[urcu.git] / urcu-static.h

#ifndef _URCU_STATIC_H
#define _URCU_STATIC_H

/*
 * urcu-static.h
 *
 * Userspace RCU header.
 *
 * TO BE INCLUDED ONLY IN LGPL-COMPATIBLE CODE. See urcu.h for linking
 * dynamically with the userspace rcu library.
 *
 * Copyright (c) 2009 Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
 * Copyright (c) 2009 Paul E. McKenney, IBM Corporation.
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 *
 * IBM's contributions to this file may be relicensed under LGPLv2 or later.
 */

#include <stdlib.h>
#include <pthread.h>
#include <syscall.h>
#include <unistd.h>

#include <urcu/compiler.h>
#include <urcu/arch.h>

/*
 * Identify a shared load. A smp_rmc() or smp_mc() should come before the load.
 */
#define _LOAD_SHARED(p)	       ACCESS_ONCE(p)

/*
 * Load a data from shared memory, doing a cache flush if required.
 */
#define LOAD_SHARED(p)			\
	({				\
		smp_rmc();		\
		_LOAD_SHARED(p);	\
	})

/*
 * Identify a shared store. A smp_wmc() or smp_mc() should follow the store.
 */
#define _STORE_SHARED(x, v)	({ ACCESS_ONCE(x) = (v); })

/*
 * Store v into x, where x is located in shared memory. Performs the required
 * cache flush after writing. Returns v.
 */
#define STORE_SHARED(x, v)		\
	({				\
		_STORE_SHARED(x, v);	\
		smp_wmc();		\
		(v);			\
	})

/**
 * _rcu_dereference - reads (copy) a RCU-protected pointer to a local variable
 * into a RCU read-side critical section. The pointer can later be safely
 * dereferenced within the critical section.
 *
 * This ensures that the pointer copy is invariant thorough the whole critical
 * section.
 *
 * Inserts memory barriers on architectures that require them (currently only
 * Alpha) and documents which pointers are protected by RCU.
 *
 * The compiler memory barrier in LOAD_SHARED() ensures that value-speculative
 * optimizations (e.g. VSS: Value Speculation Scheduling) does not perform the
 * data read before the pointer read by speculating the value of the pointer.
 * Correct ordering is ensured because the pointer is read as a volatile access.
 * This acts as a global side-effect operation, which forbids reordering of
 * dependent memory operations. Note that such concern about dependency-breaking
 * optimizations will eventually be taken care of by the "memory_order_consume"
 * addition to forthcoming C++ standard.
 *
 * Should match rcu_assign_pointer() or rcu_xchg_pointer().
 */

#define _rcu_dereference(p)     ({					\
				typeof(p) _________p1 = LOAD_SHARED(p); \
				smp_read_barrier_depends();		\
				(_________p1);				\
				})

#define futex(...)		syscall(__NR_futex, __VA_ARGS__)
#define FUTEX_WAIT		0
#define FUTEX_WAKE		1

/*
 * This code section can only be included in LGPL 2.1 compatible source code.
 * See below for the function call wrappers which can be used in code meant to
 * be only linked with the Userspace RCU library. This comes with a small
 * performance degradation on the read-side due to the added function calls.
 * This is required to permit relinking with newer versions of the library.
 */

/*
 * The signal number used by the RCU library can be overridden with
 * -DSIGURCU= when compiling the library.
 */
#ifndef SIGURCU
#define SIGURCU SIGUSR1
#endif

/*
 * If a reader is really non-cooperative and refuses to commit its
 * urcu_active_readers count to memory (there is no barrier in the reader
 * per-se), kick it after a few loops waiting for it.
 */
#define KICK_READER_LOOPS 10000

/*
 * Active attempts to check for reader Q.S. before calling futex().
 */
#define RCU_QS_ACTIVE_ATTEMPTS 100

#ifdef DEBUG_RCU
#define rcu_assert(args...)	assert(args)
#else
#define rcu_assert(args...)
#endif

#ifdef DEBUG_YIELD
#include <sched.h>
#include <time.h>
#include <pthread.h>
#include <unistd.h>

#define YIELD_READ 	(1 << 0)
#define YIELD_WRITE	(1 << 1)

/*
 * Updates without URCU_MB are much slower. Account this in
 * the delay.
 */
#ifdef URCU_MB
/* maximum sleep delay, in us */
#define MAX_SLEEP 50
#else
#define MAX_SLEEP 30000
#endif

extern unsigned int yield_active;
extern unsigned int __thread rand_yield;

static inline void debug_yield_read(void)
{
	if (yield_active & YIELD_READ)
		if (rand_r(&rand_yield) & 0x1)
			usleep(rand_r(&rand_yield) % MAX_SLEEP);
}

static inline void debug_yield_write(void)
{
	if (yield_active & YIELD_WRITE)
		if (rand_r(&rand_yield) & 0x1)
			usleep(rand_r(&rand_yield) % MAX_SLEEP);
}

static inline void debug_yield_init(void)
{
	rand_yield = time(NULL) ^ pthread_self();
}
#else
static inline void debug_yield_read(void)
{
}

static inline void debug_yield_write(void)
{
}

static inline void debug_yield_init(void)
{

}
#endif

#ifdef URCU_MB
static inline void reader_barrier()
{
	smp_mb();
}
#else
static inline void reader_barrier()
{
	barrier();
}
#endif

/*
 * The trick here is that RCU_GP_CTR_BIT must be a multiple of 8 so we can use a
 * full 8-bits, 16-bits or 32-bits bitmask for the lower order bits.
 */
#define RCU_GP_COUNT		(1UL << 0)
/* Use the amount of bits equal to half of the architecture long size */
#define RCU_GP_CTR_BIT		(1UL << (sizeof(long) << 2))
#define RCU_GP_CTR_NEST_MASK	(RCU_GP_CTR_BIT - 1)

/*
 * Global quiescent period counter with low-order bits unused.
 * Using a int rather than a char to eliminate false register dependencies
 * causing stalls on some architectures.
 */
extern long urcu_gp_ctr;

extern long __thread urcu_active_readers;

extern int gp_futex;

/*
 * Wake-up waiting synchronize_rcu(). Called from many concurrent threads.
 */
static inline void wake_up_gp(void)
{
	if (unlikely(uatomic_read(&gp_futex) == -1)) {
		uatomic_set(&gp_futex, 0);
		futex(&gp_futex, FUTEX_WAKE, 1,
		      NULL, NULL, 0);
	}
}

static inline int rcu_old_gp_ongoing(long *value)
{
	long v;

	if (value == NULL)
		return 0;
	/*
	 * Make sure both tests below are done on the same version of *value
	 * to insure consistency.
	 */
	v = LOAD_SHARED(*value);
	return (v & RCU_GP_CTR_NEST_MASK) &&
		 ((v ^ urcu_gp_ctr) & RCU_GP_CTR_BIT);
}

static inline void _rcu_read_lock(void)
{
	long tmp;

	tmp = urcu_active_readers;
	/* urcu_gp_ctr = RCU_GP_COUNT | (~RCU_GP_CTR_BIT or RCU_GP_CTR_BIT) */
	if (likely(!(tmp & RCU_GP_CTR_NEST_MASK))) {
		_STORE_SHARED(urcu_active_readers, _LOAD_SHARED(urcu_gp_ctr));
		/*
		 * Set active readers count for outermost nesting level before
		 * accessing the pointer. See force_mb_all_threads().
		 */
		reader_barrier();
	} else {
		_STORE_SHARED(urcu_active_readers, tmp + RCU_GP_COUNT);
	}
}

static inline void _rcu_read_unlock(void)
{
	long tmp;

	tmp = urcu_active_readers;
	/*
	 * Finish using rcu before decrementing the pointer.
	 * See force_mb_all_threads().
	 */
	if (likely((tmp & RCU_GP_CTR_NEST_MASK) == RCU_GP_COUNT)) {
		reader_barrier();
		_STORE_SHARED(urcu_active_readers,
			      urcu_active_readers - RCU_GP_COUNT);
		/* write urcu_active_readers before read futex */
		reader_barrier();
		wake_up_gp();
	} else {
		_STORE_SHARED(urcu_active_readers,
			      urcu_active_readers - RCU_GP_COUNT);
	}
}

/**
 * _rcu_assign_pointer - assign (publicize) a pointer to a new data structure
 * meant to be read by RCU read-side critical sections. Returns the assigned
 * value.
 *
 * Documents which pointers will be dereferenced by RCU read-side critical
 * sections and adds the required memory barriers on architectures requiring
 * them. It also makes sure the compiler does not reorder code initializing the
 * data structure before its publication.
 *
 * Should match rcu_dereference_pointer().
 */

#define _rcu_assign_pointer(p, v)			\
	({						\
		if (!__builtin_constant_p(v) || 	\
		    ((v) != NULL))			\
			wmb();				\
		STORE_SHARED(p, v);			\
	})

/**
 * _rcu_cmpxchg_pointer - same as rcu_assign_pointer, but tests if the pointer
 * is as expected by "old". If succeeds, returns the previous pointer to the
 * data structure, which can be safely freed after waiting for a quiescent state
 * using synchronize_rcu(). If fails (unexpected value), returns old (which
 * should not be freed !).
 */

#define _rcu_cmpxchg_pointer(p, old, _new)		\
	({						\
		if (!__builtin_constant_p(_new) ||	\
		    ((_new) != NULL))			\
			wmb();				\
		uatomic_cmpxchg(p, old, _new);		\
	})

/**
 * _rcu_xchg_pointer - same as rcu_assign_pointer, but returns the previous
 * pointer to the data structure, which can be safely freed after waiting for a
 * quiescent state using synchronize_rcu().
 */

#define _rcu_xchg_pointer(p, v)				\
	({						\
		if (!__builtin_constant_p(v) ||		\
		    ((v) != NULL))			\
			wmb();				\
		uatomic_xchg(p, v);			\
	})

/*
 * Exchanges the pointer and waits for quiescent state.
 * The pointer returned can be freed.
 */
#define _rcu_publish_content(p, v)			\
	({						\
		void *oldptr;				\
		oldptr = _rcu_xchg_pointer(p, v);	\
		synchronize_rcu();			\
		oldptr;					\
	})

#endif /* _URCU_STATIC_H */
Commit	Line	Data
	1	#ifndef _URCU_STATIC_H
	2	#define _URCU_STATIC_H
	3
	4	/*
	5	* urcu-static.h
	6	*
	7	* Userspace RCU header.
	8	*
	9	* TO BE INCLUDED ONLY IN LGPL-COMPATIBLE CODE. See urcu.h for linking
	10	* dynamically with the userspace rcu library.
	11	*
	12	* Copyright (c) 2009 Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
	13	* Copyright (c) 2009 Paul E. McKenney, IBM Corporation.
	14	*
	15	* This library is free software; you can redistribute it and/or
	16	* modify it under the terms of the GNU Lesser General Public
	17	* License as published by the Free Software Foundation; either
	18	* version 2.1 of the License, or (at your option) any later version.
	19	*
	20	* This library is distributed in the hope that it will be useful,
	21	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	22	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	23	* Lesser General Public License for more details.
	24	*
	25	* You should have received a copy of the GNU Lesser General Public
	26	* License along with this library; if not, write to the Free Software
	27	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
	28	*
	29	* IBM's contributions to this file may be relicensed under LGPLv2 or later.
	30	*/
	31
	32	#include <stdlib.h>
	33	#include <pthread.h>
	34	#include <syscall.h>
	35	#include <unistd.h>
	36
	37	#include <urcu/compiler.h>
	38	#include <urcu/arch.h>
	39
	40	/*
	41	* Identify a shared load. A smp_rmc() or smp_mc() should come before the load.
	42	*/
	43	#define _LOAD_SHARED(p) ACCESS_ONCE(p)
	44
	45	/*
	46	* Load a data from shared memory, doing a cache flush if required.
	47	*/
	48	#define LOAD_SHARED(p) \
	49	({ \
	50	smp_rmc(); \
	51	_LOAD_SHARED(p); \
	52	})
	53
	54	/*
	55	* Identify a shared store. A smp_wmc() or smp_mc() should follow the store.
	56	*/
	57	#define _STORE_SHARED(x, v) ({ ACCESS_ONCE(x) = (v); })
	58
	59	/*
	60	* Store v into x, where x is located in shared memory. Performs the required
	61	* cache flush after writing. Returns v.
	62	*/
	63	#define STORE_SHARED(x, v) \
	64	({ \
	65	_STORE_SHARED(x, v); \
	66	smp_wmc(); \
	67	(v); \
	68	})
	69
	70	/**
	71	* _rcu_dereference - reads (copy) a RCU-protected pointer to a local variable
	72	* into a RCU read-side critical section. The pointer can later be safely
	73	* dereferenced within the critical section.
	74	*
	75	* This ensures that the pointer copy is invariant thorough the whole critical
	76	* section.
	77	*
	78	* Inserts memory barriers on architectures that require them (currently only
	79	* Alpha) and documents which pointers are protected by RCU.
	80	*
	81	* The compiler memory barrier in LOAD_SHARED() ensures that value-speculative
	82	* optimizations (e.g. VSS: Value Speculation Scheduling) does not perform the
	83	* data read before the pointer read by speculating the value of the pointer.
	84	* Correct ordering is ensured because the pointer is read as a volatile access.
	85	* This acts as a global side-effect operation, which forbids reordering of
	86	* dependent memory operations. Note that such concern about dependency-breaking
	87	* optimizations will eventually be taken care of by the "memory_order_consume"
	88	* addition to forthcoming C++ standard.
	89	*
	90	* Should match rcu_assign_pointer() or rcu_xchg_pointer().
	91	*/
	92
	93	#define _rcu_dereference(p) ({ \
	94	typeof(p) _________p1 = LOAD_SHARED(p); \
	95	smp_read_barrier_depends(); \
	96	(_________p1); \
	97	})
	98
	99	#define futex(...) syscall(__NR_futex, __VA_ARGS__)
	100	#define FUTEX_WAIT 0
	101	#define FUTEX_WAKE 1
	102
	103	/*
	104	* This code section can only be included in LGPL 2.1 compatible source code.
	105	* See below for the function call wrappers which can be used in code meant to
	106	* be only linked with the Userspace RCU library. This comes with a small
	107	* performance degradation on the read-side due to the added function calls.
	108	* This is required to permit relinking with newer versions of the library.
	109	*/
	110
	111	/*
	112	* The signal number used by the RCU library can be overridden with
	113	* -DSIGURCU= when compiling the library.
	114	*/
	115	#ifndef SIGURCU
	116	#define SIGURCU SIGUSR1
	117	#endif
	118
	119	/*
	120	* If a reader is really non-cooperative and refuses to commit its
	121	* urcu_active_readers count to memory (there is no barrier in the reader
	122	* per-se), kick it after a few loops waiting for it.
	123	*/
	124	#define KICK_READER_LOOPS 10000
	125
	126	/*
	127	* Active attempts to check for reader Q.S. before calling futex().
	128	*/
	129	#define RCU_QS_ACTIVE_ATTEMPTS 100
	130
	131	#ifdef DEBUG_RCU
	132	#define rcu_assert(args...) assert(args)
	133	#else
	134	#define rcu_assert(args...)
	135	#endif
	136
	137	#ifdef DEBUG_YIELD
	138	#include <sched.h>
	139	#include <time.h>
	140	#include <pthread.h>
	141	#include <unistd.h>
	142
	143	#define YIELD_READ (1 << 0)
	144	#define YIELD_WRITE (1 << 1)
	145
	146	/*
	147	* Updates without URCU_MB are much slower. Account this in
	148	* the delay.
	149	*/
	150	#ifdef URCU_MB
	151	/* maximum sleep delay, in us */
	152	#define MAX_SLEEP 50
	153	#else
	154	#define MAX_SLEEP 30000
	155	#endif
	156
	157	extern unsigned int yield_active;
	158	extern unsigned int __thread rand_yield;
	159
	160	static inline void debug_yield_read(void)
	161	{
	162	if (yield_active & YIELD_READ)
	163	if (rand_r(&rand_yield) & 0x1)
	164	usleep(rand_r(&rand_yield) % MAX_SLEEP);
	165	}
	166
	167	static inline void debug_yield_write(void)
	168	{
	169	if (yield_active & YIELD_WRITE)
	170	if (rand_r(&rand_yield) & 0x1)
	171	usleep(rand_r(&rand_yield) % MAX_SLEEP);
	172	}
	173
	174	static inline void debug_yield_init(void)
	175	{
	176	rand_yield = time(NULL) ^ pthread_self();
	177	}
	178	#else
	179	static inline void debug_yield_read(void)
	180	{
	181	}
	182
	183	static inline void debug_yield_write(void)
	184	{
	185	}
	186
	187	static inline void debug_yield_init(void)
	188	{
	189
	190	}
	191	#endif
	192
	193	#ifdef URCU_MB
	194	static inline void reader_barrier()
	195	{
	196	smp_mb();
	197	}
	198	#else
	199	static inline void reader_barrier()
	200	{
	201	barrier();
	202	}
	203	#endif
	204
	205	/*
	206	* The trick here is that RCU_GP_CTR_BIT must be a multiple of 8 so we can use a
	207	* full 8-bits, 16-bits or 32-bits bitmask for the lower order bits.
	208	*/
	209	#define RCU_GP_COUNT (1UL << 0)
	210	/* Use the amount of bits equal to half of the architecture long size */
	211	#define RCU_GP_CTR_BIT (1UL << (sizeof(long) << 2))
	212	#define RCU_GP_CTR_NEST_MASK (RCU_GP_CTR_BIT - 1)
	213
	214	/*
	215	* Global quiescent period counter with low-order bits unused.
	216	* Using a int rather than a char to eliminate false register dependencies
	217	* causing stalls on some architectures.
	218	*/
	219	extern long urcu_gp_ctr;
	220
	221	extern long __thread urcu_active_readers;
	222
	223	extern int gp_futex;
	224
	225	/*
	226	* Wake-up waiting synchronize_rcu(). Called from many concurrent threads.
	227	*/
	228	static inline void wake_up_gp(void)
	229	{
	230	if (unlikely(uatomic_read(&gp_futex) == -1)) {
	231	uatomic_set(&gp_futex, 0);
	232	futex(&gp_futex, FUTEX_WAKE, 1,
	233	NULL, NULL, 0);
	234	}
	235	}
	236
	237	static inline int rcu_old_gp_ongoing(long *value)
	238	{
	239	long v;
	240
	241	if (value == NULL)
	242	return 0;
	243	/*
	244	* Make sure both tests below are done on the same version of *value
	245	* to insure consistency.
	246	*/
	247	v = LOAD_SHARED(*value);
	248	return (v & RCU_GP_CTR_NEST_MASK) &&
	249	((v ^ urcu_gp_ctr) & RCU_GP_CTR_BIT);
	250	}
	251
	252	static inline void _rcu_read_lock(void)
	253	{
	254	long tmp;
	255
	256	tmp = urcu_active_readers;
	257	/* urcu_gp_ctr = RCU_GP_COUNT \| (~RCU_GP_CTR_BIT or RCU_GP_CTR_BIT) */
	258	if (likely(!(tmp & RCU_GP_CTR_NEST_MASK))) {
	259	_STORE_SHARED(urcu_active_readers, _LOAD_SHARED(urcu_gp_ctr));
	260	/*
	261	* Set active readers count for outermost nesting level before
	262	* accessing the pointer. See force_mb_all_threads().
	263	*/
	264	reader_barrier();
	265	} else {
	266	_STORE_SHARED(urcu_active_readers, tmp + RCU_GP_COUNT);
	267	}
	268	}
	269
	270	static inline void _rcu_read_unlock(void)
	271	{
	272	long tmp;
	273
	274	tmp = urcu_active_readers;
	275	/*
	276	* Finish using rcu before decrementing the pointer.
	277	* See force_mb_all_threads().
	278	*/
	279	if (likely((tmp & RCU_GP_CTR_NEST_MASK) == RCU_GP_COUNT)) {
	280	reader_barrier();
	281	_STORE_SHARED(urcu_active_readers,
	282	urcu_active_readers - RCU_GP_COUNT);
	283	/* write urcu_active_readers before read futex */
	284	reader_barrier();
	285	wake_up_gp();
	286	} else {
	287	_STORE_SHARED(urcu_active_readers,
	288	urcu_active_readers - RCU_GP_COUNT);
	289	}
	290	}
	291
	292	/**
	293	* _rcu_assign_pointer - assign (publicize) a pointer to a new data structure
	294	* meant to be read by RCU read-side critical sections. Returns the assigned
	295	* value.
	296	*
	297	* Documents which pointers will be dereferenced by RCU read-side critical
	298	* sections and adds the required memory barriers on architectures requiring
	299	* them. It also makes sure the compiler does not reorder code initializing the
	300	* data structure before its publication.
	301	*
	302	* Should match rcu_dereference_pointer().
	303	*/
	304
	305	#define _rcu_assign_pointer(p, v) \
	306	({ \
	307	if (!__builtin_constant_p(v) \|\| \
	308	((v) != NULL)) \
	309	wmb(); \
	310	STORE_SHARED(p, v); \
	311	})
	312
	313	/**
	314	* _rcu_cmpxchg_pointer - same as rcu_assign_pointer, but tests if the pointer
	315	* is as expected by "old". If succeeds, returns the previous pointer to the
	316	* data structure, which can be safely freed after waiting for a quiescent state
	317	* using synchronize_rcu(). If fails (unexpected value), returns old (which
	318	* should not be freed !).
	319	*/
	320
	321	#define _rcu_cmpxchg_pointer(p, old, _new) \
	322	({ \
	323	if (!__builtin_constant_p(_new) \|\| \
	324	((_new) != NULL)) \
	325	wmb(); \
	326	uatomic_cmpxchg(p, old, _new); \
	327	})
	328
	329	/**
	330	* _rcu_xchg_pointer - same as rcu_assign_pointer, but returns the previous
	331	* pointer to the data structure, which can be safely freed after waiting for a
	332	* quiescent state using synchronize_rcu().
	333	*/
	334
	335	#define _rcu_xchg_pointer(p, v) \
	336	({ \
	337	if (!__builtin_constant_p(v) \|\| \
	338	((v) != NULL)) \
	339	wmb(); \
	340	uatomic_xchg(p, v); \
	341	})
	342
	343	/*
	344	* Exchanges the pointer and waits for quiescent state.
	345	* The pointer returned can be freed.
	346	*/
	347	#define _rcu_publish_content(p, v) \
	348	({ \
	349	void *oldptr; \
	350	oldptr = _rcu_xchg_pointer(p, v); \
	351	synchronize_rcu(); \
	352	oldptr; \
	353	})
	354
	355	#endif /* _URCU_STATIC_H */