khugepaged: Reduce race probability between migration and khugepaged
Subsystems (2)
Dev Jain
reviewerPATCH v3 1/1 khugepaged: Reduce race probability between migration and khugepaged
Suppose a folio is under migration, and khugepaged is also trying to collapse it. collapse_pte_mapped_thp() will retrieve the folio from the page cache via filemap_lock_folio(), thus taking a reference on the folio and sleeping on the folio lock, since the lock is held by the migration path. Migration will then fail in__folio_migrate_mapping -> folio_ref_freeze. Reduce the probability of such a race happening (leading to migration failure) by bailing out if we detect a PMD is marked with a migration entry.
This fixes the migration-shared-anon-thp testcase failure on Apple M3.
Note that, this is not a "fix" since it only reduces the chance of interference of khugepaged with migration, wherein both the kernel functionalities are deemed "best-effort".
---
v2->v3:
- Improve comment (David)
v1->v2:
- Remove SCAN_PMD_MIGRATION, merge into SCAN_PMD_MAPPED (David, Anshuman)
- Add a comment (Lorenzo)
v1:
- https://lore.kernel.org/all/[email protected]/
mm/khugepaged.c | 8 ++++++++ 1 file changed, 8 insertions(+)
mm/khugepaged.c +8 -0
@@ -941,6 +941,14 @@ static inline int check_pmd_state(pmd_t *pmd)if (pmd_none (pmde))return SCAN_PMD_NONE ; + +/* +* The folio may be under migration when khugepaged is trying to +* collapse it. Migration success or failure will eventually end +* up with a present PMD mapping a folio again. +*/ +if (is_pmd_migration_entry (pmde)) +return SCAN_PMD_MAPPED ;if (!pmd_present (pmde))return SCAN_PMD_NULL ;if (pmd_trans_huge (pmde))
Signature · 1 line
2.30.2
Quoted 15 lines · Dev Jain
Suppose a folio is under migration, and khugepaged is also trying to collapse it.
collapse_pte_mapped_thp()will retrieve the folio from the page cache viafilemap_lock_folio(), thus taking a reference on the folio and sleeping on the folio lock, since the lock is held by the migration path. Migration will then fail in__folio_migrate_mapping->folio_ref_freeze. Reduce the probability of such a race happening (leading to migration failure) by bailing out if we detect a PMD is marked with a migration entry.This fixes the migration-shared-anon-thp testcase failure on Apple M3.
Note that, this is not a "fix" since it only reduces the chance of interference of khugepaged with migration, wherein both the kernel functionalities are deemed "best-effort".
Quoted 6 lines
Acked-by Oscar Salvador <[email protected]>Reviewed-by Anshuman Khandual <[email protected]>Reviewed-by Zi Yan <[email protected]>Reviewed-by Baolin Wang <[email protected]>Signed-off-by Dev Jain <[email protected]>
LGTM, so:
Quoted 37 lines
---
v2->v3:
- Improve comment (David)
v1->v2:
- RemoveSCAN_PMD_MIGRATION, merge intoSCAN_PMD_MAPPED(David, Anshuman)
- Add a comment (Lorenzo)v1:
- https://lore.kernel.org/all/[email protected]/mm/khugepaged.c | 8 ++++++++ 1 file changed, 8 insertions(+)mm/khugepaged.c +8
-1@@ -941,6 +941,14 @@ static inline int check_pmd_state(pmd_t *pmd)if (pmd_none (pmde))return SCAN_PMD_NONE ; + +/* +* The folio may be under migration when khugepaged is trying to +* collapse it. Migration success or failure will eventually end +* up with a present PMD mapping a folio again. +*/ +if (is_pmd_migration_entry (pmde)) +return SCAN_PMD_MAPPED ;if (!pmd_present (pmde))return SCAN_PMD_NULL ;if (pmd_trans_huge (pmde)) --2.30.2
Quoted 16 lines · Dev Jain
Suppose a folio is under migration, and khugepaged is also trying to collapse it.
collapse_pte_mapped_thp()will retrieve the folio from the page cache viafilemap_lock_folio(), thus taking a reference on the folio and sleeping on the folio lock, since the lock is held by the migration path. Migration will then fail in__folio_migrate_mapping->folio_ref_freeze. Reduce the probability of such a race happening (leading to migration failure) by bailing out if we detect a PMD is marked with a migration entry.This fixes the migration-shared-anon-thp testcase failure on Apple M3.
Note that, this is not a "fix" since it only reduces the chance of interference of khugepaged with migration, wherein both the kernel functionalities are deemed "best-effort".
Quoted 6 lines
Acked-by Oscar Salvador <[email protected]>Reviewed-by Anshuman Khandual <[email protected]>Reviewed-by Zi Yan <[email protected]>Reviewed-by Baolin Wang <[email protected]>Signed-off-by Dev Jain <[email protected]>
LGTM! This is a nice check to have here :)
Quoted 37 lines
---
v2->v3:
- Improve comment (David)
v1->v2:
- RemoveSCAN_PMD_MIGRATION, merge intoSCAN_PMD_MAPPED(David, Anshuman)
- Add a comment (Lorenzo)v1:
- https://lore.kernel.org/all/[email protected]/mm/khugepaged.c | 8 ++++++++ 1 file changed, 8 insertions(+)mm/khugepaged.c +8
-1@@ -941,6 +941,14 @@ static inline int check_pmd_state(pmd_t *pmd)if (pmd_none (pmde))return SCAN_PMD_NONE ; + +/* +* The folio may be under migration when khugepaged is trying to +* collapse it. Migration success or failure will eventually end +* up with a present PMD mapping a folio again. +*/ +if (is_pmd_migration_entry (pmde)) +return SCAN_PMD_MAPPED ;if (!pmd_present (pmde))return SCAN_PMD_NULL ;if (pmd_trans_huge (pmde)) --2.30.2