> From: Jan Beulich [mailto:jbeul...@suse.com] > Sent: Wednesday, November 16, 2016 9:34 PM > > >>> On 16.11.16 at 14:01, <andrew.coop...@citrix.com> wrote: > > On 16/11/16 12:53, Jan Beulich wrote: > >>>>> On 16.11.16 at 13:31, <andrew.coop...@citrix.com> wrote: > >>> This reduces the net complexity of CPUID handling by having all > >>> adjustments in > >>> at the same place. Remove the now-unused vpmu_do_cpuid() infrastructure. > >> I have to admit that I'm not convinced this is a good idea at this point, > >> due to the added redundancy. Iirc your plan is to combine hvm_cpuid() > >> and pv_cpuid() anyway, at which point the folding done here would be > >> quite a bit more natural. > > > > Indeed, to guest_cpuid(). > > > > It is far easier to reason about the safety of both changes by first > > untangling the calltree, then merging the functions. I tried it the > > other way first, but that is far more complicated change. > > Well, let's see what others think. >
I'm fine with this change, as long as it paves a clear incremental way towards final guest_cpuid. Acked-by: Kevin Tian <kevin.t...@intel.com> Thanks Kevin _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
