2022-10-24 09:45:05

by Baolin Wang

[permalink] [raw]
Subject: [PATCH v3 1/2] mm: migrate: Fix return value if all subpages of THPs are migrated successfully

During THP migration, if THPs are not migrated but they are split and all
subpages are migrated successfully, migrate_pages() will still return the
number of THP pages that were not migrated. This will confuse the callers
of migrate_pages(). For example, the longterm pinning will failed though
all pages are migrated successfully.

Thus we should return 0 to indicate that all pages are migrated in this
case

Fixes: b5bade978e9b ("mm: migrate: fix the return value of migrate_pages()")
Signed-off-by: Baolin Wang <[email protected]>
Reviewed-by: Alistair Popple <[email protected]>
Cc: <[email protected]>
---
Changes from v2:
- Add Fixes tag suggested by Yang Shi and Huang, Ying.
- Drop 'nr_thp_split' validation suggested by Alistair.
- Add reviewed tag from Alistair.
- Update the commit message suggested by Andrew.
Changes from v1:
- Fix the return value of migrate_pages() instead of fixing the
callers' validation.
---
mm/migrate.c | 7 +++++++
1 file changed, 7 insertions(+)

diff --git a/mm/migrate.c b/mm/migrate.c
index 8e5eb6e..2eb16f8 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1582,6 +1582,13 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
*/
list_splice(&ret_pages, from);

+ /*
+ * Return 0 in case all subpages of fail-to-migrate THPs are
+ * migrated successfully.
+ */
+ if (list_empty(from))
+ rc = 0;
+
count_vm_events(PGMIGRATE_SUCCESS, nr_succeeded);
count_vm_events(PGMIGRATE_FAIL, nr_failed_pages);
count_vm_events(THP_MIGRATION_SUCCESS, nr_thp_succeeded);
--
1.8.3.1


2022-10-24 19:06:18

by Yang Shi

[permalink] [raw]
Subject: Re: [PATCH v3 1/2] mm: migrate: Fix return value if all subpages of THPs are migrated successfully

On Mon, Oct 24, 2022 at 1:34 AM Baolin Wang
<[email protected]> wrote:
>
> During THP migration, if THPs are not migrated but they are split and all
> subpages are migrated successfully, migrate_pages() will still return the
> number of THP pages that were not migrated. This will confuse the callers
> of migrate_pages(). For example, the longterm pinning will failed though
> all pages are migrated successfully.
>
> Thus we should return 0 to indicate that all pages are migrated in this
> case
>
> Fixes: b5bade978e9b ("mm: migrate: fix the return value of migrate_pages()")
> Signed-off-by: Baolin Wang <[email protected]>
> Reviewed-by: Alistair Popple <[email protected]>
> Cc: <[email protected]>

Reviewed-by: Yang Shi <[email protected]>

> ---
> Changes from v2:
> - Add Fixes tag suggested by Yang Shi and Huang, Ying.
> - Drop 'nr_thp_split' validation suggested by Alistair.
> - Add reviewed tag from Alistair.
> - Update the commit message suggested by Andrew.
> Changes from v1:
> - Fix the return value of migrate_pages() instead of fixing the
> callers' validation.
> ---
> mm/migrate.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 8e5eb6e..2eb16f8 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -1582,6 +1582,13 @@ int migrate_pages(struct list_head *from, new_page_t get_new_page,
> */
> list_splice(&ret_pages, from);
>
> + /*
> + * Return 0 in case all subpages of fail-to-migrate THPs are
> + * migrated successfully.
> + */
> + if (list_empty(from))
> + rc = 0;
> +
> count_vm_events(PGMIGRATE_SUCCESS, nr_succeeded);
> count_vm_events(PGMIGRATE_FAIL, nr_failed_pages);
> count_vm_events(THP_MIGRATION_SUCCESS, nr_thp_succeeded);
> --
> 1.8.3.1
>