[PATCH -V3 8/8] mm: migrate: Do not retry 10 times for the subpages of fail-to-migrate THP

[Huang Ying]

From: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>

If THP is failed to migrate due to -ENOSYS or -ENOMEM case, the THP will
be split, and the subpages of fail-to-migrate THP will be tried to migrate
again, so we should not account the retry counter in the second loop, since
we already accounted 'nr_thp_failed' in the first loop.

Moreover we also do not need retry 10 times for -EAGAIN case for the subpages
of fail-to-migrate THP in the second loop, since we already regarded the
THP as migration failure, and save some migration time (for the worst case,
will try 512 * 10 times) according to previous discussion [1].

[1] https://lore.kernel.org/linux-mm/87r13a7n04.fsf@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/

Tested-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
Signed-off-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
Signed-off-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
Cc: Oscar Salvador <osalvador@xxxxxxx>
Cc: Zi Yan <ziy@xxxxxxxxxx>
Cc: Yang Shi <shy828301@xxxxxxxxx>
---
 mm/migrate.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/mm/migrate.c b/mm/migrate.c
index 55fbf9669431..06a653977835 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1522,7 +1522,7 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
 			case -EAGAIN:
 				if (is_thp)
 					thp_retry++;
-				else
+				else if (!no_subpage_counting)
 					retry++;
 				nr_retry_pages += nr_subpages;
 				break;
@@ -1548,8 +1548,7 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
 			}
 		}
 	}
-	if (!no_subpage_counting)
-		nr_failed += retry;
+	nr_failed += retry;
 	nr_thp_failed += thp_retry;
 	nr_failed_pages += nr_retry_pages;
 	/*
-- 
2.30.2