Date: Tue, 9 Feb 1999 13:31:56 -0500 (EST) From: Richard Cownie <tich@ma.ikos.com> To: Thierry.Herbelot@alcatel.fr, smp@FreeBSD.ORG Subject: Re: Mixing P2 Types Message-ID: <199902091831.NAA24011@lonesome.ma.ikos.com>
next in thread | raw e-mail | index | archive | help
With the PentiumPro it was definitely true that all CPU's in an SMP
system had to be identical (same speed, same silicon stepping) -
one particular problem that would occur if you violated this was
that the cpu's would come out of reset and start their APIC bus
negotiation at different times, resulting in more than one cpu
acting as the Boot CPU and attempting to run the BIOS; this would
usually cause severe confusion. There may be other problems as well.
Since nothing much has changed in the bus architecture, boot protocol,
or indeed the CPU core between PentiumPro and Pentium-II, I believe
it's probably still true that all cpu's in an SMP system must be
identical. Not to say that a mixed-CPU system won't work at all -
but it isn't guaranteed to work, and it probably won't work reliably.
In my former incarnation at Axil Computer, I had to write the diagnostic
code that would allow us to check the silicon stepping info of each CPU
in an 8-way SMP NX801 to detect any mismatches. Believe me, I wouldn't
have done this if it wasn't a problem ...
Regards
Richard Cownie (tich@ma.ikos.com)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-smp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199902091831.NAA24011>