[Guy Harris <guy@xxxxxxxxxx>]

> 	- pass some information from Todd's DCERPC dissector down to
> 	  the subdissectors in the pinfo->private field.  We need to
> 	  know at least the call id from the DCERPC header as well as
> 	  the SMB fid in order to make a unique hash key

It looks as if there's request/response matching being done by the
SPOOLSS dissector.  Is that matching DCERPC-layer requests and
responses? If so, perhaps we need a way for the DCERPC dissector to
allow subdissectors to attach their own private data to entries in *its*
hash table, and get that information when processing a response, rather
than having the subdissectors duplicate its matching.