SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    RE: iSCSI: PDU formats



    I agree with 1 and 2, but I still prefer to overlap Status-Detail with the
    T/C/CNxSG in the same byte for the same reason behind changes 1 and 2. Both
    fields provide login response status details, with meaning interpreted
    differently based on the Status-Class.
    
    -Ayman
    
    
    > -----Original Message-----
    > From: owner-ips@ece.cmu.edu [mailto:owner-ips@ece.cmu.edu]On Behalf Of
    > Robert D. Russell
    > Sent: Wednesday, September 12, 2001 9:44 AM
    > To: Julian Satran
    > Cc: ips@ece.cmu.edu
    > Subject: RE: iSCSI: PDU formats
    >
    >
    > Julian:
    >
    > I would like to request 3 small changes in the format of some of
    > the PDUs.  One of the design features that you have employed
    > very successfully to date is to have a given field, such as the
    > "Initiator Task Tag" for example, always appear in the same position
    > (bytes 16-19) in any PDU in which it appears.  This makes it easy
    > to understand, to implement, and to debug.  However, a few small
    > inconsistencies in the application of this design principle have
    > crept in with draft 7, and I would like to propose that we fix them.
    >
    > 1.  In draft 6 the SCSI Response PDU had one Status/Response field
    > 	in byte 3 -- in draft 7-90 it now has a Status field in byte 2
    > 	and a Response field in byte 3.  In draft 6 the Data-in PDU also
    > 	had a Status field in byte 3, and in draft 7-90 it is still in
    > 	byte 3, with byte 2 unused (reserved).  Would you please either:
    >
    > 	a.	Reorder the bytes in the SCSI Response PDU so that
    > the Status
    > 		field will be in byte 3 (so it is consistent with
    > the Data-in
    > 		PDU) and the Response field will be in byte 2; or
    >
    > 	b.	Move the status field in the Data-in PDU from byte
    > 3 to byte 2
    > 		(so it remains consistent with the SCSI Response PDU).
    >
    > 	I would prefer alternative a. because it would leave the Data-in
    > 	PDU unchanged for drafts 6, 7 and 8, and the SCSI Response PDU
    > 	has to change in any case.  However, obviously either solution
    > 	would work.
    >
    > 2.	In draft 7-90, a field called "Response" appears in 3 PDUs:
    >
    > 	a.	In byte 36 of the Task Management Response PDU.
    > 	b.	In byte 36 of the Logout Response PDU.
    > 	c.	In byte 2 (if my request 1a above is taken) in the
    > 		SCSI Response PDU.  This is clearly inconsistent
    > with a and b.
    >
    > 	Since bytes 2 and 3 are currently unused (reserved) in both
    > 	the Task Management Response PDU and the Logout Response PDU,
    > 	the simplest solution would be to move the "Response" field in
    > 	those 2 PDUs to byte 2 in order to be consistent with the SCSI
    > 	Response PDU.  To keep the design clean, the new "Qualifier" field
    > 	in the Task Management Response PDU should probably also be
    > 	moved to byte 3.
    >
    > 3.	In draft 7-90 the Login and Login Response PDUs have been modified
    > 	with the introduction of the T, C, and CNxSG fields in byte 37.
    > 	However, in the Login Response PDU these fields overlay the
    > 	Status-Detail field, which is also in byte 37.  Although the way
    > 	to interpret this field is uniquely determined by the context,
    > 	it is context dependent and I believe that this will lead to a lot
    > 	of needless errors in coding, and that it also makes debugging more
    > 	difficult, because the use of this byte changes during the
    > login phase
    > 	exchanges.  This means that you can't always look at it the
    > same way.
    > 	To avoid this overlay, would you please move the new fields
    > 	(T, C, and CNxSG) to one of the currently unused bytes.  Many bytes
    > 	(2, 10-11, 20-23, 38-39, 40-47) are currently unused in
    > both of these
    > 	PDUs, so there would appear to be no urgent need to overlay the new
    > 	fields on top of an existing field in order to save space.
    >
    > Thanks,
    >
    > Bob Russell
    > InterOperability Lab
    > University of New Hampshire
    > rdr@iol.unh.edu
    > 603-862-3774
    >
    >
    
    


Home

Last updated: Wed Sep 12 19:17:08 2001
6527 messages in chronological order