The following series provides khugepaged with the capability to collapse
anonymous memory regions to mTHPs.
To achieve this we generalize the khugepaged functions to no longer depend
on PMD_ORDER. Then during the PMD scan, we use a bitmap to track individual
pages that are occupied (!none/zero). After the PMD scan is done, we use
the bitmap to find the optimal mTHP sizes for the PMD range. The
restriction on max_ptes_none is removed during the scan, to make sure we
account for the whole PMD range in the bitmap. When no mTHP size is
enabled, the legacy behavior of khugepaged is maintained.
We currently only support max_ptes_none values of 0 or HPAGE_PMD_NR - 1
(ie 511). If any other value is specified, the kernel will emit a warning
and no mTHP collapse will be attempted. If a mTHP collapse is attempted,
but contains swapped out, or shared pages, we don't perform the collapse.
It is now also possible to collapse to mTHPs without requiring the PMD THP
size to be enabled. These limitations are to prevent collapse "creep"
behavior. This prevents constantly promoting mTHPs to the next available
size, which would occur because a collapse introduces more non-zero pages
that would satisfy the promotion condition on subsequent scans.
Patch 1-2: Generalize hugepage_vma_revalidate and alloc_charge_folio
for arbitrary orders.
Patch 3: Rework max_ptes_* handling into helper functions
Patch 4-5: Generalize __collapse_huge_page_* and collapse_huge_page
Patch 6: Skip collapsing mTHP to smaller orders
Patch 7-8: Add per-order mTHP statistics and tracepoints
Patch 9: Introduce collapse_allowable_orders helper function
Patch 10-12: Introduce bitmap and mTHP collapse support, fully enabled
Patch 13: Documentation
Testing:
- Built for x86_64, aarch64, ppc64le, and s390x
- ran all arches on test suites provided by the kernel-tests project
- internal testing suites: functional testing and performance testing
- selftests mm
- I created a test script that I used to push khugepaged to its limits
while monitoring a number of stats and tracepoints. The code is
available here[1] (Run in legacy mode for these changes and set mthp
sizes to inherit)
The summary from my testings was that there was no significant
regression noticed through this test. In some cases my changes had
better collapse latencies, and was able to scan more pages in the same
amount of time/work, but for the most part the results were consistent.
- redis testing. I did some testing with these changes along with my defer
changes (see followup [2] post for more details). We've decided to get
the mTHP changes merged first before attempting the defer series.
- some basic testing on 64k page size.
- lots of general use.
V16 Changes:
- This was quite the respin, so hopefully everything is ok! I dropped
some acks/rb on patches that changed significantly. Major changes
are related to the locking cleanup in collapse_huge_page().
sorry if i missed any feedback there was quite a bit of minor requests
- New cleanup patch: introduce collapse_max_ptes_none/shared/swap()
helpers, absorbing madvise_collapse special-casing (David)
- Merge "introduce collapse_max_ptes_none helper" into the generalize
__collapse_huge_page_* patch, dropping the separate patch (David)
- Simplify collapse_huge_page() locking: move mmap_read_unlock() to
caller, remove mmap_locked tracking parameter entirely (Lorenzo)
- Use enum tva_type instead of bool is_khugepaged in
collapse_allowable_orders() (David, Lorenzo)
- collapse_allowable_orders() takes (vma, tva_flags), accesses
vma->vm_flags internally (Lorenzo)
- Rename mthp_* functions to collapse_mthp_* prefix (David, Lorenzo)
- Use MAX_PTRS_PER_PTE consistently for bitmap ops (Lorenzo)
- Downgrade BUG_ON() to WARN_ON_ONCE() in collapse_huge_page() (Lorenzo)
- Add /*expect_anon=*/true annotation to revalidate calls (Lorenzo)
- Pull common spin_lock + WARN_ON_ONCE out of PMD/mTHP branch (Lorenzo)
- Fix pteval declaration style; use end_addr variable (Lorenzo)
- Use ternary for result assignment in collapse_scan_pmd() (Lorenzo)
- Improve anon_vma locking comment for mTHP case (Lorenzo)
- Document why skip-smaller-order isn't done in scan phase (David)
- Make documentation of max_ptes_none mTHP behavior (only 0 and
HPAGE_PMD_NR-1 supported) in admin guide more apparent. (David, Lorenzo)
- Collect new Acks/RB tags from v15 review
V15: https://lore.kernel.org/all/20260226031741.230674-1-npache@redhat.com
V14: https://lore.kernel.org/all/20260122192841.128719-1-npache@redhat.com
V13: https://lore.kernel.org/all/20251201174627.23295-1-npache@redhat.com
V12: https://lore.kernel.org/all/20251022183717.70829-1-npache@redhat.com
V11: https://lore.kernel.org/all/20250912032810.197475-1-npache@redhat.com
V10: https://lore.kernel.org/all/20250819134205.622806-1-npache@redhat.com
V9 : https://lore.kernel.org/all/20250714003207.113275-1-npache@redhat.com
V8 : https://lore.kernel.org/all/20250702055742.102808-1-npache@redhat.com
V7 : https://lore.kernel.org/all/20250515032226.128900-1-npache@redhat.com
V6 : https://lore.kernel.org/all/20250515030312.125567-1-npache@redhat.com
V5 : https://lore.kernel.org/all/20250428181218.85925-1-npache@redhat.com
V4 : https://lore.kernel.org/all/20250417000238.74567-1-npache@redhat.com
V3 : https://lore.kernel.org/all/20250414220557.35388-1-npache@redhat.com
V2 : https://lore.kernel.org/all/20250211003028.213461-1-npache@redhat.com
V1 : https://lore.kernel.org/all/20250108233128.14484-1-npache@redhat.com
A big thanks to everyone that has reviewed, tested, and participated in
the development process. Its been a great experience working with all of
you on this endeavor.
[1] - https://gitlab.com/npache/khugepaged_mthp_test
[2] - https://lore.kernel.org/lkml/20250515033857.132535-1-npache@redhat.com/
Baolin Wang (1):
mm/khugepaged: run khugepaged for all orders
Dev Jain (1):
mm/khugepaged: generalize alloc_charge_folio()
Nico Pache (11):
mm/khugepaged: generalize hugepage_vma_revalidate for mTHP support
mm/khugepaged: rework max_ptes_* handling with helper functions
mm/khugepaged: generalize __collapse_huge_page_* for mTHP support
mm/khugepaged: generalize collapse_huge_page for mTHP collapse
mm/khugepaged: skip collapsing mTHP to smaller orders
mm/khugepaged: add per-order mTHP collapse failure statistics
mm/khugepaged: improve tracepoints for mTHP orders
mm/khugepaged: introduce collapse_allowable_orders helper function
mm/khugepaged: Introduce mTHP collapse support
mm/khugepaged: avoid unnecessary mTHP collapse attempts
Documentation: mm: update the admin guide for mTHP collapse
Documentation/admin-guide/mm/transhuge.rst | 81 ++-
include/linux/huge_mm.h | 5 +
include/linux/khugepaged.h | 6 +-
include/trace/events/huge_memory.h | 34 +-
mm/huge_memory.c | 13 +-
mm/khugepaged.c | 623 ++++++++++++++++-----
mm/vma.c | 6 +-
tools/testing/vma/include/stubs.h | 3 +-
8 files changed, 589 insertions(+), 182 deletions(-)
base-commit: 4569c01e84d440f1f6c79df1e13e9c86a9fb80ca
--
2.53.0