Title: RE: [t13] ATAPI Overlap and Queuing
Yes. "Old style" (ATA-7 and previous) command queuing is supported on SATA. NCQ is silent on use by ATAPI. I believe it should be prohibited for use by packet devices. (And I'm not going to comment on the wisdom of having 3rd parties define protocols for T13 devices)


From: [EMAIL PROTECTED] on behalf of Hale Landis
Sent: Tue 12/7/2004 4:53 PM
To: [EMAIL PROTECTED]
Subject: RE: [t13] ATAPI Overlap and Queuing

This message is from the T13 list server.


On Tue, 7 Dec 2004 15:42:21 -0800, Mark Overby wrote:
>This message is from the T13 list server.
>I can firmly state that there are at least two host controllers on the
>market that support command queuing for ATA devices in hardware.

OK, then I would agree that O/Q should not be removed for either PATA
ATA or ATAPI devices.

But this is such a mess...

* Old style O/Q maybe supported by some PATA ATA devices and maybe by
some PATA ATAPI devices (MKE note: The PACKET command supports old
style PATA O/Q. Note everyone: The R/W DMA QUEUED commands are the
PATA O/Q commands for for ATA devices).

* Is old style O/Q supported by SATA for SATA ATA or SATA ATAPI
devices? Is there any reason old style O/Q can not be made obsolete
for SATA?

* Is new style queuing (is that NCQ?) supported for SATA ATAPI? If
so, what changes have been made for this (new version of the PACKET
command, or new options of the PACKET command)? Is there a definition
of new style queuing (NCQ?) for SATA ATAPI? If not, why not?

Hale


*** Hale Landis *** www.ata-atapi.com ***



Reply via email to