Re: [PATCH] Revert "mm: skip CMA pages when they are not available"

[Usama Arif]

On 21/08/2024 15:35, Usama Arif wrote:
> This reverts commit 5da226dbfce3a2f44978c2c7cf88166e69a6788b.
> 
> lruvec->lru_lock is highly contended and is held when calling
> isolate_lru_folios. If the lru has a large number of CMA folios
> consecutively, while the allocation type requested is not
> MIGRATE_MOVABLE, isolate_lru_folios can hold the lock for a very long
> time while it skips those. For FIO workload, ~150million order=0
> folios were skipped to isolate a few ZONE_DMA folios [1].
> This can cause lockups [1] and high memory pressure for extended periods
> of time [2].
> 
> [1] https://lore.kernel.org/all/CAOUHufbkhMZYz20aM_3rHZ3OcK4m2puji2FGpUpn_-DevGk3Kg@xxxxxxxxxxxxxx/
> [2] https://lore.kernel.org/all/ZrssOrcJIDy8hacI@xxxxxxxxx/
> 
> Signed-off-by: Usama Arif <usamaarif642@xxxxxxxxx>
> ---
>  mm/vmscan.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 5dc96a843466..78ad4a141409 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1690,8 +1690,7 @@ static unsigned long isolate_lru_folios(unsigned long nr_to_scan,
>  		nr_pages = folio_nr_pages(folio);
>  		total_scan += nr_pages;
>  
> -		if (folio_zonenum(folio) > sc->reclaim_idx ||
> -				skip_cma(folio, sc)) {
> +		if (folio_zonenum(folio) > sc->reclaim_idx) {
>  			nr_skipped[folio_zonenum(folio)] += nr_pages;
>  			move_to = &folios_skipped;
>  			goto move;


Actually, I didn't build test this with other config options before sending this patch.
skip_cma is only neeeded with MGLRU after this patch, we need to move the definition of
the function to be guarded under CONFIG_LRU_GEN, so that we don't get 
‘skip_cma’ defined but not used error. Below patch should be good: