On Wed, Sep 27, 2000 at 01:34:02PM +0200, "O. Hartmann"
<[EMAIL PROTECTED]> wrote:
>Dear Sirs.
>
>Every step done under Linux is well documented by the press, especially here
>in Germany! When Microsoft found a bug in AMD's Athlon CPUs which causes
>Win2k to crash, the 4GB physical memory limit came into play again. Some
>discussions were made about PAE addressing modell and so on and although the
>discussion started on a M$ <-> AMD problem, the migration towards Linux's
>memory modell was done smoothly. Well, we now know, Linux has recently a
>4GB physical memory limitation due its PAE address space model. And what is
>about FreeBSD 4.1? We use now TYAN's Thunder 2500 maonboard which is capable
>to hold two CPUs and get equipted with max. 8GB main memory. This is a nice
>option because we plan to solve some memory intensive environtmental
>research calculations. So, my question seems to be stupid for those who
>understand reading kernel code, but for me, a kind of "normalo", please
>tell me: how much memory and how many CPUs is FreeBSD 4.1.1-STABLE capable to
>work with in a stable fashion?
There are some fairly complex issues that need to be resolved before FreeBSD
can support more than 4GB of physical memory (splitting vm_offset/vm_size into
virtual and physical parts like netbsd, cleanups, drivers that do dma..). But
4GB works just fine. As for processors, IIRC, it supports any amount that
the hardware can handle, but it won't scale very well with many due to the
big lock.
>Many thanks,
>
>Gruss O. Hartmann
>-------------------------------------------------------------------
>[EMAIL PROTECTED]
>
>Klimadatenserver des IPA, Universitaet Mainz
>Netzwerk- und Systembetreuung
--
Andreas Persson
[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message
