[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] Xen 4.5 development update (July update)
- To: Dario Faggioli <dario.faggioli@xxxxxxxxxx>
- From: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
- Date: Mon, 4 Aug 2014 11:44:39 -0400
- Cc: artem.mygaiev@xxxxxxxxxxxxxxx, msw@xxxxxxxxxx, Ian.Jackson@xxxxxxxxxxxxx, dongxiao.xu@xxxxxxxxx, mengxu@xxxxxxxxxxxxx, JBeulich@xxxxxxxx, feng.wu@xxxxxxxxx, zhigang.x.wang@xxxxxxxxxx, parth.dixit@xxxxxxxxxx, Paul.Skentzos@xxxxxxxxxxxxxxx, eddie.dong@xxxxxxxxx, rcojocaru@xxxxxxxxxxxxxxx, guijianfeng@xxxxxxxxxxxxxx, Vijaya.Kumar@xxxxxxxxxxxxxxxxxx, josh.whitehead@xxxxxxxxxxxxxxx, zoltan.kiss@xxxxxxxxxx, avanzini.arianna@xxxxxxxxx, yang.z.zhang@xxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, serge.broslavsky@xxxxxxxxxx, yjhyun.yoo@xxxxxxxxxxx, olaf@xxxxxxxxx, wency@xxxxxxxxxxxxxx, Ian.Campbell@xxxxxxxxxx, vijay.kilari@xxxxxxxxx, stefano.stabellini@xxxxxxxxxxxxx, mcgrof@xxxxxxxx, julien.grall@xxxxxxxxxx, dave.scott@xxxxxxxxxx, robert.vanvossen@xxxxxxxxxxxxxxx, shantong.kang@xxxxxxxxx, roy.franz@xxxxxxxxxx, anthony.perard@xxxxxxxxxxx, Paul.Durrant@xxxxxxxxxx, ufimtseva@xxxxxxxxx, bjzhang@xxxxxxxx, boris.ostrovsky@xxxxxxxxxx, andrii.tseglytskyi@xxxxxxxxxxxxxxx, jgross@xxxxxxxx, talex5@xxxxxxxxx, Wei.Liu2@xxxxxxxxxx, george.dunlap@xxxxxxxxxxxxx, andrew.cooper3@xxxxxxxxxx, Steve.VanderLeest@xxxxxxxxxxxxxxx, Kelly.Zytaruk@xxxxxxx, dslutz@xxxxxxxxxxx, ross.lagerwall@xxxxxxxxxx, Aravind.Gopalakrishnan@xxxxxxx, david.vrabel@xxxxxxxxxx, Suravee.Suthikulpanit@xxxxxxx, aravindp@xxxxxxxxx, tiejun.chen@xxxxxxxxx, malcolm.crossley@xxxxxxxxxx, yanghy@xxxxxxxxxxxxxx, daniel.kiper@xxxxxxxxxx, christoffer.dall@xxxxxxxxxx, roger.pau@xxxxxxxxxx
- Delivery-date: Mon, 04 Aug 2014 15:47:18 +0000
- List-id: Xen developer discussion <xen-devel.lists.xen.org>
On Mon, Jul 28, 2014 at 07:35:22PM +0200, Dario Faggioli wrote:
> On mar, 2014-07-22 at 20:28 -0400, konrad.wilk@xxxxxxxxxx wrote:
>
> > == x86 ==
>
> > * HT enabled, virtualization overhead is high (Xen 4.4) (none)
> > kernbench demonstrated it
> > looking and tracing it
> > - Dario Faggioli
> >
> There is a (sub-)thread sill ongoing here:
> http://www.gossamer-threads.com/lists/xen/devel/339409
>
> However, there have been found counterexamples, i.e., numbers showing
> that virt overhead is similar with and without HT, so I think this
> bullet can go away (not sure with what tracking status...)
The 'This confirms, for me, that it's an SMT balancing issue that we're seen. '
is not an issue anymore? As in we can get bad and good results depending
on the workload?
Now here is another question - did you run the guests with 'xen_nopvspin'
to see if this was an PV ticketlock (or the lack of it) issue?
Also what did the " I'll try more runs, e.g. with number of VCPUs equal
less than nr_corse/2 and see what happens." come out to be?
Thank you!
>
> Dario
>
> --
> <<This happens because I choose it to happen!>> (Raistlin Majere)
> -----------------------------------------------------------------
> Dario Faggioli, Ph.D, http://about.me/dario.faggioli
> Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
>
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel
|
