[PATCH v5 0/3] Consolidate memory policy interpretation in alloc_hugetlb_folio()
From: Ackerley Tng via B4 Relay
Date: Mon Aug 03 2026 - 09:48:28 EST
This patch series is the first 3 patches, split out from "Open HugeTLB
allocation routine for more generic use" [1].
I'm splitting these 3 patches into this separate series, and continuing as
v5 from [1] to avoid being blocked on addressing pre-existing
bugs. (Sashiko's reviews on [1] revealed quite a few pre-existing bugs,
which is separately being addressed in [2].)
These 3 patches have been reviewed and can be merged first. (Thank you
Andrew!) The other 3 patches in [1] will probably build on [2], so I'd like
to leave those to later.
In this v5, the only change is to the first patch - I improved the wording
of the comment just above the call to dequeue_hugetlb_folio_vma() in
alloc_hugetlb_folio().
[1] https://lore.kernel.org/all/20260702-hugetlb-open-up-v4-0-d53cefcccf34@xxxxxxxxxx/T/
[2] https://lore.kernel.org/all/20260722-hugetlb-alloc-failure-fixes-v4-0-88e8b81970dc@xxxxxxxxxx/T/
Testing:
+ libhugetlbfs tests pass
+ tools/testing/selftests/mm/ksft_hugetlb.sh passes
RFC v1: https://lore.kernel.org/all/bb35a69a-5be9-45f5-a557-1902487a1bc2@xxxxxxxxx/
v2: https://lore.kernel.org/r/20260506-hugetlb-open-up-v2-0-826a0c5f28fc@xxxxxxxxxx
v3: https://lore.kernel.org/r/20260518-hugetlb-open-up-v3-0-e14b302477f8@xxxxxxxxxx
v4: https://lore.kernel.org/r/20260702-hugetlb-open-up-v4-0-d53cefcccf34@xxxxxxxxxx
Signed-off-by: Ackerley Tng <ackerleytng@xxxxxxxxxx>
---
Ackerley Tng (3):
mm: hugetlb: Consolidate interpretation of gbl_chg within alloc_hugetlb_folio()
mm: hugetlb: Move mpol interpretation out of alloc_buddy_hugetlb_folio_with_mpol()
mm: hugetlb: Move mpol interpretation out of dequeue_hugetlb_folio_vma()
include/uapi/linux/mempolicy.h | 2 +-
mm/hugetlb.c | 110 +++++++++++++++++++++++------------------
2 files changed, 62 insertions(+), 50 deletions(-)
---
base-commit: 075b74841bd0065a3bda3440873c747938e69b68
change-id: 20260803-hugetlb-mpol-interpretation-1c61e03073b4
Best regards,
--
Ackerley Tng <ackerleytng@xxxxxxxxxx>