On Fri, Jul 30 2021 at 04:19:21 PM -0400, Peter Xu <peterx@xxxxxxxxxx>
wrote:
separate patch. At the meantime I also don't understand why memcpy()
here,
which seems to be even slower..
Alright, I've now had a chance to compare the object code generated
before
my patch is applied, with what is generated after it is applied and the
same object code is generated for arch/x86/kvm/mmu/mmu.c in both cases
(at
least when compiling with clang, however I suspect other optimizing
compilers would behave similarly).
