On Tue, Jun 10, 2014 at 07:21:12PM +0200, Michal Privoznik wrote:
> There are two places where you'll find info on huge pages. The first
> one is under <cpu/> element, where all supported huge page sizes are
> listed. Then the second one is under each <cell/> element which refers
> to concrete NUMA node. At this place, the size of huge page's pool is
> reported. So the capabilities XML looks something like this:
>
> <capabilities>
>
> <host>
> <uuid>01281cda-f352-cb11-a9db-e905fe22010c</uuid>
> <cpu>
> <arch>x86_64</arch>
> <model>Westmere</model>
> <vendor>Intel</vendor>
> <topology sockets='1' cores='1' threads='1'/>
> ...
> <pages unit='KiB' size='1048576'/>
> <pages unit='KiB' size='2048'/>
Should have normal sized pages (ie 4k on x86) too, to avoid
apps having to special case small pages.
> </cpu>
> ...
> <topology>
> <cells num='4'>
> <cell id='0'>
> <memory unit='KiB'>4054408</memory>
> <pages unit='KiB' size='1048576'>1</pages>
> <pages unit='KiB' size='2048'>3</pages>
Should have normal sized pages here too.
We should also declare whether '<memory>' refers to total memory
across all page sizes, or just total memory of smallest page
sizes. I'd say it should refer to all memory, since conceptually
this is really about telling you how much RAM DIMMS are in each
node.
> <distances/>
> <cpus num='1'>
> <cpu id='0' socket_id='0' core_id='0' siblings='0'/>
> </cpus>
> </cell>
> <cell id='1'>
> <memory unit='KiB'>4071072</memory>
> <pages unit='KiB' size='1048576'>2</pages>
> <pages unit='KiB' size='2048'>1024</pages>
> <distances/>
> <cpus num='1'>
> <cpu id='1' socket_id='0' core_id='0' siblings='1'/>
> </cpus>
> </cell>
> ...
> </cells>
> </topology>
> ...
> </host>
>
> <guest/>
>
> </capabilities>
>
> Signed-off-by: Michal Privoznik <mprivozn@xxxxxxxxxx>
> ---
> docs/schemas/capability.rng | 21 +++++++++++++++++++++
> src/conf/capabilities.c | 25 ++++++++++++++++++++++---
> src/conf/capabilities.h | 15 ++++++++++++++-
> src/libxl/libxl_conf.c | 1 +
> src/nodeinfo.c | 41 ++++++++++++++++++++++++++++++++++++++++-
> src/qemu/qemu_capabilities.c | 29 ++++++++++++++++++++++++++++-
> src/test/test_driver.c | 2 +-
> src/xen/xend_internal.c | 1 +
> tests/vircaps2xmltest.c | 3 ++-
> tests/vircapstest.c | 1 +
> 10 files changed, 131 insertions(+), 8 deletions(-)
> diff --git a/src/conf/capabilities.h b/src/conf/capabilities.h
> index 53a83c9..384e256 100644
> --- a/src/conf/capabilities.h
> +++ b/src/conf/capabilities.h
> @@ -102,6 +102,13 @@ struct _virCapsHostNUMACellSiblingInfo {
> unsigned int distance; /* distance to the node */
> };
>
> +typedef struct _virCapsHostNUMACellHugePageInfo virCapsHostNUMACellHugePageInfo;
> +typedef virCapsHostNUMACellHugePageInfo *virCapsHostNUMACellHugePageInfoPtr;
> +struct _virCapsHostNUMACellHugePageInfo {
> + unsigned int size; /* huge page size in kibibytes */
> + size_t avail; /* the size of pool */
> +};
s/Huge//. since this should be used to report on all page sizes
> +
> typedef struct _virCapsHostNUMACell virCapsHostNUMACell;
> typedef virCapsHostNUMACell *virCapsHostNUMACellPtr;
> struct _virCapsHostNUMACell {
> @@ -111,6 +118,8 @@ struct _virCapsHostNUMACell {
> virCapsHostNUMACellCPUPtr cpus;
> int nsiblings;
> virCapsHostNUMACellSiblingInfoPtr siblings;
> + int nhugepages;
> + virCapsHostNUMACellHugePageInfoPtr hugepages;
Better named as 'pageinfo' rather than 'hugepages'
Regards,
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list