This message is from the T13 list server.

I have an ATAPI DVD drive that is < 3 months old that exhibits this
behavior. (Actually I have two, both from different manufacturers)

Conveniently, since I have the host and interrupt controller under my
chipset I can probe both lines. The INTRQ line on the ATA interface
fires, the interrupt fires in the chipset, and then the status is read
with BSY still asserted and the INTRQ becomes unasserted again. I'll dig
out the hard drive I have that exhibits this behavior, I don't remember
how old it was.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Hale
Landis
Sent: Wednesday, November 17, 2004 3:27 PM
To: [EMAIL PROTECTED]
Subject: Re: [t13] BSY and INTRQ at end of command

This message is from the T13 list server.


Mark Overby wrote:

 >I agree with you, Hale, that if a SATA host does this while emulating
 >PATA, it is a non-compliant host. However, on the PATA side I have
seen
 >multiple devices that do this. The host has to be tolerant of that
 >behavior and deal with it (most certainly in SW).


Hmmm... I have never seen a device (even the most poorly implemented 
ATAPI devices) assert INTRQ *before* setting BSY=0. This would be a very

  badly broken device and not likely to work on very many systems.

Are you talking about a device you saw 15 years ago or something more 
recent?

And are you sure you where not seeing false INTRQ assertions detected by

a host's interrupt controller due to noise on the INTRQ signal line? 
I've seen this. But this isn't the problem we are talking about here.

Hale

-- 

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

Reply via email to