Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!osuunx.ucc.okstate.edu!moe.ksu.ksu.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!agate!agate.berkeley.edu!cgd
From: cgd@eden.CS.Berkeley.EDU (Chris G. Demetriou)
Newsgroups: comp.os.386bsd.misc
Subject: Re: Status on discussed merge between NetBSD and FreeBSD
Date: 14 Nov 93 08:56:27
Organization: Kernel Hackers 'r' Us
Lines: 56
Message-ID: <CGD.93Nov14085627@eden.CS.Berkeley.EDU>
References: <JKH.93Nov13222001.2@whisker.lotus.ie> <crt.753292942@tiamat.umd.umich.edu>
NNTP-Posting-Host: eden.cs.berkeley.edu
In-reply-to: crt@tiamat.umd.umich.edu's message of 14 Nov 1993 11:06:35 -0500
this is being posted for exactly one reason: "myth dispulsion."
I'm sick of seeing this rumor; it's simply *NOT TRUE*. doubts? ask
current-users@sun-lamp.cs.berkeley.edu, or post news asking the many
people who had 80+ and 100+ day uptimes under 0.8 and 0.9 to send
you some mail...
In article <crt.753292942@tiamat.umd.umich.edu> crt@tiamat.umd.umich.edu (Rob Shady) writes:
>Here is something I've been thinking about for some time now. I *love* the
>new features and splitting edge technology that NetBSD is currently offering,
>but I don't like having something that is 'iffy' or 'unstable'.
you've not tried NetBSD-current lately on a 'serious' system, have you?
we're more stable than they are, and we *have been* for a while. (I'd
say we were "more stable" when 0.9 came out, but that's not a fair
comparison; they'd not done a release yet! however, we killed a couple of
real *killers* right before 0.9 that i know they've not killed yet, because
i get get their commit messages via e-mail, in the same way that many of
them get ours.)
sure, if you pick up -current *EVERY NIGHT* occasionally you'll lose, but:
we've had perfectly working shared libs running for going
on 2 weeks; they still don't have it right. We have
shared XFree86. They have linker problems.
we can still reliably run on 4M machines; they can't --
they claim it's a bug from Net/2, but i've done serious
development on 4M machines from 386BSD 0.0 day one (because
the original machine i had was a 386 with 4M RAM), and never
been bitten by it.
we have a real buffer cache, no longer done out of kernel
malloc memory. this leads to more speed, and greater
reliability (because there's less kernel map fragmentation).
we've fixed *so* many machine-dependencies and chunks of
bogus code it's unbelievable; many of those areas they've
not *touched*.
we've fixed *so* many bugs that they've not -- and that
they don't even know are there. we've found them
by stress-testing the hell out of NetBSD; they've
not even come close to doing that.
"NetBSD is not stable or well-tested" is a myth that started long
ago, and it just isn't true. at one point, it arguably might have been,
and even today, if you pick up a new release every day, you can still
get slightly burned. But if you're careful, you won't.
And i'm not even going to talk seriously about the fact that NetBSD
is your *only* choice if you want to run on ... a sparc, an hp300,
an amiga, a mac, &c.
chris
--
chris g. demetriou cgd@cs.berkeley.edu
smarter than your average clam.