Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 11 Jan 2010 11:40:28 -0800
From:      Chuck Swiger <cswiger@mac.com>
To:        Paul Halliday <paul.halliday@gmail.com>
Cc:        questions@freebsd.org
Subject:   Re: FreeBSD 8.0, HyperV and non-uniform processors.
Message-ID:  <B6D1958E-746E-499C-B498-AFBDBF096C72@mac.com>
In-Reply-To: <2dab70a31001111043kd6bb6b4u5062a3a4b4106b5@mail.gmail.com>
References:  <2dab70a31001111043kd6bb6b4u5062a3a4b4106b5@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi--

On Jan 11, 2010, at 10:43 AM, Paul Halliday wrote:
> Is this warning as harmful as it sounds:
> 
> WARNING: Non-uniform processors.
> WARNING: Using suboptimal topology.
> 
> More info:
> 
> CPU: Intel(R) Xeon(R) CPU           E7330  @ 2.40GHz (2304.83-MHz 686-class CPU)
> ACPI APIC Table: <VRTUAL MICROSFT>
> FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
> FreeBSD/SMP: 0 package(s) x 4 core(s)
> cpu0 (BSP): APIC ID:  0
> cpu1 (AP): APIC ID:  1
> WARNING: Non-uniform processors.
> WARNING: Using suboptimal topology.
> 
> Unfortunately I am forced to use this setup. Is there anything I can
> do? Should I even be worried?

This comes from the SMP probing code in i386/i386/mp_machdep.c (and similar for amd64):

        if (mp_ncpus % (cpu_cores * cpu_logical) != 0) {
                printf("WARNING: Non-uniform processors.\n");
                printf("WARNING: Using suboptimal topology.\n");
                return (smp_topo_none());
        }

smp_topo_none() means that the system assumes none of the L1/L2 cache levels are shared; for a virtual machine, this is probably correct, so you should not be unduly concerned.

-- 
-Chuck




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?B6D1958E-746E-499C-B498-AFBDBF096C72>