This patchset makes it possible for MGLRU to consult secondary MMUs
while doing aging, not just during eviction. This allows for more
accurate reclaim decisions, which is especially important for proactive
reclaim.
This series includes:
1. Cleanup, add support for locklessly memslot walks in KVM (patches
1-2).
2. Support for lockless aging for x86 TDP MMU (patches 3-4).
3. Further small optimizations (patches 5-6).
4. Support for lockless harvesting of access information for the x86
shadow MMU (patches 7-10).
5. Some mm cleanup (patch 11).
6. Add fast-only aging MMU notifiers (patches 12-13).
7. Support fast-only aging in KVM/x86 (patches 14-16).
8. Have KVM participate in MGLRU aging (patch 17).
9. Updates to the access_tracking_perf_test to verify MGLRU
functionality (patch 18).
Patches 1-10 are pure optimizations and could be applied without the
rest of the series, though the lockless shadow MMU lockless patches
become more useful in the context of MGLRU aging.
Please note that mmu_notifier_test_young_fast_only() is added but not
used in this series. I am happy to remove it if that would be
appropriate.
The fast-only notifiers serve a particular purpose: for aging, we
neither want to delay other operations (e.g. unmapping for eviction)
nor do we want to be delayed by these other operations ourselves. By
default, the implementations of test_young() and clear_young() are meant
to be *accurate*, not fast. The fast-only notifiers will only give age
information that can be gathered fast.
The fast-only notifiers are non-trivially implemented for only x86. The
TDP MMU and the shadow MMU are both supported, but the shadow MMU will
not actually age sptes locklessly if A/D bits in the spte have been
disabled (i.e., if L1 disables them).
access_tracking_perf_test now has a mode (-p) to check performance of
MGLRU aging while the VM is faulting memory in.
This series has been tested with access_tracking_perf_test and Sean's
mmu_stress_test[6], both with tdp_mmu=0 and tdp_mmu=1.
=== Previous Versions ===
Since v6[1]:
- Rebased on top of kvm-x86/next and Sean's lockless rmap walking
changes[6].
- Removed HAVE_KVM_MMU_NOTIFIER_YOUNG_FAST_ONLY (thanks DavidM).
- Split up kvm_age_gfn() / kvm_test_age_gfn() optimizations (thanks
DavidM and Sean).
- Improved new MMU notifier documentation (thanks DavidH).
- Dropped arm64 locking change.
- No longer retry for CAS failure in TDP MMU non-A/D case (thanks
Sean).
- Added some R-bys and A-bys.
Since v5[2]:
- Reworked test_clear_young_fast_only() into a new parameter for the
existing notifiers (thanks Sean).
- Added mmu_notifier.has_fast_aging to tell mm if calling fast-only
notifiers should be done.
- Added mm_has_fast_young_notifiers() to inform users if calling
fast-only notifier helpers is worthwhile (for look-around to use).
- Changed MGLRU to invoke a single notifier instead of two when
aging and doing look-around (thanks Yu).
- For KVM/x86, check indirect_shadow_pages > 0 instead of
kvm_memslots_have_rmaps() when collecting age information
(thanks Sean).
- For KVM/arm, some fixes from Oliver.
- Small fixes to access_tracking_perf_test.
- Added missing !MMU_NOTIFIER version of mmu_notifier_clear_young().
Since v4[3]:
- Removed Kconfig that controlled when aging was enabled. Aging will
be done whenever the architecture supports it (thanks Yu).
- Added a new MMU notifier, test_clear_young_fast_only(), specifically
for MGLRU to use.
- Add kvm_fast_{test_,}age_gfn, implemented by x86.
- Fix locking for clear_flush_young().
- Added KVM_MMU_NOTIFIER_YOUNG_LOCKLESS to clean up locking changes
(thanks Sean).
- Fix WARN_ON and other cleanup for the arm64 locking changes
(thanks Oliver).
Since v3[4]:
- Vastly simplified the series (thanks David). Removed mmu notifier
batching logic entirely.
- Cleaned up how locking is done for mmu_notifier_test/clear_young
(thanks David).
- Look-around is now only done when there are no secondary MMUs
subscribed to MMU notifiers.
- CONFIG_LRU_GEN_WALKS_SECONDARY_MMU has been added.
- Fixed the lockless implementation of kvm_{test,}age_gfn for x86
(thanks David).
- Added MGLRU functional and performance tests to
access_tracking_perf_test (thanks Axel).
- In v3, an mm would be completely ignored (for aging) if there was a
secondary MMU but support for secondary MMU walking was missing. Now,
missing secondary MMU walking support simply skips the notifier
calls (except for eviction).
- Added a sanity check for that range->lockless and range->on_lock are
never both provided for the memslot walk.
For the changes since v2[5], see v3.
Based on latest kvm-x86/next.
[1]: https://lore.kernel.org/linux-mm/20240724011037.3671523-1-jthoughton@xxxxxxxxxx/
[2]: https://lore.kernel.org/linux-mm/20240611002145.2078921-1-jthoughton@xxxxxxxxxx/
[3]: https://lore.kernel.org/linux-mm/20240529180510.2295118-1-jthoughton@xxxxxxxxxx/
[4]: https://lore.kernel.org/linux-mm/20240401232946.1837665-1-jthoughton@xxxxxxxxxx/
[5]: https://lore.kernel.org/kvmarm/20230526234435.662652-1-yuzhao@xxxxxxxxxx/
[6]: https://lore.kernel.org/kvm/20240809194335.1726916-1-seanjc@xxxxxxxxxx/
James Houghton (14):
KVM: Remove kvm_handle_hva_range helper functions
KVM: Add lockless memslot walk to KVM
KVM: x86/mmu: Factor out spte atomic bit clearing routine
KVM: x86/mmu: Relax locking for kvm_test_age_gfn and kvm_age_gfn
KVM: x86/mmu: Rearrange kvm_{test_,}age_gfn
KVM: x86/mmu: Only check gfn age in shadow MMU if
indirect_shadow_pages > 0
mm: Add missing mmu_notifier_clear_young for !MMU_NOTIFIER
mm: Add has_fast_aging to struct mmu_notifier
mm: Add fast_only bool to test_young and clear_young MMU notifiers
KVM: Pass fast_only to kvm_{test_,}age_gfn
KVM: x86/mmu: Locklessly harvest access information from shadow MMU
KVM: x86/mmu: Enable has_fast_aging
mm: multi-gen LRU: Have secondary MMUs participate in aging
KVM: selftests: Add multi-gen LRU aging to access_tracking_perf_test
Sean Christopherson (4):
KVM: x86/mmu: Refactor low level rmap helpers to prep for walking w/o
mmu_lock
KVM: x86/mmu: Add infrastructure to allow walking rmaps outside of
mmu_lock
KVM: x86/mmu: Add support for lockless walks of rmap SPTEs
KVM: x86/mmu: Support rmap walks without holding mmu_lock when aging
gfns
Documentation/admin-guide/mm/multigen_lru.rst | 6 +-
arch/x86/include/asm/kvm_host.h | 4 +-
arch/x86/kvm/Kconfig | 1 +
arch/x86/kvm/mmu/mmu.c | 355 ++++++++++++----
arch/x86/kvm/mmu/tdp_iter.h | 27 +-
arch/x86/kvm/mmu/tdp_mmu.c | 57 ++-
include/linux/kvm_host.h | 2 +
include/linux/mmu_notifier.h | 82 +++-
include/linux/mmzone.h | 6 +-
include/trace/events/kvm.h | 19 +-
mm/damon/vaddr.c | 2 -
mm/mmu_notifier.c | 38 +-
mm/rmap.c | 9 +-
mm/vmscan.c | 148 +++++--
tools/testing/selftests/kvm/Makefile | 1 +
.../selftests/kvm/access_tracking_perf_test.c | 369 +++++++++++++++--
.../selftests/kvm/include/lru_gen_util.h | 55 +++
.../testing/selftests/kvm/lib/lru_gen_util.c | 391 ++++++++++++++++++
virt/kvm/Kconfig | 3 +
virt/kvm/kvm_main.c | 124 +++---
20 files changed, 1451 insertions(+), 248 deletions(-)
create mode 100644 tools/testing/selftests/kvm/include/lru_gen_util.h
create mode 100644 tools/testing/selftests/kvm/lib/lru_gen_util.c
base-commit: 3cc25d5adcfd2a2c33baa0b2a1979c2dbc9b990b
--
2.46.0.792.g87dc391469-goog
