In unmap_and_move(), if the new THP cannot be allocated, -ENOMEM will
be returned, and migrate_pages() will try to split the THP unless
"reason" is MR_NUMA_MISPLACED (that is, nosplit == true). But when
nosplit == true, the THP migration failure will not be counted. This
is incorrect. So in this patch, the THP migration failure will be
counted for -ENOMEM regardless of nosplit is true or false. The
nr_failed counting isn't fixed because it's not used. Added some
comments for it per Baolin's suggestion.
Signed-off-by: "Huang, Ying" <[email protected]>
Fixes: 5984fabb6e82 ("mm: move_pages: report the number of non-attempted pages")
Reviewed-by: Baolin Wang <[email protected]>
Cc: Zi Yan <[email protected]>
Cc: Yang Shi <[email protected]>
---
mm/migrate.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/mm/migrate.c b/mm/migrate.c
index 794312072eb3..38e2c789a9c3 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1426,11 +1426,11 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
/*
* When memory is low, don't bother to try to migrate
* other pages, just exit.
- * THP NUMA faulting doesn't split THP to retry.
*/
- if (is_thp && !nosplit) {
+ if (is_thp) {
nr_thp_failed++;
- if (!try_split_thp(page, &thp_split_pages)) {
+ /* THP NUMA faulting doesn't split THP to retry. */
+ if (!nosplit && !try_split_thp(page, &thp_split_pages)) {
nr_thp_split++;
goto retry;
}
@@ -1446,6 +1446,7 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
* the caller otherwise the page refcnt will be leaked.
*/
list_splice_init(&thp_split_pages, from);
+ /* nr_failed isn't updated for not used */
nr_thp_failed += thp_retry;
goto out;
case -EAGAIN:
--
2.30.2
On Mon, Jul 11, 2022 at 04:49:44PM +0800, Huang Ying wrote:
> In unmap_and_move(), if the new THP cannot be allocated, -ENOMEM will
> be returned, and migrate_pages() will try to split the THP unless
> "reason" is MR_NUMA_MISPLACED (that is, nosplit == true). But when
> nosplit == true, the THP migration failure will not be counted. This
> is incorrect. So in this patch, the THP migration failure will be
> counted for -ENOMEM regardless of nosplit is true or false. The
> nr_failed counting isn't fixed because it's not used. Added some
> comments for it per Baolin's suggestion.
>
> Signed-off-by: "Huang, Ying" <[email protected]>
> Fixes: 5984fabb6e82 ("mm: move_pages: report the number of non-attempted pages")
> Reviewed-by: Baolin Wang <[email protected]>
> Cc: Zi Yan <[email protected]>
> Cc: Yang Shi <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
> ---
> mm/migrate.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 794312072eb3..38e2c789a9c3 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -1426,11 +1426,11 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
> /*
> * When memory is low, don't bother to try to migrate
> * other pages, just exit.
> - * THP NUMA faulting doesn't split THP to retry.
> */
> - if (is_thp && !nosplit) {
> + if (is_thp) {
> nr_thp_failed++;
> - if (!try_split_thp(page, &thp_split_pages)) {
> + /* THP NUMA faulting doesn't split THP to retry. */
> + if (!nosplit && !try_split_thp(page, &thp_split_pages)) {
> nr_thp_split++;
> goto retry;
> }
> @@ -1446,6 +1446,7 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
> * the caller otherwise the page refcnt will be leaked.
> */
> list_splice_init(&thp_split_pages, from);
> + /* nr_failed isn't updated for not used */
> nr_thp_failed += thp_retry;
> goto out;
> case -EAGAIN:
> --
> 2.30.2
>
>
--
Oscar Salvador
SUSE Labs