One thing to keep in mind is that the minutes reflect conversation, not specification...
Pat
OK, at the meeting I forgot about the bit that can inform the host about a 16 byte CDB. That would just make things worse! What I was trying to communicate is that this does not allow the host to transfer multiple size CDB's to the device. If the 12 byte flag is on then everything must be 12 bytes. If you want to send a 6 byte CDB you have to pad it to 12 bytes. That bit of inefficiency is my point...
-----------------------
Curtis E. Stevens
Pacific Digital Corp.
2052 Alton Parkway
Irvine, CA 92606
Phone (949) 477-5713
Fax (949) 252-9397
E-Mail: [EMAIL PROTECTED]
WEB: www.PacificDigital.com
When you put the 2 words "The" and "IRS" together it spells "THEIRS"...
-----Original Message-----
From: Pat LaVarre [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, April 24, 2002 8:39 AM
To: [EMAIL PROTECTED]
Subject: [t13] April 23 - Atapi Cdb lengths
This message is from the T13 list server.
> [t13] Minutes of the April 23 plenary meeting
> From: Mclean, Pete [mailto:[EMAIL PROTECTED]]
> Sent: Wed 4/24/2002 8:42 AM
...
> 7.5.8 Reserved op-codes for serial ATA (e01145r1) [McLean]
> ...
> Curtis Stevens asked about using
> the current ATAPI command for this capability.
> Hale Landis noted that current ATAPI only allows
> for 12 byte CDB's.
> Several people expressed that
> this would be a big modification to the document
> and it was more than just a pass through.
> Tony Goodfellow suggested
> that using ATAPI could allow for any size CDB.
Eh? I thought ...
T13 Atapi says in the first few "word"s of op xA1 IdentifyPacketDevice data whether the host should always copy Out a 12 or 16 byte "packet" at BSY DRQ C/D I/O = 0 1 command out time.
T13 has no other comment on how long the Cdb is, except that by implication it can't be longer than 16 bytes.
No?
Thanks in advance,
Pat LaVarre
