Xen project Mailing List

Re: [Xen-users] Bug or cofiguration error?

To: xen-users <xen-users@xxxxxxxxxxxxxxxxxxx>

From: Trolle Selander <trolle.selander@xxxxxxxxx>

Date: Thu, 1 Mar 2012 12:36:59 -0500

Authentication-results: mr.google.com; spf=pass (google.com: domain of trolle.selander@xxxxxxxxx designates 10.42.156.7 as permitted sender) smtp.mail=trolle.selander@xxxxxxxxx; dkim=pass header.i=trolle.selander@xxxxxxxxx

Delivery-date: Thu, 01 Mar 2012 17:37:38 +0000

List-id: Xen user discussion <xen-users.lists.xen.org>

It is neither a bug or a configuration error; XCP, like current versions of XenServer, only assigns 4 vCPUs to dom0 by default. There are very few scenarios where more dom0 vCPUs are beneficial.

Regards,

Trolle

On Thu, Mar 1, 2012 at 11:57 AM, Leonardo Pereira <leoudimg@xxxxxxxxx> wrote:

Hello,

We are students of Computer Science and we are doing some
experiments with XCP, we have a server with 2 xeon processors each
one containing 4 physical cores with HT, but the XCP recognizes 16
cores (8 physical + 8 virtual) but uses only 4 (cpu 0,1,2 and 3).
When accessing "/ proc / cpuinfo" are listed only 4 cores, but
"/ sys / devices / system / cpu" lists 16 cores but the cpu file
"online" for 4 to 15 cpu's is set to "0".

making:
echo 1> / sys/devices/system/cpu/cpu4/online
.
.
.
echo 1> / sys/devices/system/cpu/cpu15/online

all cores become activated.
But when the server is restarted only cpus 0,1,2,3 are active.

Is this a bug or a config error?

Thanks.

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users

_______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxx http://lists.xen.org/xen-users