[cifs-protocol] RE: erroneous references to little-endian

Richard Guthrie rguthrie at microsoft.com
Tue May 5 20:25:01 GMT 2009


Andrew,

Thank you for the feedback.  I wanted to provide you with a status update at this time.  We have completed our investigation and will be updating the documentation to reflect that NRPC allows RPC to negotiate the endianess on fields with the exception of a subset of byte arrays which NRPC handles explicitly.  We will keep the identifying text for those fields in which NRPC explicitly sets the endianess.  We are working to correct the documentation at this time and once the final changes are complete I will send you the updated NRPC documentation.  Please let us know if you have any further comments/questions.

Richard Guthrie
Support Escalation Engineer
Open Protocols Support Team
Tel: +1 (469) 775-7794
E-mail: rguthrie at microsoft.com 

-----Original Message-----
From: Andrew Bartlett [mailto:abartlet at samba.org] 
Sent: Wednesday, April 15, 2009 7:13 AM
To: Interoperability Documentation Help
Cc: pfif at tridgell.net; cifs-protocol at samba.org
Subject: erroneous references to little-endian

Before many (but oddly, not all) of the impossible-to-parse bitfield diagrams in MS-NRPC is the statement:

> A set of bit flags in little-endian format ... A flag is TRUE (or set) 
> if its value is equal to 1. The value is constructed from zero or more bit flags from the following table.

(search for little-endian)

These refer to bitfields are transferred over DCE/RPC, and as such are in negotiated bit order, as chosen by the client and server.  Therefore the reference to their bit order should be removed.  (This is doubly confusing because the table itself is in bit-endian order. :-)

There are exceptions where the use is actually correct - the IP address and credentials calculations, or in conjunction with non-RPC structures, but the rest appear to be a copy-and-pasted template reproduced though the entire document.

Can you please review this (and other RPC protocols) to ensure that these references are corrected?

Thanks,

Andrew Bartlett

-- 
Andrew Bartlett                                http://samba.org/~abartlet/
Authentication Developer, Samba Team           http://samba.org
Samba Developer, Red Hat Inc.                  http://redhat.com



More information about the cifs-protocol mailing list