SAS does count bytes?

Pat LaVarre LAVARRE at iomega.com
Fri Nov 1 08:31:20 PST 2002


* From the T10 Reflector (t10 at t10.org), posted by:
* "Pat LaVarre" <LAVARRE at iomega.com>
*
"Elliott, Robert (Server Storage)" wrote:
 
> A residual is not provided in SAS response IUs for
> the same reason as parallel SCSI - the initiator is
> lockstepped with the target and its count of data
> received should match the target's count of data
> transmitted.  We're not worried about losing entire
> data frames in a vast store-and-forward fabric.
 
This is new with SAS?
 
In classic Ata/pi, the device & the host can fall out
of sync if data cross the bus in any mode other than
Atapi Pio?
 
The first trouble is that we clock data in byte pairs.
If you look at nothing but data, you can't tell the
difference between copying 5 bytes or 6, for example.
 
Atapi Pio differs from the other Ata/pi data transfer
modes by including an option to preface the burst of
data with a copy of the count of bytes.
 
But with Atapi Dma, the pair issue reappears.  And as
you increase burst rate, it gets worse, because the
device cannot unambigously signal where data stops.
 
The system works only because the host & device so
commonly agree over how many bytes should be
copied, and much of the rest of the time we can
copy a little extra data and not hurt anyone.
 
You mean to say SAS does count bytes?
 
Curiously yours in awesome ignorance, Pat LaVarre
*
* 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