Return-Path: X-Sieve: cmu-sieve 2.0 Return-Path: Received: (from majordom@localhost) by ece.cmu.edu (8.11.0/8.10.2) id h1L6WNR27066 for ips-outgoing; Fri, 21 Feb 2003 01:32:23 -0500 (EST) X-Authentication-Warning: ece.cmu.edu: majordom set sender to owner-ips@ece.cmu.edu using -f Received: from d12lmsgate-4.de.ibm.com (d12lmsgate-4.de.ibm.com [194.196.100.237]) by ece.cmu.edu (8.11.0/8.10.2) with ESMTP id h1L6WLW27062 for ; Fri, 21 Feb 2003 01:32:21 -0500 (EST) Received: from d12relay02.de.ibm.com (d12relay02.de.ibm.com [9.165.215.23]) by d12lmsgate-4.de.ibm.com (8.12.3/8.12.3) with ESMTP id h1L6W1AN076440; Fri, 21 Feb 2003 07:32:01 +0100 Received: from d10ml001.telaviv.ibm.com (d10ml001.telaviv.ibm.com [9.148.216.55]) by d12relay02.de.ibm.com (8.12.3/NCO/VER6.5) with ESMTP id h1L6VxcP259634; Fri, 21 Feb 2003 07:31:59 +0100 In-Reply-To: <7D17958F8536B640885BE12A446D548F02E6F1@hq-ex-4.brocade.com> To: Dennis Young Cc: "'Buck Landry'" , ips@ece.cmu.edu, Lakshmi Ramasubramanian , owner-ips@ece.cmu.edu Subject: RE: iSCSI: Data In PDU with S bit set MIME-Version: 1.0 X-Mailer: Lotus Notes Release 6.0.1 February 07, 2003 From: "Julian Satran" Message-ID: Date: Fri, 21 Feb 2003 08:31:56 +0200 X-MIMETrack: Serialize by Router on D10ML001/10/M/IBM(Release 5.0.9a |January 7, 2002) at 21/02/2003 08:31:59, Serialize complete at 21/02/2003 08:31:59 Content-Type: multipart/alternative; boundary="=_alternative 001EE2E8C2256CD4_=" Sender: owner-ips@ece.cmu.edu Precedence: bulk This is a multipart message in MIME format. --=_alternative 001EE2E8C2256CD4_= Content-Type: text/plain; charset="US-ASCII" Dennis, Target has to send a single status (either in the data or in a separate response). The last statement says only that initiator should be handle both types of responses but it does not say that it has to support them on a single command. As Eddy has already pointed out the task ceases to exist as soon as the status is received. Sending a second status is WRONG and may inadvertently "close" at the initiator a subsequent command using the same ITT. Julo Dennis Young Sent by: owner-ips@ece.cmu.edu 20/02/03 04:47 To "'Buck Landry'" , Lakshmi Ramasubramanian cc ips@ece.cmu.edu Subject RE: iSCSI: Data In PDU with S bit set In section 10.7 (draft 20): Status can accompany the last Data-in PDU if the command did not end with an exception (i.e., the status is "good status" - GOOD, CONDITION MET or INTERMEDIATE CONDITION MET). The presence of status (and of a residual count) is signaled though the S flag bit. Although targets MAY choose to send even non-exception status in separate responses, initiators MUST support non-exception status in Data-In PDUs. Based on the last sentence of the above paragraph, the initiator has to be able to handle both cases, while the target has the option to not send a separate response if there is no exception. -Dennis >-----Original Message----- >From: Buck Landry [mailto:blandry@crossroads.com] >Sent: Wednesday, February 19, 2003 5:18 PM >To: Lakshmi Ramasubramanian >Cc: ips@ece.cmu.edu; Julian_Satran@il.ibm.com >Subject: RE: iSCSI: Data In PDU with S bit set > > >Well, I don't see any explicit text (too bad), but.. > >3.5.1.5 "the last SCSI Data-in PDU passed for a command may >also contain the status if the status indicates termination >with no exceptions (no sense or response involved)" > >.. implies (maybe just to me) that the SCSI task is terminated >by a Data-In PDU w/S bit set. A further response to a >terminated task would be inappropriate methinks, unless you >want to start talking about error recovery.. > >Hope this helps, --buck > > -----Original Message----- >From: Lakshmi Ramasubramanian [mailto:nramas@windows.microsoft.com] >Sent: Wednesday, February 19, 2003 4:55 PM >To: ips@ece.cmu.edu >Subject: iSCSI: Data In PDU with S bit set > > >If the target sends a Data In PDU with S bit set, >can it also send a SCSI Response PDU for the same command? > >If the answer is no, there is no text in the spec that >says a target should not send scsi response PDU. > >thanks! > -lakshmi > > > > > --=_alternative 001EE2E8C2256CD4_= Content-Type: text/html; charset="US-ASCII"
Dennis,

Target has to send a single status (either in the data or in a separate response). The last statement says only that initiator should be handle both types of responses
but it does not say that it has to support them on a single command. As Eddy has already pointed out the task ceases to exist as soon as the status is received.

Sending a second status is WRONG and may inadvertently "close" at the initiator a subsequent command using the same ITT.


Julo



Dennis Young <DYoung@Brocade.COM>
Sent by: owner-ips@ece.cmu.edu

20/02/03 04:47

To
"'Buck Landry'" <blandry@crossroads.com>, Lakshmi Ramasubramanian <nramas@windows.microsoft.com>
cc
ips@ece.cmu.edu
Subject
RE: iSCSI: Data In PDU with S bit set





In section 10.7 (draft 20):

   Status can accompany the last Data-in PDU if the command did not end
  with an exception (i.e., the status is "good status" - GOOD,
  CONDITION MET or INTERMEDIATE CONDITION MET).  The presence of
  status (and of a residual count) is signaled though the S flag bit.
  Although targets MAY choose to send even non-exception status in
  separate responses, initiators MUST support non-exception status in
  Data-In PDUs.

Based on the last sentence of the above paragraph, the initiator has
to be able to handle both cases, while the target has the option to
not send a separate response if there is no exception.

-Dennis

>-----Original Message-----
>From: Buck Landry [mailto:blandry@crossroads.com]
>Sent: Wednesday, February 19, 2003 5:18 PM
>To: Lakshmi Ramasubramanian
>Cc: ips@ece.cmu.edu; Julian_Satran@il.ibm.com
>Subject: RE: iSCSI: Data In PDU with S bit set
>
>
>Well, I don't see any explicit text (too bad), but..
>
>3.5.1.5 "the last SCSI Data-in PDU passed for a command may
>also contain the status if the status indicates termination
>with no exceptions (no sense or response involved)"
>
>.. implies (maybe just to me) that the SCSI task is terminated
>by a Data-In PDU w/S bit set.  A further response to a
>terminated task would be inappropriate methinks, unless you
>want to start talking about error recovery..
>
>Hope this helps, --buck
>
> -----Original Message-----
>From: Lakshmi Ramasubramanian [mailto:nramas@windows.microsoft.com]
>Sent: Wednesday, February 19, 2003 4:55 PM
>To: ips@ece.cmu.edu
>Subject: iSCSI: Data In PDU with S bit set
>
>
>If the target sends a Data In PDU with S bit set,
>can it also send a SCSI Response PDU for the same command?
>
>If the answer is no, there is no text in the spec that
>says a target should not send scsi response PDU.
>
>thanks!
> -lakshmi
>
>
>
>
>

--=_alternative 001EE2E8C2256CD4_=--