> Not sure how to answer that - using XP, it comes up with a standard
> (for XP) crash dialogue that has a Debug button to try and attach to
> it and debug, that's what caused that invalid PID problem. Could that
> message possibly be caused if the crash occurred in the worker thread
> rather than the main thread?

Quite so, but why should it complain about an invalid PID, that's what
puzzles me.  The only thing that comes to my mind is the process must have
been terminated somehow...

> Curious - it does break into the code, but it seems it has gone awry
> in the middle of DllUnregisterServer()! Why would it end up there?!
>
> Here's the stack:
>
> MyCommsDLL! DllUnregisterServer + 2708 bytes MyCommsDLL!
> DllUnregisterServer + 65966 bytes KERNEL32! 7c80b50b()
>
> and here's the offending assembly code:
>
> 02CC1E22   xor         ebx,ebx
> 02CC1E24   mov         ecx,dword ptr [esi+0Ch]        <==== Here
> 02CC1E27   push        edi
> 02CC1E28   push        eax
> 02CC1E29   push        ecx

Probably it doesn't have anything to do with DllUnregisterServer( ).  Have
you built your DLL with debug symbols enabled?  Is it a Debug build, or a
debug-enabled release build?

Also, please look further down into the call stack.  If necessary, post a
complete stack trace here so we can take a look.

-------------
Ehsan Akhgari

Farda Technology (http://www.farda-tech.com/)

List Owner: [email protected]

[ Email: [EMAIL PROTECTED] ]
[ WWW: http://www.beginthread.com/Ehsan ]

Whoever fights monsters should see to it that in the process he does not
become a monster. And when you look long into the abyss, the abyss also
looks into you.
-Beyond Good And Evil, F. W. Nietzsche




_______________________________________________
msvc mailing list
[email protected]
See http://beginthread.com/mailman/listinfo/msvc_beginthread.com for 
subscription changes, and list archive.

Reply via email to