From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f50.google.com (mail-wm0-f50.google.com [74.125.82.50]) by dpdk.org (Postfix) with ESMTP id 32DE6C5AE for ; Thu, 23 Jun 2016 18:53:45 +0200 (CEST) Received: by mail-wm0-f50.google.com with SMTP id f126so134656681wma.1 for ; Thu, 23 Jun 2016 09:53:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=6wind-com.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=WCthhE8h8/yfsMVG4cQXk/v9/aNjjFNyWH8SvNIUVdE=; b=hmAqE5XlbU83SgjYRUzN/uKpmJO+1AGOVnmeLAlkL31PB2z89YLuNZ7KLy3XQ06aF7 NFNw4X2A0Ds4C5qFQUL7yhaUDI276OQo420O15hq2pcP+Hi2tC67GBb6f/LVP/s2q33q TgWlCtQIj778qlle+BE8Psou4In1PFzhVdUHm/gFFV2/v932sDEzsrdClrxaTGd+Z8od MZwKFf+Om/hDrECDCNcso0nSHXreiAnwGWy/54rtHjjbVY3YAatUI/o/Q78XoVYOnrIy vBzS0XM1OAqbxFHWXYD16DpbxHdzBoHCwK9kFv8gQ1wlq55kRJsQ/KTzKk8EMAO464gM j+Hg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=WCthhE8h8/yfsMVG4cQXk/v9/aNjjFNyWH8SvNIUVdE=; b=fLB3LVUUbVh/wM+MtGipcjNjWvzONKtaQjm1IbxLCd9roiC/hrAF85USsm6PpAgXEY cKFibsnC48eGrNrWPqtiw9Ztm73l9UZbnfsEJl5StJyejUiUnyWwKvh/0QfNgNjIbUsO Yb6MvjzEAA8Wb1DdYe+0nbtBv5s0UuNZB5Ivqazbs6WIHoqCpBSR2s82RDxHUsKNQEYk OrF7St26fut1PLGADnFr4iSjG4sK2ZTj7czA2A8SAACQVBMAXrcsZolzFvXRMNCY0cv8 fEusNs4NR5DfY8NXVg7IY2Jc4pRwQCUGINIaas+mYGPYfAYdyd5cVD4m5NHl3iKQxywS f6Dg== X-Gm-Message-State: ALyK8tJ6LqJheLlMEL0qqED4K21rhXwSxVkNZEuQVWQaIZGnweDTNqH0gVWGweICTQb+zX85 X-Received: by 10.194.235.66 with SMTP id uk2mr7294721wjc.79.1466700824192; Thu, 23 Jun 2016 09:53:44 -0700 (PDT) Received: from ping.vm.6wind.com (guy78-3-82-239-227-177.fbx.proxad.net. [82.239.227.177]) by smtp.gmail.com with ESMTPSA id ze10sm992509wjb.2.2016.06.23.09.53.41 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Thu, 23 Jun 2016 09:53:42 -0700 (PDT) From: Nelio Laranjeiro To: dev@dpdk.org Cc: Ferruh Yigit , Adrien Mazarguil Date: Thu, 23 Jun 2016 18:52:58 +0200 Message-Id: <1466700801-10383-3-git-send-email-nelio.laranjeiro@6wind.com> X-Mailer: git-send-email 2.1.4 In-Reply-To: <1466700801-10383-1-git-send-email-nelio.laranjeiro@6wind.com> References: <1466586355-30777-1-git-send-email-nelio.laranjeiro@6wind.com> <1466700801-10383-1-git-send-email-nelio.laranjeiro@6wind.com> Subject: [dpdk-dev] [PATCH v5 02/25] mlx5: split memory registration function X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 23 Jun 2016 16:53:45 -0000 Except for the first time when memory registration occurs, the lkey is always cached. Since memory registration is slow and performs system calls, performance can be improved by moving that code to its own function outside of the data path so only the lookup code is left in the original inlined function. Signed-off-by: Nelio Laranjeiro Signed-off-by: Adrien Mazarguil --- drivers/net/mlx5/Makefile | 1 + drivers/net/mlx5/mlx5_mr.c | 280 +++++++++++++++++++++++++++++++++++++++++++ drivers/net/mlx5/mlx5_rxtx.c | 209 ++------------------------------ drivers/net/mlx5/mlx5_rxtx.h | 8 +- 4 files changed, 298 insertions(+), 200 deletions(-) create mode 100644 drivers/net/mlx5/mlx5_mr.c diff --git a/drivers/net/mlx5/Makefile b/drivers/net/mlx5/Makefile index 82558aa..999ada5 100644 --- a/drivers/net/mlx5/Makefile +++ b/drivers/net/mlx5/Makefile @@ -47,6 +47,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_vlan.c SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_stats.c SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_rss.c SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_fdir.c +SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_mr.c # Dependencies. DEPDIRS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += lib/librte_ether diff --git a/drivers/net/mlx5/mlx5_mr.c b/drivers/net/mlx5/mlx5_mr.c new file mode 100644 index 0000000..bb44041 --- /dev/null +++ b/drivers/net/mlx5/mlx5_mr.c @@ -0,0 +1,280 @@ +/*- + * BSD LICENSE + * + * Copyright 2016 6WIND S.A. + * Copyright 2016 Mellanox. + * + * 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 6WIND S.A. 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. + */ + +/* Verbs header. */ +/* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */ +#ifdef PEDANTIC +#pragma GCC diagnostic ignored "-pedantic" +#endif +#include +#ifdef PEDANTIC +#pragma GCC diagnostic error "-pedantic" +#endif + +/* DPDK headers don't like -pedantic. */ +#ifdef PEDANTIC +#pragma GCC diagnostic ignored "-pedantic" +#endif +#include +#ifdef PEDANTIC +#pragma GCC diagnostic error "-pedantic" +#endif + +#include "mlx5.h" +#include "mlx5_rxtx.h" + +struct mlx5_check_mempool_data { + int ret; + char *start; + char *end; +}; + +/* Called by mlx5_check_mempool() when iterating the memory chunks. */ +static void +mlx5_check_mempool_cb(struct rte_mempool *mp, + void *opaque, struct rte_mempool_memhdr *memhdr, + unsigned int mem_idx) +{ + struct mlx5_check_mempool_data *data = opaque; + + (void)mp; + (void)mem_idx; + + /* It already failed, skip the next chunks. */ + if (data->ret != 0) + return; + /* It is the first chunk. */ + if (data->start == NULL && data->end == NULL) { + data->start = memhdr->addr; + data->end = data->start + memhdr->len; + return; + } + if (data->end == memhdr->addr) { + data->end += memhdr->len; + return; + } + if (data->start == (char *)memhdr->addr + memhdr->len) { + data->start -= memhdr->len; + return; + } + /* Error, mempool is not virtually contiguous. */ + data->ret = -1; +} + +/** + * Check if a mempool can be used: it must be virtually contiguous. + * + * @param[in] mp + * Pointer to memory pool. + * @param[out] start + * Pointer to the start address of the mempool virtual memory area + * @param[out] end + * Pointer to the end address of the mempool virtual memory area + * + * @return + * 0 on success (mempool is virtually contiguous), -1 on error. + */ +static int mlx5_check_mempool(struct rte_mempool *mp, uintptr_t *start, + uintptr_t *end) +{ + struct mlx5_check_mempool_data data; + + memset(&data, 0, sizeof(data)); + rte_mempool_mem_iter(mp, mlx5_check_mempool_cb, &data); + *start = (uintptr_t)data.start; + *end = (uintptr_t)data.end; + + return data.ret; +} + +/** + * Register mempool as a memory region. + * + * @param pd + * Pointer to protection domain. + * @param mp + * Pointer to memory pool. + * + * @return + * Memory region pointer, NULL in case of error. + */ +struct ibv_mr * +mlx5_mp2mr(struct ibv_pd *pd, struct rte_mempool *mp) +{ + const struct rte_memseg *ms = rte_eal_get_physmem_layout(); + uintptr_t start; + uintptr_t end; + unsigned int i; + + if (mlx5_check_mempool(mp, &start, &end) != 0) { + ERROR("mempool %p: not virtually contiguous", + (void *)mp); + return NULL; + } + + DEBUG("mempool %p area start=%p end=%p size=%zu", + (void *)mp, (void *)start, (void *)end, + (size_t)(end - start)); + /* Round start and end to page boundary if found in memory segments. */ + for (i = 0; (i < RTE_MAX_MEMSEG) && (ms[i].addr != NULL); ++i) { + uintptr_t addr = (uintptr_t)ms[i].addr; + size_t len = ms[i].len; + unsigned int align = ms[i].hugepage_sz; + + if ((start > addr) && (start < addr + len)) + start = RTE_ALIGN_FLOOR(start, align); + if ((end > addr) && (end < addr + len)) + end = RTE_ALIGN_CEIL(end, align); + } + DEBUG("mempool %p using start=%p end=%p size=%zu for MR", + (void *)mp, (void *)start, (void *)end, + (size_t)(end - start)); + return ibv_reg_mr(pd, + (void *)start, + end - start, + IBV_ACCESS_LOCAL_WRITE); +} + +/** + * Register a Memory Region (MR) <-> Memory Pool (MP) association in + * txq->mp2mr[]. If mp2mr[] is full, remove an entry first. + * + * This function should only be called by txq_mp2mr(). + * + * @param txq + * Pointer to TX queue structure. + * @param[in] mp + * Memory Pool for which a Memory Region lkey must be returned. + * @param idx + * Index of the next available entry. + * + * @return + * mr->lkey on success, (uint32_t)-1 on failure. + */ +uint32_t +txq_mp2mr_reg(struct txq *txq, struct rte_mempool *mp, unsigned int idx) +{ + struct ibv_mr *mr; + + /* Add a new entry, register MR first. */ + DEBUG("%p: discovered new memory pool \"%s\" (%p)", + (void *)txq, mp->name, (void *)mp); + mr = mlx5_mp2mr(txq->priv->pd, mp); + if (unlikely(mr == NULL)) { + DEBUG("%p: unable to configure MR, ibv_reg_mr() failed.", + (void *)txq); + return (uint32_t)-1; + } + if (unlikely(idx == RTE_DIM(txq->mp2mr))) { + /* Table is full, remove oldest entry. */ + DEBUG("%p: MR <-> MP table full, dropping oldest entry.", + (void *)txq); + --idx; + claim_zero(ibv_dereg_mr(txq->mp2mr[0].mr)); + memmove(&txq->mp2mr[0], &txq->mp2mr[1], + (sizeof(txq->mp2mr) - sizeof(txq->mp2mr[0]))); + } + /* Store the new entry. */ + txq->mp2mr[idx].mp = mp; + txq->mp2mr[idx].mr = mr; + txq->mp2mr[idx].lkey = mr->lkey; + DEBUG("%p: new MR lkey for MP \"%s\" (%p): 0x%08" PRIu32, + (void *)txq, mp->name, (void *)mp, txq->mp2mr[idx].lkey); + return txq->mp2mr[idx].lkey; +} + +struct txq_mp2mr_mbuf_check_data { + int ret; +}; + +/** + * Callback function for rte_mempool_obj_iter() to check whether a given + * mempool object looks like a mbuf. + * + * @param[in] mp + * The mempool pointer + * @param[in] arg + * Context data (struct txq_mp2mr_mbuf_check_data). Contains the + * return value. + * @param[in] obj + * Object address. + * @param index + * Object index, unused. + */ +static void +txq_mp2mr_mbuf_check(struct rte_mempool *mp, void *arg, void *obj, + uint32_t index __rte_unused) +{ + struct txq_mp2mr_mbuf_check_data *data = arg; + struct rte_mbuf *buf = obj; + + /* + * Check whether mbuf structure fits element size and whether mempool + * pointer is valid. + */ + if (sizeof(*buf) > mp->elt_size || buf->pool != mp) + data->ret = -1; +} + +/** + * Iterator function for rte_mempool_walk() to register existing mempools and + * fill the MP to MR cache of a TX queue. + * + * @param[in] mp + * Memory Pool to register. + * @param *arg + * Pointer to TX queue structure. + */ +void +txq_mp2mr_iter(struct rte_mempool *mp, void *arg) +{ + struct txq *txq = arg; + struct txq_mp2mr_mbuf_check_data data = { + .ret = 0, + }; + unsigned int i; + + /* Register mempool only if the first element looks like a mbuf. */ + if (rte_mempool_obj_iter(mp, txq_mp2mr_mbuf_check, &data) == 0 || + data.ret == -1) + return; + for (i = 0; (i != RTE_DIM(txq->mp2mr)); ++i) { + if (unlikely(txq->mp2mr[i].mp == NULL)) { + /* Unknown MP, add a new MR for it. */ + break; + } + if (txq->mp2mr[i].mp == mp) + return; + } + txq_mp2mr_reg(txq, mp, i); +} diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c index 9cb1dfa..616cf7a 100644 --- a/drivers/net/mlx5/mlx5_rxtx.c +++ b/drivers/net/mlx5/mlx5_rxtx.c @@ -140,121 +140,6 @@ txq_complete(struct txq *txq) return 0; } -struct mlx5_check_mempool_data { - int ret; - char *start; - char *end; -}; - -/* Called by mlx5_check_mempool() when iterating the memory chunks. */ -static void mlx5_check_mempool_cb(struct rte_mempool *mp, - void *opaque, struct rte_mempool_memhdr *memhdr, - unsigned mem_idx) -{ - struct mlx5_check_mempool_data *data = opaque; - - (void)mp; - (void)mem_idx; - - /* It already failed, skip the next chunks. */ - if (data->ret != 0) - return; - /* It is the first chunk. */ - if (data->start == NULL && data->end == NULL) { - data->start = memhdr->addr; - data->end = data->start + memhdr->len; - return; - } - if (data->end == memhdr->addr) { - data->end += memhdr->len; - return; - } - if (data->start == (char *)memhdr->addr + memhdr->len) { - data->start -= memhdr->len; - return; - } - /* Error, mempool is not virtually contigous. */ - data->ret = -1; -} - -/** - * Check if a mempool can be used: it must be virtually contiguous. - * - * @param[in] mp - * Pointer to memory pool. - * @param[out] start - * Pointer to the start address of the mempool virtual memory area - * @param[out] end - * Pointer to the end address of the mempool virtual memory area - * - * @return - * 0 on success (mempool is virtually contiguous), -1 on error. - */ -static int mlx5_check_mempool(struct rte_mempool *mp, uintptr_t *start, - uintptr_t *end) -{ - struct mlx5_check_mempool_data data; - - memset(&data, 0, sizeof(data)); - rte_mempool_mem_iter(mp, mlx5_check_mempool_cb, &data); - *start = (uintptr_t)data.start; - *end = (uintptr_t)data.end; - - return data.ret; -} - -/* For best performance, this function should not be inlined. */ -struct ibv_mr *mlx5_mp2mr(struct ibv_pd *, struct rte_mempool *) - __attribute__((noinline)); - -/** - * Register mempool as a memory region. - * - * @param pd - * Pointer to protection domain. - * @param mp - * Pointer to memory pool. - * - * @return - * Memory region pointer, NULL in case of error. - */ -struct ibv_mr * -mlx5_mp2mr(struct ibv_pd *pd, struct rte_mempool *mp) -{ - const struct rte_memseg *ms = rte_eal_get_physmem_layout(); - uintptr_t start; - uintptr_t end; - unsigned int i; - - if (mlx5_check_mempool(mp, &start, &end) != 0) { - ERROR("mempool %p: not virtually contiguous", - (void *)mp); - return NULL; - } - - DEBUG("mempool %p area start=%p end=%p size=%zu", - (void *)mp, (void *)start, (void *)end, - (size_t)(end - start)); - /* Round start and end to page boundary if found in memory segments. */ - for (i = 0; (i < RTE_MAX_MEMSEG) && (ms[i].addr != NULL); ++i) { - uintptr_t addr = (uintptr_t)ms[i].addr; - size_t len = ms[i].len; - unsigned int align = ms[i].hugepage_sz; - - if ((start > addr) && (start < addr + len)) - start = RTE_ALIGN_FLOOR(start, align); - if ((end > addr) && (end < addr + len)) - end = RTE_ALIGN_CEIL(end, align); - } - DEBUG("mempool %p using start=%p end=%p size=%zu for MR", - (void *)mp, (void *)start, (void *)end, - (size_t)(end - start)); - return ibv_reg_mr(pd, - (void *)start, - end - start, - IBV_ACCESS_LOCAL_WRITE); -} - /** * Get Memory Pool (MP) from mbuf. If mbuf is indirect, the pool from which * the cloned mbuf is allocated is returned instead. @@ -273,6 +158,10 @@ txq_mb2mp(struct rte_mbuf *buf) return buf->pool; } +static inline uint32_t +txq_mp2mr(struct txq *txq, struct rte_mempool *mp) + __attribute__((always_inline)); + /** * Get Memory Region (MR) <-> Memory Pool (MP) association from txq->mp2mr[]. * Add MP to txq->mp2mr[] if it's not registered yet. If mp2mr[] is full, @@ -286,11 +175,11 @@ txq_mb2mp(struct rte_mbuf *buf) * @return * mr->lkey on success, (uint32_t)-1 on failure. */ -static uint32_t +static inline uint32_t txq_mp2mr(struct txq *txq, struct rte_mempool *mp) { unsigned int i; - struct ibv_mr *mr; + uint32_t lkey = (uint32_t)-1; for (i = 0; (i != RTE_DIM(txq->mp2mr)); ++i) { if (unlikely(txq->mp2mr[i].mp == NULL)) { @@ -300,89 +189,13 @@ txq_mp2mr(struct txq *txq, struct rte_mempool *mp) if (txq->mp2mr[i].mp == mp) { assert(txq->mp2mr[i].lkey != (uint32_t)-1); assert(txq->mp2mr[i].mr->lkey == txq->mp2mr[i].lkey); - return txq->mp2mr[i].lkey; + lkey = txq->mp2mr[i].lkey; + break; } } - /* Add a new entry, register MR first. */ - DEBUG("%p: discovered new memory pool \"%s\" (%p)", - (void *)txq, mp->name, (void *)mp); - mr = mlx5_mp2mr(txq->priv->pd, mp); - if (unlikely(mr == NULL)) { - DEBUG("%p: unable to configure MR, ibv_reg_mr() failed.", - (void *)txq); - return (uint32_t)-1; - } - if (unlikely(i == RTE_DIM(txq->mp2mr))) { - /* Table is full, remove oldest entry. */ - DEBUG("%p: MR <-> MP table full, dropping oldest entry.", - (void *)txq); - --i; - claim_zero(ibv_dereg_mr(txq->mp2mr[0].mr)); - memmove(&txq->mp2mr[0], &txq->mp2mr[1], - (sizeof(txq->mp2mr) - sizeof(txq->mp2mr[0]))); - } - /* Store the new entry. */ - txq->mp2mr[i].mp = mp; - txq->mp2mr[i].mr = mr; - txq->mp2mr[i].lkey = mr->lkey; - DEBUG("%p: new MR lkey for MP \"%s\" (%p): 0x%08" PRIu32, - (void *)txq, mp->name, (void *)mp, txq->mp2mr[i].lkey); - return txq->mp2mr[i].lkey; -} - -struct txq_mp2mr_mbuf_check_data { - int ret; -}; - -/** - * Callback function for rte_mempool_obj_iter() to check whether a given - * mempool object looks like a mbuf. - * - * @param[in] mp - * The mempool pointer - * @param[in] arg - * Context data (struct txq_mp2mr_mbuf_check_data). Contains the - * return value. - * @param[in] obj - * Object address. - * @param index - * Object index, unused. - */ -static void -txq_mp2mr_mbuf_check(struct rte_mempool *mp, void *arg, void *obj, - uint32_t index __rte_unused) -{ - struct txq_mp2mr_mbuf_check_data *data = arg; - struct rte_mbuf *buf = obj; - - /* Check whether mbuf structure fits element size and whether mempool - * pointer is valid. */ - if (sizeof(*buf) > mp->elt_size || buf->pool != mp) - data->ret = -1; -} - -/** - * Iterator function for rte_mempool_walk() to register existing mempools and - * fill the MP to MR cache of a TX queue. - * - * @param[in] mp - * Memory Pool to register. - * @param *arg - * Pointer to TX queue structure. - */ -void -txq_mp2mr_iter(struct rte_mempool *mp, void *arg) -{ - struct txq *txq = arg; - struct txq_mp2mr_mbuf_check_data data = { - .ret = 0, - }; - - /* Register mempool only if the first element looks like a mbuf. */ - if (rte_mempool_obj_iter(mp, txq_mp2mr_mbuf_check, &data) == 0 || - data.ret == -1) - return; - txq_mp2mr(txq, mp); + if (unlikely(lkey == (uint32_t)-1)) + lkey = txq_mp2mr_reg(txq, mp, i); + return lkey; } /** diff --git a/drivers/net/mlx5/mlx5_rxtx.h b/drivers/net/mlx5/mlx5_rxtx.h index 47f6299..462eddf 100644 --- a/drivers/net/mlx5/mlx5_rxtx.h +++ b/drivers/net/mlx5/mlx5_rxtx.h @@ -337,12 +337,16 @@ uint16_t mlx5_tx_burst_secondary_setup(void *, struct rte_mbuf **, uint16_t); /* mlx5_rxtx.c */ -struct ibv_mr *mlx5_mp2mr(struct ibv_pd *, struct rte_mempool *); -void txq_mp2mr_iter(struct rte_mempool *, void *); uint16_t mlx5_tx_burst(void *, struct rte_mbuf **, uint16_t); uint16_t mlx5_rx_burst_sp(void *, struct rte_mbuf **, uint16_t); uint16_t mlx5_rx_burst(void *, struct rte_mbuf **, uint16_t); uint16_t removed_tx_burst(void *, struct rte_mbuf **, uint16_t); uint16_t removed_rx_burst(void *, struct rte_mbuf **, uint16_t); +/* mlx5_mr.c */ + +struct ibv_mr *mlx5_mp2mr(struct ibv_pd *, struct rte_mempool *); +void txq_mp2mr_iter(struct rte_mempool *, void *); +uint32_t txq_mp2mr_reg(struct txq *, struct rte_mempool *, unsigned int); + #endif /* RTE_PMD_MLX5_RXTX_H_ */ -- 2.1.4