Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ukma!rutgers!att!lzaz!hcj
From: hcj@lzaz.ATT.COM (HC Johnson)
Newsgroups: comp.os.minix
Subject: Re: MINIX
Summary: I don't think is fair to blame AST, PH. just blame Atari,
	 everyone else does.
Message-ID: <491@lzaz.ATT.COM>
Date: 23 Mar 89 20:41:24 GMT
References: <11440@louie.udel.EDU>
Organization: AT&T ISL Lincroft NJ USA
Lines: 31

In article <11440@louie.udel.EDU>, PRINS%HLERUL5.BITNET@cunyvm.cuny.edu (Ryko Prins, CRI, Leiden, Netherlands) writes:
> I am giving up.
> I don't have the time to debug the Minix driver so it works for
> my Megafile 60. I dont'have the time to backup my whole hard disk each time
> before I try out Minix and restore my hard disk every time Minix destroys it.
> I am writing off the Minix system and the Tanenbaum book than goes with it
> as a loss. 

There are likely one of two problems behind Ryko's problems.
1. His Megafile 60 is sick.  Remember that TOS is polled status, Minix is 
   interrupt driven.  Just because TOS works doesn't mean the Minix will.
   [ Unix(r) systems have been showing up bad computer hardware for years].

2. There are 3 major approches to writing a ST DMA driver for TOS.  And
   when using interrupts it gets even trickier.  I know my SCSI Streaming
   tape requires a different pacing from the hard disk.  Supra is experiencing
   this, in that there seem to be three flavors of problems(old don't work, 
   medium, posted a driver change to work around timing problems, and
   new- just work.

So as we bid Ryko a found farewell, we are left wondering just what is in his
Megafile 60, and whether it is only his that fails.

I thing AST, and Prentice-Hall did a good job in getting minix up on a ST, 
and should be praised, not condemed.


Howard C. Johnson
ATT Bell Labs
att!lzaz!hcj
hcj@lzaz.att.com