Re: [storm] iSER update request from Alex Nezhinsky on AHS size

[<[e-mail]>]

WG chair hat off:
 
That sounds like a reasonable solution.  What's the typical value
that IB implementations use?  That value will want to be the default
for this new key.

Thanks,
--David


 


________________________________

	From: [e-mail] [mailto:[e-mail]] On
Behalf Of Mike Ko
	Sent: Thursday, August 20, 2009 7:26 PM
	To: STORM
	Subject: [storm] iSER update request from Alex Nezhinsky on AHS
size
	
	
	Alex sent me the following iSER update request:
	 
	> In the Infiniband implementation ISER has to post recv-buffers
for
	> the maximal possible message length. If a PDU contains AHS,
then no
	> upper size limit can be set beforehand, because AHS size is
not fixed
	> (it is determined by the contents of BHS, TotalAHSLength
field). Then
	> our posted buffer may prove insufficient.
	>
	> Suggested solution: to add a hard limit on AHS length as an
	> operational parameter, per session. This is in a sense similar
to
	> MaxOutstandingUnexpectedPDUs key which has no
	> meaning in a stream-based protocol.
	 
	It sounds reasonable to me.  If there is no objection, we can
add the MaxAHSLength key to the iSER update as Alex suggested.
	 
	Mike