Re: [Xen-devel] [PATCH v3 11/14] libxl: get and set soft affinity

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Wed, 2013-11-20 at 13:59 +0000, George Dunlap wrote:
> On 20/11/13 11:30, Ian Campbell wrote:
> > On Tue, 2013-11-19 at 19:58 +0100, Dario Faggioli wrote:
> >> As per the mechanism used to amend the current API, I don't have a real
> >> preference. Both me and George asked (in a previous thread) what
> >> mechanism we should use, but we did not get much feedback at that time
> >> (not complain, just summing up what has happened).
> > Yes, sorry about that.
> >
> >> Right now, I think I just need for someone authoritative enough to
> >> express his preference. You certainly qualify as such, but, if possible,
> >> I'd like to see what the other Ian thinks too, as he's been quite deeply
> >> involved in reviewing this interface.
> > I think the major argument in favour of the LIBXL_API_VERSION based
> > approach is that this is how we have dealt with all the previous such
> > changes. Having a mixture of this and fooN interfaces seems to me to be
> > worse that either option by itself.
> 
> Just to be clear -- if we bump the API number, then we have to do some 
> #ifdef-ery to allow programs compiled against the old API to compile, right?

Yes.

I expect that the libxl_domain_create_restore machinery is the template
which would be desired here.

This does not remove the need for a LIBXL_HAVE_FOO either -- they are
somewhat orthogonal. (In practice at least libvirt prefers the
LIBXL_HAVE mechanism)

Ian.


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