Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 06 May 1997 14:04:53 -0700
From:      Jason Thorpe <thorpej@nas.nasa.gov>
To:        Steve Passe <smp@csn.net>
Cc:        "Jordan K. Hubbard" <jkh@time.cdrom.com>, "Jonathan M. Bresler" <jmb@freebsd.org>, hackers@freebsd.org
Subject:   Re: One last call for a show of hands on the ALPHA port... 
Message-ID:  <199705062104.OAA21144@lestat.nas.nasa.gov>

next in thread | raw e-mail | index | archive | help
On Tue, 06 May 1997 14:31:23 -0600 
 Steve Passe <smp@csn.net> wrote:

 > does anyone know if there is a digital part equivilant to the APIC?
 > is there a standard for SMP with Alpha chips?  if so there must be something
 > published which the non-digital MB manufactures refer to, or is digital
 > the only one to make SMP/Alpha boards?

Are there even any multi-processor alpha workstations?  I know of the
servers... and they are very non-PC-like (although NetBSD does run
on the 8200 and 8400 :-)

Anyhow, the standard Alpha architecture references should describe how
multiprocessor systems are supposed to behave.  I'm pretty sure it's
nothing like the Intel way of doing things.  (Note, I say that partially
because I don't know of _any_ multiprocessor alpha PC-like workstations,
and the PC-like alphas are different enough from PCs in a bunch of
respects, anyhow...)

In general, though, if FreeBSD is going to even bother with the alpha:

	(1) just make it run first (you have a lot of work ahead of you,
	    especially if you even want to consider running on the alphas
	    that don't look like PCs)

	(2) change your "SMP" model... the current FreeBSD SMP code
	    seems to be the wrong approach to me.  It feels like there
	    is a bunch of "this is a PC operating system" assumptions
	    in code that's in sys/kern (i.e. init_smp.c).  Also,
	    the kernel isn't multi-threaded... I think you've done
	    this backwards, because the current model will mean more
	    work to multithread your kernel later.

	(3) *then* think about supporting multiple processors on
	    other architectures.

...if you do things in the wrong order (which I think you already have,
but whatever), you only end up creating more work for yourself later,
which tends to lead to kludgy solutions to problems.

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199705062104.OAA21144>