|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: iSCSI: SCSI Response Hdr - Sense Data LengthI
believe you misunderstood the point I was emphasizing. The
current format requires the initiator to parse two headers before it knows
where all the messages begin. Assuming an initiator is trying to be
efficient and avoid data copies as much as possible, it would do something like
this:
After
step #3, the initiator can setup the scatter/gather list of where the
sense-data and receive data should be placed.
Ideally, you want as few "stops" as possible.
Once the length of all segments are known, it's generally trivial to program
where the hardware should put everything. If you allow several
intermediate steps (the worst-case scenario), then some sort of logic
(software/hardware) will have to decide how to handle the PDU during a
stop.
By
putting the sense-length in the BHS, you avoid the additional
stop-and-check-the-header step of #3. I really see no advantage to putting
the sense-length out of the BHS! Once you know where everything
belongs, it's not unusual to be able to program hardware to handle the rest of
the message segments:
I also
don't like seeing statements like: "in most of the cases". By putting
extra complexity in the design, it means that all software and hardware MUST
support it. By saying that most of the time you won't use the feature or
message, just says the extra complexity was a
waste.
"In most of the cases the data segment will contain
nothing."
I see
nothing in the draft that prevents targets from loading sense and/or data
the SCSI Response (0x21) message. In fact, it seems to make sense that
an SCSI Command (0x01) might be responded to with a SCSI Response (0x21) by
default with some implementations.
Thanks.
Home Last updated: Mon Apr 22 19:18:19 2002 9753 messages in chronological order |