rte_ring.h

Go to the documentation of this file.

/*-
 *   BSD LICENSE
 *
 *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
 *   All rights reserved.
 *
 *   Redistribution and use in source and binary forms, with or without
 *   modification, are permitted provided that the following conditions
 *   are met:
 *
 *     * Redistributions of source code must retain the above copyright
 *       notice, this list of conditions and the following disclaimer.
 *     * 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.
 *     * Neither the name of Intel Corporation nor the names of its
 *       contributors may be used to endorse or promote products derived
 *       from this software without specific prior written permission.
 *
 *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 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 COPYRIGHT
 *   OWNER 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.
 */

/*
 * Derived from FreeBSD's bufring.h
 *
 **************************************************************************
 *
 * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
 * 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. The name of Kip Macy nor the names of other
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 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 COPYRIGHT OWNER 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.
 *
 ***************************************************************************/

#ifndef _RTE_RING_H_
#define _RTE_RING_H_

#ifdef __cplusplus
extern "C" {
#endif

#include <stdio.h>
#include <stdint.h>
#include <sys/queue.h>
#include <errno.h>
#include <rte_common.h>
#include <rte_memory.h>
#include <rte_lcore.h>
#include <rte_atomic.h>
#include <rte_branch_prediction.h>
#include <rte_memzone.h>

#define RTE_TAILQ_RING_NAME "RTE_RING"

enum rte_ring_queue_behavior {
        RTE_RING_QUEUE_FIXED = 0, /* Enq/Deq a fixed number of items from a ring */
        RTE_RING_QUEUE_VARIABLE   /* Enq/Deq as many items as possible from ring */
};

#ifdef RTE_LIBRTE_RING_DEBUG

struct rte_ring_debug_stats {
        uint64_t enq_success_bulk; 
        uint64_t enq_success_objs; 
        uint64_t enq_quota_bulk;   
        uint64_t enq_quota_objs;   
        uint64_t enq_fail_bulk;    
        uint64_t enq_fail_objs;    
        uint64_t deq_success_bulk; 
        uint64_t deq_success_objs; 
        uint64_t deq_fail_bulk;    
        uint64_t deq_fail_objs;    
} __rte_cache_aligned;
#endif

#define RTE_RING_MZ_PREFIX "RG_"

#define RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
                           sizeof(RTE_RING_MZ_PREFIX) + 1)

#ifndef RTE_RING_PAUSE_REP_COUNT
#define RTE_RING_PAUSE_REP_COUNT 0 
#endif

struct rte_memzone; /* forward declaration, so as not to require memzone.h */

struct rte_ring {
        /*
         * Note: this field kept the RTE_MEMZONE_NAMESIZE size due to ABI
         * compatibility requirements, it could be changed to RTE_RING_NAMESIZE
         * next time the ABI changes
         */
        char name[RTE_MEMZONE_NAMESIZE];    
        int flags;                       
        const struct rte_memzone *memzone;
        struct prod {
                uint32_t watermark;      
                uint32_t sp_enqueue;     
                uint32_t size;           
                uint32_t mask;           
                volatile uint32_t head;  
                volatile uint32_t tail;  
        } prod __rte_cache_aligned;

        struct cons {
                uint32_t sc_dequeue;     
                uint32_t size;           
                uint32_t mask;           
                volatile uint32_t head;  
                volatile uint32_t tail;  
#ifdef RTE_RING_SPLIT_PROD_CONS
        } cons __rte_cache_aligned;
#else
        } cons;
#endif

#ifdef RTE_LIBRTE_RING_DEBUG
        struct rte_ring_debug_stats stats[RTE_MAX_LCORE];
#endif

        void *ring[] __rte_cache_aligned;   
};

#define RING_F_SP_ENQ 0x0001 
#define RING_F_SC_DEQ 0x0002 
#define RTE_RING_QUOT_EXCEED (1 << 31)  
#define RTE_RING_SZ_MASK  (unsigned)(0x0fffffff) 
#ifdef RTE_LIBRTE_RING_DEBUG
#define __RING_STAT_ADD(r, name, n) do {                        \
                unsigned __lcore_id = rte_lcore_id();           \
                if (__lcore_id < RTE_MAX_LCORE) {               \
                        r->stats[__lcore_id].name##_objs += n;  \
                        r->stats[__lcore_id].name##_bulk += 1;  \
                }                                               \
        } while(0)
#else
#define __RING_STAT_ADD(r, name, n) do {} while(0)
#endif

ssize_t rte_ring_get_memsize(unsigned count);

int rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
        unsigned flags);

struct rte_ring *rte_ring_create(const char *name, unsigned count,
                                 int socket_id, unsigned flags);
void rte_ring_free(struct rte_ring *r);

int rte_ring_set_water_mark(struct rte_ring *r, unsigned count);

void rte_ring_dump(FILE *f, const struct rte_ring *r);

/* the actual enqueue of pointers on the ring.
 * Placed here since identical code needed in both
 * single and multi producer enqueue functions */
#define ENQUEUE_PTRS() do { \
        const uint32_t size = r->prod.size; \
        uint32_t idx = prod_head & mask; \
        if (likely(idx + n < size)) { \
                for (i = 0; i < (n & ((~(unsigned)0x3))); i+=4, idx+=4) { \
                        r->ring[idx] = obj_table[i]; \
                        r->ring[idx+1] = obj_table[i+1]; \
                        r->ring[idx+2] = obj_table[i+2]; \
                        r->ring[idx+3] = obj_table[i+3]; \
                } \
                switch (n & 0x3) { \
                        case 3: r->ring[idx++] = obj_table[i++]; \
                        case 2: r->ring[idx++] = obj_table[i++]; \
                        case 1: r->ring[idx++] = obj_table[i++]; \
                } \
        } else { \
                for (i = 0; idx < size; i++, idx++)\
                        r->ring[idx] = obj_table[i]; \
                for (idx = 0; i < n; i++, idx++) \
                        r->ring[idx] = obj_table[i]; \
        } \
} while(0)

/* the actual copy of pointers on the ring to obj_table.
 * Placed here since identical code needed in both
 * single and multi consumer dequeue functions */
#define DEQUEUE_PTRS() do { \
        uint32_t idx = cons_head & mask; \
        const uint32_t size = r->cons.size; \
        if (likely(idx + n < size)) { \
                for (i = 0; i < (n & (~(unsigned)0x3)); i+=4, idx+=4) {\
                        obj_table[i] = r->ring[idx]; \
                        obj_table[i+1] = r->ring[idx+1]; \
                        obj_table[i+2] = r->ring[idx+2]; \
                        obj_table[i+3] = r->ring[idx+3]; \
                } \
                switch (n & 0x3) { \
                        case 3: obj_table[i++] = r->ring[idx++]; \
                        case 2: obj_table[i++] = r->ring[idx++]; \
                        case 1: obj_table[i++] = r->ring[idx++]; \
                } \
        } else { \
                for (i = 0; idx < size; i++, idx++) \
                        obj_table[i] = r->ring[idx]; \
                for (idx = 0; i < n; i++, idx++) \
                        obj_table[i] = r->ring[idx]; \
        } \
} while (0)

static inline int __attribute__((always_inline))
__rte_ring_mp_do_enqueue(struct rte_ring *r, void * const *obj_table,
                         unsigned n, enum rte_ring_queue_behavior behavior)
{
        uint32_t prod_head, prod_next;
        uint32_t cons_tail, free_entries;
        const unsigned max = n;
        int success;
        unsigned i, rep = 0;
        uint32_t mask = r->prod.mask;
        int ret;

        /* Avoid the unnecessary cmpset operation below, which is also
         * potentially harmful when n equals 0. */
        if (n == 0)
                return 0;

        /* move prod.head atomically */
        do {
                /* Reset n to the initial burst count */
                n = max;

                prod_head = r->prod.head;
                cons_tail = r->cons.tail;
                /* The subtraction is done between two unsigned 32bits value
                 * (the result is always modulo 32 bits even if we have
                 * prod_head > cons_tail). So 'free_entries' is always between 0
                 * and size(ring)-1. */
                free_entries = (mask + cons_tail - prod_head);

                /* check that we have enough room in ring */
                if (unlikely(n > free_entries)) {
                        if (behavior == RTE_RING_QUEUE_FIXED) {
                                __RING_STAT_ADD(r, enq_fail, n);
                                return -ENOBUFS;
                        }
                        else {
                                /* No free entry available */
                                if (unlikely(free_entries == 0)) {
                                        __RING_STAT_ADD(r, enq_fail, n);
                                        return 0;
                                }

                                n = free_entries;
                        }
                }

                prod_next = prod_head + n;
                success = rte_atomic32_cmpset(&r->prod.head, prod_head,
                                              prod_next);
        } while (unlikely(success == 0));

        /* write entries in ring */
        ENQUEUE_PTRS();
        rte_smp_wmb();

        /* if we exceed the watermark */
        if (unlikely(((mask + 1) - free_entries + n) > r->prod.watermark)) {
                ret = (behavior == RTE_RING_QUEUE_FIXED) ? -EDQUOT :
                                (int)(n | RTE_RING_QUOT_EXCEED);
                __RING_STAT_ADD(r, enq_quota, n);
        }
        else {
                ret = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : n;
                __RING_STAT_ADD(r, enq_success, n);
        }

        /*
         * If there are other enqueues in progress that preceded us,
         * we need to wait for them to complete
         */
        while (unlikely(r->prod.tail != prod_head)) {
                rte_pause();

                /* Set RTE_RING_PAUSE_REP_COUNT to avoid spin too long waiting
                 * for other thread finish. It gives pre-empted thread a chance
                 * to proceed and finish with ring dequeue operation. */
                if (RTE_RING_PAUSE_REP_COUNT &&
                    ++rep == RTE_RING_PAUSE_REP_COUNT) {
                        rep = 0;
                        sched_yield();
                }
        }
        r->prod.tail = prod_next;
        return ret;
}

static inline int __attribute__((always_inline))
__rte_ring_sp_do_enqueue(struct rte_ring *r, void * const *obj_table,
                         unsigned n, enum rte_ring_queue_behavior behavior)
{
        uint32_t prod_head, cons_tail;
        uint32_t prod_next, free_entries;
        unsigned i;
        uint32_t mask = r->prod.mask;
        int ret;

        prod_head = r->prod.head;
        cons_tail = r->cons.tail;
        /* The subtraction is done between two unsigned 32bits value
         * (the result is always modulo 32 bits even if we have
         * prod_head > cons_tail). So 'free_entries' is always between 0
         * and size(ring)-1. */
        free_entries = mask + cons_tail - prod_head;

        /* check that we have enough room in ring */
        if (unlikely(n > free_entries)) {
                if (behavior == RTE_RING_QUEUE_FIXED) {
                        __RING_STAT_ADD(r, enq_fail, n);
                        return -ENOBUFS;
                }
                else {
                        /* No free entry available */
                        if (unlikely(free_entries == 0)) {
                                __RING_STAT_ADD(r, enq_fail, n);
                                return 0;
                        }

                        n = free_entries;
                }
        }

        prod_next = prod_head + n;
        r->prod.head = prod_next;

        /* write entries in ring */
        ENQUEUE_PTRS();
        rte_smp_wmb();

        /* if we exceed the watermark */
        if (unlikely(((mask + 1) - free_entries + n) > r->prod.watermark)) {
                ret = (behavior == RTE_RING_QUEUE_FIXED) ? -EDQUOT :
                        (int)(n | RTE_RING_QUOT_EXCEED);
                __RING_STAT_ADD(r, enq_quota, n);
        }
        else {
                ret = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : n;
                __RING_STAT_ADD(r, enq_success, n);
        }

        r->prod.tail = prod_next;
        return ret;
}

static inline int __attribute__((always_inline))
__rte_ring_mc_do_dequeue(struct rte_ring *r, void **obj_table,
                 unsigned n, enum rte_ring_queue_behavior behavior)
{
        uint32_t cons_head, prod_tail;
        uint32_t cons_next, entries;
        const unsigned max = n;
        int success;
        unsigned i, rep = 0;
        uint32_t mask = r->prod.mask;

        /* Avoid the unnecessary cmpset operation below, which is also
         * potentially harmful when n equals 0. */
        if (n == 0)
                return 0;

        /* move cons.head atomically */
        do {
                /* Restore n as it may change every loop */
                n = max;

                cons_head = r->cons.head;
                prod_tail = r->prod.tail;
                /* The subtraction is done between two unsigned 32bits value
                 * (the result is always modulo 32 bits even if we have
                 * cons_head > prod_tail). So 'entries' is always between 0
                 * and size(ring)-1. */
                entries = (prod_tail - cons_head);

                /* Set the actual entries for dequeue */
                if (n > entries) {
                        if (behavior == RTE_RING_QUEUE_FIXED) {
                                __RING_STAT_ADD(r, deq_fail, n);
                                return -ENOENT;
                        }
                        else {
                                if (unlikely(entries == 0)){
                                        __RING_STAT_ADD(r, deq_fail, n);
                                        return 0;
                                }

                                n = entries;
                        }
                }

                cons_next = cons_head + n;
                success = rte_atomic32_cmpset(&r->cons.head, cons_head,
                                              cons_next);
        } while (unlikely(success == 0));

        /* copy in table */
        DEQUEUE_PTRS();
        rte_smp_rmb();

        /*
         * If there are other dequeues in progress that preceded us,
         * we need to wait for them to complete
         */
        while (unlikely(r->cons.tail != cons_head)) {
                rte_pause();

                /* Set RTE_RING_PAUSE_REP_COUNT to avoid spin too long waiting
                 * for other thread finish. It gives pre-empted thread a chance
                 * to proceed and finish with ring dequeue operation. */
                if (RTE_RING_PAUSE_REP_COUNT &&
                    ++rep == RTE_RING_PAUSE_REP_COUNT) {
                        rep = 0;
                        sched_yield();
                }
        }
        __RING_STAT_ADD(r, deq_success, n);
        r->cons.tail = cons_next;

        return behavior == RTE_RING_QUEUE_FIXED ? 0 : n;
}

static inline int __attribute__((always_inline))
__rte_ring_sc_do_dequeue(struct rte_ring *r, void **obj_table,
                 unsigned n, enum rte_ring_queue_behavior behavior)
{
        uint32_t cons_head, prod_tail;
        uint32_t cons_next, entries;
        unsigned i;
        uint32_t mask = r->prod.mask;

        cons_head = r->cons.head;
        prod_tail = r->prod.tail;
        /* The subtraction is done between two unsigned 32bits value
         * (the result is always modulo 32 bits even if we have
         * cons_head > prod_tail). So 'entries' is always between 0
         * and size(ring)-1. */
        entries = prod_tail - cons_head;

        if (n > entries) {
                if (behavior == RTE_RING_QUEUE_FIXED) {
                        __RING_STAT_ADD(r, deq_fail, n);
                        return -ENOENT;
                }
                else {
                        if (unlikely(entries == 0)){
                                __RING_STAT_ADD(r, deq_fail, n);
                                return 0;
                        }

                        n = entries;
                }
        }

        cons_next = cons_head + n;
        r->cons.head = cons_next;

        /* copy in table */
        DEQUEUE_PTRS();
        rte_smp_rmb();

        __RING_STAT_ADD(r, deq_success, n);
        r->cons.tail = cons_next;
        return behavior == RTE_RING_QUEUE_FIXED ? 0 : n;
}

static inline int __attribute__((always_inline))
rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
                         unsigned n)
{
        return __rte_ring_mp_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED);
}

static inline int __attribute__((always_inline))
rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
                         unsigned n)
{
        return __rte_ring_sp_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED);
}

static inline int __attribute__((always_inline))
rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
                      unsigned n)
{
        if (r->prod.sp_enqueue)
                return rte_ring_sp_enqueue_bulk(r, obj_table, n);
        else
                return rte_ring_mp_enqueue_bulk(r, obj_table, n);
}

static inline int __attribute__((always_inline))
rte_ring_mp_enqueue(struct rte_ring *r, void *obj)
{
        return rte_ring_mp_enqueue_bulk(r, &obj, 1);
}

static inline int __attribute__((always_inline))
rte_ring_sp_enqueue(struct rte_ring *r, void *obj)
{
        return rte_ring_sp_enqueue_bulk(r, &obj, 1);
}

static inline int __attribute__((always_inline))
rte_ring_enqueue(struct rte_ring *r, void *obj)
{
        if (r->prod.sp_enqueue)
                return rte_ring_sp_enqueue(r, obj);
        else
                return rte_ring_mp_enqueue(r, obj);
}

static inline int __attribute__((always_inline))
rte_ring_mc_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned n)
{
        return __rte_ring_mc_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED);
}

static inline int __attribute__((always_inline))
rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned n)
{
        return __rte_ring_sc_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED);
}

static inline int __attribute__((always_inline))
rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned n)
{
        if (r->cons.sc_dequeue)
                return rte_ring_sc_dequeue_bulk(r, obj_table, n);
        else
                return rte_ring_mc_dequeue_bulk(r, obj_table, n);
}

static inline int __attribute__((always_inline))
rte_ring_mc_dequeue(struct rte_ring *r, void **obj_p)
{
        return rte_ring_mc_dequeue_bulk(r, obj_p, 1);
}

static inline int __attribute__((always_inline))
rte_ring_sc_dequeue(struct rte_ring *r, void **obj_p)
{
        return rte_ring_sc_dequeue_bulk(r, obj_p, 1);
}

static inline int __attribute__((always_inline))
rte_ring_dequeue(struct rte_ring *r, void **obj_p)
{
        if (r->cons.sc_dequeue)
                return rte_ring_sc_dequeue(r, obj_p);
        else
                return rte_ring_mc_dequeue(r, obj_p);
}

static inline int
rte_ring_full(const struct rte_ring *r)
{
        uint32_t prod_tail = r->prod.tail;
        uint32_t cons_tail = r->cons.tail;
        return ((cons_tail - prod_tail - 1) & r->prod.mask) == 0;
}

static inline int
rte_ring_empty(const struct rte_ring *r)
{
        uint32_t prod_tail = r->prod.tail;
        uint32_t cons_tail = r->cons.tail;
        return !!(cons_tail == prod_tail);
}

static inline unsigned
rte_ring_count(const struct rte_ring *r)
{
        uint32_t prod_tail = r->prod.tail;
        uint32_t cons_tail = r->cons.tail;
        return (prod_tail - cons_tail) & r->prod.mask;
}

static inline unsigned
rte_ring_free_count(const struct rte_ring *r)
{
        uint32_t prod_tail = r->prod.tail;
        uint32_t cons_tail = r->cons.tail;
        return (cons_tail - prod_tail - 1) & r->prod.mask;
}

void rte_ring_list_dump(FILE *f);

struct rte_ring *rte_ring_lookup(const char *name);

static inline unsigned __attribute__((always_inline))
rte_ring_mp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
                         unsigned n)
{
        return __rte_ring_mp_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE);
}

static inline unsigned __attribute__((always_inline))
rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
                         unsigned n)
{
        return __rte_ring_sp_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE);
}

static inline unsigned __attribute__((always_inline))
rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
                      unsigned n)
{
        if (r->prod.sp_enqueue)
                return rte_ring_sp_enqueue_burst(r, obj_table, n);
        else
                return rte_ring_mp_enqueue_burst(r, obj_table, n);
}

static inline unsigned __attribute__((always_inline))
rte_ring_mc_dequeue_burst(struct rte_ring *r, void **obj_table, unsigned n)
{
        return __rte_ring_mc_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE);
}

static inline unsigned __attribute__((always_inline))
rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table, unsigned n)
{
        return __rte_ring_sc_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE);
}

static inline unsigned __attribute__((always_inline))
rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table, unsigned n)
{
        if (r->cons.sc_dequeue)
                return rte_ring_sc_dequeue_burst(r, obj_table, n);
        else
                return rte_ring_mc_dequeue_burst(r, obj_table, n);
}

#ifdef __cplusplus
}
#endif

#endif /* _RTE_RING_H_ */