Re: Checking SCO is using multiple CPUs

From: David McCormack (nospam@profitmaster.co.uk)
Date: 04/08/03 

From: "David McCormack" <nospam@profitmaster.co.uk>
Date: Tue, 8 Apr 2003 10:59:46 +0100

"Stephen M. Dunn" <stephen@bokonon.stevedunn.ca> wrote in message
news:HCzxC4.3Eo@bokonon.stevedunn.ca...

> In article <b6s8bt$fcr$1$8300dec7@news.demon.co.uk> "David McCormack"
<nospam@profitmaster.co.uk> writes:

[ snip - cpusar states second processor not being used ]
[ on dual xeon P4 system ]

> Officially, 5.0.5 doesn't support P4s due to some potential issues
> with CPU overheating. For the most part, this is a theoretical
> problem - it's possible, but in practice it's apparently unlikely
> (so far I've only seen one person post to the newsgroup stating that
> 5.0.5 burned up a P4, and I can't imagine that they're the only
> person who's ever tried this combination).
>
> As long as your CPUs don't catch fire :-), you should be able
> to use both of them with SMP.

I would have thought that the inbuilt thermal protection on the P4 would
kick in before that happens... least I hope so! :-)

[snip - 100% idle might not mean 100% idle, might be 99.6% instead ]

> Run two CPU-intensive jobs at once and see what happens. For
> example, yes | compress >/dev/null should be enough to keep a
> CPU busy (it will run forever, until you interrupt it, producing
> no output on the screen and causing no physical I/O to slow it
> down), so run that on two different screens at the same time
> and see what cpusar reports.

I left the systems running 200 odd 'yes | compress >/dev/null' commands.
Using 'w' reported that the load average was just over 250, and sar reported
0% idle time. Even though I have maxed the system, 'cpusar -P 2' still
reports that the second processor is not being used. Idle time is 100%.

Looking though /usr/adm/syslog I have spotted the following line which
doesn't look too good.

    WARNING: CPU2 did not finish boot code

Does this mean that the second CPU has been spotted but will not be used ?

David McCormack

Relevant Pages

  • Re: Follow up to the problem.
    ... Manager svchost.exe was using a 100% CPU. ... thread it reports is ntdll.dll!RtlUserThreadStart and suspending this makes ... The sysinternals process explorer has shown itâ??s a single thread causing the ... Post Originated from http://www.VistaForums.com Vista Support Forums ...
    (microsoft.public.windows.vista.performance_maintenance)
  • Dell OptiPlex 320 - Core 2 Duo
    ... de-sync, SATA controller detection, etc), this particular model doesn't ... seem to correctly recognize the CPU, which I'm not even sure what it is ... However, /proc/cpuinfo only reports 1 processor, 1.8Ghz. ... DMA zone: ...
    (Linux-Kernel)
  • Re: Where to find statistics for fetched programs?
    ... will one of the reports display the CPU usage within each module? ... None of the reports you listed will show CPU consumption ...
    (bit.listserv.ibm-main)
  • Linux reports incorrect CPU speed
    ... Anyway it's all working but KDE feels very slow, and looked up the CPU ... BIOS reports. ... Intel machine check architecture supported. ... Intel Pentium III (Coppermine) stepping 0a ...
    (Debian-User)
  • Re: 2.6.24-rc6-git12: Reported regressions from 2.6.23
    ... Somewhere i saw it mentioned that "the CPU throws out of C mode". ... What does that mean - does it mean we try to idle again and again, ... as several other reports; Mark is by far not the only one) is something hardware ... and one of them is a pending interrupt of some kind, which is the most likely thing going on here; ...
    (Linux-Kernel)