Minutes of SAS Protocol Teleconference
lohmeyer at t10.org
Tue May 14 11:27:03 PDT 2002
* From the T10 Reflector (t10 at t10.org), posted by:
* John Lohmeyer <lohmeyer at t10.org>
A PDF version of these minutes is available at:
Minutes of SAS Protocol Teleconference - May 14, 2002
The teleconference was held from 10:00 am to 12:50 pm CDT. At least
the following people were present on the call:
The agenda was:
Scrambler bit order [Coomes]
CRC code sample [Coomes]
SAS OOB Timing [Elliott]
SAS Wide Link Rules (02-170) [Galloway]
SAS XFER_RDY Rules (02-169) [Galloway]
Review of SAS document (02-157) [Elliott]
SAS PHY MARGIN CONTROL request (02-179) [Roberts]
Items added at the meeting:
BREAK impact to transport layer [Hall]
Define behavior if extra ACK or NACK received [Hall]
Primitive to say all credit has been sent [Galloway]
Error cases in AEN [Galloway]
Hash in or out? [Galloway]
Fairness timing issue? [Hoglund]
What does it mean to ignore or discard a frame? [Galloway]
** SAS XFER_RDY Rules (02-169) [Galloway]
Bill Galloway reviewed 02-169r1, SAS XFER_RDY Rules. He included a
Relative Offset definition in the transfer ready frame anticipating
adding it in a future version of SAS. The group agreed to leave this
field reserved. Bill will prepare rev 2, backing out this change.
In absence of objection, the group agreed to recommend that 02-169r2
be accepted for inclusion in the SAS working draft.
** SAS Wide Link Rules (02-170) [Galloway]
Bill Galloway reviewed 02-170r0, SAS Wide Link Rules. He received
In absence of objection, the group agreed to recommend that 02-170r1
be accepted for inclusion in the SAS working draft.
** BREAK impact to transport layer [Hall]
Dana Hall asked about the impact of receiving a BREAK primitive on the
transport layer. Receiving a BREAK while the protocol-specific state
machine is not idle should cause the associated I/O to be aborted. We
need to work out exactly how this is accomplished between the various
state machines. Dana Hall will write a proposal.
** Define behavior if extra ACK or NACK received [Hall]
Dana Hall requested that we define the behavior to be taken when a
device detects that an extra ACK or NACK was received. There was a
discussion of how to handle this situation.
Rob suggested that the target should report this error as a unit
attention condition with a new ASC/ASCQ. The initiator should send
Tim Hoglund suggested that the link layer should send BREAK and notify
the higher layers of the error so it can be reported.
Dana Hall agreed to write a more detailed proposal.
** Primitive to say all credit has been sent [Galloway]
Bill Galloway noted that this topic had originated at LSI with a
suggestion that another reason code be created for the DONE primitive.
Bill suggested that DONE would probably have already been sent, so he
suggested that some new primitive would be a better approach.
Bill said there are two cases, 1) there will never be more credit for
this connection no matter how long the other device waits, and 2)
there won't be more credit for at least a millisecond. The group
agreed that we could merge these two cases.
Bill Galloway agreed to write a more detailed proposal.
** Error cases in AEN [Galloway]
Bill Galloway noted that we need to remember to document the AEN error
cases. He agreed to send George an email to this effect.
** Hash in or out? [Galloway]
In Nashua, the question came up of do we really need the hashed 24-bit
SID and DID addresses in frames? The CRC protection on the open frames
renders this feature largely unnecessary. Should the hash values be
checked at the link layer or the transport layer? No decision was
made. We need to close on this question at the June meeting.
Bill Galloway agreed to write up two proposals, one proposing that
hashing be eliminated and the other that hashing be checked at the
** Fairness timing issue? [Hoglund]
Tim Hoglund said there is a potential ambiguity problem with the AWT
timers in that the values may be different from expander to expander.
Bill said that if you compare what you sent to what you received, then
there should not be ambiguity. Tim was concerned that the actual
wording in the working draft might not be clear. Tim will review the
SAS working draft wording and propose changes, if necessary.
** What does it mean to ignore or discard a frame? [Galloway]
Bill Galloway wanted to explicitly say for each error case whether a
frame is ACK'd or NACK'd, is RRDY sent, and is the command aborted.
Bill Galloway agreed to write a proposal.
** Recommendations to T10:
that 02-169r2 be accepted for inclusion in the SAS working draft
that 02-170r1 be accepted for inclusion in the SAS working draft
** Meeting Schedule
A SAS Protocol teleconference is scheduled for May 21, 2002:
Tuesday May 21, 2002
11:00 am -- 2:00 pm EDT
10:00 am -- 1:00 pm CDT
9:00 am -- noon MDT
8:00 am -- 11:00 am PDT
USA Toll Free Number: 877-417-9828
Toll Number: +1-706-679-9028
Conference Code: 719 533 7560 (same as my phone number)
WebEx Session: http://seagate.webex.com/
Meeting Name: SAS Protocol teleconference
Password: to be announced on T10 reflector
Face-to-face SAS Protocol meetings are scheduled for:
June 5-7, 2002 in Minneapolis, MN. See
http://www.t10.org/ftp/t10/sas0605.txt for meeting information. There
will also be a concurrent SAS Physical meeting.
June 24-26, 2002 in Denver, CO. See
http://www.t10.org/ftp/t10/sas0624.txt for meeting information. (No
concurrent Physical meeting.)
July 15-16, 2002 in Colorado Springs, CO with T10. See
http://www.t10.org/ftp/t10/announce/ann-m050.pdf for meeting
information and http://www.t10.org/ftp/t10/mtg_map.txt for meeting
times. There will also be a concurrent SAS Physical meeting.
John Lohmeyer Email: lohmeyer at t10.org
LSI Logic Corp. Voice: +1-719-533-7560
4420 ArrowsWest Dr. Fax: +1-719-533-7183
Colo Spgs, CO 80907
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org
More information about the T10