal output. I'm using LMDE 6, kernel 6.1, on Lenovo V15-G2, with Wine 8.17.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #1 from luca.finizio@protonmail.com --- Created attachment 75231 --> https://bugs.winehq.org/attachment.cgi?id=75231 Backtrace - Wine 8.17
https://bugs.winehq.org/show_bug.cgi?id=55736
Thomas SMA.TFG@gmx.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |SMA.TFG@gmx.de
--- Comment #2 from Thomas SMA.TFG@gmx.de --- May be I have the same Problem, so i deliver additional Information.
For me it seams to be a Problem in or with Mono as my Backtrace crashes threre.
I attach the backtrace and Solid Edge Log also with the Commandline Output.
Tested with -> wine-8.19 (Staging) -> wine-8.19 (Developemnt) Problem seams to be the same.
OS: Ubuntu 23.10 with wine-8.19 (staging), DELL Precision 3580 i7-1370P 32GB/1TB, Intel and Nvidia RTX A500
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #3 from Thomas SMA.TFG@gmx.de --- Created attachment 75363 --> https://bugs.winehq.org/attachment.cgi?id=75363 wine-8-19_backtrace
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #4 from Thomas SMA.TFG@gmx.de --- Created attachment 75364 --> https://bugs.winehq.org/attachment.cgi?id=75364 wine-8-19_crashlogf
This Logfile could be sent to Siemens for analyses
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #5 from Thomas SMA.TFG@gmx.de --- Created attachment 75365 --> https://bugs.winehq.org/attachment.cgi?id=75365 wine-8-19_env (Loaded Env Vars)
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #6 from Thomas SMA.TFG@gmx.de --- Created attachment 75366 --> https://bugs.winehq.org/attachment.cgi?id=75366 wine-8-19_run (Console Output STD and STDERR)
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #7 from Thomas SMA.TFG@gmx.de --- Created attachment 75367 --> https://bugs.winehq.org/attachment.cgi?id=75367 wine-8-19_sysinfo (OS Hardware Info)
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #8 from Thomas SMA.TFG@gmx.de --- Seams to be a regression Bug as last time I tested with 8.0.1 the Problem haven't occured. See App DB https://appdb.winehq.org/objectManager.php?sClass=version&iId=41292
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #9 from Thomas SMA.TFG@gmx.de --- Tested with Wine8-20 -> same result.
https://bugs.winehq.org/show_bug.cgi?id=55736
Bernhard Übelacker bernhardu@mailbox.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bernhardu@mailbox.org
--- Comment #10 from Bernhard Übelacker bernhardu@mailbox.org --- Created attachment 75558 --> https://bugs.winehq.org/attachment.cgi?id=75558 solid-edge-crash-notes-lldb-backtrace-with-debug-symbols-wine-8.19.txt
Hello, I tried to get a symolized backtrace from this crash. Unfortunately I could just test with a newser Solid Edge version, but I think it shows the same issue.
Attached file contains some unsorted notes how to reproduce and parts of the logs.
Having mono dbgsyms in place in combination with a LLDB this is where the crash happens:
(lldb) up 0x6fffd365f8ea libmono-2.0-x86_64.dll`mono_guid_signature_append_type (res=0x00007fc1a9067090, type=0x00007fc1a87ab080) at metadata.c:8208 8205 for (i = 0; i < type->data.array->rank; ++i) 8206 { 8207 if (i > 0) g_string_append_c (res, ','); -> 8208 if (type->data.array->sizes[i] == 0 || type->data.array->lobounds[i] == 0) continue; 8209 g_string_append_printf (res, "%d", type->data.array->lobounds[i]); 8210 g_string_append (res, "..."); 8211 g_string_append_printf (res, "%d", type->data.array->lobounds[i] + type->data.array->sizes[i] + 1);
https://github.com/madewokherd/mono/blob/ca48d46d3a000a661c746ece82bf5df746b...
The crash happens because "type->data.array->sizes[i]" is accessed, but "type->data.array->sizes" is NULL.
Therefore is this something that has to be fixed in mono?
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #11 from Thomas SMA.TFG@gmx.de --- I've downloaded de and us Version of 2024_2310 both are Idendical with the md5sum. So we have to fix this only once.
Solid Edge 2024 2310 seams to start on my system and crashes if I create a ISO Metric Part. After loading the Screen of ISO Metric Part SolidEdge freeses the first time my whole System freezed on the second try SolidEge freezed
The Crash of 2023 and 2024 is different, 2024 seams to freeze and doesn't react anymore. 2023 Crashes with a Crash dialog and exits.
I would be finde with any of the both Versions running on my Linux System.
I'm sorry that I don't have so much experience to debug the output as @Bernhard Übelacker
It would be realy nice if we could fix this so that a Version is running because then I could eleminate my Win System
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #12 from Bernhard Übelacker bernhardu@mailbox.org --- Created attachment 75562 --> https://bugs.winehq.org/attachment.cgi?id=75562 Patch for wine-mono to avoid the crash if array is "empty".
I got now to build wine-mono myself to continue some experiments.
And it looks like in this case array->sizes is still NULL because array->numsizes is still zero.
Attached patch does a continue in the for loop in case numsizes or numlobounds is zero. A libmono-2.0-x86_64.dll built with this patch does no longer show the crash and Solid Edge stays running.
So I guess this bug should get switched to component mscoree?
https://bugs.winehq.org/show_bug.cgi?id=55736
Bernhard Übelacker bernhardu@mailbox.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |mscoree
--- Comment #13 from Bernhard Übelacker bernhardu@mailbox.org --- Setting component to mscoree.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #14 from Thomas SMA.TFG@gmx.de --- How do we get any resonsible Person that can add the fix to the official Source Rep0?
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #15 from Esme Povirk madewokherd@gmail.com --- This might make sense, but it definitely shouldn't be a hard-coded check for 0. We should probably compare i to numsizes.
Also, I think it's important to check what the spec has to say about these two fields.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #16 from Esme Povirk madewokherd@gmail.com --- Per II.23.2.13 in the spec, this is a valid array type definition. Both numsizes and numlobounds are specified as 0 or more. So accesses of sizes and lobounds should be bounds-checked independently of rank.
I did a quick search of Mono code, and it seems other places working with MonoArrayType do this.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #17 from Thomas SMA.TFG@gmx.de --- Esme Povirk thank you for your engagement.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #18 from Esme Povirk madewokherd@gmail.com --- Skipping lobounds of 0 also seems wrong to me (that's the typical case), so I looked up where this code came from and arrived at https://github.com/mono/mono/pull/20499
So we're actually duplicating the logic from coreclr, therefore we should check there for the correct behavior. The link to the original code is broken, but I think it's this: https://github.com/dotnet/runtime/blob/435e9aa552a3f4fa7348b2a18e12fd5615791...
I'm struggling to find the code that does this specific string conversion though.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #19 from Esme Povirk madewokherd@gmail.com --- I think this is the original code: https://github.com/dotnet/runtime/blob/main/src/coreclr/utilcode/prettyprint...
It looks much the same as Mono's version, complete with the lack of bounds checking, but they allocate the arrays with a size of rank, not the actual sizes, zero the array, and then only assign the values that are in range. So *effectively* this is the same as bounds-checking i within the loop.
Also, it seems they do ignore a lower-bound set explicitly to 0, which seems like a bug to me, but since this is getting hashed to produce a COM interface GUID, we have to preserve the bug.
I made a pull request: https://github.com/madewokherd/mono/pull/57
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #20 from Esme Povirk madewokherd@gmail.com --- Pushed to Wine Mono as https://github.com/madewokherd/wine-mono/commit/1f80e8fe37e103b4ae6e9d085f43...
Should be fixed when Wine Mono 9.1.0 is released.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #21 from Thomas SMA.TFG@gmx.de --- Today I had the Time to compile wine-mono on my own with the current GIT Repo. Now Solid Edge 2024 is running and the Crash is gone.
I tried the following
export WINEPREFIX=~/.winepfx/SE_2024/; winecfg -> Win 10 export WINEPREFIX=~/.winepfx/SE_2024/; wine64 uninstaller #> Unisntall WineMono wineserver -k export WINEPREFIX=~/.winepfx/SE_2024/; wine64 msiexec /i /home/thomas/wine-mono/wine-mono/wine-mono-9.0.99-x86.msi wineserver -k export WINEPREFIX=~/.winepfx/SE_2024/; wine64 /home/thomas/Downloads/VC_redist.x64.exe export WINEPREFIX=~/.winepfx/SE_2024/; wine64 /home/thomas/Downloads/de-DE_Solid_Edge_Community_2024_2310.exe wineserver -k
--> Start of Solid Edge. in Solid Edge
New -> ISO Metric Part -> (Sequential Mode)
Klicked on the "Eye Symbol" on Basis Reference Layers and Pressed STRG+I for ISO Metric View -> now the Layers apper.
Now I klicked on "Sketch" in the Iconbar -> and then on the upper XY Layer -> Sketch View opens
Now I Klicked on the Circle Symbol (Circle over Middle Point) -> and then on the Mittle Point on the Layers. I Pulled it up a bit so that the circle is good to see.
then I tried to close the Sketch view with the green Check Button
.... but nothing happens only the Kontext Menu disapears. Normaly the RibbonBar Icons should change back to the initial view. Even the button Close Sketch on the RibbonBar does nothing
Output at my console:
0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:ole:OleSetMenuDescriptor (0000000000000000, 00000000000100A8, 0000000000000000, 0000000000000000, 000000000A7B03D0), Context sensitive help filtering not implemented! 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:ole:OleSetMenuDescriptor (0000000000000000, 00000000000100A8, 0000000000000000, 0000000000000000, 000000000A7B03D0), Context sensitive help filtering not implemented! 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:appbar:SHAppBarMessage unknown msg: 4 0050:fixme:appbar:handle_appbarmessage SHAppBarMessage(ABM_GETSTATE): stub 0180:fixme:ieframe:ClientSite_GetContainer (000000002741AD20)->(0000000000A1C6A8) 0180:fixme:ieframe:DocHostUIHandler_GetDropTarget (000000002741AD20) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A4FC2B0)->(000000002C38A090) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A4FC380)->(000000002C38A1E0) 0180:fixme:ieframe:PropertyNotifySink_OnChanged unimplemented dispid 1005 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A4FC5F0)->(000000002C34EF40) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A4FC790)->(000000002C34F090) 0180:fixme:mshtml:nsChannel_GetContentLength (000000000A4FC450)->(0000000000A1B6E8) 0180:fixme:wincodecs:FormatConverter_CanConvert Unsupported destination format {6fddc324-4e03-4bfe-b185-3d77768dc900} 0180:fixme:wincodecs:mqw_SetMetadataByName iface 000000001314D160, name L"/dc:title", value 0000000000A1C890 stub. 0180:fixme:mshtml:HTMLElement2_getAdjacentText (000000002C5EABC0)->(L"beforeEnd" 0000000000A1CCF8) 0180:fixme:mshtml:HTMLElement2_replaceAdjacentText (000000002C5EABC0)->(L"beforeEnd" L" Skizze schlie\00dfen" 0000000000A1CCF8) 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C2B24E0)->(0000000000A1C878) 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C2B2520)->(0000000000A1C878) 0180:fixme:mshtml:HTMLDocument_createStyleSheet semi-stub for href L"C:\users\thomas\AppData\Roaming\Siemens\Solid Edge\Version 224\ui_content\theme.css" 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C2B2660)->(0000000000A1C878) 0180:fixme:mshtml:HTMLStyleSheet_put_href (000000002C2B26A0)->(L"") 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C2B26A0)->(0000000000A1C878) 0180:fixme:mshtml:HTMLBodyElement_get_bottomMargin (000000002BDAC9E0)->(0000000000A1CE38) 0180:fixme:mshtml:HTMLBodyElement_get_topMargin (000000002BDAC9E0)->(0000000000A1CE20) 0180:fixme:mshtml:HTMLBodyElement_get_leftMargin (000000002BDAC9E0)->(0000000000A1CE50) 0180:fixme:mshtml:HTMLBodyElement_get_rightMargin (000000002BDAC9E0)->(0000000000A1CE08) 0180:fixme:mshtml:nsChannel_GetContentLength (000000000A4AAA40)->(0000000000A1B898) 0180:fixme:ieframe:ClientSite_GetContainer (000000002741AD20)->(0000000000A1C6A8) 0180:fixme:ieframe:DocHostUIHandler_GetDropTarget (000000002741AD20) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A454030)->(000000002C34F950) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A454100)->(000000002C34FAA0) 0180:fixme:ieframe:PropertyNotifySink_OnChanged unimplemented dispid 1005 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A454440)->(000000002C34FB10) 0180:fixme:mshtml:nsChannel_AsyncOpen2 (000000000A454510)->(000000002C616060) 0180:fixme:mshtml:nsChannel_GetContentLength (000000000A4542A0)->(0000000000A1B6E8) 0180:fixme:wincodecs:FormatConverter_CanConvert Unsupported destination format {6fddc324-4e03-4bfe-b185-3d77768dc900} 0180:fixme:wincodecs:mqw_SetMetadataByName iface 000000002BDBC0D0, name L"/dc:title", value 0000000000A1C890 stub. 0180:fixme:mshtml:HTMLElement2_getAdjacentText (000000000AA7C620)->(L"beforeEnd" 0000000000A1CCF8) 0180:fixme:mshtml:HTMLElement2_replaceAdjacentText (000000000AA7C620)->(L"beforeEnd" L" Skizze schlie\00dfen" 0000000000A1CCF8) 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C67DF60)->(0000000000A1C878) 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C67DFA0)->(0000000000A1C878) 0180:fixme:mshtml:HTMLDocument_createStyleSheet semi-stub for href L"C:\users\thomas\AppData\Roaming\Siemens\Solid Edge\Version 224\ui_content\theme.css" 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C67E0E0)->(0000000000A1C878) 0180:fixme:mshtml:HTMLStyleSheet_put_href (000000002C67E120)->(L"") 0180:fixme:mshtml:HTMLStyleSheet_get_imports (000000002C67E120)->(0000000000A1C878) 0180:fixme:mshtml:HTMLBodyElement_get_bottomMargin (000000002BDAD160)->(0000000000A1CE38) 0180:fixme:mshtml:HTMLBodyElement_get_topMargin (000000002BDAD160)->(0000000000A1CE20) 0180:fixme:mshtml:HTMLBodyElement_get_leftMargin (000000002BDAD160)->(0000000000A1CE50) 0180:fixme:mshtml:HTMLBodyElement_get_rightMargin (000000002BDAD160)->(0000000000A1CE08) 0180:fixme:mshtml:nsChannel_GetContentLength (000000002C4E2A30)->(0000000000A1B898)
@Bernhard Übelacker do you have the same Problem? PS a big Thank you for debugging the initial Crash with Mono.
my current wine is 9.1 testing
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #22 from Bernhard Übelacker bernhardu@mailbox.org --- Hello Thomas, I am sorry but I know Solid Edge just from this bug report. Thanks for testing and reporting back. Please put different issues into their own bug reports, and please put logging output into an attached file.
https://bugs.winehq.org/show_bug.cgi?id=55736
--- Comment #23 from Thomas SMA.TFG@gmx.de --- I'll open a new Bug Report for the Issue.
This Bug can be closed in my opinion.
https://bugs.winehq.org/show_bug.cgi?id=55736
Esme Povirk madewokherd@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |e00762b961b78b46be3a9ddaf0c | |fb74450de1427 Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED
--- Comment #24 from Esme Povirk madewokherd@gmail.com --- Resolving with Wine Mono 9.1.0 release.
https://bugs.winehq.org/show_bug.cgi?id=55736
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #25 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 9.8.