search for: cpusets

On 09/14/2018 03:36 PM, Lukas Hejtmanek wrote: > Hello, > > ok, I found that cpu pinning was wrong, so I corrected it to be 1:1. The issue > with iozone remains the same. > > The spec is running, however, it runs slower than 1-NUMA case. > > The corrected XML looks like follows: [Reformated XML for better reading] <cpu mode="host-passthrough">

Hello, so the current domain configuration: <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3' memory='62000000' /><cell cpus='4-7' memory='62000000' /><cell cpus='8-11' memory='62000000' /><cell cpus='12-15'

Hello, I have cluster with AMD EPYC 7351 cpu. Two CPUs per node. I have performance 8-NUMA configuration: This is from hypervizor: [root@hde10 ~]# lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 64 On-line CPU(s) list: 0-63 Thread(s) per core: 2 Core(s) per socket: 16 Socket(s): 2 NUMA

Hello again, when the iozone writes slow. This is how slabtop looks like: 62476752 62476728 0% 0.10K 1601968 39 6407872K buffer_head 1000678 999168 0% 0.56K 142954 7 571816K radix_tree_node 132184 125911 0% 0.03K 1066 124 4264K kmalloc-32 118496 118224 0% 0.12K 3703 32 14812K kmalloc-node 73206 56467 0% 0.19K 3486 21

Hello, ok, I found that cpu pinning was wrong, so I corrected it to be 1:1. The issue with iozone remains the same. The spec is running, however, it runs slower than 1-NUMA case. The corrected XML looks like follows: <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3'

Hello All, Since 4.5.0-23.el7 version (Red Hat 7.7), when I launch pinned VM, libvirtd reset CPU affinity to all enabled in host CPUs, if it runs in custom cpuset. I can't reproduce this behavior with 4.5.0-10.el7_6.12 with the same kernel version (Red Hat 7.7). Libvirt runs in a custom cpuset 'libvirt', where the number of available cpus is restricted to 0,2,4,6,8. And this

On 09/17/2018 04:59 PM, Lukas Hejtmanek wrote: > Hello, > > so the current domain configuration: > <cpu mode='host-passthrough'><topology sockets='8' cores='4' threads='1'/><numa><cell cpus='0-3' memory='62000000' /><cell cpus='4-7' memory='62000000' /><cell cpus='8-11'

Hi, At least on 0.10.2 setting a cpuset doesn`t match a real process placement - VM still consumes all available cores. VM config: .snip. <vcpu placement='static' cpuset='0-5,12-17'>12</vcpu> .snip. for cpuset in $(find /cgroup/cpuset/libvirt/qemu/vmid/ -name cpuset.cpus) ; do grep 0-5 $cpuset ; done got: empty responce, e.g. 0-23 in my setup expected: at least

Hi. I have following errors in my log file: Mar 22 05:33:11 xentest libvirtd: 05:33:11.518: error : failed Xen syscall topology cpuset syntax error Mar 22 05:33:21 xentest libvirtd: 05:33:21.418: error : failed Xen syscall topology cpuset syntax error Mar 22 05:33:41 xentest libvirtd: 05:33:41.274: error : failed Xen syscall topology cpuset syntax error Mar 22 05:33:41 xentest libvirtd:

Hi there, I have a question regarding the shares option of the cputune section. I want to illustrate my question with the following example. Let's assume I have two virtual machines like the following on four dedicated core with two threads each: VM1: <cputune> <shares>512</shares> <vcpupin vcpu="0" cpuset="0"/> <vcpupin

Dear all I allocate only one vcpu for the container by the following statement, that is, I want to pin the vcpu to physical core "2". <vcpu placement='static' cpuset="2" >1</vcpu> My host has 4 physical cores. Before test, all the 4 cores are idle. After I run 4 processes in the container, I found all the 4 cores in the host are 100% used. That is, the

Hi, I'm using libvirt and qemu on Debian Wheezy. I'm having a strange behavior. Guests can't connect to each other when they're on the same host. On the host I'm using bonding (in active / backup mode) and vlan. It looks like this : eth0 \ / macvtap0 bond0 --- vlan222 eth1 / \ macvtap1 So I've got two guests, let's say A and B. When

Hello, According to the docs, vcpupin will use either cgroups or sched_setaffinity to pin vcpu threads to cpus. How is this decision made? I observe differences even on different hosts featuring the same version of libvirtd (1.3.1): on one host vcpupin affects cpuset.cpus (cgroup), and on the other it affects vcpu threads affinity (observed through taskset). Thanks, Nikos -------------- next

...puset attribute of the CPU Allocation element in a domain ? <domain> ... <vcpu placement='static' cpuset="1-4,^3,6" current="1">2</vcpu> ... </domain> I have seen that virsh vcpupin and virsh emulatorpin can be used to query and set the cpusets of the <vcpupin> and <emulatorpin> children elements of <cputune> which override the cpuset of <vcpu>. If i did not have to pin to different PCPUs, I would execute just one command for the whole domain rather than one command per VCPU + one per the emulator threads. Th...

...echo 1 > cpuset.cpu_exclusive for one VM. Now, I am launching another VM using virt-install and I have given the option --cpuset 1. There is no overlapping of cpus here, but still libvirt is showing error that it's unable to create cgroup for the new VM. Is it possible to have two exclusive cpusets inside a same hierarchy?   ~Peeyush Gupta

Hi all, I have been trying to run a VM on some particular cpu cores. Now the thing is I can set cpuset with virt-install and that will reflect in VM's XML and it will define the affinity. But what about cpuset.cpus? That is inherited directly from the parent. Is there any way to set it when launching the VM? Again, if I have given cpuset during virt-install, why doesn't cpuset.cpus

Hi there, Do you know if there is a way to modify how libvirt interacts with the cgroup? Because, I successfully add the /dev/net/tun support in my LXC container by doing: echo c 10:200 rwm >> /cgroup/libvirt/lxc/instance-00000005/devices.allow But when I restart the instance/LXC container, this option has gone. How can I make this persistant? Is there a configuration file? Thanks

Hi all, I have been trying to use cpuset with virt-install. I am using the following command: $ virt-install --virt-type kvm --name Ubuntu13 --ram 1024 --cdrom=/home/peeyush/Downloads/ubuntu-13.04-desktop-i386.iso --cpuset=0,2 --disk path=/var/lib/libvirt/guest.qcow2,size=10,format=qcow2 Now, according to virt-install man page, my VM's cpu affinity should be set to pcpu 0 and 2 . But when

On Wed, Jan 15, 2014 at 05:49:23AM +0000, WANG Cheng D wrote: > Dear all > > I allocate only one vcpu for the container by the following > statement, that is, I want to pin the vcpu to physical core "2". > <vcpu placement='static' cpuset="2" >1</vcpu> > My host has 4 physical cores. Before test, all the 4 cores are > idle. After I run 4

We're running the current version of torque. On our small supercomputer (an SGI), no updates to torque since July, but just recently - someone may be trying something new - /var/log/messages is on-and-off being spammed with Oct 15 18:02:04 servername pbs_mom: LOG_INFO::create_job_cpuset, creating cpuset for job 1971[656].york.cit.nih.gov: 1 cpus (12), 1 mems (1) and I mean thousands of lines.