This message is from the T13 list server.

On Mon, 08 Apr 2002 17:13:24 -0600, Pat LaVarre wrote:
>This message is from the T13 list server.
>Isn't it evil that the specification doesn't even 
>mention these cases?  It doesn't even say the host and
>the device may do as they please - not out loud anywhere.

Well... I'm not sure "evil" is the right word... But Pat has a
point... And maybe it is time for me to follow up on my threat to
rewrite the first few paragraphs of Clause 7. The one thing that
ATA/ATAPI doesn't describe well is this:

* the ATA interface has two states: PIO and DMA.

* a DMA data transfer command may have one or more DMA bursts.

* some hints about why a host or device should use DMA pause or
termination.

Note that we have received very similar comments before (before U-DMA
existed) concerning how MW DMA operates. We did enhance the MW DMA
timing diagrams but there is still a lot of information that is "folk
lore" and known only to people with years of ATA interface
experience.



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



Reply via email to