Subject: Re: FreeBSD Bus DMA
To: None <>
From: Perry E. Metzger <>
List: tech-kern
Date: 06/11/1998 18:28:59
"John S. Dyson" writes:
> Funny, that is why our build times are 3X faster on FreeBSD, for NetBSD
> based products than on NetBSD, right?  We even have a development
> team avoiding *BSD all together because of NetBSD/X86 performance
> problems.


You are claiming that FreeBSD is three times faster than NetBSD on
identical hardware? THREE TIMES?

My last check determined that, so far as I could tell, NetBSD and
FreeBSD were more or less within a few percent of each other in
performance. NetBSD is faster at some things, slower at others --
certainly no real measurable difference.

If you feel there is a difference, though, feel free to present a

> Of course all of the platforms have many of the same, architectural
> performance glitches.  Gessh, even Linux outperforms NetBSD on many
> platforms, and that is embarassing (even to me.)


Please, feel free to show us the numbers.

> > I made a specific claim, to whit, that the FreeBSD development team
> > does not have anywhere near the experience with multi-platform
> > portability that the NetBSD team has, and thus doesn't have any place
> > making claims about portable architecture UNTIL it has managed to port
> > its system to at least *one* other architecture other than
> > Wintel. Hell, you don't even have the experience that the Linux people
> > have at this point.

> Who do you mean when you say "you."  Neither Linux nor NetBSD "know"
> anything, it is the team members who posses "knowledge", and we
> do have very competent people, and you have absolutely no monopoly on that.
> (We have had Linux, QNX and other people working with us from
> time to time -- it is really weird to somehow say that we are
> clueless.)

A specific claim was made by your guys that you had a clean/portable
set of constructs. Whether your guys have general knowledge on seventy
other systems isn't going to tell us anything about whether this
specific set of constructs is portable. The proof is in the
pudding. Will FreeBSD's constructs port easily to new hardware.

Might I ask, btw, why the "FreeBSD" Alpha folks are using a NetBSD

> > If you are going to claim that you have a better architectural design
> > than us for multi-platform portability, you'd better be able to back
> > that up with at least one non-Wintel box that your design is running
> > on.

> FreeBSD isn't perfect, but also claiming somehow that only you have
> magical and more correct solutions seems to show that working with
> you isn't going to happen.

I'm not claiming we have a monopoly on brains. I'm claiming FreeBSD
has never been ported to anything other architecture from its Wintel
base. Do you dispute this?

> We certainly don't want to be constrained
> by the limitations of an OS that has lots of ports...  sort of...

Sort of?

Lets see...

Vax: working
PC: working
Arm32, including Acorns, DNARDS, etc: Working
HP300: Working
Mac68k: Working
MVME68k: Working
x68k: working
PMAX: Working
News/PMAX: working
Amiga: Working
Atari: Working
Sparc: Working
Sun3: Working
NeXT: just finished first bootstraps
macppc, bebox: near working

> This is a quality vs. quantity argument.  I suggest that there
> is a difference between "working" and "WORKING."

Our code base is of extremely high quality. That has cost us over the
long run, but it has been worth it.