Re: [PATCH 21/23] hugetlb/userfaultfd: Only drop uffd-wp special pte if required

From: Peter Xu
Date: Mon Apr 26 2021 - 18:05:51 EST


On Mon, Apr 26, 2021 at 02:36:26PM -0700, Mike Kravetz wrote:
> On 4/26/21 2:16 PM, Peter Xu wrote:
> > On Fri, Apr 23, 2021 at 01:33:08PM -0700, Mike Kravetz wrote:
> >> On 3/22/21 5:50 PM, Peter Xu wrote:
> >>> diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
> >>> index 92710600596e..4047fa042782 100644
> >>> --- a/include/linux/hugetlb.h
> >>> +++ b/include/linux/hugetlb.h
> >>> @@ -121,14 +121,15 @@ long follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *,
> >>> unsigned long *, unsigned long *, long, unsigned int,
> >>> int *);
> >>> void unmap_hugepage_range(struct vm_area_struct *,
> >>> - unsigned long, unsigned long, struct page *);
> >>> + unsigned long, unsigned long, struct page *,
> >>> + unsigned long);
> >>> void __unmap_hugepage_range_final(struct mmu_gather *tlb,
> >>> struct vm_area_struct *vma,
> >>> unsigned long start, unsigned long end,
> >>> - struct page *ref_page);
> >>> + struct page *ref_page, unsigned long zap_flags);
> >>> void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
> >>> unsigned long start, unsigned long end,
> >>> - struct page *ref_page);
> >>> + struct page *ref_page, unsigned long zap_flags);
> >>
> >> Nothing introduced with your patch, but it seems __unmap_hugepage_range_final
> >> does not need to be in the header and can be static in hugetlb.c.
> >
> > It seems to be used in unmap_single_vma() of mm/memory.c?
>
> Sorry, that should have been __unmap_hugepage_range. No need for you to
> address in this series.

Ah yes; I'd rather add a patch if you won't object, since my series will touch
that definition.

>
> >>
> >> Everything else looks good,
> >>
> >> Reviewed-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
> >
> > Please let me know whether you want my extra paragraph in the commit message,
> > then I'll coordinate accordingly with the R-b. Thanks!
>
> Yes, please do add the extra paragraph.

Will do.

--
Peter Xu