freebsd-nq/sys/ia64/include/atomic.h
John Baldwin ccbdd9ee59 - Expand the set of atomic operations to optionally include memory barriers
in most of the atomic operations.  Now for these operations, you can
  use the normal atomic operation, you can use the operation with a read
  barrier, or you can use the operation with a write barrier.  The function
  names follow the same semantics used in the ia64 instruction set.  An
  atomic operation with a read barrier has the extra suffix 'acq', due to
  it having "acquire" semantics.  An atomic operation with a write barrier
  has the extra suffix 'rel'.  These suffixes are inserted between the
  name of the operation to perform and the typename.  For example, the
  atomic_add_int() function now has 3 variants:
  - atomic_add_int() - this is the same as the previous function
  - atomic_add_acq_int() - this function combines the add operation with a
    read memory barrier
  - atomic_add_rel_int() - this function combines the add operation with a
    write memory barrier
- Add 'ptr' to the list of types that we can perform atomic operations
  on.  This allows one to do atomic operations on uintptr_t's.  This is
  useful in the mutex code, for example, because the actual mutex lock is
  a pointer.
- Add two new operations for doing loads and stores with memory barriers.
  The new load operations use a read barrier before the load, and the
  new store operations use a write barrier after the load.  For example,
  atomic_load_acq_int() will atomically load an integer as well as
  enforcing a read barrier.
2000-10-20 07:00:48 +00:00

364 lines
11 KiB
C

/*-
* Copyright (c) 1998 Doug Rabson
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
#ifndef _MACHINE_ATOMIC_H_
#define _MACHINE_ATOMIC_H_
/*
* Various simple arithmetic on memory which is atomic in the presence
* of interrupts and SMP safe.
*/
/*
* Everything is built out of cmpxchg.
*/
#define IA64_CMPXCHG(sz, sem, type, p, cmpval, newval) \
({ \
type _cmpval = cmpval; \
type _newval = newval; \
volatile type *_p = (volatile type *) p; \
type _ret; \
\
__asm __volatile ( \
"mov ar.ccv=%2;;\n\t" \
"cmpxchg" #sz "." #sem " %0=%4,%3,ar.ccv\n\t" \
: "=r" (_ret), "=m" (*_p) \
: "r" (_cmpval), "r" (_newval), "m" (*_p) \
: "memory"); \
_ret; \
})
/*
* Some common forms of cmpxch.
*/
static __inline u_int32_t
ia64_cmpxchg_acq_32(volatile u_int32_t* p, u_int32_t cmpval, u_int32_t newval)
{
return IA64_CMPXCHG(4, acq, u_int32_t, p, cmpval, newval);
}
static __inline u_int32_t
ia64_cmpxchg_rel_32(volatile u_int32_t* p, u_int32_t cmpval, u_int32_t newval)
{
return IA64_CMPXCHG(4, rel, u_int32_t, p, cmpval, newval);
}
static __inline u_int64_t
ia64_cmpxchg_acq_64(volatile u_int64_t* p, u_int64_t cmpval, u_int64_t newval)
{
return IA64_CMPXCHG(8, acq, u_int64_t, p, cmpval, newval);
}
static __inline u_int64_t
ia64_cmpxchg_rel_64(volatile u_int64_t* p, u_int64_t cmpval, u_int64_t newval)
{
return IA64_CMPXCHG(8, rel, u_int64_t, p, cmpval, newval);
}
#define ATOMIC_STORE_LOAD(type, width, size) \
static __inline u_int##width##_t \
ia64_ld_acq_##width(volatile u_int##width##_t* p) \
{ \
u_int##width_t v; \
\
__asm __volatile ("ld" size ".acq %0=%1" \
: "r" (v) \
: "=m" (*p) \
: "memory"); \
return (v); \
} \
\
static __inline u_int##width##_t \
atomic_load_acq_##width(volatile u_int##width##_t* p) \
{ \
u_int##width_t v; \
\
__asm __volatile ("ld" size ".acq %0=%1" \
: "r" (v) \
: "=m" (*p) \
: "memory"); \
return (v); \
} \
\
static __inline u_int##width##_t \
atomic_load_acq_##type(volatile u_int##width##_t* p) \
{ \
u_int##width_t v; \
\
__asm __volatile ("ld" size ".acq %0=%1" \
: "r" (v) \
: "=m" (*p) \
: "memory"); \
return (v); \
} \
\
static __inline void \
ia64_st_rel_##width(volatile u_int##width##_t* p, u_int##width##_t v)\
{ \
__asm __volatile ("st" size ".rel %0=%1" \
: "=m" (*p) \
: "r" (v) \
: "memory"); \
} \
\
static __inline void \
atomic_store_rel_##width(volatile u_int##width##_t* p, u_int##width##_t v)\
{ \
__asm __volatile ("st" size ".rel %0=%1" \
: "=m" (*p) \
: "r" (v) \
: "memory"); \
} \
\
static __inline void \
atomic_store_rel_##type(volatile u_int##width##_t* p, u_int##width##_t v)\
{ \
__asm __volatile ("st" size ".rel %0=%1" \
: "=m" (*p) \
: "r" (v) \
: "memory"); \
}
ATOMIC_STORE_LOAD(char, 8, "1")
ATOMIC_STORE_LOAD(short, 16, "2")
ATOMIC_STORE_LOAD(int, 32, "4")
ATOMIC_STORE_LOAD(long, 64, "8")
#undef ATOMIC_STORE_LOAD
#define IA64_ATOMIC(sz, type, name, width, op) \
\
static __inline void \
atomic_##name##_acq_##width(volatile type *p, type v) \
{ \
type old; \
do { \
old = *p; \
} while (IA64_CMPXCHG(sz, acq, type, p, old, old op v) != old); \
} \
\
static __inline void \
atomic_##name##_rel_##width(volatile type *p, type v) \
{ \
type old; \
do { \
old = *p; \
} while (IA64_CMPXCHG(sz, rel, type, p, old, old op v) != old); \
}
IA64_ATOMIC(1, u_int8_t, set, 8, |)
IA64_ATOMIC(2, u_int16_t, set, 16, |)
IA64_ATOMIC(4, u_int32_t, set, 32, |)
IA64_ATOMIC(8, u_int64_t, set, 64, |)
IA64_ATOMIC(1, u_int8_t, clear, 8, &~)
IA64_ATOMIC(2, u_int16_t, clear, 16, &~)
IA64_ATOMIC(4, u_int32_t, clear, 32, &~)
IA64_ATOMIC(8, u_int64_t, clear, 64, &~)
IA64_ATOMIC(1, u_int8_t, add, 8, +)
IA64_ATOMIC(2, u_int16_t, add, 16, +)
IA64_ATOMIC(4, u_int32_t, add, 32, +)
IA64_ATOMIC(8, u_int64_t, add, 64, +)
IA64_ATOMIC(1, u_int8_t, subtract, 8, -)
IA64_ATOMIC(2, u_int16_t, subtract, 16, -)
IA64_ATOMIC(4, u_int32_t, subtract, 32, -)
IA64_ATOMIC(8, u_int64_t, subtract, 64, -)
#undef IA64_ATOMIC
#undef IA64_CMPXCHG
#define atomic_set_8 atomic_set_acq_8
#define atomic_clear_8 atomic_clear_acq_8
#define atomic_add_8 atomic_add_acq_8
#define atomic_subtract_8 atomic_subtract_acq_8
#define atomic_set_16 atomic_set_acq_16
#define atomic_clear_16 atomic_clear_acq_16
#define atomic_add_16 atomic_add_acq_16
#define atomic_subtract_16 atomic_subtract_acq_16
#define atomic_set_32 atomic_set_acq_32
#define atomic_clear_32 atomic_clear_acq_32
#define atomic_add_32 atomic_add_acq_32
#define atomic_subtract_32 atomic_subtract_acq_32
#define atomic_set_64 atomic_set_acq_64
#define atomic_clear_64 atomic_clear_acq_64
#define atomic_add_64 atomic_add_acq_64
#define atomic_subtract_64 atomic_subtract_acq_64
#define atomic_set_char atomic_set_8
#define atomic_clear_char atomic_clear_8
#define atomic_add_char atomic_add_8
#define atomic_subtract_char atomic_subtract_8
#define atomic_set_acq_char atomic_set_acq_8
#define atomic_clear_acq_char atomic_clear_acq_8
#define atomic_add_acq_char atomic_add_acq_8
#define atomic_subtract_acq_char atomic_subtract_acq_8
#define atomic_set_rel_char atomic_set_rel_8
#define atomic_clear_rel_char atomic_clear_rel_8
#define atomic_add_rel_char atomic_add_rel_8
#define atomic_subtract_rel_char atomic_subtract_rel_8
#define atomic_set_short atomic_set_16
#define atomic_clear_short atomic_clear_16
#define atomic_add_short atomic_add_16
#define atomic_subtract_short atomic_subtract_16
#define atomic_set_acq_short atomic_set_acq_16
#define atomic_clear_acq_short atomic_clear_acq_16
#define atomic_add_acq_short atomic_add_acq_16
#define atomic_subtract_acq_short atomic_subtract_acq_16
#define atomic_set_rel_short atomic_set_rel_16
#define atomic_clear_rel_short atomic_clear_rel_16
#define atomic_add_rel_short atomic_add_rel_16
#define atomic_subtract_rel_short atomic_subtract_rel_16
#define atomic_set_int atomic_set_32
#define atomic_clear_int atomic_clear_32
#define atomic_add_int atomic_add_32
#define atomic_subtract_int atomic_subtract_32
#define atomic_set_acq_int atomic_set_acq_32
#define atomic_clear_acq_int atomic_clear_acq_32
#define atomic_add_acq_int atomic_add_acq_32
#define atomic_subtract_acq_int atomic_subtract_acq_32
#define atomic_set_rel_int atomic_set_rel_32
#define atomic_clear_rel_int atomic_clear_rel_32
#define atomic_add_rel_int atomic_add_rel_32
#define atomic_subtract_rel_int atomic_subtract_rel_32
#define atomic_set_long atomic_set_64
#define atomic_clear_long atomic_clear_64
#define atomic_add_long atomic_add_64
#define atomic_subtract_long atomic_subtract_64
#define atomic_set_acq_long atomic_set_acq_64
#define atomic_clear_acq_long atomic_clear_acq_64
#define atomic_add_acq_long atomic_add_acq_64
#define atomic_subtract_acq_long atomic_subtract_acq_64
#define atomic_set_rel_long atomic_set_rel_64
#define atomic_clear_rel_long atomic_clear_rel_64
#define atomic_add_rel_long atomic_add_rel_64
#define atomic_subtract_rel_long atomic_subtract_rel_64
/*
* Atomically compare the value stored at *p with cmpval and if the
* two values are equal, update the value of *p with newval. Returns
* zero if the compare failed, nonzero otherwise.
*/
static __inline int
atomic_cmpset_acq_32(volatile u_int32_t* p, u_int32_t cmpval, u_int32_t newval)
{
return ia64_cmpxchg_acq_32(p, cmpval, newval) == cmpval;
}
static __inline int
atomic_cmpset_rel_32(volatile u_int32_t* p, u_int32_t cmpval, u_int32_t newval)
{
return ia64_cmpxchg_rel_32(p, cmpval, newval) == cmpval;
}
/*
* Atomically compare the value stored at *p with cmpval and if the
* two values are equal, update the value of *p with newval. Returns
* zero if the compare failed, nonzero otherwise.
*/
static __inline int
atomic_cmpset_acq_64(volatile u_int64_t* p, u_int64_t cmpval, u_int64_t newval)
{
return ia64_cmpxchg_acq_64(p, cmpval, newval) == cmpval;
}
static __inline int
atomic_cmpset_rel_64(volatile u_int64_t* p, u_int64_t cmpval, u_int64_t newval)
{
return ia64_cmpxchg_rel_64(p, cmpval, newval) == cmpval;
}
#define atomic_cmpset_32 atomic_cmpset_acq_32
#define atomic_cmpset_64 atomic_cmpset_acq_64
#define atomic_cmpset_int atomic_cmpset_32
#define atomic_cmpset_long atomic_cmpset_64
#define atomic_cmpset_acq_int atomic_cmpset_acq_32
#define atomic_cmpset_rel_int atomic_cmpset_rel_32
#define atomic_cmpset_acq_long atomic_cmpset_acq_64
#define atomic_cmpset_rel_long atomic_cmpset_rel_64
static __inline int
atomic_cmpset_acq_ptr(volatile void *dst, void *exp, void *src)
{
return atomic_cmpset_acq_long((volatile u_long *)dst,
(u_long)exp, (u_long)src);
}
static __inline int
atomic_cmpset_rel_ptr(volatile void *dst, void *exp, void *src)
{
return atomic_cmpset_rel_long((volatile u_long *)dst,
(u_long)exp, (u_long)src);
}
#define atomic_cmpset_ptr atomic_cmpset_acq_ptr
static __inline void *
atomic_load_acq_ptr(volatile void *p)
{
return (void *)atomic_load_acq_long((volatile u_long *)p);
}
static __inline void
atomic_store_rel_ptr(volatile void *p, void *v)
{
atomic_store_rel_long((volatile u_long *)p, (u_long)v);
}
static __inline u_int32_t
atomic_readandclear_32(volatile u_int32_t* p)
{
u_int32_t val;
do {
val = *p;
} while (!atomic_cmpset_32(p, val, 0));
return val;
}
static __inline u_int64_t
atomic_readandclear_64(volatile u_int64_t* p)
{
u_int64_t val;
do {
val = *p;
} while (!atomic_cmpset_64(p, val, 0));
return val;
}
#define atomic_readandclear_int atomic_readandclear_32
#define atomic_readandclear_long atomic_readandclear_64
#endif /* ! _MACHINE_ATOMIC_H_ */