Re: secur32: First small test for a Kerberos SSP. (try 2)

[Kai Blin]

On Tuesday 15 July 2008 14:23:29 Kai Blin wrote:
> -
> - ret = pAcquireCredentialsHandleA(NULL, sec_pkg_name,
>
> SECPKG_CRED_INBOUND, +            NULL, NULL, NULL, NULL, &cred, &ttl);
> - pFreeCredentialsHandle(&cred);
> -

Of course I missed the ok() check, and of course it's failing. Seems like the 
patch won't be that small after all. Disregard for now, I'll resend if I have 
a full patch that works on windows for both server and client side.

Cheers,
Kai

uhandled exception error

I have finally tracked this down and have the patch for an unhandled exception error.

The error occurs in? provider.c in routine :

static HRESULT DXDiag_InitDXDiagDirectShowFiltersContainer(IDxDiagContainer* pSubCont) which is in provider.c in dxdiagn dll.

I won't post the whole routine as I rewrote it and cleaned it up considerably and removed the GOTO's and so forth which are not part of the fix..

The fix is down at the bottom of the routine there is a call to IEnumMoniker_Release(pEnum) and when the call is done for some reason pEnum is still 
NULL and is not checked for so there is the following :

fixme:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer        ClassEnumerator for clsid({cc7bfb41-f175-11d1-a392-00e0291f3959}) pEnum((nil))
trace:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer Enum Moniker release
trace:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer pEnum <(null)>
wine: Unhandled page fault on read access to 0x00000000 at address 0x61091d2b (thread 0009), starting debugger...
Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0x61091d2b).

by putting in a check for NULL before the call to IEnumMoniker_Release the pagefault is now gone.

Now mind you this should be done. However, it doesnt explain why the pEnum is NULL.

Since I cleaned up the routine and added this fix to the pagefault how would I submit the cleaned up routine?

I know with the bug I will put the patch on the Bugzilla page for the page fault and also put it on the patch list as well. But how do we 
submit cleaned up code?

I have this thing over using goto's and started clearing the code up and found that it was actually smaller when I was done and alot cleaner to follow. 
Goto's tend to make debugging very difficult.

Chris

uhandled exception error

[Austin English]

On Tue, Jul 15, 2008 at 4:21 PM,   wrote:
> I have finally tracked this down and have the patch for an unhandled
> exception error.
>
> The error occurs in  provider.c in routine :
>
> static HRESULT DXDiag_InitDXDiagDirectShowFiltersContainer(IDxDiagContainer*
> pSubCont) which is in provider.c in dxdiagn dll.
>
> I won't post the whole routine as I rewrote it and cleaned it up
> considerably and removed the GOTO's and so forth which are not part of the
> fix..
>
> The fix is down at the bottom of the routine there is a call to
> IEnumMoniker_Release(pEnum) and when the call is done for some reason pEnum
> is still
>
> NULL and is not checked for so there is the following :
>
> fixme:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer
> ClassEnumerator for clsid({cc7bfb41-f175-11d1-a392-00e0291f3959})
> pEnum((nil))
>
> trace:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer Enum Moniker
> release
>
> trace:dxdiag:DXDiag_InitDXDiagDirectShowFiltersContainer pEnum <(null)>
>
> wine: Unhandled page fault on read access to 0x00000000 at address
> 0x61091d2b (thread 0009), starting debugger...
>
> Unhandled exception: page fault on read access to 0x00000000 in 32-bit code
> (0x61091d2b).
>
> by putting in a check for NULL before the call to IEnumMoniker_Release the
> pagefault is now gone.
>
> Now mind you this should be done. However, it doesnt explain why the pEnum
> is NULL.
>
> Since I cleaned up the routine and added this fix to the pagefault how would
> I submit the cleaned up routine?
>
> I know with the bug I will put the patch on the Bugzilla page for the page
> fault and also put it on the patch list as well. But how do we
>
> submit cleaned up code?
>
> I have this thing over using goto's and started clearing the code up and
> found that it was actually smaller when I was done and alot cleaner to
> follow.
>
> Goto's tend to make debugging very difficult.
>
> Chris
>
> The Famous, the Infamous, the Lame - in your browser. Get the TMZ Toolbar
> Now!

Patches go to [e-mail]

Also, please don't use HTML e-mail...plain text is preferred.