openwrt/target/linux/generic/backport-5.15/020-v6.1-14-mm-multi-gen-LRU-retry-pages-written-back-while-isol.patch

From 6f315879ad750391a0b1fab8c9170bc054a5f5d7 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yuzhao@google.com>
Date: Tue, 15 Nov 2022 18:38:07 -0700
Subject: [PATCH 14/29] mm: multi-gen LRU: retry pages written back while
 isolated

The page reclaim isolates a batch of pages from the tail of one of the
LRU lists and works on those pages one by one.  For a suitable
swap-backed page, if the swap device is async, it queues that page for
writeback.  After the page reclaim finishes an entire batch, it puts back
the pages it queued for writeback to the head of the original LRU list.

In the meantime, the page writeback flushes the queued pages also by
batches.  Its batching logic is independent from that of the page reclaim.
For each of the pages it writes back, the page writeback calls
rotate_reclaimable_page() which tries to rotate a page to the tail.

rotate_reclaimable_page() only works for a page after the page reclaim
has put it back.  If an async swap device is fast enough, the page
writeback can finish with that page while the page reclaim is still
working on the rest of the batch containing it.  In this case, that page
will remain at the head and the page reclaim will not retry it before
reaching there.

This patch adds a retry to evict_pages().  After evict_pages() has
finished an entire batch and before it puts back pages it cannot free
immediately, it retries those that may have missed the rotation.

Before this patch, ~60% of pages swapped to an Intel Optane missed
rotate_reclaimable_page().  After this patch, ~99% of missed pages were
reclaimed upon retry.

This problem affects relatively slow async swap devices like Samsung 980
Pro much less and does not affect sync swap devices like zram or zswap at
all.

Link: https://lkml.kernel.org/r/20221116013808.3995280-1-yuzhao@google.com
Fixes: ac35a4902374 ("mm: multi-gen LRU: minimal implementation")
Signed-off-by: Yu Zhao <yuzhao@google.com>
Cc: "Yin, Fengwei" <fengwei.yin@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 mm/vmscan.c | 48 +++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 37 insertions(+), 11 deletions(-)

--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4723,10 +4723,13 @@ static int evict_pages(struct lruvec *lr
 	int scanned;
 	int reclaimed;
 	LIST_HEAD(list);
+	LIST_HEAD(clean);
 	struct page *page;
+	struct page *next;
 	enum vm_event_item item;
 	struct reclaim_stat stat;
 	struct lru_gen_mm_walk *walk;
+	bool skip_retry = false;
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
 
@@ -4743,20 +4746,37 @@ static int evict_pages(struct lruvec *lr
 
 	if (list_empty(&list))
 		return scanned;
-
+retry:
 	reclaimed = shrink_page_list(&list, pgdat, sc, &stat, false);
+	sc->nr_reclaimed += reclaimed;
 
-	list_for_each_entry(page, &list, lru) {
-		/* restore LRU_REFS_FLAGS cleared by isolate_page() */
-		if (PageWorkingset(page))
-			SetPageReferenced(page);
+	list_for_each_entry_safe_reverse(page, next, &list, lru) {
+		if (!page_evictable(page)) {
+			list_del(&page->lru);
+			putback_lru_page(page);
+			continue;
+		}
 
-		/* don't add rejected pages to the oldest generation */
 		if (PageReclaim(page) &&
-		    (PageDirty(page) || PageWriteback(page)))
-			ClearPageActive(page);
-		else
-			SetPageActive(page);
+		    (PageDirty(page) || PageWriteback(page))) {
+			/* restore LRU_REFS_FLAGS cleared by isolate_page() */
+			if (PageWorkingset(page))
+				SetPageReferenced(page);
+			continue;
+		}
+
+		if (skip_retry || PageActive(page) || PageReferenced(page) ||
+		    page_mapped(page) || PageLocked(page) ||
+		    PageDirty(page) || PageWriteback(page)) {
+			/* don't add rejected pages to the oldest generation */
+			set_mask_bits(&page->flags, LRU_REFS_MASK | LRU_REFS_FLAGS,
+				      BIT(PG_active));
+			continue;
+		}
+
+		/* retry pages that may have missed rotate_reclaimable_page() */
+		list_move(&page->lru, &clean);
+		sc->nr_scanned -= thp_nr_pages(page);
 	}
 
 	spin_lock_irq(&lruvec->lru_lock);
@@ -4778,7 +4798,13 @@ static int evict_pages(struct lruvec *lr
 	mem_cgroup_uncharge_list(&list);
 	free_unref_page_list(&list);
 
-	sc->nr_reclaimed += reclaimed;
+	INIT_LIST_HEAD(&list);
+	list_splice_init(&clean, &list);
+
+	if (!list_empty(&list)) {
+		skip_retry = true;
+		goto retry;
+	}
 
 	if (need_swapping && type == LRU_GEN_ANON)
 		*need_swapping = true;