Re: [PATCH v6 4/4] mm: swap: filter swap allocation by memcg tier mask

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

On Wed, May 27, 2026 at 09:42:54AM +0800, Baoquan He wrote:
> On 04/21/26 at 02:53pm, Youngjun Park wrote:
> > Apply memcg tier effective mask during swap slot allocation to
> > enforce per-cgroup swap tier restrictions.
> > 
> > In the fast path, check the percpu cached swap_info's tier_mask
> > against the folio's effective mask. If it does not match, fall
> > through to the slow path. In the slow path, skip swap devices
> > whose tier_mask is not covered by the folio's effective mask.
> > 
> > This works correctly when there is only one non-rotational
> > device in the system and no devices share the same priority.
> > However, there are known limitations:
> > 
> >  - When non-rotational devices are distributed across multiple
> >    tiers, and different memcgs are configured to use those
> >    distinct tiers, they may constantly overwrite the shared
> >    percpu swap cache. This cache thrashing leads to frequent
> >    fast path misses.
> > 
> >  - Combined with the above issue, if same-priority devices exist
> >    among them, a percpu cache miss (overwritten by another memcg)
> >    forces the allocator to round-robin to the next device
> >    prematurely, even if the current cluster is not fully
> >    exhausted.
> > 
> > These edge cases do not affect the primary use case of
> > directing swap traffic per cgroup. Further optimization is
> > planned for future work.
> > 
> > Signed-off-by: Youngjun Park <youngjun.park@lge.com>
> > ---
> >  mm/swapfile.c | 13 ++++++++++++-
> >  1 file changed, 12 insertions(+), 1 deletion(-)
> > 
> > diff --git a/mm/swapfile.c b/mm/swapfile.c
> > index d5abc831cde7..8734e5d26b08 100644
> > --- a/mm/swapfile.c
> > +++ b/mm/swapfile.c
> > @@ -1352,15 +1352,22 @@ static bool swap_alloc_fast(struct folio *folio)
> >  	struct swap_cluster_info *ci;
> >  	struct swap_info_struct *si;
> >  	unsigned int offset;
> > +	int mask = folio_tier_effective_mask(folio);
> >  
> >  	/*
> >  	 * Once allocated, swap_info_struct will never be completely freed,
> >  	 * so checking it's liveness by get_swap_device_info is enough.
> >  	 */
> >  	si = this_cpu_read(percpu_swap_cluster.si[order]);
> > +	if (!si || !swap_tiers_mask_test(si->tier_mask, mask) ||
> > +		!get_swap_device_info(si))
> > +		return false;
> > +
> >  	offset = this_cpu_read(percpu_swap_cluster.offset[order]);
> > -	if (!si || !offset || !get_swap_device_info(si))
> > +	if (!offset) {
> > +		put_swap_device(si);
> >  		return false;
> > +	}
> 
> The whole patch looks good to me except of one nitpick. Is it a lille
> cleaner with below tiny adjustment?
> 
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 2864cd8c2da9..cdf453bf6b80 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1359,15 +1359,12 @@ static bool swap_alloc_fast(struct folio *folio)
>  	 * so checking it's liveness by get_swap_device_info is enough.
>  	 */
>  	si = this_cpu_read(percpu_swap_cluster.si[order]);
> -	if (!si || !swap_tiers_mask_test(si->tier_mask, mask) ||
> -		!get_swap_device_info(si))
> +	if (!si || !swap_tiers_mask_test(si->tier_mask, mask))
>  		return false;
>  
>  	offset = this_cpu_read(percpu_swap_cluster.offset[order]);
> -	if (!offset) {
> -		put_swap_device(si);
> +	if (!offset || !get_swap_device_info(si))
>  		return false;
> -	}

Thanks!

Your suggested version of code is simpler than privious one.
I will apply it.