Re: [PATCH] mm/swapfile.c: select the swap device with default priority round robin

[YoungJun Park <youngjun.park@lge.com>]

On Wed, Sep 24, 2025 at 05:17:46PM +0800, Baoquan He wrote:
> -	int i;
> -
> -	if (prio >= 0)
> -		si->prio = prio;
> -	else
> -		si->prio = --least_priority;
> +	si->prio = prio;


> -     if (p->prio < 0) {
> -             struct swap_info_struct *si = p;
> -             int nid;
> -
> -             plist_for_each_entry_continue(si, &swap_active_head, list) {
> -                     si->prio++;
> -                     si->list.prio--;
> -                     for_each_node(nid) {
> -                             if (si->avail_lists[nid].prio != 1)
> -                                     si->avail_lists[nid].prio--;
> -                     }
> -             }
> -             least_priority++;
> -     }
 
Hi Baoquan He,

I noticed that in `_enable_swap_info()` there is still a comment describing
the behavior that this patch removes(auto assigning priority). 
Shouldn’t that part also be updated accordingly?

static void _enable_swap_info(struct swap_info_struct *si)
{
        ...
        /*
         * both lists are plists, and thus priority ordered.
         * swap_active_head needs to be priority ordered for swapoff(),
         * which on removal of any swap_info_struct with an auto-assigned
         * (i.e. negative) priority increments the auto-assigned priority
         * of any lower-priority swap_info_structs.
         * swap_avail_head needs to be priority ordered for folio_alloc_swap(),
         * which allocates swap pages from the highest available priority
         * swap_info_struct.
         */

Best regards,
Youngjun Park