dpdk/dpdk-17.02_patches/0007-add-hardware-checksum-offload-for-tunnel-pa.patch

From f0dda2ab16635894b1e3836d0b960b9270a3b491 Mon Sep 17 00:00:00 2001
From: Shahaf Shuler <shahafs@mellanox.com>
Date: Thu, 2 Mar 2017 11:05:44 +0200
Subject: [PATCH] net/mlx5: add hardware checksum offload for tunnel packets

Prior to this commit Tx checksum offload was supported only for the
inner headers.
This commit adds support for the hardware to compute the checksum for the
outer headers as well.

The support is for tunneling protocols GRE and VXLAN.

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
---
 doc/guides/nics/features/mlx5.ini |  2 ++
 doc/guides/nics/mlx5.rst          |  3 ++-
 drivers/net/mlx5/mlx5.c           |  7 +++++++
 drivers/net/mlx5/mlx5.h           |  2 ++
 drivers/net/mlx5/mlx5_ethdev.c    |  2 ++
 drivers/net/mlx5/mlx5_prm.h       |  6 ++++++
 drivers/net/mlx5/mlx5_rxtx.c      | 14 +++++++++++++-
 drivers/net/mlx5/mlx5_rxtx.h      |  2 ++
 drivers/net/mlx5/mlx5_txq.c       |  2 ++
 9 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/doc/guides/nics/features/mlx5.ini b/doc/guides/nics/features/mlx5.ini
index 8df25ce..1814f82 100644
--- a/doc/guides/nics/features/mlx5.ini
+++ b/doc/guides/nics/features/mlx5.ini
@@ -27,6 +27,8 @@ CRC offload          = Y
 VLAN offload         = Y
 L3 checksum offload  = Y
 L4 checksum offload  = Y
+Inner L3 checksum    = Y
+Inner L4 checksum    = Y
 Packet type parsing  = Y
 Basic stats          = Y
 Stats per queue      = Y
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 9b0ba29..41f3a47 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -91,13 +91,14 @@ Features
 - KVM and VMware ESX SR-IOV modes are supported.
 - RSS hash result is supported.
 - Hardware TSO.
+- Hardware checksum TX offload for VXLAN and GRE.
 
 Limitations
 -----------
 
 - Inner RSS for VXLAN frames is not supported yet.
 - Port statistics through software counters only.
-- Hardware checksum offloads for VXLAN inner header are not supported yet.
+- Hardware checksum RX offloads for VXLAN inner header are not supported yet.
 - Secondary process RX is not supported.
 
 Configuration
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index 03ed3b3..6f42948 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -375,6 +375,7 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
 	struct ibv_device_attr device_attr;
 	unsigned int sriov;
 	unsigned int mps;
+	unsigned int tunnel_en;
 	int idx;
 	int i;
 
@@ -429,12 +430,17 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
 		 * as all ConnectX-5 devices.
 		 */
 		switch (pci_dev->id.device_id) {
+		case PCI_DEVICE_ID_MELLANOX_CONNECTX4:
+			tunnel_en = 1;
+			mps = 0;
+			break;
 		case PCI_DEVICE_ID_MELLANOX_CONNECTX4LX:
 		case PCI_DEVICE_ID_MELLANOX_CONNECTX5:
 		case PCI_DEVICE_ID_MELLANOX_CONNECTX5VF:
 		case PCI_DEVICE_ID_MELLANOX_CONNECTX5EX:
 		case PCI_DEVICE_ID_MELLANOX_CONNECTX5EXVF:
 			mps = 1;
+			tunnel_en = 1;
 			break;
 		default:
 			mps = 0;
@@ -539,6 +545,7 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
 		priv->mtu = ETHER_MTU;
 		priv->mps = mps; /* Enable MPW by default if supported. */
 		priv->cqe_comp = 1; /* Enable compression by default. */
+		priv->tunnel_en = tunnel_en;
 		err = mlx5_args(priv, pci_dev->device.devargs);
 		if (err) {
 			ERROR("failed to process device arguments: %s",
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 93f129b..870e01f 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -127,6 +127,8 @@ struct priv {
 	unsigned int cqe_comp:1; /* Whether CQE compression is enabled. */
 	unsigned int pending_alarm:1; /* An alarm is pending. */
 	unsigned int tso:1; /* Whether TSO is supported. */
+	unsigned int tunnel_en:1;
+	/* Whether Tx offloads for tunneled packets are supported. */
 	unsigned int max_tso_payload_sz; /* Maximum TCP payload for TSO. */
 	unsigned int txq_inline; /* Maximum packet size for inlining. */
 	unsigned int txqs_inline; /* Queue number threshold for inlining. */
diff --git a/drivers/net/mlx5/mlx5_ethdev.c b/drivers/net/mlx5/mlx5_ethdev.c
index 5542193..8be9e77 100644
--- a/drivers/net/mlx5/mlx5_ethdev.c
+++ b/drivers/net/mlx5/mlx5_ethdev.c
@@ -695,6 +695,8 @@ mlx5_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
 			 DEV_TX_OFFLOAD_TCP_CKSUM);
 	if (priv->tso)
 		info->tx_offload_capa |= DEV_TX_OFFLOAD_TCP_TSO;
+	if (priv->tunnel_en)
+		info->tx_offload_capa |= DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM;
 	if (priv_get_ifname(priv, &ifname) == 0)
 		info->if_index = if_nametoindex(ifname);
 	/* FIXME: RETA update/query API expects the callee to know the size of
diff --git a/drivers/net/mlx5/mlx5_prm.h b/drivers/net/mlx5/mlx5_prm.h
index 3318668..0a77f5b 100644
--- a/drivers/net/mlx5/mlx5_prm.h
+++ b/drivers/net/mlx5/mlx5_prm.h
@@ -120,6 +120,12 @@
 /* Tunnel packet bit in the CQE. */
 #define MLX5_CQE_RX_TUNNEL_PACKET (1u << 0)
 
+/* Inner L3 checksum offload (Tunneled packets only). */
+#define MLX5_ETH_WQE_L3_INNER_CSUM (1u << 4)
+
+/* Inner L4 checksum offload (Tunneled packets only). */
+#define MLX5_ETH_WQE_L4_INNER_CSUM (1u << 5)
+
 /* INVALID is used by packets matching no flow rules. */
 #define MLX5_FLOW_MARK_INVALID 0
 
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 98889f6..c2eb891 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -443,7 +443,19 @@ mlx5_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
 		/* Should we enable HW CKSUM offload */
 		if (buf->ol_flags &
 		    (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM)) {
-			cs_flags = MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM;
+			const uint64_t is_tunneled = buf->ol_flags &
+						     (PKT_TX_TUNNEL_GRE |
+						      PKT_TX_TUNNEL_VXLAN);
+
+			if (is_tunneled && txq->tunnel_en) {
+				cs_flags = MLX5_ETH_WQE_L3_INNER_CSUM |
+					   MLX5_ETH_WQE_L4_INNER_CSUM;
+				if (buf->ol_flags & PKT_TX_OUTER_IP_CKSUM)
+					cs_flags |= MLX5_ETH_WQE_L3_CSUM;
+			} else {
+				cs_flags = MLX5_ETH_WQE_L3_CSUM |
+					   MLX5_ETH_WQE_L4_CSUM;
+			}
 		}
 		raw = ((uint8_t *)(uintptr_t)wqe) + 2 * MLX5_WQE_DWORD_SIZE;
 		/* Replace the Ethernet type by the VLAN if necessary. */
diff --git a/drivers/net/mlx5/mlx5_rxtx.h b/drivers/net/mlx5/mlx5_rxtx.h
index 6b328cf..9669564 100644
--- a/drivers/net/mlx5/mlx5_rxtx.h
+++ b/drivers/net/mlx5/mlx5_rxtx.h
@@ -256,6 +256,8 @@ struct txq {
 	uint16_t max_inline; /* Multiple of RTE_CACHE_LINE_SIZE to inline. */
 	uint16_t inline_en:1; /* When set inline is enabled. */
 	uint16_t tso_en:1; /* When set hardware TSO is enabled. */
+	uint16_t tunnel_en:1;
+	/* When set TX offload for tunneled packets are supported. */
 	uint32_t qp_num_8s; /* QP number shifted by 8. */
 	volatile struct mlx5_cqe (*cqes)[]; /* Completion queue. */
 	volatile void *wqes; /* Work queue (use volatile to write into). */
diff --git a/drivers/net/mlx5/mlx5_txq.c b/drivers/net/mlx5/mlx5_txq.c
index 995b763..9d0c00f 100644
--- a/drivers/net/mlx5/mlx5_txq.c
+++ b/drivers/net/mlx5/mlx5_txq.c
@@ -356,6 +356,8 @@ txq_ctrl_setup(struct rte_eth_dev *dev, struct txq_ctrl *txq_ctrl,
 					      max_tso_inline);
 		tmpl.txq.tso_en = 1;
 	}
+	if (priv->tunnel_en)
+		tmpl.txq.tunnel_en = 1;
 	tmpl.qp = ibv_exp_create_qp(priv->ctx, &attr.init);
 	if (tmpl.qp == NULL) {
 		ret = (errno ? errno : EINVAL);
-- 
2.7.4