Date   : Sat, 11 Jun 1983 04:32:44-PDT (Sat)
From   : helge@ucb-vax
Subject: s100 unix and bios hackers

Newsgroups: net.micro.cpm
Subject: Re:  S-100 Unix
References: <1770@sri-arpa.UUCP> (From: goldfarb.ucf-cs%rand-relay@sri-unix.UUCP)

   I have no doubt that a good BIOS hacker could figure out how to modify
   a bootstrap and write device drivers for Unix.  And while working with
   the kernel requires a firm educational foundation in operating systems
   theory, most of the routines in there were deliberately kept simple,
   with complex algorithms being used only when they would result in great
   gains.  But, then, there is usually little reason to go poking around
   in the kernel, anyway, once Unix has been ported to your specific
   machine (how often do you modify the BDOS?), nor do most Unix
   installations have the luxury of a source license.


Well, I really doubt it. It takes a helluva cp/m hacker to write
disk device driver for unix. Not only will he have to now how to
deal with the device itself (which he probably knows, being a
BIOS hacker), but he will have to know a lot about the 
internals of Unix since the drivers take care of a lot of buffer
handling. If the system source is not readily available, this may
make the task impossible since the buffer management may change from 
one implementation to the next thus invalidating any existing 
documentation. But if a sample driver is supplied with the system,
yes, it could probably be done by a "good BIOS hacker".

helge@berkeley