Hi Guy,
see my comments in-line.
Best regards
Michael
On Mar 26, 2005, at 22:31 Uhr, guy@xxxxxxxxxxxx wrote:
User: guy
Date: 2005/03/26 03:31 PM
Log:
Update the URL for draft-stewart-sctp-pktdrprep to the -02 draft.
Clean up indentation.
Have the main loop for dissecting chunks check that the chunk size is
large enough for the chunk header, and have it pass the chunk size,
minus the size of the chunk header, to dissectors for particular chunk
types. Make those dissectors check that value to make sure it's large
enough for any fixed-length portion before subtracting the length of
that portion and using the result as a remaining data length.
Why?
The principle I based the dissector on was that if the tvb is too short
I will call proto_tree_add_item() with an index out of bound. This
raised
the 'malformed frame' exception and that is fine.
The old style was used with some thinking to handle the different cases
of chunk length values and the length of the tvbs (which is not always
equal)
correctly.
Directory: /trunk/epan/dissectors/
Changes Path Action
+122 -93 packet-sctp.c Modified
http://anonsvn.ethereal.com/viewcvs/viewcvs.py?rev=13924&view=rev
_______________________________________________
Ethereal-cvs mailing list
Ethereal-cvs@xxxxxxxxxxxx
http://www.ethereal.com/mailman/listinfo/ethereal-cvs
