Re: [PATCH 2/2] Linux support for hugepages as a Xen PV guest

[Dave McCracken <dcm@mccr.org>]

On Friday 10 October 2008, Keir Fraser wrote:
> hugetlbfs should only be enabled at runtime if supported by the hypervisor.
> Perhaps we can enable the PSE CPUID flag if the hypervisor supports
> superpages?

I'll look into it.

> > --- linux-2.6.18-xen//./mm/hugetlb.c 2008-07-17 09:54:19.000000000 -0500
> > +++ linux-hpage/./mm/hugetlb.c 2008-10-02 15:07:54.000000000 -0500
> > @@ -294,12 +294,14 @@ static pte_t make_huge_pte(struct vm_are
> > int writable)
> >  {
> > pte_t entry;
> > + pgprot_t pgprot;
> >
> > + pgprot = __pgprot(pgprot_val(vma->vm_page_prot) | _PAGE_PRESENT);
> > if (writable) {
> > entry =
> > -      pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
> > +      pte_mkwrite(pte_mkdirty(mk_pte(page, pgprot)));
> > } else {
> > -  entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
> > +  entry = pte_wrprotect(mk_pte(page, pgprot));
> > }
> > entry = pte_mkyoung(entry);
> > entry = pte_mkhuge(entry);
>
> Why do we need to do something different for Xen here?

In the original implementation _PAGE_PRESENT is set in later macros.  Xen 
needs it early to trigger the proper phys_to_machine() translation.

Dave McCracken