PAC Progress

[Stefan (metze) Metzmacher]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Stefan (metze) Metzmacher schrieb:
> Luke Howard schrieb:
> 
>>>>I'm not sure it is worth the pidl mangling to 'peek into' that byte
>>>>buffer, and it is the management of that byte buffer that I'm having the
>>>>most trouble with.  To me, it looks like the pointed-to BUFFERS on the
>>>>wire are padded behind to force the next structure to be 8 byte
>>>>aligned. 
>>>
>>>
>>>Have you read the MSDN specification at:
>>>
>>>http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnkerb/html/msdn_pac.asp
>>>
>>>It specifies that all buffer offsets are multiples of 8.
> 
> 
> yep, but that doesn't explain the padding to at the end of the last buffer,
> so this means that the buffersize should be also a multiple of 8 not only the offset

so we should maybe use something like this:

	typedef [nodiscriminant,gensize,flag(NDR_ALIGN8)] union {
		[case(PAC_TYPE_LOGON_INFO)]	PAC_LOGON_INFO logon_info;
		[case(PAC_TYPE_SRV_CHECKSUM)]	PAC_SIGNATURE_DATA srv_cksum;
		[case(PAC_TYPE_KDC_CHECKSUM)]	PAC_SIGNATURE_DATA kdc_cksum;
		[case(PAC_TYPE_LOGON_NAME)]	PAC_LOGON_NAME logon_name;
	} PAC_INFO;

	typedef struct {
		uint32 type;
		[value(ndr_size_PAC_INFO_rounded(info,ndr->flags))] uint32 size;
		[relative,switch_is(type),subcontext(0),subcontext_size(size)] PAC_INFO *info;
		uint32 _pad; /* Top half of a 64 bit pointer? */
	} PAC_BUFFER;

and a wrapper function like this

uint32_t ndr_size_PAC_INFO_rounded(struct PAC_INFO *info, uint32_t flags)
{
	uint32_t s;
	s = ndr_size_PAC_INFO(info, flags);
	s = (s+7) & ~7;
	return s;
}

- --
metze

Stefan Metzmacher <metze at samba.org> www.samba.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3-nr1 (Windows XP)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCyQS1m70gjA5TCD8RAtoRAJ4nx2/r6Ce1KP/sObaF0fOl0B8DWgCgjYWt
fQhrMkQbUP2QPgdXyi8sZ+8=
=hC0y
-----END PGP SIGNATURE-----