Xen project Mailing List

RE: [Xen-devel] Re: [Xen-users] What does xm top mean by the following:

To: "rickey berkeley" <rickey.berkeley@xxxxxxxxx>, "pv" <vishnubhatt@xxxxxxxxx>

From: "Petersson, Mats" <Mats.Petersson@xxxxxxx>

Date: Tue, 4 Jul 2006 15:17:10 +0200

Cc: xen-devel@xxxxxxxxxxxxxxxxxxx, xen-users@xxxxxxxxxxxxxxxxxxx

Delivery-date: Tue, 04 Jul 2006 06:17:53 -0700

List-id: Xen developer discussion <xen-devel.lists.xensource.com>

Thread-index: AcafaovTAB4GHcqzTAGNKao6FZEeEgAARSqA

Thread-topic: [Xen-devel] Re: [Xen-users] What does xm top mean by the following:

From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx [mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of rickey berkeley
Sent: 04 July 2006 14:05
To: pv
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx; xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-devel] Re: [Xen-users] What does xm top mean by the following:

I think we can send this also to the developing mail list.

from pv

I'm compiling some tarballs in both the guest domains below (fedora1 and fedora2), they're quite compute/cpu intensive, when I do a xm top, I get the following, can someone tell me what does it mean to have CPU at 98.6% on one guest domain while 84.9% in the other when the host/domain-0 is at 3.2%? Thx in advance.

xentop - 20:38:02   Xen 3.0-unstable
3 domains: 2 running, 0 blocked, 0 paused, 0 crashed, 0 dying, 0 shutdown
Mem: 1038488k total, 1028408k used, 10080k free    CPUs: 2 @ 3391MHz
      NAME STATE   CPU(sec) CPU(%)     MEM(k) MEM(%) MAXMEM(k) MAXMEM(%) VCPUS NETS NETTX(k) NETRX(k) SSID
Domain-0 -----r         41    3.2     131100   12.6   no limit       n/a     2    8      879      378    0
   fedora1 ------         91   98.6     437652   42.1     442368      42.6     1    2        8       34    0
   fedora2 -----r         60   84.9     437564   42.1     442368      42.6     1    2        7

Are you asking "Where does the rest of the CPU time (out of 200%) go?"

Then I can give you a rough answer: It is "lost" in the Hypervisor... It's probably used up in part to handle the disk accesses that your compile will do. Interrupts are passed through the hypervisor, as does page-fault handling and a few other processor exceptions. I doubt that the time spent dealing with for example page-faults is accounted in the correct domain (and it may not be accounted at all).

If this is not what you were asking, please clarify what your question is...

Mats

_______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-devel