[vfio-users] Memory allocation with numatune

Zachary Boley zboley00 at gmail.com
Fri Jan 27 19:51:07 UTC 2017


I don't understand numa but I have about the same setup (2x 5540s) which
I've pinned respectively and everything seems alright as far as I can help
perhaps, but have you looked into hugepages if that's your main issue?

On Jan 27, 2017 1:21 PM, "sL1pKn07 SpinFlo" <sl1pkn07 at gmail.com> wrote:

> Can you share the New vcpu config?
>
> Im intereses because have, more or less, the same configuration (2x xeon
> x5650)
>
> Greetings
>
> El 27 ene. 2017 6:04 p. m., "Jan Wiele" <jan at wiele.org> escribió:
>
>> Am Freitag, 27. Januar 2017, 08:52:50 CET schrieb Alex Williamson:
>> > On Fri, 27 Jan 2017 11:58:05 +0100
>> >
>> > Jan Wiele <jan at wiele.org> wrote:
>> > > Hi,
>> > > I've two E5-2670 CPUs and want to pin a VM to a node. CPU pinning
>> works
>> > > fine, however memory allocation seems to happen randomly. Sometimes
>> it is
>> > > done on node 0, sometimes on node 1.
>> > >
>> > > My VM config: http://pastebin.com/raw/J6trq7gJ
>> > > I'm on Arch Linux Kernel 4.8.13-1 with libvirt 2.4.0-2
>> > >
>> > > Any ideas?
>> >
>> > Looks like it should work to me.  I've certainly tested this sort of
>> > setup on RHEL and Fedora in the past and it did the right thing.
>>
>> > Your
>> > vcpupin actually looks a little off though, just like on the host, the
>> > VM is going to enumerate cores then threads, this is why your numactl
>> > info shows node0 w/ {0-7,16-23}, node1 w/ {8-15,24-31}, ie.
>> > {cores,threads}.  By interleaving the node1 pinning as you've done, the
>> > mapping is off in the VM.  For instance, the guest will think CPU0 and
>> > CPU6 are threads on the same core, but they're mapped to CPU8 and CPU11
>> > on the host, which are unrelated.
>> Thanks! I will fix that. (I think I did this because of a user, who
>> reported
>> slightly better performance, in the old archlinux-vfio-thread)
>>
>> > I know that doesn't help your
>> > numatune issue though.
>>
>> Do you know which part of the system will actually allocate the memory on
>> the
>> specified node? Should I see some information about this in the qemu
>> command
>> line parameters?
>>
>> _______________________________________________
>> vfio-users mailing list
>> vfio-users at redhat.com
>> https://www.redhat.com/mailman/listinfo/vfio-users
>>
>
> _______________________________________________
> vfio-users mailing list
> vfio-users at redhat.com
> https://www.redhat.com/mailman/listinfo/vfio-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/vfio-users/attachments/20170127/2d24d241/attachment.htm>


More information about the vfio-users mailing list