Re: [PATCH] mm: fix shrink nr.unqueued_dirty counter issue

[Andrew Morton]

On Fri, 12 Jan 2024 09:23:52 +0800 Zhiguo Jiang <justinjiang@xxxxxxxx> wrote:

> It is needed to ensure sc->nr.unqueued_dirty > 0, which can avoid to
> set PGDAT_DIRTY flag when sc->nr.unqueued_dirty and sc->nr.file_taken
> are both zero at the same time.
> 
> ...
>
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -5957,7 +5957,8 @@ static void shrink_node(pg_data_t *pgdat, struct scan_control *sc)
>  			set_bit(PGDAT_WRITEBACK, &pgdat->flags);
>  
>  		/* Allow kswapd to start writing pages during reclaim.*/
> -		if (sc->nr.unqueued_dirty == sc->nr.file_taken)
> +		if (sc->nr.unqueued_dirty &&
> +			sc->nr.unqueued_dirty == sc->nr.file_taken)
>  			set_bit(PGDAT_DIRTY, &pgdat->flags);
>  

Seems sensible.  Was this discovered by code inspection, or is there
some observable runtime effect?  If the latter, can you please describe
that effect?