Subject: Re: FreeBSD Bus DMA
To: David Edelsohn <email@example.com>
From: John S. Dyson <firstname.lastname@example.org>
Date: 06/12/1998 15:32:39
David Edelsohn said:
> entire product including header files, the working set of the toolchain,
> and the OS all fit within your physical memory.
> I think that we could have avoided a lot of this discussion if you
> had mentioned those details from the beginning. It also would have been
> nice if you had mentioned specific OS releases, compiler options,
> filesystem mount options, system tuning parameters, ... as well.
This is a case where the messenger gets killed, and assumed
wrong? The problem is that I made a statement that was
very accurate (AFAIK), and true. Strong assertions that
it *cannot* be true appear without provocation.
I truely forget at times all of the differeces between
Free and Net.
If you do tune NBUFS upwards on NetBSD, you will have performance
consequences that just do not happen on FreeBSD (unless you
tune our legacy buffer cache.) It isn't just an issue of self
tuning, but there are some aspects of tuning involved.
To me, a machine with 256MB is very desirable, because FreeBSD
makes very effective use of the memory. I really forget
what the effect of memory is on NetBSD (both MACH and UVM),
but I suspect that it doesn't help very much (only to cache