Re: [PATCH v5 3/3] mm/vmscan: reduce lru_lock contention via vmstat-derived scan-balance cost

On 7/27/26 18:23, Usama Arif wrote:
> The anon/file scan balance in get_scan_count() is driven by two scalars
> in struct lruvec, anon_cost and file_cost, accumulated by every reclaim
> producer under lruvec->lru_lock. The acquisition sites for cost work
> specifically are:
> 
>   - shrink_inactive_list() re-takes lru_lock at function exit purely
>     to call lru_note_cost_unlock_irq() with (nr_pageout, nr_scanned -
>     nr_reclaimed). One acquisition per inactive shrink.
>   - shrink_active_list() does the same with (0, nr_rotated). One
>     acquisition per active shrink.
>   - workingset_refault() takes the lock via folio_lruvec_lock_irq()
>     purely to record the refault cost. One acquisition per refault.
>   - prepare_scan_control() takes lru_lock just to snapshot the two
>     scalars into sc->{anon,file}_cost.
>   - lru_note_cost_unlock_irq() itself walks parent_lruvec and
>     re-acquires lru_lock on each ancestor to propagate the update,
>     adding O(memcg-depth) acquisitions per producer call.
> 
> This hurts because lru_lock is already a heavy contention point on
> memory-heavy workloads: every isolate_lru_folios(), move_folios_to_lru()
> and folio_add_lru() takes it. The cost work itself is trivial (two
> scalar bumps and one comparison), but it contends with and causes
> contention for actual LRU manipulation. The parent_lruvec() walk also
> multiplies cost-update overhead by memcg hierarchy depth.
> 
> The balance formula for anon and file, respectively, is this:
> 
>     cost = nr_io * SWAP_CLUSTER_MAX + nr_rotated
> 
> Instead of recording cost and running averaging logic directly when
> these events occur, snapshot running vmstat counters once per reclaim
> cycle and derive the balance from event deltas since the last run.
> 
> Use PGROTATE_* from the preceding patch for the rotation input.
> WORKINGSET_RESTORE_* and NR_VMSCAN_WRITE provide the remaining event
> counters. Charge NR_VMSCAN_WRITE through lruvec stats so all inputs can
> be sampled per lruvec and aggregated through the memcg hierarchy. This
> is overall cheaper and has fewer lock acquisition sites.
> 
> Moving accumulation and decay to the reclaim side also improves the cost
> model across reclaim gaps. With producer-side decay, events that happen
> while reclaim is idle still age each other before reclaim ever samples
> the costs. If a workload refaults a large anon set and then a smaller
> file set before reclaim runs again, the later file activity can age the
> earlier anon activity out of the cost model. The new scheme observes the
> whole between-reclaim delta and decays anon and file proportionally, so
> the scan-balance history better represents what happened since the last
> reclaim pass.
> 
> A dedicated per-lruvec spinlock, cost_lock, serialises the delta
> extraction, the cost->count update and the halving loop against
> concurrent reclaimers in the same memcg+node.
> 
> NR_VMSCAN_WRITE is accounted at writeout(), so reclaim_stat.nr_pageout is
> no longer needed and is removed.
> 
> memcg-v1's memory.stat anon_cost/file_cost is now sourced from
> cost[].count instead of the removed lruvec anon_cost/file_cost fields.
> The reported values only refresh when prepare_scan_control() runs and
> are bounded at ~lrusize/4 by the halving loop; the scan-balance signal
> they express is unchanged.
> 
> Under pure MGLRU the scan-balance signal itself is not consumed (both
> prepare_scan_control() and get_scan_count() are short-circuited on the
> MGLRU paths, and MGLRU's own type/tier selection comes from read_ctrl_pos()
> on lrugen->{avg_refaulted,avg_total,refaulted,evicted}, not from
> anon_cost/file_cost). NR_VMSCAN_WRITE naturally covers writeout from
> either reclaim implementation. The preceding patch also bumps
> PGROTATE_{ANON,FILE} from evict_folios(), so rotation-driven reclaim
> work is accounted consistently across both implementations.
> 
> Acked-by: Shakeel Butt <[email protected]>
> Acked-by: Johannes Weiner <[email protected]>
> Signed-off-by: Usama Arif <[email protected]>

Acked-by: Vlastimil Babka (SUSE) <[email protected]>