linux_dsm_epyc7002/arch/s390/include/asm/spinlock.h
Peter Zijlstra 726328d92a locking/spinlock, arch: Update and fix spin_unlock_wait() implementations
This patch updates/fixes all spin_unlock_wait() implementations.

The update is in semantics; where it previously was only a control
dependency, we now upgrade to a full load-acquire to match the
store-release from the spin_unlock() we waited on. This ensures that
when spin_unlock_wait() returns, we're guaranteed to observe the full
critical section we waited on.

This fixes a number of spin_unlock_wait() users that (not
unreasonably) rely on this.

I also fixed a number of ticket lock versions to only wait on the
current lock holder, instead of for a full unlock, as this is
sufficient.

Furthermore; again for ticket locks; I added an smp_rmb() in between
the initial ticket load and the spin loop testing the current value
because I could not convince myself the address dependency is
sufficient, esp. if the loads are of different sizes.

I'm more than happy to remove this smp_rmb() again if people are
certain the address dependency does indeed work as expected.

Note: PPC32 will be fixed independently

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: chris@zankel.net
Cc: cmetcalf@mellanox.com
Cc: davem@davemloft.net
Cc: dhowells@redhat.com
Cc: james.hogan@imgtec.com
Cc: jejb@parisc-linux.org
Cc: linux@armlinux.org.uk
Cc: mpe@ellerman.id.au
Cc: ralf@linux-mips.org
Cc: realmz6@gmail.com
Cc: rkuo@codeaurora.org
Cc: rth@twiddle.net
Cc: schwidefsky@de.ibm.com
Cc: tony.luck@intel.com
Cc: vgupta@synopsys.com
Cc: ysato@users.sourceforge.jp
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-06-14 11:55:15 +02:00

281 lines
6.2 KiB
C

/*
* S390 version
* Copyright IBM Corp. 1999
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*
* Derived from "include/asm-i386/spinlock.h"
*/
#ifndef __ASM_SPINLOCK_H
#define __ASM_SPINLOCK_H
#include <linux/smp.h>
#include <asm/barrier.h>
#include <asm/processor.h>
#define SPINLOCK_LOCKVAL (S390_lowcore.spinlock_lockval)
extern int spin_retry;
static inline int
_raw_compare_and_swap(unsigned int *lock, unsigned int old, unsigned int new)
{
return __sync_bool_compare_and_swap(lock, old, new);
}
/*
* Simple spin lock operations. There are two variants, one clears IRQ's
* on the local processor, one does not.
*
* We make no fairness assumptions. They have a cost.
*
* (the type definitions are in asm/spinlock_types.h)
*/
void arch_lock_relax(unsigned int cpu);
void arch_spin_lock_wait(arch_spinlock_t *);
int arch_spin_trylock_retry(arch_spinlock_t *);
void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags);
static inline void arch_spin_relax(arch_spinlock_t *lock)
{
arch_lock_relax(lock->lock);
}
static inline u32 arch_spin_lockval(int cpu)
{
return ~cpu;
}
static inline int arch_spin_value_unlocked(arch_spinlock_t lock)
{
return lock.lock == 0;
}
static inline int arch_spin_is_locked(arch_spinlock_t *lp)
{
return ACCESS_ONCE(lp->lock) != 0;
}
static inline int arch_spin_trylock_once(arch_spinlock_t *lp)
{
barrier();
return likely(arch_spin_value_unlocked(*lp) &&
_raw_compare_and_swap(&lp->lock, 0, SPINLOCK_LOCKVAL));
}
static inline void arch_spin_lock(arch_spinlock_t *lp)
{
if (!arch_spin_trylock_once(lp))
arch_spin_lock_wait(lp);
}
static inline void arch_spin_lock_flags(arch_spinlock_t *lp,
unsigned long flags)
{
if (!arch_spin_trylock_once(lp))
arch_spin_lock_wait_flags(lp, flags);
}
static inline int arch_spin_trylock(arch_spinlock_t *lp)
{
if (!arch_spin_trylock_once(lp))
return arch_spin_trylock_retry(lp);
return 1;
}
static inline void arch_spin_unlock(arch_spinlock_t *lp)
{
typecheck(unsigned int, lp->lock);
asm volatile(
"st %1,%0\n"
: "+Q" (lp->lock)
: "d" (0)
: "cc", "memory");
}
static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
{
while (arch_spin_is_locked(lock))
arch_spin_relax(lock);
smp_acquire__after_ctrl_dep();
}
/*
* Read-write spinlocks, allowing multiple readers
* but only one writer.
*
* NOTE! it is quite common to have readers in interrupts
* but no interrupt writers. For those circumstances we
* can "mix" irq-safe locks - any writer needs to get a
* irq-safe write-lock, but readers can get non-irqsafe
* read-locks.
*/
/**
* read_can_lock - would read_trylock() succeed?
* @lock: the rwlock in question.
*/
#define arch_read_can_lock(x) ((int)(x)->lock >= 0)
/**
* write_can_lock - would write_trylock() succeed?
* @lock: the rwlock in question.
*/
#define arch_write_can_lock(x) ((x)->lock == 0)
extern int _raw_read_trylock_retry(arch_rwlock_t *lp);
extern int _raw_write_trylock_retry(arch_rwlock_t *lp);
#define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
#define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
static inline int arch_read_trylock_once(arch_rwlock_t *rw)
{
unsigned int old = ACCESS_ONCE(rw->lock);
return likely((int) old >= 0 &&
_raw_compare_and_swap(&rw->lock, old, old + 1));
}
static inline int arch_write_trylock_once(arch_rwlock_t *rw)
{
unsigned int old = ACCESS_ONCE(rw->lock);
return likely(old == 0 &&
_raw_compare_and_swap(&rw->lock, 0, 0x80000000));
}
#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
#define __RAW_OP_OR "lao"
#define __RAW_OP_AND "lan"
#define __RAW_OP_ADD "laa"
#define __RAW_LOCK(ptr, op_val, op_string) \
({ \
unsigned int old_val; \
\
typecheck(unsigned int *, ptr); \
asm volatile( \
op_string " %0,%2,%1\n" \
"bcr 14,0\n" \
: "=d" (old_val), "+Q" (*ptr) \
: "d" (op_val) \
: "cc", "memory"); \
old_val; \
})
#define __RAW_UNLOCK(ptr, op_val, op_string) \
({ \
unsigned int old_val; \
\
typecheck(unsigned int *, ptr); \
asm volatile( \
op_string " %0,%2,%1\n" \
: "=d" (old_val), "+Q" (*ptr) \
: "d" (op_val) \
: "cc", "memory"); \
old_val; \
})
extern void _raw_read_lock_wait(arch_rwlock_t *lp);
extern void _raw_write_lock_wait(arch_rwlock_t *lp, unsigned int prev);
static inline void arch_read_lock(arch_rwlock_t *rw)
{
unsigned int old;
old = __RAW_LOCK(&rw->lock, 1, __RAW_OP_ADD);
if ((int) old < 0)
_raw_read_lock_wait(rw);
}
static inline void arch_read_unlock(arch_rwlock_t *rw)
{
__RAW_UNLOCK(&rw->lock, -1, __RAW_OP_ADD);
}
static inline void arch_write_lock(arch_rwlock_t *rw)
{
unsigned int old;
old = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR);
if (old != 0)
_raw_write_lock_wait(rw, old);
rw->owner = SPINLOCK_LOCKVAL;
}
static inline void arch_write_unlock(arch_rwlock_t *rw)
{
rw->owner = 0;
__RAW_UNLOCK(&rw->lock, 0x7fffffff, __RAW_OP_AND);
}
#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */
extern void _raw_read_lock_wait(arch_rwlock_t *lp);
extern void _raw_write_lock_wait(arch_rwlock_t *lp);
static inline void arch_read_lock(arch_rwlock_t *rw)
{
if (!arch_read_trylock_once(rw))
_raw_read_lock_wait(rw);
}
static inline void arch_read_unlock(arch_rwlock_t *rw)
{
unsigned int old;
do {
old = ACCESS_ONCE(rw->lock);
} while (!_raw_compare_and_swap(&rw->lock, old, old - 1));
}
static inline void arch_write_lock(arch_rwlock_t *rw)
{
if (!arch_write_trylock_once(rw))
_raw_write_lock_wait(rw);
rw->owner = SPINLOCK_LOCKVAL;
}
static inline void arch_write_unlock(arch_rwlock_t *rw)
{
typecheck(unsigned int, rw->lock);
rw->owner = 0;
asm volatile(
"st %1,%0\n"
: "+Q" (rw->lock)
: "d" (0)
: "cc", "memory");
}
#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */
static inline int arch_read_trylock(arch_rwlock_t *rw)
{
if (!arch_read_trylock_once(rw))
return _raw_read_trylock_retry(rw);
return 1;
}
static inline int arch_write_trylock(arch_rwlock_t *rw)
{
if (!arch_write_trylock_once(rw) && !_raw_write_trylock_retry(rw))
return 0;
rw->owner = SPINLOCK_LOCKVAL;
return 1;
}
static inline void arch_read_relax(arch_rwlock_t *rw)
{
arch_lock_relax(rw->owner);
}
static inline void arch_write_relax(arch_rwlock_t *rw)
{
arch_lock_relax(rw->owner);
}
#endif /* __ASM_SPINLOCK_H */