summaryrefslogtreecommitdiff
path: root/mm/swap.c
diff options
context:
space:
mode:
authorJohannes Weiner <hannes@cmpxchg.org>2022-10-13 15:31:13 -0400
committerAndrew Morton <akpm@linux-foundation.org>2022-11-08 17:37:11 -0800
commit0538a82c39e94d49fa6985c6a0101ca819be11ee (patch)
tree9ceaff53dbf4cd39680617274f5856b0bdb853ae /mm/swap.c
parent57a196a58421a4b0c45949ae7309f21829aaa77f (diff)
mm: vmscan: make rotations a secondary factor in balancing anon vs file
We noticed a 2% webserver throughput regression after upgrading from 5.6. This could be tracked down to a shift in the anon/file reclaim balance (confirmed with swappiness) that resulted in worse reclaim efficiency and thus more kswapd activity for the same outcome. The change that exposed the problem is aae466b0052e ("mm/swap: implement workingset detection for anonymous LRU"). By qualifying swapins based on their refault distance, it lowered the cost of anon reclaim in this workload, in turn causing (much) more anon scanning than before. Scanning the anon list is more expensive due to the higher ratio of mmapped pages that may rotate during reclaim, and so the result was an increase in %sys time. Right now, rotations aren't considered a cost when balancing scan pressure between LRUs. We can end up with very few file refaults putting all the scan pressure on hot anon pages that are rotated en masse, don't get reclaimed, and never push back on the file LRU again. We still only reclaim file cache in that case, but we burn a lot CPU rotating anon pages. It's "fair" from an LRU age POV, but doesn't reflect the real cost it imposes on the system. Consider rotations as a secondary factor in balancing the LRUs. This doesn't attempt to make a precise comparison between IO cost and CPU cost, it just says: if reloads are about comparable between the lists, or rotations are overwhelmingly different, adjust for CPU work. This fixed the regression on our webservers. It has since been deployed to the entire Meta fleet and hasn't caused any problems. Link: https://lkml.kernel.org/r/20221013193113.726425-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Rik van Riel <riel@surriel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Diffstat (limited to 'mm/swap.c')
-rw-r--r--mm/swap.c22
1 files changed, 17 insertions, 5 deletions
diff --git a/mm/swap.c b/mm/swap.c
index 955930f41d20..2f12a2ee1d3a 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -295,8 +295,20 @@ void folio_rotate_reclaimable(struct folio *folio)
}
}
-void lru_note_cost(struct lruvec *lruvec, bool file, unsigned int nr_pages)
+void lru_note_cost(struct lruvec *lruvec, bool file,
+ unsigned int nr_io, unsigned int nr_rotated)
{
+ unsigned long cost;
+
+ /*
+ * Reflect the relative cost of incurring IO and spending CPU
+ * time on rotations. This doesn't attempt to make a precise
+ * comparison, it just says: if reloads are about comparable
+ * between the LRU lists, or rotations are overwhelmingly
+ * different between them, adjust scan balance for CPU work.
+ */
+ cost = nr_io * SWAP_CLUSTER_MAX + nr_rotated;
+
do {
unsigned long lrusize;
@@ -310,9 +322,9 @@ void lru_note_cost(struct lruvec *lruvec, bool file, unsigned int nr_pages)
spin_lock_irq(&lruvec->lru_lock);
/* Record cost event */
if (file)
- lruvec->file_cost += nr_pages;
+ lruvec->file_cost += cost;
else
- lruvec->anon_cost += nr_pages;
+ lruvec->anon_cost += cost;
/*
* Decay previous events
@@ -335,10 +347,10 @@ void lru_note_cost(struct lruvec *lruvec, bool file, unsigned int nr_pages)
} while ((lruvec = parent_lruvec(lruvec)));
}
-void lru_note_cost_folio(struct folio *folio)
+void lru_note_cost_refault(struct folio *folio)
{
lru_note_cost(folio_lruvec(folio), folio_is_file_lru(folio),
- folio_nr_pages(folio));
+ folio_nr_pages(folio), 0);
}
static void folio_activate_fn(struct lruvec *lruvec, struct folio *folio)