This message is from the T13 list server.

Sheffield, Robert L wrote:
> [...]
> 
> Now - to be fully consistent with existing SCSI models, we recognize
> that the SCSI/ATA passthrough CDB has a SCSI op-code. All SCSI op-codes
> defined so far indicate a transfer direction as part of the opcode (if
> there is a transfer at all). That being the case, it might be prudent to
> [...]

I'm totally agnostic on the SCSI opcode issue. Whatever
sits best with the SCSI community works for me. I was
just coding up yet another set of pio_in/pio_out cdbs
while also retaining the read/write info (needed for
DMA operations) and was struck by the redundancy.

If read/write SCSI opcodes are used, we can drop the
T_Dir bit, I assume.

For any command where the transfer length is zero,
(Hard Reset, SRST, Non-data, Device Diagnostic, Device
Reset), either read or write SCSI op-codes should be
valid.

Is there any need for a unique ASC/ASCQ combo for
bridges that may be capable of sanity checking the SCSI
op-code/transfer length/ATA command/protocol. Any checking
would be strictly optional.
-- 
[EMAIL PROTECTED]

Andy Warner             Voice: (612) 801-8549   Fax: (208) 575-5634

Reply via email to