On Thu, Oct 11, 2018 at 12:03:44PM -0400, John Ferlan wrote:
On 9/29/18 6:58 AM, Peng Hao wrote:
> sometimes we hoped that the memory of vm can be evenly distributed in
> all nodes according to interleave mode. But different hosts
> has different node number. So we add nodeset='all' for interleave mode.
>
> Signed-off-by: Peng Hao <peng.hao2(a)zte.com.cn>
> ---
> src/conf/numa_conf.c | 77 ++++++++++++++++++++++++++++++++++++-------
> 1 files changed, 64 insertions(+), 13 deletions(-)
>
Anyway, back to the cpuset/nodeset presentation. If today it's "A,C,E"
or "A-E,^B,^D", then why not devise a way to use "0,~0" meaning 0
and
the one's complement of 0 (e.g. everything else). That way if I wanted
to, I could "0,~0,^4" meaning everything except 4. Of course the
downside is that I need to know what the guest maximum is. The only way
to know that is in post parse processing. So doing this may involve
moving the translation of @nodeset and @cpuset...
Adding ~0 will break applications parsing the attribute in the same
way "all" will, as they will reject '~' as an invalid value.
Regards,
Daniel
--
|:
https://berrange.com -o-
https://www.flickr.com/photos/dberrange :|
|:
https://libvirt.org -o-
https://fstop138.berrange.com :|
|:
https://entangle-photo.org -o-
https://www.instagram.com/dberrange :|