[t13] T10/04-262r0

Jeff Garzik jgarzik at pobox.com
Thu Aug 12 22:08:25 PDT 2004

* From the T10 Reflector (t10 at t10.org), posted by:
* Jeff Garzik <jgarzik at pobox.com>

A note regarding PATA<->SATA hardware bridges, and vendor-reserved commands.

As Nathan M and Pat noted, there is a question of what happens for a 
vendor-reserved ATA command that must traverse a PATA<->SATA bridge.

For those unfamiliar with the current hardware landscape, many early 
SATA host controllers were simply pre-existing PATA solutions with a 
SATA bridge.  I call these "first gen SATA", to distinguish them from 
the non-bridged, usually FIS-based SATA controllers coming out today.

First gen SATA controllers will typically -not- be able to control the 
command protocol associated with a vendor-reserved ATA command.  As I 
noted in my response to Pat, my guess is that the SATA bridges will work 
for non-data and PIO[-mult] protocols only, but I have no test results 
to back this up.

PATA controllers and "second generation" SATA controllers shouldn't have 
any problem with a vendor-reserved command's protocol requirements, 
however esoteric.


* 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