Re: [PATCH RFC v2 00/15] hazptr: batch synchronize operations through a shared scan

[Bradley Morgan]

On 2 October 2026 18:08:32 BST, Kunwu Chan <kunwu.chan@xxxxxxxxx> wrote:
>This series extends Paul McKenney's v3 hazptr implementation [1]
>with a shared scan path for concurrent hazptr_synchronize() callers,
>and adapts the lockdep dynamic-key hashlist use case from Boqun
>Feng's 2025 shazptr series [2] to the current hazptr API.
>
>The series also adds rcuscale support and torture coverage for
>the hazptr implementation.
>
>The lockdep conversion replaces the expedited RCU wait in
>lockdep_unregister_key() with hazptr_synchronize().  This limits
>the wait to hazard pointers protecting the target hash bucket
>instead of waiting for a system-wide expedited RCU grace period.
>
>[1]
>https://lore.kernel.org/all/20260919000056.3132131-26-paulmck@xxxxxxxxxx/
>[2]
>https://lore.kernel.org/lkml/20250625031101.12555-1-boqun.feng@xxxxxxxxx/
>


This is well needed. Thanks for the patch, I'll maybe review maybe test if
I have time, cough cough, school!

>Performance data
>================
>
>All measurements are on an ARM64 KVM guest with
>HAZPTR_SCALE_NR_OBJS=8 and round-robin updater selection; unless
>noted otherwise, nreaders=0.  Scale tests run with
>PROVE_LOCKING=n.
>
>rcuscale synchronize latency (96 CPUs, nw=1):
>
>              nreaders=0      nreaders=1       nreaders=4      nreaders=96
>    scale       avg              avg             avg              avg
>    --------------------------------------------------------------------
>    hazptr     46 us            47 us          102 us            106 ms
>    RCU       8.3 ms           8.5 ms         16.6 ms         54.4 ms
>    SRCU      8.2 ms           8.0 ms         11.7 ms         11.8 ms
>
>Hazptr single-writer latency remains below 110 us with up to
>four readers on this 96-CPU guest, but rises to 106 ms when
>all 96 CPUs hold hazard pointers.
>
>Under 16 concurrent synchronize callers (nw=16), per-writer
>latency at four CPU counts:
>
>              hazptr nw=16      RCU nw=16     SRCU nw=16
>    CPUs       avg               avg             avg
>    -----------------------------------------------------------
>      24       8.0 ms          13.6 ms          9.2 ms
>      96       8.0 ms          13.3 ms          8.6 ms
>     128       7.9 ms          13.0 ms         11.7 ms
>     256       8.0 ms          20.8 ms          9.6 ms
>
>Hazptr remains around 8.0 ms across these CPU counts, with the
>scan-kthread retry interval contributing to the latency.
>RCU rises to ~21 ms at 256 CPUs while SRCU stays around 9-12 ms.
>
>Reader-side overhead (refscale, 96 readers on the 96-CPU guest,
>3 runs each):
>
>    hazptr     25.7 ns/op
>    RCU        96.7 ns/op
>    SRCU      134.6 ns/op
>
>lockdep workload -- tc qdisc mq x100, ARM64 KVM, PROVE_LOCKING=y,
>96 background hazptr readers, function-call IPIs per 100 ops:
>
>    CPUs       hazptr              exp RCU            reduction
>    ------------------------------------------------------------
>      24       196                 207                 5%
>      96       189                 265                29%
>     128       193                 266                27%
>     256       222                 384                42%
>
>Hazptr issues fewer function-call IPIs at each CPU count, with
>the difference reaching 42% at 256 CPUs.
>
>The hazptr.sh test suite passes, including the lockdep scenarios
>and the 8-to-256 CPU sweep, with no lockdep warnings, deadlocks,
>or crashes.
>Additional x86 server testing with Lian Wang is planned(maybe after
>LPC).
>
>Changes since RFC/WIP
>=====================
>
>- RFC/WIP: https://lore.kernel.org/all/20260922070950.4173245-1-kunwu.chan@xxxxxxxxx/
>
>- Split the original 4-patch RFC/WIP into smaller commits covering
>  shared scanning, correctness, API support, lockdep, scaling, and
>  torture testing.
>
>- Incorporated Boqun Feng's review feedback: use a Bloom filter to
>  avoid per-waiter allocation, add scoped_guard() support, and add
>  a debug option to force the hazptr acquire slow path.
>
>- Fixed scan ordering around backup-slot promotion by scanning all
>  per-CPU slots before the overflow lists, with a separate
>  overflow-list phase.
>
>- Simplified the scan cycle to flip first and drain only the old
>  wildcard generation, with herd7-verified LKMM tests for both the
>  in-flight and resolved publication cases.
>
>- Extended rcuscale and hazptrtorture coverage, added a selftest
>  script for the torture configurations, and fixed the
>  hazptr_release() kernel-doc.
>
>Kunwu Chan (15):
>  hazptr: add shared scan kthread
>  hazptr: use Bloom filter for shared scan waiters
>  hazptr: scan all per-CPU slots before overflow lists
>  hazptr: add scoped_guard() support
>  hazptr: add debug option to force the acquire slow path
>  hazptr: elide redundant first drain pass
>  Documentation/litmus-tests: add hazptr wildcard-flip escape test
>  locking/lockdep: use hazptr to wait for dynamic key lookups
>  rcuscale: add hazptr scale type
>  hazptr: fix kernel-doc of hazptr_release()
>  Documentation/litmus-tests: add hazptr acquire-before-scan test
>  hazptrtorture: add slowpath and lockdep scenarios
>  hazptrtorture: add READERS4 and READERS0 torture configs
>  hazptrtorture: add 128- and 256-CPU configs
>  selftests/rcutorture: add hazptr torture test script
>
> Documentation/litmus-tests/README             |  13 +
> .../hazptr/hazptr-acquire-before-scan.litmus  |  45 +++
> .../hazptr/hazptr-wildcard-flip-escape.litmus |  46 +++
> include/linux/hazptr.h                        |  56 ++-
> kernel/hazptr.c                               | 336 +++++++++++++++++-
> kernel/locking/lockdep.c                      |  25 +-
> kernel/rcu/Kconfig.debug                      |  10 +
> kernel/rcu/hazptrtorture.c                    |  57 ++-
> kernel/rcu/rcuscale.c                         |  70 +++-
> .../selftests/rcutorture/bin/hazptr.sh        | 146 ++++++++
> .../rcutorture/configs/hazptr/CFLIST          |   6 +
> .../rcutorture/configs/hazptr/CPU128          |  16 +
> .../rcutorture/configs/hazptr/CPU128.boot     |   1 +
> .../rcutorture/configs/hazptr/CPU256          |  16 +
> .../rcutorture/configs/hazptr/CPU256.boot     |   1 +
> .../rcutorture/configs/hazptr/LOCKDEP         |  17 +
> .../rcutorture/configs/hazptr/LOCKDEP.boot    |   1 +
> .../rcutorture/configs/hazptr/READERS0        |  16 +
> .../rcutorture/configs/hazptr/READERS0.boot   |   2 +
> .../rcutorture/configs/hazptr/READERS4        |  16 +
> .../rcutorture/configs/hazptr/READERS4.boot   |   2 +
> .../rcutorture/configs/hazptr/SLOWPATH        |  16 +
> 22 files changed, 881 insertions(+), 33 deletions(-)
> create mode 100644
> Documentation/litmus-tests/hazptr/hazptr-acquire-before-scan.litmus
> create mode 100644
> Documentation/litmus-tests/hazptr/hazptr-wildcard-flip-escape.litmus
> create mode 100755 tools/testing/selftests/rcutorture/bin/hazptr.sh
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/CPU128
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/CPU128.boot
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/CPU256
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/CPU256.boot
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/LOCKDEP
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/LOCKDEP.boot
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/READERS0
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/READERS0.boot
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/READERS4
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/READERS4.boot
> create mode 100644
> tools/testing/selftests/rcutorture/configs/hazptr/SLOWPATH
>
>

--- Thanks!
"I'm not a very positive person" - Linus torvalds