[v7,1/4] net/i40e: cleanup Tx buffers

Message ID 20200109103822.89011-2-chenxux.di@intel.com (mailing list archive)
State Superseded, archived
Delegated to: xiaolong ye
Headers
Series drivers/net: cleanup Tx buffers |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/iol-testing success Testing PASS
ci/iol-nxp-Performance success Performance Testing PASS
ci/iol-mellanox-Performance success Performance Testing PASS
ci/iol-intel-Performance success Performance Testing PASS
ci/Intel-compilation success Compilation OK

Commit Message

Chenxu Di Jan. 9, 2020, 10:38 a.m. UTC
  Add support to the i40e driver for the API rte_eth_tx_done_cleanup
to force free consumed buffers on Tx ring.

Signed-off-by: Chenxu Di <chenxux.di@intel.com>
---
 drivers/net/i40e/i40e_ethdev.c    |   3 +
 drivers/net/i40e/i40e_ethdev_vf.c |   3 +
 drivers/net/i40e/i40e_rxtx.c      | 151 ++++++++++++++++++++++++++++++
 drivers/net/i40e/i40e_rxtx.h      |   8 ++
 4 files changed, 165 insertions(+)
  

Patch

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 5999c964b..e0b071891 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -522,6 +522,7 @@  static const struct eth_dev_ops i40e_eth_dev_ops = {
 	.mac_addr_set                 = i40e_set_default_mac_addr,
 	.mtu_set                      = i40e_dev_mtu_set,
 	.tm_ops_get                   = i40e_tm_ops_get,
+	.tx_done_cleanup              = i40e_tx_done_cleanup,
 };
 
 /* store statistics names and its offset in stats structure */
@@ -1358,6 +1359,8 @@  eth_i40e_dev_init(struct rte_eth_dev *dev, void *init_params __rte_unused)
 	dev->tx_pkt_burst = i40e_xmit_pkts;
 	dev->tx_pkt_prepare = i40e_prep_pkts;
 
+	i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_scalar);
+
 	/* for secondary processes, we don't initialise any further as primary
 	 * has already done this work. Only check we don't need a different
 	 * RX function */
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c b/drivers/net/i40e/i40e_ethdev_vf.c
index 5dba0928b..3dcc9434c 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/drivers/net/i40e/i40e_ethdev_vf.c
@@ -215,6 +215,7 @@  static const struct eth_dev_ops i40evf_eth_dev_ops = {
 	.rss_hash_conf_get    = i40evf_dev_rss_hash_conf_get,
 	.mtu_set              = i40evf_dev_mtu_set,
 	.mac_addr_set         = i40evf_set_default_mac_addr,
+	.tx_done_cleanup      = i40e_tx_done_cleanup,
 };
 
 /*
@@ -1473,6 +1474,8 @@  i40evf_dev_init(struct rte_eth_dev *eth_dev)
 	eth_dev->rx_pkt_burst = &i40e_recv_pkts;
 	eth_dev->tx_pkt_burst = &i40e_xmit_pkts;
 
+	i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_scalar);
+
 	/*
 	 * For secondary processes, we don't initialise any further as primary
 	 * has already done this work.
diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
index 17dc8c78f..dfbca06b6 100644
--- a/drivers/net/i40e/i40e_rxtx.c
+++ b/drivers/net/i40e/i40e_rxtx.c
@@ -2455,6 +2455,154 @@  i40e_tx_queue_release_mbufs(struct i40e_tx_queue *txq)
 	}
 }
 
+static i40e_tx_done_cleanup_t i40e_tx_done_cleanup_op;
+
+int
+i40e_tx_done_cleanup_scalar(struct i40e_tx_queue *txq,
+			uint32_t free_cnt)
+{
+	uint32_t pkt_cnt;
+	uint16_t i;
+	uint16_t tx_last;
+	uint16_t tx_id;
+	uint16_t nb_tx_to_clean;
+	uint16_t nb_tx_free_last;
+	struct i40e_tx_entry *swr_ring = txq->sw_ring;
+
+	/* Start free mbuf from the next of tx_tail */
+	tx_last = txq->tx_tail;
+	tx_id  = swr_ring[tx_last].next_id;
+
+	if (txq->nb_tx_free == 0)
+		if (i40e_xmit_cleanup(txq))
+			return 0;
+
+	nb_tx_to_clean = txq->nb_tx_free;
+	nb_tx_free_last = txq->nb_tx_free;
+	if (!free_cnt)
+		free_cnt = txq->nb_tx_desc;
+
+	/* Loop through swr_ring to count the amount of
+	 * freeable mubfs and packets.
+	 */
+	for (pkt_cnt = 0; pkt_cnt < free_cnt; ) {
+		for (i = 0; i < nb_tx_to_clean &&
+			pkt_cnt < free_cnt &&
+			tx_id != tx_last; i++) {
+			if (swr_ring[tx_id].mbuf != NULL) {
+				rte_pktmbuf_free_seg(swr_ring[tx_id].mbuf);
+				swr_ring[tx_id].mbuf = NULL;
+
+				/*
+				 * last segment in the packet,
+				 * increment packet count
+				 */
+				pkt_cnt += (swr_ring[tx_id].last_id == tx_id);
+			}
+
+			tx_id = swr_ring[tx_id].next_id;
+		}
+
+		if (tx_id == tx_last || txq->tx_rs_thresh
+			> txq->nb_tx_desc - txq->nb_tx_free)
+			break;
+
+		if (pkt_cnt < free_cnt) {
+			if (i40e_xmit_cleanup(txq))
+				break;
+
+			nb_tx_to_clean = txq->nb_tx_free - nb_tx_free_last;
+			nb_tx_free_last = txq->nb_tx_free;
+		}
+	}
+
+	PMD_TX_FREE_LOG(DEBUG,
+		"Free %u Packets successfully "
+		"(port=%d queue=%d)",
+		pkt_cnt, txq->port_id, txq->queue_id);
+
+	return (int)pkt_cnt;
+}
+
+int
+i40e_tx_done_cleanup_simple(struct i40e_tx_queue *txq,
+			uint32_t free_cnt)
+{
+	uint16_t i;
+	uint16_t tx_first;
+	uint16_t tx_id;
+	uint32_t pkt_cnt;
+	struct i40e_tx_entry *swr_ring = txq->sw_ring;
+
+	/* Start free mbuf from tx_first */
+	tx_first = txq->tx_next_dd - (txq->tx_rs_thresh - 1);
+	tx_id  = tx_first;
+
+	/* while free_cnt is 0,
+	 * suppose one mbuf per packet,
+	 * try to free packets as many as possible
+	 */
+	if (free_cnt == 0)
+		free_cnt = txq->nb_tx_desc;
+
+	/* Loop through swr_ring to count freeable packets */
+	for (pkt_cnt = 0; pkt_cnt < free_cnt; ) {
+		if (txq->nb_tx_desc - txq->nb_tx_free < txq->tx_rs_thresh)
+			break;
+
+		if (!i40e_tx_free_bufs(txq))
+			break;
+
+		for (i = 0; i != txq->tx_rs_thresh &&
+			tx_id != tx_first; i++) {
+			/* last segment in the packet,
+			 * increment packet count
+			 */
+			pkt_cnt += (tx_id == swr_ring[tx_id].last_id);
+			tx_id = swr_ring[tx_id].next_id;
+		}
+
+		if (tx_id == tx_first)
+			break;
+	}
+
+	PMD_TX_FREE_LOG(DEBUG,
+		"Free %u packets successfully "
+		"(port=%d queue=%d)",
+		pkt_cnt, txq->port_id, txq->queue_id);
+
+	return (int)pkt_cnt;
+}
+
+int
+i40e_tx_done_cleanup_vec(struct i40e_tx_queue *txq __rte_unused,
+			uint32_t free_cnt __rte_unused)
+{
+	return -ENOTSUP;
+}
+int
+i40e_tx_done_cleanup(void *txq, uint32_t free_cnt)
+{
+	i40e_tx_done_cleanup_t func = i40e_get_tx_done_cleanup_func();
+
+	if (!func)
+		return -ENOTSUP;
+
+	return func(txq, free_cnt);
+}
+
+void
+i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_t fn)
+{
+	i40e_tx_done_cleanup_op = fn;
+}
+
+i40e_tx_done_cleanup_t
+i40e_get_tx_done_cleanup_func(void)
+{
+	return i40e_tx_done_cleanup_op;
+}
+
 void
 i40e_reset_tx_queue(struct i40e_tx_queue *txq)
 {
@@ -3139,15 +3287,18 @@  i40e_set_tx_function(struct rte_eth_dev *dev)
 			else
 				dev->tx_pkt_burst =
 					i40e_get_recommend_tx_vec();
+			i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_vec);
 		} else {
 			PMD_INIT_LOG(DEBUG, "Simple tx finally be used.");
 			dev->tx_pkt_burst = i40e_xmit_pkts_simple;
+			i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_simple);
 		}
 		dev->tx_pkt_prepare = NULL;
 	} else {
 		PMD_INIT_LOG(DEBUG, "Xmit tx finally be used.");
 		dev->tx_pkt_burst = i40e_xmit_pkts;
 		dev->tx_pkt_prepare = i40e_prep_pkts;
+		i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_scalar);
 	}
 }
 
diff --git a/drivers/net/i40e/i40e_rxtx.h b/drivers/net/i40e/i40e_rxtx.h
index 2106bb355..ab2c0ffd0 100644
--- a/drivers/net/i40e/i40e_rxtx.h
+++ b/drivers/net/i40e/i40e_rxtx.h
@@ -173,6 +173,8 @@  union i40e_tx_offload {
 		uint64_t outer_l3_len:16; /**< outer L3 Header Length */
 	};
 };
+typedef int (*i40e_tx_done_cleanup_t)(struct i40e_tx_queue *txq,
+				uint32_t free_cnt);
 
 int i40e_dev_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id);
 int i40e_dev_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id);
@@ -212,6 +214,12 @@  void i40e_dev_free_queues(struct rte_eth_dev *dev);
 void i40e_reset_rx_queue(struct i40e_rx_queue *rxq);
 void i40e_reset_tx_queue(struct i40e_tx_queue *txq);
 void i40e_tx_queue_release_mbufs(struct i40e_tx_queue *txq);
+void i40e_set_tx_done_cleanup_func(i40e_tx_done_cleanup_t fn);
+i40e_tx_done_cleanup_t i40e_get_tx_done_cleanup_func(void);
+int i40e_tx_done_cleanup(void *txq, uint32_t free_cnt);
+int i40e_tx_done_cleanup_scalar(struct i40e_tx_queue *txq, uint32_t free_cnt);
+int i40e_tx_done_cleanup_vec(struct i40e_tx_queue *txq, uint32_t free_cnt);
+int i40e_tx_done_cleanup_simple(struct i40e_tx_queue *txq, uint32_t free_cnt);
 int i40e_alloc_rx_queue_mbufs(struct i40e_rx_queue *rxq);
 void i40e_rx_queue_release_mbufs(struct i40e_rx_queue *rxq);