Date: Sun, 19 Mar 2000 11:00:47 -0600 (CST) From: Joe Greco <jgreco@ns.sol.net> To: blk@skynet.be Cc: FreeBSD-STABLE@FreeBSD.ORG Subject: Re: SMP oddity Message-ID: <200003191700.LAA15601@aurora.sol.net>
next in thread | raw e-mail | index | archive | help
> > This behavior is specific to SMP kernels.
>
> Okay, even to this non-programmer, that source seems pretty
> clear. So, can anyone tell me why it is this way? I think it's
> highly unlikely that a machine could successfully get past POST if
> the two CPUs weren't running at the same speed, so I don't think it
> can be a mismatch issue, can it?
There was, at one point, a technical issue regarding the possibility of the
values being wrong. Check the mailing list archives.
Incidentally: I will let you imagine what I did to a Sun SS10 that had a
Sun FE nearly hysterical once... got called in to replace a disk and he
just HAD to open up the machine to find out why the banner reported _3_
CPU's (I don't think there's a blessed configuration with 3, maybe there
is)... found a dual CPU card plus one rather faster single CPU, and swore
that the disk problem ("the disk is DEAD, Jim") might be caused by my
creative CPU engineering.
Note that Solaris didn't have any idea how to deal with the CPU's from a
scheduling point of view: it scheduled them all fairly equally, which
wasn't exactly ideal, but better than it would have been with only two
CPU's.
I would think that it might be possible to have a similar situation in a
PC unless the BIOS explicitly checks it.
... Joe
-------------------------------------------------------------------------------
Joe Greco - Systems Administrator jgreco@ns.sol.net
Solaria Public Access UNIX - Milwaukee, WI 414/342-4847
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200003191700.LAA15601>