linux/init
Peter Zijlstra 56180dd20c futex: Use RCU-based per-CPU reference counting instead of rcuref_t
The use of rcuref_t for reference counting introduces a performance bottleneck
when accessed concurrently by multiple threads during futex operations.

Replace rcuref_t with special crafted per-CPU reference counters. The
lifetime logic remains the same.

The newly allocate private hash starts in FR_PERCPU state. In this state, each
futex operation that requires the private hash uses a per-CPU counter (an
unsigned int) for incrementing or decrementing the reference count.

When the private hash is about to be replaced, the per-CPU counters are
migrated to a atomic_t counter mm_struct::futex_atomic.
The migration process:
- Waiting for one RCU grace period to ensure all users observe the
  current private hash. This can be skipped if a grace period elapsed
  since the private hash was assigned.

- futex_private_hash::state is set to FR_ATOMIC, forcing all users to
  use mm_struct::futex_atomic for reference counting.

- After a RCU grace period, all users are guaranteed to be using the
  atomic counter. The per-CPU counters can now be summed up and added to
  the atomic_t counter. If the resulting count is zero, the hash can be
  safely replaced. Otherwise, active users still hold a valid reference.

- Once the atomic reference count drops to zero, the next futex
  operation will switch to the new private hash.

call_rcu_hurry() is used to speed up transition which otherwise might be
delay with RCU_LAZY. There is nothing wrong with using call_rcu(). The
side effects would be that on auto scaling the new hash is used later
and the SET_SLOTS prctl() will block longer.

[bigeasy: commit description + mm get/ put_async]

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250710110011.384614-3-bigeasy@linutronix.de
2025-07-11 16:02:00 +02:00
..
.gitignore kbuild: build init/built-in.a just once 2022-09-29 04:40:15 +09:00
.kunitconfig initramfs_test: kunit tests for initramfs unpacking 2025-03-08 12:13:04 +01:00
calibrate.c
do_mounts_initrd.c treewide: const qualify ctl_tables where applicable 2025-01-28 13:48:37 +01:00
do_mounts_rd.c
do_mounts.c init: replace deprecated strncpy with strscpy_pad 2024-04-30 10:34:29 -07:00
do_mounts.h init: flush async file closing 2024-02-08 18:41:03 +01:00
init_task.c signal: Provide ignored_posix_timers list 2024-11-07 02:14:45 +01:00
initramfs_internal.h init: add initramfs_internal.h 2025-03-04 09:52:36 +01:00
initramfs_test.c initramfs_test: kunit tests for initramfs unpacking 2025-03-08 12:13:04 +01:00
initramfs.c init: fix build warnings about export.h 2025-06-11 22:42:36 -07:00
Kconfig futex: Use RCU-based per-CPU reference counting instead of rcuref_t 2025-07-11 16:02:00 +02:00
main.c init: fix build warnings about export.h 2025-06-11 22:42:36 -07:00
Makefile initramfs_test: kunit tests for initramfs unpacking 2025-03-08 12:13:04 +01:00
noinitramfs.c
version-timestamp.c init/version-timestamp.c: remove unneeded #include <linux/version.h> 2023-01-09 07:34:38 +09:00
version.c init/version.c: Replace strlcpy with strscpy 2023-09-22 09:50:56 -07:00