DPDK patches and discussions
 help / color / mirror / Atom feed
From: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
To: dev@dpdk.org
Cc: Bruce Richardson <bruce.richardson@intel.com>,
	Ferruh Yigit <ferruh.yigit@intel.com>,
	Adrien Mazarguil <adrien.mazarguil@6wind.com>
Subject: [dpdk-dev] [PATCH v7 02/25] mlx5: split memory registration function
Date: Fri, 24 Jun 2016 15:17:41 +0200	[thread overview]
Message-ID: <1466774284-20932-3-git-send-email-nelio.laranjeiro@6wind.com> (raw)
In-Reply-To: <1466774284-20932-1-git-send-email-nelio.laranjeiro@6wind.com>

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 <nelio.laranjeiro@6wind.com>
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
---
 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 <infiniband/verbs.h>
+#ifdef PEDANTIC
+#pragma GCC diagnostic error "-pedantic"
+#endif
+
+/* DPDK headers don't like -pedantic. */
+#ifdef PEDANTIC
+#pragma GCC diagnostic ignored "-pedantic"
+#endif
+#include <rte_mempool.h>
+#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

  parent reply	other threads:[~2016-06-24 13:18 UTC|newest]

Thread overview: 211+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-08  9:47 [dpdk-dev] [PATCH 00/24] Refactor mlx5 to improve performance Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 01/24] mlx5: split memory registration function for better performance Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 02/24] mlx5: remove TX gather support Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 03/24] mlx5: remove RX scatter support Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 04/24] mlx5: remove configuration variable for maximum number of segments Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 05/24] mlx5: remove inline TX support Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 06/24] mlx5: split TX queue structure Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 07/24] mlx5: split RX " Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 08/24] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 09/24] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 10/24] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 11/24] mlx5: add TX/RX burst function selection wrapper Nelio Laranjeiro
2016-06-08  9:47 ` [dpdk-dev] [PATCH 12/24] mlx5: refactor RX data path Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 13/24] mlx5: refactor TX " Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 14/24] mlx5: handle RX CQE compression Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 15/24] mlx5: replace countdown with threshold for TX completions Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 16/24] mlx5: add support for inline send Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 17/24] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 18/24] mlx5: add debugging information about TX queues capabilities Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 19/24] mlx5: check remaining space while processing TX burst Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 20/24] mlx5: resurrect TX gather support Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 21/24] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 22/24] mlx5: remove redundant RX queue initialization code Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 23/24] mlx5: make RX queue reinitialization safer Nelio Laranjeiro
2016-06-08  9:48 ` [dpdk-dev] [PATCH 24/24] mlx5: resurrect RX scatter support Nelio Laranjeiro
2016-06-13 18:50 ` [dpdk-dev] [PATCH 00/24] Refactor mlx5 to improve performance Javier Blazquez
2016-06-14  6:57   ` Nélio Laranjeiro
2016-06-17 16:09 ` Ferruh Yigit
2016-06-20  7:38   ` Nélio Laranjeiro
2016-06-20 15:03     ` Ferruh Yigit
2016-06-20 15:11       ` Nélio Laranjeiro
2016-06-20 16:10 ` [dpdk-dev] [PATCH v2 00/25] " Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 02/25] mlx5: split memory registration function Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-20 16:10   ` [dpdk-dev] [PATCH v2 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-20 17:01   ` [dpdk-dev] [PATCH v2 00/25] Refactor mlx5 to improve performance Ferruh Yigit
2016-06-21  6:44     ` Nélio Laranjeiro
2016-06-21  7:23 ` [dpdk-dev] [PATCH v3 " Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 02/25] mlx5: split memory registration function Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-21 16:42     ` Ferruh Yigit
2016-06-22  7:30       ` Nélio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-21  7:23   ` [dpdk-dev] [PATCH v3 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-21  7:43   ` [dpdk-dev] [PATCH v3 00/25] Refactor mlx5 to improve performance Yuanhan Liu
2016-06-21  8:00     ` Nélio Laranjeiro
2016-06-21  8:05       ` Yuanhan Liu
2016-06-21  8:49         ` Nélio Laranjeiro
2016-06-21 10:44     ` Ferruh Yigit
2016-06-21 12:26       ` Thomas Monjalon
2016-06-21 16:42   ` Ferruh Yigit
2016-06-22  8:20     ` Adrien Mazarguil
2016-06-22  9:19       ` Bruce Richardson
2016-06-22  9:30         ` Adrien Mazarguil
2016-06-23 15:14           ` Adrien Mazarguil
2016-06-22  9:05   ` [dpdk-dev] [PATCH " Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 02/25] mlx5: split memory registration function Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-22  9:05     ` [dpdk-dev] [PATCH v4 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-23 16:52     ` [dpdk-dev] [PATCH v5 00/25] Refactor mlx5 to improve performance Nelio Laranjeiro
2016-06-23 16:52       ` [dpdk-dev] [PATCH v5 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-23 16:52       ` [dpdk-dev] [PATCH v5 02/25] mlx5: split memory registration function Nelio Laranjeiro
2016-06-23 16:52       ` [dpdk-dev] [PATCH v5 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-23 16:53       ` [dpdk-dev] [PATCH v5 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-23 17:05         ` [dpdk-dev] [PATCH v5 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-23 17:11       ` [dpdk-dev] [PATCH v5 00/25] Refactor mlx5 to improve performance Nélio Laranjeiro
2016-06-24  8:50       ` [dpdk-dev] [PATCH v6 " Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 02/25] mlx5: split memory registration function Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-24  8:50         ` [dpdk-dev] [PATCH v6 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-24  8:51         ` [dpdk-dev] [PATCH v6 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-24  8:51         ` [dpdk-dev] [PATCH v6 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-24 13:04         ` [dpdk-dev] [PATCH v6 00/25] Refactor mlx5 to improve performance Nélio Laranjeiro
2016-06-24 13:17         ` [dpdk-dev] [PATCH v7 " Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 01/25] drivers: fix PCI class id support Nelio Laranjeiro
2016-06-24 13:17           ` Nelio Laranjeiro [this message]
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 03/25] mlx5: remove Tx gather support Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 04/25] mlx5: remove Rx scatter support Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 05/25] mlx5: remove configuration variable Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 06/25] mlx5: remove inline Tx support Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 07/25] mlx5: split Tx queue structure Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 08/25] mlx5: split Rx " Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 09/25] mlx5: update prerequisites for upcoming enhancements Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 10/25] mlx5: add definitions for data path without Verbs Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 11/25] mlx5: add support for configuration through kvargs Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 12/25] mlx5: add Tx/Rx burst function selection wrapper Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 13/25] mlx5: refactor Rx data path Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 14/25] mlx5: refactor Tx " Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 15/25] mlx5: handle Rx CQE compression Nelio Laranjeiro
2016-06-27 12:03             ` Bruce Richardson
2016-06-27 12:22               ` Nélio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 16/25] mlx5: replace countdown with threshold for Tx completions Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 17/25] mlx5: add support for inline send Nelio Laranjeiro
2016-06-27 12:17             ` Bruce Richardson
2016-06-27 12:24               ` Nélio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 18/25] mlx5: add support for multi-packet send Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 19/25] mlx5: add debugging information about Tx queues capabilities Nelio Laranjeiro
2016-06-24 13:17           ` [dpdk-dev] [PATCH v7 20/25] mlx5: check remaining space while processing Tx burst Nelio Laranjeiro
2016-06-24 13:18           ` [dpdk-dev] [PATCH v7 21/25] mlx5: resurrect Tx gather support Nelio Laranjeiro
2016-06-24 13:18           ` [dpdk-dev] [PATCH v7 22/25] mlx5: work around spurious compilation errors Nelio Laranjeiro
2016-06-27 12:12             ` Bruce Richardson
2016-06-27 12:27               ` Adrien Mazarguil
2016-06-24 13:18           ` [dpdk-dev] [PATCH v7 23/25] mlx5: remove redundant Rx queue initialization code Nelio Laranjeiro
2016-06-24 13:18           ` [dpdk-dev] [PATCH v7 24/25] mlx5: make Rx queue reinitialization safer Nelio Laranjeiro
2016-06-24 13:18           ` [dpdk-dev] [PATCH v7 25/25] mlx5: resurrect Rx scatter support Nelio Laranjeiro
2016-06-27 12:31           ` [dpdk-dev] [PATCH v7 00/25] Refactor mlx5 to improve performance Bruce Richardson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1466774284-20932-3-git-send-email-nelio.laranjeiro@6wind.com \
    --to=nelio.laranjeiro@6wind.com \
    --cc=adrien.mazarguil@6wind.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).