[Bug 53181] New: xinput1_3:xinput has a rare crash in Wine

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=53181

            Bug ID: 53181
           Summary: xinput1_3:xinput has a rare crash in Wine
           Product: Wine
           Version: unspecified
          Hardware: x86-64
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: xinput
          Assignee: wine-bugs at winehq.org
          Reporter: fgouget at codeweavers.com
      Distribution: ---

xinput1_3:xinput has a rare crash in Wine:

xinput.c:820: Testing xinput9_1_0.dll
Unhandled exception: page fault on execute access to 0x651c3ae2 in 32-bit code
(0x651c3ae2).
Register dump:
 CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b
 EIP:651c3ae2 ESP:0103fe04 EBP:0103ff28 EFLAGS:00010246(  R- --  I  Z- -P- )
 EAX:00000001 EBX:00000002 ECX:0103fe00 EDX:00000000
 ESI:00000001 EDI:00000001
Stack dump:
0x0103fe04:  651c0000 000007d0 00001cff 00000002
0x0103fe14:  00000000 00000000 00000000 fffffffd
0x0103fe24:  00000000 00000000 00000000 00000000
0x0103fe34:  00000000 00000000 7b60da30 001439d0
0x0103fe44:  0011051e 00000000 00000000 00000000
0x0103fe54:  00000000 00000000 00000000 00000000
Backtrace:
=>0 0x651c3ae2 (0x0103ff28)
  1 0x7b62a1e0 in kernel32 (+0x2a1e0) (0x0103ff48)
  2 0x7bc59b27 in ntdll (+0x59b27) (0x0103ff5c)
  3 0x7bc5a1e0 RtlCreateUserThread(entry=651C3300, arg=0x000000000)
[Z:\home\fgouget\wine\winetest\src\dlls\ntdll\thread.c:261] in ntdll
(0x0103ffec)
0x651c3ae2: -- no code accessible --
[...]
PE    63480000-6349b000    Deferred        version
PE    666c0000-666e8000    Deferred        hid

https://test.winehq.org/data/patterns.html#xinput1_3:xinput

The first known instance of this crash happened on 2022-06-01: there was no
crash from Apr 19 to May 31, but 6 crashes occurred from Jun 01 to Jun 17 (~9%
crash rate). So this looks like a new issue.

The crash address is different every time but always falls in the same range
between version and hid. However I have not been able to reproduce the crash
even over 50 runs. It is possible there is a bad interaction with a previous
test.

-- 
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.