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

Re: [Xen-devel] [PATCH 2/5] allow domain heap allocations to specify more than one NUMA node



>>> On 02.03.15 at 18:12, <ian.campbell@xxxxxxxxxx> wrote:
> On Thu, 2015-02-26 at 13:53 +0000, Jan Beulich wrote:
>> ... using struct domain as a container for passing the respective
>> affinity mask: Quite a number of allocations are domain specific, yet
>> not to be accounted for that domain. Introduce a flag suppressing the
>> accounting altogether (i.e. going beyond MEMF_no_refcount) and use it
>> right away in common code (x86 and IOMMU code will get adjusted
>> subsequently).
>> 
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> 
> Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>

Thanks.

> Does this patch constitute all the "not just"(x86) from the initial
> mail? I'll assume so unless I hear otherwise.

No, this part of patch 1

"Note that this gives meaning to MEMF_exact_node specified alone (i.e.
 implicitly combined with NUMA_NO_NODE): In such a case any node inside
 the domain's node mask is acceptable, but no other node. This changed
 behavior is (implicitly) being exposed through the memop hypercalls."

does too. (Patches 4 and 5 are only indirectly x86-specific, as the
IOMMU code touched there is used on x86 only. But that's of no
concern to you anyway.)

Jan


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


 


Rackspace

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