From 84990b169557428c318df87b7836cd15f65b62dc Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Sat, 26 Nov 2022 19:59:16 +1000 Subject: powerpc/qspinlock: add mcs queueing for contended waiters This forms the basis of the qspinlock slow path. Like generic qspinlocks and unlike the vanilla MCS algorithm, the lock owner does not participate in the queue, only waiters. The first waiter spins on the lock word, then when the lock is released it takes ownership and unqueues the next waiter. This is how qspinlocks can be implemented with the spinlock API -- lock owners don't need a node, only waiters do. Signed-off-by: Nicholas Piggin Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/20221126095932.1234527-2-npiggin@gmail.com --- arch/powerpc/include/asm/qspinlock.h | 10 +++++++--- arch/powerpc/include/asm/qspinlock_types.h | 23 +++++++++++++++++++++++ 2 files changed, 30 insertions(+), 3 deletions(-) (limited to 'arch/powerpc/include') diff --git a/arch/powerpc/include/asm/qspinlock.h b/arch/powerpc/include/asm/qspinlock.h index 5e6257313557..6946dba5d087 100644 --- a/arch/powerpc/include/asm/qspinlock.h +++ b/arch/powerpc/include/asm/qspinlock.h @@ -19,12 +19,12 @@ static __always_inline int queued_spin_value_unlocked(struct qspinlock lock) static __always_inline int queued_spin_is_contended(struct qspinlock *lock) { - return 0; + return !!(atomic_read(&lock->val) & _Q_TAIL_CPU_MASK); } static __always_inline int queued_spin_trylock(struct qspinlock *lock) { - return atomic_cmpxchg_acquire(&lock->val, 0, 1) == 0; + return atomic_cmpxchg_acquire(&lock->val, 0, _Q_LOCKED_VAL) == 0; } void queued_spin_lock_slowpath(struct qspinlock *lock); @@ -37,7 +37,11 @@ static __always_inline void queued_spin_lock(struct qspinlock *lock) static inline void queued_spin_unlock(struct qspinlock *lock) { - atomic_set_release(&lock->val, 0); + for (;;) { + int val = atomic_read(&lock->val); + if (atomic_cmpxchg_release(&lock->val, val, val & ~_Q_LOCKED_VAL) == val) + return; + } } #define arch_spin_is_locked(l) queued_spin_is_locked(l) diff --git a/arch/powerpc/include/asm/qspinlock_types.h b/arch/powerpc/include/asm/qspinlock_types.h index 59606bc0c774..20a36dfb14e2 100644 --- a/arch/powerpc/include/asm/qspinlock_types.h +++ b/arch/powerpc/include/asm/qspinlock_types.h @@ -10,4 +10,27 @@ typedef struct qspinlock { #define __ARCH_SPIN_LOCK_UNLOCKED { .val = ATOMIC_INIT(0) } +/* + * Bitfields in the lock word: + * + * 0: locked bit + * 1-16: unused bits + * 17-31: tail cpu (+1) + */ +#define _Q_SET_MASK(type) (((1U << _Q_ ## type ## _BITS) - 1)\ + << _Q_ ## type ## _OFFSET) +/* 0x00000001 */ +#define _Q_LOCKED_OFFSET 0 +#define _Q_LOCKED_BITS 1 +#define _Q_LOCKED_VAL (1U << _Q_LOCKED_OFFSET) + +/* 0xfffe0000 */ +#define _Q_TAIL_CPU_OFFSET 17 +#define _Q_TAIL_CPU_BITS 15 +#define _Q_TAIL_CPU_MASK _Q_SET_MASK(TAIL_CPU) + +#if CONFIG_NR_CPUS >= (1U << _Q_TAIL_CPU_BITS) +#error "qspinlock does not support such large CONFIG_NR_CPUS" +#endif + #endif /* _ASM_POWERPC_QSPINLOCK_TYPES_H */ -- cgit v1.3.1