Re: [PATCH 4/7] khugepaged: Allow to callapse a page shared across fork

[Yang Shi]

On Fri, Mar 27, 2020 at 11:20 AM Zi Yan <zi.yan@xxxxxxxx> wrote:
>
> On 27 Mar 2020, at 13:05, Kirill A. Shutemov wrote:
>
> > The page can be included into collapse as long as it doesn't have extra
> > pins (from GUP or otherwise).
> >
> > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
> > ---
> >  mm/khugepaged.c | 28 ++++++++++++++++------------
> >  1 file changed, 16 insertions(+), 12 deletions(-)
> >
> > diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> > index 39e0994abeb8..b47edfe57f7b 100644
> > --- a/mm/khugepaged.c
> > +++ b/mm/khugepaged.c
> > @@ -581,18 +581,26 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
> >               }
> >
> >               /*
> > -              * cannot use mapcount: can't collapse if there's a gup pin.
> > -              * The page must only be referenced by the scanned process
> > -              * and page swap cache.
> > +              * Check if the page has any GUP (or other external) pins.
> > +              *
> > +              * The page table that maps the page has been already unlinked
> > +              * from the page table tree and this process cannot get
> > +              * additinal pin on the page.
> > +              *
> > +              * New pins can come later if the page is shared across fork,
> > +              * but not for the this process. It is fine. The other process
> > +              * cannot write to the page, only trigger CoW.
> >                */
> > -             if (page_count(page) != 1 + PageSwapCache(page)) {
> > +             if (total_mapcount(page) + PageSwapCache(page) !=
> > +                             page_count(page)) {
>
> Do you think having a function for this check would be better? Since the check is used three times.
>
> >                       /*
> >                        * Drain pagevec and retry just in case we can get rid
> >                        * of the extra pin, like in swapin case.
> >                        */
> >                       lru_add_drain();
> >               }
> > -             if (page_count(page) != 1 + PageSwapCache(page)) {
> > +             if (total_mapcount(page) + PageSwapCache(page) !=
> > +                             page_count(page)) {
> >                       unlock_page(page);
> >                       result = SCAN_PAGE_COUNT;
> >                       goto out;
> > @@ -680,7 +688,6 @@ static void __collapse_huge_page_copy(pte_t *pte, struct page *page,
> >               } else {
> >                       src_page = pte_page(pteval);
> >                       copy_user_highpage(page, src_page, address, vma);
> > -                     VM_BUG_ON_PAGE(page_mapcount(src_page) != 1, src_page);
>
> Maybe replace it with this?
>
> VM_BUG_ON_PAGE(page_mapcount(src_page) + PageSwapCache(src_page) != page_count(src_page), src_page);

I don't think this is correct either. If a THP is PTE mapped its
refcount would be bumped by the number of PTE mapped subpages. But
page_mapcount() would just return the mapcount of that specific
subpage. So, total_mapcount() should be used, but the same check has
been done before reaching here.

>
>
> >                       release_pte_page(src_page);
> >                       /*
> >                        * ptl mostly unnecessary, but preempt has to
> > @@ -1209,12 +1216,9 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
> >                       goto out_unmap;
> >               }
> >
> > -             /*
> > -              * cannot use mapcount: can't collapse if there's a gup pin.
> > -              * The page must only be referenced by the scanned process
> > -              * and page swap cache.
> > -              */
> > -             if (page_count(page) != 1 + PageSwapCache(page)) {
> > +             /* Check if the page has any GUP (or other external) pins */
> > +             if (total_mapcount(page) + PageSwapCache(page) !=
> > +                             page_count(page)) {
> >                       result = SCAN_PAGE_COUNT;
> >                       goto out_unmap;
> >               }
> > --
> > 2.26.0
>
> Thanks.
>
> â
> Best Regards,
> Yan Zi