Number of ECC bytes

Rod Van Meter rdv at alumni.caltech.edu
Sun Sep 21 14:09:07 PDT 2008


* From the T10 Reflector (t10 at t10.org), posted by:
* Rod Van Meter <rdv at alumni.caltech.edu>
*
On Thu, 2008-09-18 at 13:57 -0500, Gerry.Houlder at seagate.com wrote:
> * From the T10 Reflector (t10 at t10.org), posted by:
> * Gerry.Houlder at seagate.com
> *
> The READ LONG mechanism is the only standard method that even hints at the
> number of ECC bytes per LBA. Its a very good bet that [READ LONG returned
> data length] - [logical block length] - [protection information length if
> implemented] = [number of ECC bytes]. Even if you know the number of bytes
> you still don't know what error correction capability is indicated because
> this varies according to the algorithm, interleaving, and other
> characteristics of ECC.
> 
Not to mention the actual error processes, and their relationship to the
error model assumed in the error management design (which, as noted, is
a lot more than just choosing an ECC code).
		--Rod
*
* 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