Re: iSCSI: MaxCmdSN and the I bit

["Julian Satran" <Julian_Satran@il.ibm.com>]

Eddy,

Comments in text.

Julo

Eddy Quicksall <Eddy_Quicksall@ivivity.com> Sent by: owner-ips@ece.cmu.edu 19-03-02 02:22 Please respond to Eddy Quicksall

To:        "ips@ece. cmu. edu (E-mail)" <ips@ece.cmu.edu>         cc:                 Subject:        iSCSI: MaxCmdSN and the I bit

1) Section 2.2.2.1 Command Numbering and Acknowledging says

    MaxCmdSN - the maximum number to be shipped.
 
But can't anything with the I bit set take on MaxCmdSN+1? If so, should this say "for non-immediate commands, the maximum number ..."? If so, there are a couple of other places to update as well (like 9.4.10).

+++ I'll fix the wording. +++
 
2) Is there any reason why the "I" bit would be used with a SCSI Command? I ask this because all SCSI commands have to pass to the SCSI layer and there would be no way to throttle them when the "I" bit is set.
 
+++ There is no reason to stop an implementor doing it as long as it know that it can neither throttle no recover them.
E.G. a device may require an "urgent mechanical recalibration".  As a rule we did not introduce limitation unless they where mandated by the protocol logic.
+++

3) Section 9.4.9 ExpCmdSN - Next Expected CmdSN from this Initiator says:

    An ExpCmdSN equal to MaxCmdSN+1
   indicates that the target cannot accept new commands.

But it could if the "I" bit was set and your answer to #2 above is "yes", correct?

+++ correct - I'll say non-immediate commands - +++

Eddy