Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: A3000 and 2090
Message-ID: <13230@cbmvax.commodore.com>
Date: 16 Jul 90 18:01:14 GMT
References: <2886@isc-br.ISC-BR.COM>
Reply-To: daveh@cbmvax (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 28

In article <2886@isc-br.ISC-BR.COM> ricks@odin..isc-br.com (Rick Schaeffer 99) writes:

>I HAVE put "hddisk" and "hddisk.info" into the
>expansion drawer...and Binddrivers succeeds as long as the 2090 isn't
>mounted in the 3000.  However...if the 2090 IS mounted, then the system
>hangs on BindDrivers.  

The 2090 driver doesn't know how to handle the 32 bit memory in the 3000.
You can probably get the 2090 to work if you run NoFastMem, then run 
Binddrivers.  As long as the partition you're after has a Mask value of
0x00fffffe or something similar, you should be able to turn Fast memory
back on once the 2090 driver is up.

The problem is this.  The 2090 driver, and many other drivers for DMA devices,
will allocate a private memory buffer to use in the case where DMA can't be
done directly, such as in the case of memory above the 24 bit mark, or memory
that's not properly aligned (based on the DMA controller's requirements).
The problem here is that the first Fast memory the driver gets on the 3000 is
above 24 bit, and this wasn't something considered when the driver was written.
Prior to the 3000, all memory above 24 bits had to be added in by AddMem, after
booting and after Binddrivers had run.

>Rick Schaeffer          UUCP:  uunet!isc-br.isc-br.com!ricks

-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
	"I have been given the freedom to do as I see fit" -REM