Re: iSCSI: Data-In PDU having MaxBurstSize restriction for a sequence

["Mallikarjun C." <cbm@rose.hp.com>]

Sanjay,

There were earlier messages on this in the archive.

One of the reasons we decided to require F-bit for every
MaxBurstSize of Data-In is to ensure that initiator may use that
as a hint to send Data-Out in the case of bidirectional commands.
The truth is that we don't know if such a hint is required, but 
it is architected for a possible use if implementations need it.

Note that A-bit is not related to F-bit really - except both bits
are somehow related to MaxBurstSize ("at most once" Vs "at least
once").

Hope that helps.
--
Mallikarjun

Mallikarjun Chadalapaka
Networked Storage Architecture
Network Storage Solutions
Hewlett-Packard MS 5668 
Roseville CA 95747
cbm@rose.hp.com


----- Original Message ----- 
From: "Sanjay Goyal" <sanjay_goyal@ivivity.com>
To: <ips@ece.cmu.edu>
Sent: Thursday, May 23, 2002 2:01 PM
Subject: iSCSI: Data-In PDU having MaxBurstSize restriction for a sequence


> Hi
>  What is the rational behind having MaxBurstSize constraining a Sequence for
> Data-In PDUs?
>  As Initiator has to receive all the data from Target as Data-In PDUs
> anyway, what good is it to set F-bit every MaxBurstSize? It rather should be
> set, after whatever amount of data Target has in its buffer has been
> transmitted.
>  Would somebody suggest a good example when Initiator can take advantage of
> F-bit being set every MaxBurstSize.
>  I understand that A-bit is used in conjuction with MaxBurstSize, however my
> context of the question is when A-bit is not used.
> 
> Regards
> Sanjay Goyal
>