WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] Re: 25% cpu usage during idle

To: Bastian Blank <waldi@xxxxxxxxxx>, xen-devel@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-devel] Re: 25% cpu usage during idle
From: Jeremy Fitzhardinge <jeremy@xxxxxxxx>
Date: Wed, 28 Jul 2010 12:34:03 -0700
Cc:
Delivery-date: Wed, 28 Jul 2010 12:35:14 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <20100728192608.GB21304@xxxxxxxxxxxxxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <20100728192608.GB21304@xxxxxxxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.7) Gecko/20100720 Fedora/3.1.1-1.fc13 Lightning/1.0b2pre Thunderbird/3.1.1
 On 07/28/2010 12:26 PM, Bastian Blank wrote:
A new pvops kernel (78b55f90e72348e231092dbe3e50ac7414b9e1af) needs around 25%
cpu while being idle with 32 guest domains in event and ksoftirq kernel
threads:

My guesses at filling in the blanks:

   * this is dom0
   * this is a 12 core host
   * you mean 25% of one core, not 25% of all available cpu power
   * no other dom0 processes are using any cpu time
   * the guest domains are PV, with one disk and net device each
   * they're all idle, with no cpu, net or disk traffic
   * this behaviour is new to this kernel

Is that correct?

What was the last kernel which didn't have this behaviour?

Does it get better or worse when you have fewer vcpus in dom0?

How do the interrupt counts in /proc/interrupts change?

Thanks,
    J

| root         4  3.6  0.0      0     0 ?        S    21:08   0:26 [ksoftirqd/0]
| root         7  8.6  0.0      0     0 ?        S    21:08   1:03 [ksoftirqd/1]
| root        10  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/2]
| root        13  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/3]
| root        16  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/4]
| root        19  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/5]
| root        22  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/6]
| root        25  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/7]
| root        28  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/8]
| root        31  0.0  0.0      0     0 ?        S    21:08   0:00 [ksoftirqd/9]
| root        34  0.0  0.0      0     0 ?        S    21:08   0:00 
[ksoftirqd/10]
| root        37  0.0  0.0      0     0 ?        S    21:08   0:00 
[ksoftirqd/11]
| root        39  1.3  0.0      0     0 ?        S    21:08   0:09 [events/0]
| root        40 14.0  0.0      0     0 ?        S    21:08   1:44 [events/1]
| root        41  0.0  0.0      0     0 ?        S    21:08   0:00 [events/2]
| root        42  0.0  0.0      0     0 ?        S    21:08   0:00 [events/3]
| root        43  0.1  0.0      0     0 ?        S    21:08   0:00 [events/4]
| root        44  0.0  0.0      0     0 ?        S    21:08   0:00 [events/5]
| root        45  0.0  0.0      0     0 ?        S    21:08   0:00 [events/6]
| root        46  0.0  0.0      0     0 ?        S    21:08   0:00 [events/7]
| root        47  0.0  0.0      0     0 ?        S    21:08   0:00 [events/8]
| root        48  0.0  0.0      0     0 ?        S    21:08   0:00 [events/9]
| root        49  0.0  0.0      0     0 ?        S    21:08   0:00 [events/10]
| root        50  0.0  0.0      0     0 ?        S    21:08   0:00 [events/11]

Bastian



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

<Prev in Thread] Current Thread [Next in Thread>