[Bug 55809] New: FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Bug ID: 55809
           Summary: FEMM 4.2 mesh creation won't run, running an analysis
                    is impossible, works on wine-stable
           Product: Wine
           Version: 8.18
          Hardware: x86-64
                OS: Linux
            Status: UNCONFIRMED
          Severity: major
          Priority: P2
         Component: -unknown
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

Created attachment 75299
  --> https://bugs.winehq.org/attachment.cgi?id=75299
backtrace and test file

FEMM 4.2 built in tool triangle.exe crashes every time an analysis (analysis ->
analyze) or mesh creation (mesh -> create mesh) is ran.

the backtrace is attached, a test file (.FEE) - drawn terribly but enough for a
test, crashes with others too - is attached too.

Everything works on the stable version.

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Bernhard Übelacker  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #1 from Bernhard Übelacker  ---
Hello,
I was able to reproduce a crash, without using native dlls.

https://www.femm.info/wiki/Download
https://www.femm.info/wiki/Files/files.xml?action=download&file=femm42bi...

A git bisect leads here:
a81c53504ae32715e6e91bd020fdebd5bef20d48 is the first broken commit
commit a81c53504ae32715e6e91bd020fdebd5bef20d48
Author: Alexandre Julliard 
Date:   Thu Sep 28 16:30:32 2023 +0200
    ntdll: Release the low address space reservation for 64-bit apps.

The partial revert of it like in the other bugs about a81c53504a makes
the crash here also go away:
--- a/dlls/ntdll/unix/virtual.c
+++ b/dlls/ntdll/unix/virtual.c
@@ -4400,3 +4400,3 @@ void virtual_set_large_address_space(void)
         else
-            free_reserved_memory( 0, (char *)0x7ffe0000 );
+            ;//free_reserved_memory( 0, (char *)0x7ffe0000 );
 #endif

Looks like another application relying on pointers stay below 32bit.
So the bug might be in the application, but does not cause a crash in windows.

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Fabian Maurer  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]
             Status|UNCONFIRMED                 |RESOLVED
    Regression SHA1|                            |a81c53504ae32715e6e91bd020f
                   |                            |debd5bef20d48
         Resolution|---                         |DUPLICATE

--- Comment #2 from Fabian Maurer  ---
Thanks for the analysis. Marking as dupe of 55833 to collect similar issues.

*** This bug has been marked as a duplicate of bug 55833 ***

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Alistair Leslie-Hughes  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #3 from Alistair Leslie-Hughes  ---
Closing duplicate.

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

--- Comment #4 from Bernhard Übelacker  ---
Confirming commit f03c3a16 fixes this crash.

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Fabian Maurer  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|DUPLICATE                   |FIXED
      Fixed by SHA1|                            |f03c3a167c2e4abe92b1e1bf2ea
                   |                            |5f7c31a07fc3b

--- Comment #5 from Fabian Maurer  ---
Marking fixed by
https://source.winehq.org/git/wine.git/commitdiff/f03c3a167c2e4abe92b1e1bf2e...

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|CLOSED                      |RESOLVED

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

--- Comment #6 from Alexandre Julliard  ---
Confirming application bug, it's using unsigned long (32-bit) for pointer
arithmetic:

/* Move to the new block. */
      pool->nowblock = (VOID **) *(pool->nowblock);
      /* Find the first item in the block.    */
      /*   Increment by the size of (VOID *). */
      alignptr = (unsigned long) (pool->nowblock + 1);
      /* Align the item on an `alignbytes'-byte boundary. */
      pool->nextitem = (VOID *)
        (alignptr + (unsigned long) pool->alignbytes -
         (alignptr % (unsigned long) pool->alignbytes));

(in triangle/triangle.c:poolalloc)

[Bug 55809] FEMM 4.2 mesh creation won't run, running an analysis is impossible, works on wine-stable

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=55809

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #7 from Alexandre Julliard  ---
Closing bugs fixed in 9.0-rc2.