Re: [PATCH] mm: swap: remove duplicate nr_swap_pages decrement in get_swap_page_of_type()

[Chris Li <chrisl@kernel.org>]

On Sat, Nov 1, 2025 at 6:42 AM Youngjun Park <youngjun.park@lge.com> wrote:
>
> After commit 4f78252da887, nr_swap_pages is decremented in
> swap_range_alloc(). Since cluster_alloc_swap_entry() calls
> swap_range_alloc() internally, the decrement in get_swap_page_of_type()
> causes double-decrementing.
>
> Remove the duplicate decrement.

Acked-by: Chris Li <chrisl@kernel.org>

Thanks for the good catch. Indeed there is duplication, but only
affects the hibernation usage of swap as far as I can tell.
This bug is in v6.17-rc1 upwards including v6.17 as well. You might
want to add the stable tag:

Cc: stable@vger.kernel.org # v6.17-rc1

Hi Andrew,

Please apply. Can you add the above line or Youngjun needs to send out
a new version to include the stable line?

Chris

>
> Fixes: 4f78252da887 ("mm: swap: move nr_swap_pages counter decrement from folio_alloc_swap() to swap_range_alloc()")
> Signed-off-by: Youngjun Park <youngjun.park@lge.com>
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 543f303f101d..66a502cd747b 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -2020,10 +2020,8 @@ swp_entry_t get_swap_page_of_type(int type)
>                         local_lock(&percpu_swap_cluster.lock);
>                         offset = cluster_alloc_swap_entry(si, 0, 1);
>                         local_unlock(&percpu_swap_cluster.lock);
> -                       if (offset) {
> +                       if (offset)
>                                 entry = swp_entry(si->type, offset);
> -                               atomic_long_dec(&nr_swap_pages);
> -                       }
>                 }
>                 put_swap_device(si);
>         }
> --
> 2.34.1
>