Newsgroups: comp.os.minix
Path: utzoo!utgpu!cunews!bnrgate!bmerh2!bmerh107!dgraham
From: dgraham@bmerh107.bnr.ca (Douglas Graham)
Subject: Re: (shared libraries)
Message-ID: <1991May19.181215.8939@bmerh2.bnr.ca>
Sender: news@bmerh2.bnr.ca (Usenet News Admin)
Organization: Bell Northern Research, Ottawa, Canada
References: <53470@nigel.ee.udel.edu> <3142@krafla.rhi.hi.is> <1991May18.232615.9112@watserv1.waterloo.edu>
Date: Sun, 19 May 91 18:12:15 GMT

In article <1991May18.232615.9112@watserv1.waterloo.edu> hbetel@watserv1.waterloo.edu (Heather Betel) writes:

>	I don't understand this attitude. Part of the design
>principles of minix is KISS (keep it simple, Stupid!)
>	Take a good look at the kernel structure. he real kernel is
>about 100 lines of source. It traps interrupts, and passes messages,
>and chooses which process to run next. THAT IS VERY SIMPLE.

There's no need to shout, we all hear you just fine thanks.

> 	The rest are just processes. FS gets a simple message, and
>sends out simple messages. When it gets all the needed replies, it
>replies to the original message. SIMPLE. MM is not much different.

Simple?  Maybe.  It seems like every time somebody proposes a useful
addition to Minix, somebody else has to shoot down the idea by screaming
"kernel bloat".  Well, if simplicity is the only objective, why do we
need Minix at all?  MSDOS is simple, let's just stick with that.  Multitasking
is such a bother, and complicates the kernel tremendously.  Throw the
multitasking out of Minix.  Interrupts cause all kinds of problems.  Let's
just do a CLI right from the start, and then Minix could be simplified
even more.

You're right. FS and MM *are* simple.  The problem is that they are
too simple.  Yes, there is such a thing as being too simple.  Oh, they
sort of do the job, but if you consider that reasonable performance under
various operating conditions should be part of the specifications for a
usable O/S, they don't cut it.  What would your reaction be if Sun starting
hipping Minix instead of SUNOS to run on the workstations you use at
work or school?  I'd bet you wouldn't be pleased.

>	All this avoids kernel bloat. Addind a "print formatted of
>every imaginable data type" is the worst kernel bloat I have heard of.
>It is worse than having 3 distinct ways of doing a
>signal(signum,BLOCK) as in SUNOS. To sum it up, it is stupid. 

Relax.  There are already 3 copies of a printf like routine bloating
up your Minix kernel/mm/fs today.  Shared libraries of some sort would
allow you to get rid of 2 of those copies.
---
Doug Graham   dgraham@bnr.ca   Bell-Northern Research, Ottawa Ontario Canada