Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 21 May 2006 21:54:19 -0400
From:      "m m" <needacoder@gmail.com>
To:        kmacy@fsmware.com, current@freebsd.org
Subject:   Re: FreeBSD is now self-hosting on the UltraSPARC T1
Message-ID:  <1e4841eb0605211854i44c4aa4cm9dfc72506c2232ea@mail.gmail.com>

next in thread | raw e-mail | index | archive | help
On 5/21/06, Kip Macy <kip.macy@gmail.com> wrote:
> I can't find the original e-mail, but someone was suggesting I post a
> dmesg to link to.
>
> http://www.fsmware.com/sun4v/dmesg_latest.txt

...
FreeBSD/SMP: Multiprocessor System Detected: 32 CPUs
...
SMP: AP CPU #31 Launched!
SMP: AP CPU #30 Launched!
SMP: AP CPU #29 Launched!
SMP: AP CPU #28 Launched!
...

Some phylosophical questions - is this machine really an SMP?  Can we
have an "SMP" when there's only one chip? (it's CMT/CMP, isn't it?)
Can we perhaps stop calling any MP an "SMP" one of these days?  While
on topic, the Opterons aren't SMP either, and neither are the
ht-Xeons...  but we somehow keep lumping them into the "SMP" category.
 Maybe we should fix this once and for all?  Won't it be weird to
write page-allocation code for NUMA machines and put the code into an
SMP directory? What about coloring algorithms on the T1000 to improve
locality in it's funky cache hierarchy, are we going to put that under
"SMP" category too?  Who was it that decided that all the world that
has more than core is an SMP?

(please pardon the format of this mail - but I really only have
questions, no answers...)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1e4841eb0605211854i44c4aa4cm9dfc72506c2232ea>