FCP-4: FCP_RSP IU Note

Dave.Baldwin at Emulex.Com Dave.Baldwin at Emulex.Com
Wed Sep 22 17:30:32 PDT 2010


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r100922e_f.htm">HTML-formatted message</a>
Attachment #1: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r100922e_image001.gif">image001.gif</a>

Neil,
I agree with you. We see targets doing this often in the environments we plug
into. Removing the note would not be a good thing to do.
Dave
________________________________
Dave Baldwin
Senior Director, Engineering
3333 Susan Street
Costa Mesa, CA 92626
714.885.3547 Office
[cid:image001.gif at 01CB5A7B.DB9DA4D0]<http://www.emulex.com/&gt;
This message contains Emulex confidential information intended only for
specific recipients and is not to be forwarded to anyone else. If you have
received this message in error, please delete it immediately. Thank you.
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Neil
Wanamaker
Sent: Wednesday, September 22, 2010 1:54 PM
To: David Peterson; t10 at t10.org
Subject: RE: FCP-4: FCP_RSP IU Note
But... the note is for implementors of FCP-4-compliant initiators, and
suggests that they should tolerate the old targets. The word 'shall' does not
appear in this note.
If someone new to FCP picks up FCP-4 and implements an initiator, he would
otherwise have no hint that some old targets had this behaviour.
________________________________
From: David Peterson [mailto:dpeterso at Brocade.COM]
Sent: Wednesday, September 22, 2010 3:33 PM
To: Neil Wanamaker; t10 at t10.org
Subject: RE: FCP-4: FCP_RSP IU Note
That is fine, those old devices can then remain compliant to FCP, FCP-2, or
FCP-3...
From: Neil Wanamaker [mailto:Neil_Wanamaker at pmc-sierra.com]
Sent: Wednesday, September 22, 2010 1:29 PM
To: David Peterson; t10 at t10.org
Subject: RE: FCP-4: FCP_RSP IU Note
It isn't really relevant whether current implementations behave this way -
the old devices never seem to completely go away.
________________________________
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of David
Peterson
Sent: Wednesday, September 22, 2010 1:54 PM
To: t10 at t10.org
Subject: FCP-4: FCP_RSP IU Note
FCP-4 still has the following note:
NOTE 7 - Some early implementations presented the FCP_RSP IU without the
FCP_RESID field, FCP_SNS_LEN field, and FCP_RSP_LEN field if the
FCP_RESID_UNDER bit, FCP_RESID_OVER bit, FCP_SNS_LEN_VALID bit, and
FCP_RSP_LEN_VALID bit were all set to zero. This non-standard behavior should
be tolerated.
Are there still implementations that present the FCP_RSP IU without the
FCP_RESID field, FCP_SNS_LEN field, and FCP_RSP_LEN field if the
FCP_RESID_UNDER bit, FCP_RESID_OVER bit, FCP_SNS_LEN_VALID bit, and
FCP_RSP_LEN_VALID bit were all set to zero ?
I would really hope that we have moved past this non-standard behavior and
would like to remove this note.
Thanks...Dave



More information about the T10 mailing list