[Bug 51800] UserBenchmark Crash

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=51800

Bernhard Übelacker <bernhardu at mailbox.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bernhardu at mailbox.org

--- Comment #1 from Bernhard Übelacker <bernhardu at mailbox.org> ---
In the "Logs" attachement is the instruction "int $0x29" shown.
This special interrupt seems related to _KiRaiseSecurityCheckFailure.

Short before this "int $0x29" there is also a function call
IsProcessorFeaturePresent(feature=23/*PF_FASTFAIL_AVAILABLE*/) done.

Searching the internet mainly returns results for the
"Buffer Security Check" in the Microsoft compilers.

I am not sure if this is related here too, but looking a little further
in the executed function shows following:
  d3d11_device_context_QueryInterface {b2daad8b-03d4-4dbf-95eb-32ab4b63d0ab}
  not implemented, returning E_NOINTERFACE.

This leads to this bug: https://bugs.winehq.org/show_bug.cgi?id=51354
And there is a "stub" patch attached, which
starts implementing ID3DUserDefinedAnnotation.

With this patch applied the "UBMSkillBench.exe" seems to work
and shows its user interface.

(
 Unfortunately to reach this fault the whole benchmark has to be run,
 the single executable just "waits for something".
 Therefore starting with following WINEDLLOVERRIDES does speedup
 things to reach the issue:
 "UBMCPUBench.exe=b;UBMRAMBench.exe=b;SHADOW.exe=b;UBMGPUStats.exe=b"
)

-- 
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.