SPI question - IU Transfers Enabled

Gerry.Houlder at seagate.com Gerry.Houlder at seagate.com
Mon Sep 22 11:20:41 PDT 2003

* From the T10 Reflector (t10 at t10.org), posted by:
* Gerry.Houlder at seagate.com

The original reason for this is that the group intended to specify the
"power on default" transfer rate, which is asynchronous narrow. This is the
usual default when major confusion about transfer rate agreement has been

Another interesting item about this paragraph is whether the drive is
required to send SYNCHRONOUS DATA TRANSFER REQUEST message after the WDTR
message exchange. WDTR description states that an asynchronous transfer
rate shall be implied after successful WDTR exchange, but there is another
spot in the standard that requires targets to always send SDTR after
successfully sending WDTR. I am aware of target implementations that go bus
free both with and without the SDTR message after WDTR exchange. Most
initiators seem to tolerate both cases.

                      (HP-UnitedKingdom        To:       "T10 Reflector (t10 at t10.org)" <t10 at t10.org>                      
                      ,ex2)"                   cc:                                                                        
                      <michael.banther@        Subject:  SPI question - IU Transfers Enabled                              
                      Sent by:                                                                                            
                      owner-t10 at t10.org                                                                                   
                      No Phone Info                                                                                       
                      09/22/2003 08:23                                                                                    

Sub-clause  of SPI-5r05, states:

If the first message received by the  SCSI target port during theMESSAGE
OUT  phase is not a TARGET RESET message or a PPR message the SCSI target
port  shallchange to a MESSAGE IN phase and  issue a MESSAGE REJECT message
then originate WDTRnegotiation (see with the TRANSFER  WIDTH
EXPONENT field set to 00h.

If the initiator has been reset or has otherwise forgotten the negotiated
transfer parameters, it makes sense for the target to go into wide
negotiation,  but what is the reasoning behind forcing the result of the
negotiation to narrow  (Transfer Width Exponent field equal zero)?  By the
way, this text exists  at least as far back as SPI-3.
Michael Banther
Hewlett-Packard Ltd.
Telephone +44 (117)  312-9503

<<File BANTHER,MICHAEL (HP-UnitedKingdom,ex2).vcf (354 Bytes) has been
removed. >>

* 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 mailing list