[dpdk-stable] patch 'net/mlx5: optimize Tx external memory registration' has been queued to stable release 18.08.1

[Kevin Traynor]

Hi,

FYI, your patch has been queued to stable release 18.08.1

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 12/08/18. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the patch applied
to the branch. If the code is different (ie: not only metadata diffs), due for example to
a change in context or macro names, please double check it.

Thanks.

Kevin Traynor

---
>From 5c354a0f6dbcd7841296acc99e66454130d55d2a Mon Sep 17 00:00:00 2001
From: Yongseok Koh <yskoh at mellanox.com>
Date: Thu, 15 Nov 2018 10:29:17 +0000
Subject: [PATCH] net/mlx5: optimize Tx external memory registration

[ upstream commit feddd5d243b7b43d51d2e9655b9143c462384ea8 ]

There's some performance drop due to extra condition checks on the
datapath. Checking for external memory registration should be consolidated
to the existing bottom-half.

Fixes: 7e43a32ee060 ("net/mlx5: support externally allocated static memory")

Signed-off-by: Yongseok Koh <yskoh at mellanox.com>
Acked-by: Shahaf Shuler <shahafs at mellanox.com>
---
 drivers/net/mlx5/mlx5_mr.c   | 28 +++++++++++++++++++++++++++-
 drivers/net/mlx5/mlx5_rxtx.h | 26 ++++++--------------------
 2 files changed, 33 insertions(+), 21 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_mr.c b/drivers/net/mlx5/mlx5_mr.c
index 6d7653d7d..442b2d232 100644
--- a/drivers/net/mlx5/mlx5_mr.c
+++ b/drivers/net/mlx5/mlx5_mr.c
@@ -1046,5 +1046,5 @@ mlx5_rx_addr2mr_bh(struct mlx5_rxq_data *rxq, uintptr_t addr)
  *   Searched LKey on success, UINT32_MAX on no match.
  */
-uint32_t
+static uint32_t
 mlx5_tx_addr2mr_bh(struct mlx5_txq_data *txq, uintptr_t addr)
 {
@@ -1060,4 +1060,30 @@ mlx5_tx_addr2mr_bh(struct mlx5_txq_data *txq, uintptr_t addr)
 }
 
+/**
+ * Bottom-half of LKey search on Tx. If it can't be searched in the memseg
+ * list, register the mempool of the mbuf as externally allocated memory.
+ *
+ * @param txq
+ *   Pointer to Tx queue structure.
+ * @param mb
+ *   Pointer to mbuf.
+ *
+ * @return
+ *   Searched LKey on success, UINT32_MAX on no match.
+ */
+uint32_t
+mlx5_tx_mb2mr_bh(struct mlx5_txq_data *txq, struct rte_mbuf *mb)
+{
+	uintptr_t addr = (uintptr_t)mb->buf_addr;
+	uint32_t lkey;
+
+	lkey = mlx5_tx_addr2mr_bh(txq, addr);
+	if (lkey == UINT32_MAX && rte_errno == ENXIO) {
+		/* Mempool may have externally allocated memory. */
+		return mlx5_tx_update_ext_mp(txq, addr, mlx5_mb2mp(mb));
+	}
+	return lkey;
+}
+
 /**
  * Flush all of the local cache entries.
diff --git a/drivers/net/mlx5/mlx5_rxtx.h b/drivers/net/mlx5/mlx5_rxtx.h
index 1db468c3b..e1dc6db21 100644
--- a/drivers/net/mlx5/mlx5_rxtx.h
+++ b/drivers/net/mlx5/mlx5_rxtx.h
@@ -363,5 +363,5 @@ uint16_t mlx5_rx_burst_vec(void *dpdk_txq, struct rte_mbuf **pkts,
 void mlx5_mr_flush_local_cache(struct mlx5_mr_ctrl *mr_ctrl);
 uint32_t mlx5_rx_addr2mr_bh(struct mlx5_rxq_data *rxq, uintptr_t addr);
-uint32_t mlx5_tx_addr2mr_bh(struct mlx5_txq_data *txq, uintptr_t addr);
+uint32_t mlx5_tx_mb2mr_bh(struct mlx5_txq_data *txq, struct rte_mbuf *mb);
 uint32_t mlx5_tx_update_ext_mp(struct mlx5_txq_data *txq, uintptr_t addr,
 			       struct rte_mempool *mp);
@@ -619,5 +619,5 @@ mlx5_tx_complete(struct mlx5_txq_data *txq)
  *   Memory pool where data is located for given mbuf.
  */
-static struct rte_mempool *
+static inline struct rte_mempool *
 mlx5_mb2mp(struct rte_mbuf *buf)
 {
@@ -668,7 +668,8 @@ mlx5_rx_addr2mr(struct mlx5_rxq_data *rxq, uintptr_t addr)
  */
 static __rte_always_inline uint32_t
-mlx5_tx_addr2mr(struct mlx5_txq_data *txq, uintptr_t addr)
+mlx5_tx_mb2mr(struct mlx5_txq_data *txq, struct rte_mbuf *mb)
 {
 	struct mlx5_mr_ctrl *mr_ctrl = &txq->mr_ctrl;
+	uintptr_t addr = (uintptr_t)mb->buf_addr;
 	uint32_t lkey;
 
@@ -681,21 +682,6 @@ mlx5_tx_addr2mr(struct mlx5_txq_data *txq, uintptr_t addr)
 	if (likely(lkey != UINT32_MAX))
 		return lkey;
-	/* Take slower bottom-half (binary search) on miss. */
-	return mlx5_tx_addr2mr_bh(txq, addr);
-}
-
-static __rte_always_inline uint32_t
-mlx5_tx_mb2mr(struct mlx5_txq_data *txq, struct rte_mbuf *mb)
-{
-	uintptr_t addr = (uintptr_t)mb->buf_addr;
-	uint32_t lkey = mlx5_tx_addr2mr(txq, addr);
-
-	if (likely(lkey != UINT32_MAX))
-		return lkey;
-	if (rte_errno == ENXIO) {
-		/* Mempool may have externally allocated memory. */
-		lkey = mlx5_tx_update_ext_mp(txq, addr, mlx5_mb2mp(mb));
-	}
-	return lkey;
+	/* Take slower bottom-half on miss. */
+	return mlx5_tx_mb2mr_bh(txq, mb);
 }
 
-- 
2.19.0

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2018-11-29 13:11:36.270910483 +0000
+++ 0047-net-mlx5-optimize-Tx-external-memory-registration.patch	2018-11-29 13:11:34.000000000 +0000
@@ -1,14 +1,15 @@
-From feddd5d243b7b43d51d2e9655b9143c462384ea8 Mon Sep 17 00:00:00 2001
+From 5c354a0f6dbcd7841296acc99e66454130d55d2a Mon Sep 17 00:00:00 2001
 From: Yongseok Koh <yskoh at mellanox.com>
 Date: Thu, 15 Nov 2018 10:29:17 +0000
 Subject: [PATCH] net/mlx5: optimize Tx external memory registration
 
+[ upstream commit feddd5d243b7b43d51d2e9655b9143c462384ea8 ]
+
 There's some performance drop due to extra condition checks on the
 datapath. Checking for external memory registration should be consolidated
 to the existing bottom-half.
 
 Fixes: 7e43a32ee060 ("net/mlx5: support externally allocated static memory")
-Cc: stable at dpdk.org
 
 Signed-off-by: Yongseok Koh <yskoh at mellanox.com>
 Acked-by: Shahaf Shuler <shahafs at mellanox.com>
@@ -60,24 +61,24 @@
  /**
   * Flush all of the local cache entries.
 diff --git a/drivers/net/mlx5/mlx5_rxtx.h b/drivers/net/mlx5/mlx5_rxtx.h
-index 1b6200f6c..59fb43fef 100644
+index 1db468c3b..e1dc6db21 100644
 --- a/drivers/net/mlx5/mlx5_rxtx.h
 +++ b/drivers/net/mlx5/mlx5_rxtx.h
-@@ -364,5 +364,5 @@ uint16_t mlx5_rx_burst_vec(void *dpdk_txq, struct rte_mbuf **pkts,
+@@ -363,5 +363,5 @@ uint16_t mlx5_rx_burst_vec(void *dpdk_txq, struct rte_mbuf **pkts,
  void mlx5_mr_flush_local_cache(struct mlx5_mr_ctrl *mr_ctrl);
  uint32_t mlx5_rx_addr2mr_bh(struct mlx5_rxq_data *rxq, uintptr_t addr);
 -uint32_t mlx5_tx_addr2mr_bh(struct mlx5_txq_data *txq, uintptr_t addr);
 +uint32_t mlx5_tx_mb2mr_bh(struct mlx5_txq_data *txq, struct rte_mbuf *mb);
  uint32_t mlx5_tx_update_ext_mp(struct mlx5_txq_data *txq, uintptr_t addr,
  			       struct rte_mempool *mp);
-@@ -620,5 +620,5 @@ mlx5_tx_complete(struct mlx5_txq_data *txq)
+@@ -619,5 +619,5 @@ mlx5_tx_complete(struct mlx5_txq_data *txq)
   *   Memory pool where data is located for given mbuf.
   */
 -static struct rte_mempool *
 +static inline struct rte_mempool *
  mlx5_mb2mp(struct rte_mbuf *buf)
  {
-@@ -669,7 +669,8 @@ mlx5_rx_addr2mr(struct mlx5_rxq_data *rxq, uintptr_t addr)
+@@ -668,7 +668,8 @@ mlx5_rx_addr2mr(struct mlx5_rxq_data *rxq, uintptr_t addr)
   */
  static __rte_always_inline uint32_t
 -mlx5_tx_addr2mr(struct mlx5_txq_data *txq, uintptr_t addr)
@@ -87,7 +88,7 @@
 +	uintptr_t addr = (uintptr_t)mb->buf_addr;
  	uint32_t lkey;
  
-@@ -682,21 +683,6 @@ mlx5_tx_addr2mr(struct mlx5_txq_data *txq, uintptr_t addr)
+@@ -681,21 +682,6 @@ mlx5_tx_addr2mr(struct mlx5_txq_data *txq, uintptr_t addr)
  	if (likely(lkey != UINT32_MAX))
  		return lkey;
 -	/* Take slower bottom-half (binary search) on miss. */