On Fri, 20 Oct 2000, Darren Reed wrote:

# Date: Fri, 20 Oct 2000 23:50:41 +1100 (EST)
# From: Darren Reed <darrenr@reed.wattle.id.au>
# To: greywolf@starwolf.com
# Cc: ragge@ludd.luth.se, tech-kern@netbsd.org, joel.lindholm@telia.com
# Subject: Re: Proposal: In-kernel loader + symbol table handler.
# 
# In some email I received from Greywolf, sie wrote:
# > On Thu, 19 Oct 2000, Anders Magnusson wrote:
# > 
# > # Date: Thu, 19 Oct 2000 23:45:44 +0200 (MET DST)
# > # From: Anders Magnusson <ragge@ludd.luth.se>
# > # To: tech-kern@netbsd.org
# > # Cc: joel.lindholm@telia.com
# > # Subject: Proposal: In-kernel loader + symbol table handler.
# > # 
# > # 
# > # Background:
# > # The current kernel module system is quite bad. It lacks many of the 
# > # features that are needed to do something useful with it.
# > # 
# > # Vision:
# > # To add full functionality that makes it possible to have a complete
# > # dynamic linked kernel. How this will work and its implications will be
# > # described in a special paper.
# > 
# > Will this dynamic loading perform better than the excuse that Solamis
# > uses (especially on the SPARC)?
# 
# IMHO, the Solaris one is actually quite good.  The only problem I have
# with it is the number of miscellaneous files in /etc required to boot.

IMHO, the Solaris one is actually quite SLOW.  If it can run at a speed
approximating current reality, I don't have a problem with it, but if
it's going to pull in drivers as slowly as Solaris does, I'll stick
with static configuration, thanks.

# 
# Darren
# 


				--*greywolf;
--
*BSD: What do you want to serve today?