Re: [PATCH 7/7] mm/thp: remap_page() is only needed on anonymous THP

[Yang Shi <shy828301@gmail.com>]

On Tue, Jun 1, 2021 at 2:17 PM Hugh Dickins <hughd@google.com> wrote:
>
> THP splitting's unmap_page() only sets TTU_SPLIT_FREEZE when PageAnon,
> and migration entries are only inserted when TTU_MIGRATION (unused here)
> or TTU_SPLIT_FREEZE is set: so it's just a waste of time for remap_page()
> to search for migration entries to remove when !PageAnon.
>
> Fixes: baa355fd3314 ("thp: file pages support for split_huge_page()")
> Signed-off-by: Hugh Dickins <hughd@google.com>

Reviewed-by: Yang Shi <shy828301@gmail.com>

> ---
>  mm/huge_memory.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 305f709a7aca..e4a83e310452 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -2354,6 +2354,7 @@ static void unmap_page(struct page *page)
>
>         VM_BUG_ON_PAGE(!PageHead(page), page);
>
> +       /* If TTU_SPLIT_FREEZE is ever extended to file, update remap_page() */
>         if (PageAnon(page))
>                 ttu_flags |= TTU_SPLIT_FREEZE;
>
> @@ -2368,6 +2369,10 @@ static void unmap_page(struct page *page)
>  static void remap_page(struct page *page, unsigned int nr)
>  {
>         int i;
> +
> +       /* If TTU_SPLIT_FREEZE is ever extended to file, remove this check */
> +       if (!PageAnon(page))
> +               return;
>         if (PageTransHuge(page)) {
>                 remove_migration_ptes(page, page, true);
>         } else {
> --
> 2.32.0.rc0.204.g9fa02ecfa5-goog
>