Re: [tip:x86/mm] x86: Convert the rest of the code to support p4d_t

From: Kirill A. Shutemov
Date: Tue Apr 25 2017 - 05:05:04 EST


On Tue, Apr 25, 2017 at 08:38:42AM +0200, Ingo Molnar wrote:
>
> * Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> wrote:
>
> > On Mon, Apr 24, 2017 at 11:42:34AM -0700, Dan Williams wrote:
> > > >
> > > > I cannot reproduce the issue.
> > > >
> > > > Could you share your kernel kernel config?
> > >
> > > Kernel config available here:
> > > https://gist.github.com/djbw/2b99a014396cca9f4406586573be98c0
> >
> > Patch below should help. Please check.
> >
> > diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
> > index a242139df8fe..745e5e183169 100644
> > --- a/arch/x86/mm/init_64.c
> > +++ b/arch/x86/mm/init_64.c
> > @@ -962,7 +962,7 @@ remove_pud_table(pud_t *pud_start, unsigned long addr, unsigned long end,
> > continue;
> > }
> >
> > - pmd_base = (pmd_t *)pud_page_vaddr(*pud);
> > + pmd_base = pmd_offset(pud, 0);
> > remove_pmd_table(pmd_base, addr, next, direct);
> > free_pmd_table(pmd_base, pud);
> > }
> > @@ -988,7 +988,7 @@ remove_p4d_table(p4d_t *p4d_start, unsigned long addr, unsigned long end,
> >
> > BUILD_BUG_ON(p4d_large(*p4d));
> >
> > - pud_base = (pud_t *)p4d_page_vaddr(*p4d);
> > + pud_base = pud_offset(p4d, 0);
> > remove_pud_table(pud_base, addr, next, direct);
> > free_pud_table(pud_base, p4d);
> > }
> > @@ -1013,7 +1013,7 @@ remove_pagetable(unsigned long start, unsigned long end, bool direct)
> > if (!pgd_present(*pgd))
> > continue;
> >
> > - p4d = (p4d_t *)pgd_page_vaddr(*pgd);
> > + p4d = p4d_offset(pgd, 0);
> > remove_p4d_table(p4d, addr, next, direct);
> > }
>
> Type casts are dangerous, chapter 378. I missed them when applying the patches.
>
> Could you please double-check all other type casts introduced by the 5-level
> paging patches, to make sure it's all using natural type conversions?

I looked though at the rest of (p4d_t) and (p4d_t *) casts and they look
okay: under pgd_none()/!pgd_preset() or within #if PTRS_PER_P4d > 1.

I'll prepare proper patch.

--
Kirill A. Shutemov