SPI rev 15a Question

George Penokie GOP at rchvmp3.vnet.ibm.COM
Thu Mar 7 14:43:23 PST 1996

* From the SCSI Reflector, posted by:
* "George Penokie" <GOP at RCHVMP3.VNET.IBM.COM>
 SCSI-3 Parallel Interface, Rev 15a.

 In table 9 on page 27, the column for DB15-8 DB(P1) of the P cable
 signals shows the signal sources for command, status, message in, and
 message out phases should be NONE.  None is defined as:  The signal
 SHALL be released; that is, not driven by any SCSI device.

 Section 10.11.3 on page 42 states:
 During 8-bit data .... bus.  The DB(15-8,P1) signals are undefined
 and parity MAY not be valid.

 1) If the signals shall be released, isn't parity always invalid?
    Which is correct?

 2) If the device is set up for 16 bit data transfers, should DB(P1)
    be driven to ensure good parity on the upper 8 bits during the
    non-data phases (status, message, etc)?  If it doesn't matter,
    what is recommended?

Bye for now,

More information about the T10 mailing list