Groups > linux.kernel > #1652001

Re: [PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime

From	"Kirill A. Shutemov" <kirill@shutemov.name>
Newsgroups	linux.kernel
Subject	Re: [PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime
Date	2017-05-28 00:50 +0200
Message-ID	<tLSGZ-1L7-1@gated-at.bofh.it> (permalink)
References	<tL7I5-4wz-3@gated-at.bofh.it> <tL7I5-4wz-13@gated-at.bofh.it> <tLLFv-5HI-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Sat, May 27, 2017 at 11:09:54AM -0400, Brian Gerst wrote:
> >  static inline int pgd_none(pgd_t pgd)
> >  {
> > +       if (p4d_folded)
> > +               return 0;
> >         /*
> >          * There is no need to do a workaround for the KNL stray
> >          * A/D bit erratum here.  PGDs only point to page tables
> 
> These should use static_cpu_has(X86_FEATURE_LA57), so that it gets
> patched by alternatives.

Right, eventually we would likely need something like this. But at this
point I'm more worried about correctness than performance. Performance
will be the next step.

And I haven't tried it yet, but I would expect direct use of alternatives
wouldn't be possible. If I read code correctly, we enable paging way
before we apply alternatives. But we need to have something functional in
between.

I guess it will be fun :)

-- 
 Kirill A. Shutemov

Thread

[PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> - 2017-05-25 22:40 +0200
  Re: [PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime Brian Gerst <brgerst@gmail.com> - 2017-05-27 17:20 +0200
    Re: [PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-05-28 00:50 +0200
      Re: [PATCHv1, RFC 5/8] x86/mm: Fold p4d page table layer at runtime Brian Gerst <brgerst@gmail.com> - 2017-05-28 01:00 +0200

csiph-web