[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH RFC v2 1/7] xen/vNUMA: vNUMA support for PV guests.

Hello Li, Dario! :)

On Thu, Oct 3, 2013 at 11:17 AM, Dario Faggioli
<dario.faggioli@xxxxxxxxxx> wrote:
> On gio, 2013-10-03 at 20:27 +0800, Li Yechen wrote:
>> Hi Elena,
>> Thank you for your great work first  :)

>> In your patch, guest can use a hypercall name: "XENMEM_get_vnuma_info"
>> to get the structure below:
>> +struct vnuma_topology_info {
>> +    domid_t domid;
>> +    uint16_t nr_vnodes;
>> +    uint32_t _pad;
>> +    XEN_GUEST_HANDLE_64(uint) vdistance;
>> +    XEN_GUEST_HANDLE_64(uint) vcpu_to_vnode;
>> +    XEN_GUEST_HANDLE_64(vnuma_
>> memblk_t) vnuma_memblks;
>> +};
>> I see that Xen have save a vnode_to_pnode[] array for mapping
> Yes it does. We spoke about it quite a bit, and Xen seems to be the best
> place where to store such info, considering how easy it is to keep it
> there, and that it may turn out useful in future (and ballooning is a
> lcear example of that).
>> However, this hypercall won't give this mapping to Guest VM.
> Not with this hypercall, no. Actually, did most of the people, when
> reviewing your RFC patches, said that they prefer the mapping not
> exposed at all to the guest?

Yes Dario, correct. I think someone, possibly Konrad, was saying that
it would be an interesting possibility.
But I dont think there was a change in this.
So fo Li I think I can provide hypercall that will return the mappings.
Li, do you think it work for 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



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.