Re: [PATCH 1/2] mm/zsmalloc.c: Migration can leave pages in ZS_EMPTY indefinitely

[Jonathan Adams <jwadams@google.com>]

On Thu, Aug 1, 2019 at 6:53 PM Henry Burns <henryburns@google.com> wrote:
>
> In zs_page_migrate() we call putback_zspage() after we have finished
> migrating all pages in this zspage. However, the return value is ignored.
> If a zs_free() races in between zs_page_isolate() and zs_page_migrate(),
> freeing the last object in the zspage, putback_zspage() will leave the page
> in ZS_EMPTY for potentially an unbounded amount of time.
>
> To fix this, we need to do the same thing as zs_page_putback() does:
> schedule free_work to occur.  To avoid duplicated code, move the
> sequence to a new putback_zspage_deferred() function which both
> zs_page_migrate() and zs_page_putback() call.
>
> Signed-off-by: Henry Burns <henryburns@google.com>

Reviewed-by: Jonathan Adams <jwadams@google.com>
>
> ---
>  mm/zsmalloc.c | 30 ++++++++++++++++++++----------
>  1 file changed, 20 insertions(+), 10 deletions(-)
>
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index 1cda3fe0c2d9..efa660a87787 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -1901,6 +1901,22 @@ static void dec_zspage_isolation(struct zspage *zspage)
>         zspage->isolated--;
>  }
>
> +static void putback_zspage_deferred(struct zs_pool *pool,
> +                                   struct size_class *class,
> +                                   struct zspage *zspage)
> +{
> +       enum fullness_group fg;
> +
> +       fg = putback_zspage(class, zspage);
> +       /*
> +        * Due to page_lock, we cannot free zspage immediately
> +        * so let's defer.
> +        */
> +       if (fg == ZS_EMPTY)
> +               schedule_work(&pool->free_work);
> +
> +}
> +
>  static void replace_sub_page(struct size_class *class, struct zspage *zspage,
>                                 struct page *newpage, struct page *oldpage)
>  {
> @@ -2070,7 +2086,7 @@ static int zs_page_migrate(struct address_space *mapping, struct page *newpage,
>          * the list if @page is final isolated subpage in the zspage.
>          */
>         if (!is_zspage_isolated(zspage))
> -               putback_zspage(class, zspage);
> +               putback_zspage_deferred(pool, class, zspage);
>
>         reset_page(page);
>         put_page(page);
> @@ -2115,15 +2131,9 @@ static void zs_page_putback(struct page *page)
>
>         spin_lock(&class->lock);
>         dec_zspage_isolation(zspage);
> -       if (!is_zspage_isolated(zspage)) {
> -               fg = putback_zspage(class, zspage);
> -               /*
> -                * Due to page_lock, we cannot free zspage immediately
> -                * so let's defer.
> -                */
> -               if (fg == ZS_EMPTY)
> -                       schedule_work(&pool->free_work);
> -       }
> +       if (!is_zspage_isolated(zspage))
> +               putback_zspage_deferred(pool, class, zspage);
> +
>         spin_unlock(&class->lock);
>  }
>
> --
> 2.22.0.770.g0f2c4a37fd-goog
>