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

Re: [Xen-devel] [PATCH v10 06/11] libxl/xl: push VCPU affinity pinning down to libxl

On Fri, 2014-06-20 at 18:19 +0200, Dario Faggioli wrote:
> From: Wei Liu <wei.liu2@xxxxxxxxxx>
> This patch introduces an array of libxl_bitmap called "vcpu_hard_affinity"
> in libxl IDL to preserve VCPU to PCPU mapping. This is necessary for libxl
> to preserve all information to construct a domain.
> The array accommodates at most max_vcpus elements, each containing the
> affinity of the respective VCPU. If less than max_vcpus bitmaps are
> present, the VCPUs associated to the missing elements will just stay with
> their default affinity (they'll be free to execute on every PCPU).
> In case both this new field, and the already existing cpumap field are
> used, the content of the array will override what's set in cpumap. (In
> xl, we make sure that this never happens in xl, by using only one of the
> two at any given time.)
> The proper macro to mark the API change (called
> LIBXL_HAVE_BUILDINFO_VCPU_AFFINITY_ARRAYS) is added but it is commented.
> It will be uncommented by the patch in the series that completes the
> process, by adding the "vcpu_soft_affinity" array. This is because, after
> all, these two fields are being added sort-of together, and are very
> very similar, in both meaning and usage, so it makes sense for them to
> share the same marker.
> This patch was originally part of Wei's series about pushing as much
> information as possible on domain configuration in libxl, rather than
> xl. See here, for more details:
>   http://lists.xen.org/archives/html/xen-devel/2014-06/msg01026.html
>   http://lists.xen.org/archives/html/xen-devel/2014-06/msg01031.html
> Signed-off-by: Wei Liu <wei.liu2@xxxxxxxxxx>
> Signed-off-by: Dario Faggioli <dario.faggioli@xxxxxxxxxx>>

Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>

Xen-devel mailing list



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