mm/migrate: make migrate_pages_batch() stats consistent.
Subsystems (2)
PATCH v2 1/1 mm/migrate: make migrate_pages_batch() stats consistent.
From: Zi Yan <[email protected]>
As Ying pointed out in [1], stats->nr_thp_failed needs to be updated to avoid stats inconsistency between MIGRATE_SYNC and MIGRATE_ASYNC when calling migrate_pages_batch().
Because if not, when migrate_pages_batch() is called via migrate_pages(MIGRATE_ASYNC), nr_thp_failed will not be increased and when migrate_pages_batch() is called via migrate_pages(MIGRATE_SYNC*), nr_thp_failed will be increase in migrate_pages_sync() by stats->nr_thp_failed += astats.nr_thp_split.
[1] https://lore.kernel.org/linux-mm/[email protected]/
--- mm/migrate.c | 5 +++++ 1 file changed, 5 insertions(+)
mm/migrate.c +5 -0
@@ -1645,6 +1645,10 @@ static int migrate_pages_batch(struct list_head *from,* migrate_pages() may report success with (split but * unmigrated) pages still on its fromlist; whereas it * always reports success when its fromlist is empty. +* stats->nr_thp_failed should be increased too, +* otherwise stats inconsistency will happen when +* migrate_pages_batch is called via migrate_pages() +* with MIGRATE_SYNC and MIGRATE_ASYNC. * * Only check it without removing it from the list. * Since the folio can be on deferred_split_scan()
@@ -1661,6 +1665,7 @@ static int migrate_pages_batch(struct list_head *from, !list_empty(&folio->_deferred_list)) {if (try_split_folio (folio, split_folios) ==0 ) { nr_failed++; + stats->nr_thp_failed += is_thp; stats->nr_thp_split += is_thp; stats->nr_split++;continue ;
base-commit: 7c4c5a2ebbcea9031dbb130bb529c8eba025b16a
Signature · 1 line
2.43.0
Quoted 12 lines · Zi Yan
From: Zi Yan <[email protected]>
As Ying pointed out in [1],
stats->nr_thp_failedneeds to be updated to avoid stats inconsistency betweenMIGRATE_SYNCandMIGRATE_ASYNCwhen callingmigrate_pages_batch().Because if not, when
migrate_pages_batch()is called viamigrate_pages(MIGRATE_ASYNC),nr_thp_failedwill not be increased and whenmigrate_pages_batch()is called viamigrate_pages(MIGRATE_SYNC*),nr_thp_failedwill be increase inmigrate_pages_sync()bystats->nr_thp_failed+= astats.nr_thp_split.
Quoted 4 lines
[1] https://lore.kernel.org/linux-mm/[email protected]/
Suggested-by "Huang, Ying" <[email protected]>Signed-off-by Zi Yan <[email protected]>
LGTM, Thanks!
Quoted 29 lines
--- mm/migrate.c | 5 +++++ 1 file changed, 5 insertions(+)mm/migrate.c +5
-0@@ -1645,6 +1645,10 @@ static int migrate_pages_batch(struct list_head *from,* migrate_pages() may report success with (split but * unmigrated) pages still on its fromlist; whereas it * always reports success when its fromlist is empty. +* stats->nr_thp_failed should be increased too, +* otherwise stats inconsistency will happen when +* migrate_pages_batch is called via migrate_pages() +* with MIGRATE_SYNC and MIGRATE_ASYNC. * * Only check it without removing it from the list. * Since the folio can be on deferred_split_scan() @@ -1661,6 +1665,7 @@ static int migrate_pages_batch(struct list_head *from, !list_empty(&folio->_deferred_list)) {if (try_split_folio (folio, split_folios) ==0 ) { nr_failed++; + stats->nr_thp_failed += is_thp; stats->nr_thp_split += is_thp; stats->nr_split++;continue ;base-commit:
7c4c5a2ebbcea9031dbb130bb529c8eba025b16a