Re: [PATCH 1/2] mm: Allow architectures to request 'old' entries when prefaulting

[Kirill A. Shutemov]

On Sat, Dec 19, 2020 at 12:34:17PM -0800, Linus Torvalds wrote:
> On Sat, Dec 19, 2020 at 4:41 AM Kirill A. Shutemov <kirill@xxxxxxxxxxxxx> wrote:
> >
> > @@ -2884,19 +2966,18 @@ void filemap_map_pages(struct vm_fault *vmf,
> >                 if (vmf->pte)
> >                         vmf->pte += xas.xa_index - last_pgoff;
> >                 last_pgoff = xas.xa_index;
> > -               if (alloc_set_pte(vmf, page))
> > -                       goto unlock;
> > +               if (pte_none(*vmf->pte))
> > +                       do_set_pte(vmf, page);
> > +               /* no need to invalidate: a not-present page won't be cached */
> > +               update_mmu_cache(vma, vmf->address, vmf->pte);
> >                 unlock_page(head);
> > -               goto next;
> > +               continue;
> 
> This can't be right.
> 
> Look at what happens if "pte_none()" is not true.. It won't install
> the new pte, but it also won't drop the ref to the page.

Ouch.

> So I think it needs to be
> 
> -               if (alloc_set_pte(vmf, page))
> +               if (!pte_none(*vmf->pte))
>                         goto unlock;
> +               do_set_pte(vmf, page);
> 
> instead, so that the "if somebody else already filled the page table"
> case gets handled right.

Updated patch is below.