*BSD News Article 24229


Return to BSD News archive

Xref: sserve comp.periphs.scsi:15497 comp.os.386bsd.questions:7032 comp.unix.osf.osf1:1450
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!elroy.jpl.nasa.gov!swrinde!emory!news-feed-2.peachnet.edu!umn.edu!msc.edu!mckinley.cit.macalstr.edu!sferris
From: sferris@budapest.math.macalstr.edu (Scott Ferris)
Newsgroups: comp.periphs.scsi,comp.os.386bsd.questions,comp.unix.osf.osf1
Subject: Re: Driver for Adaptec 274xT
Followup-To: comp.periphs.scsi,comp.os.386bsd.questions,comp.unix.osf.osf1
Date: 19 Nov 1993 23:42:29 GMT
Organization: Macalester College, St. Paul Mn	
Lines: 48
Distribution: world
Message-ID: <2cjll5$lcm@mckinley.cit.macalstr.edu>
References: <1993Nov9.163202.2387@chpc.org> <2ce69c53.jonas@jonas.ars.muc.de> <CGIqME.48K@rahul.net> <2ceaa12f.jonas@jonas.ars.muc.de>
NNTP-Posting-Host: sydney.math.macalstr.edu
X-Newsreader: TIN [version 1.2 PL0]

Kai Uwe Rommel (rommel@ars.muc.de) wrote:
> lps@rahul.net (Kevin Martinez) writes in article <CGIqME.48K@rahul.net>:
> >In <2ce69c53.jonas@jonas.ars.muc.de> rommel@ars.muc.de (Kai Uwe Rommel) writes:
> >
> >>heas@chpc.org (Heas) writes in article <1993Nov9.163202.2387@chpc.org>:
> >>>hi all,
> >>>I am looking for a PD driver for an Adaptec 274x[T] EISA SCSI card. 
> >>>This is the new 2 channel + floppy SCSI card which replaces the 174x.  It
> >>>features a new 29 bit RISC based processor.
> >>                ^^^^^^
> >>Are you sure about this? :-)
> >
> >According to Adaptec docs, it is a 29 bit sequencer or state machine.

> However, that's not a 29-bit processor, or am I wrong?

  The AIC-7770 chip has 448 29-bit words for the sequencer (RISC chip)
program memory (effectively microcode).  The rest of the chip is
like a standard PC card.  The sequencer is used to offload
processing from the host CPU, speeding operation of the card
by reducing the number of host interrupts that must be generated,
as well as optimizing the instruction set for the operations most
likely to be needed for controlling the card.  I'm not sure why they 
went with 29-bit instructions.  I guess they had all the opcodes they 
wanted and didn't want to cram more bits on the chip for the program 
memory.

  The register set of the card is *not* 29-bit, only the
sequencer program RAM.  In every other respect the card is
like the a normal PC scsi controller.


> Kai Uwe Rommel

> --
> /* Kai Uwe Rommel                                      Muenchen, Germany *
>  * rommel@ars.muc.de                              CompuServe 100265,2651 *
>  * rommel@informatik.tu-muenchen.de                  Fax +49 89 324 4524 */

> DOS ... is still a real mode only non-reentrant interrupt
> handler, and always will be.                -Russell Williams

--
Scott Ferris,             sferris@macalstr.edu
LaserMaster Inc,
Macalester College,
and points in between.