net/mlx4: restore Tx checksum offloads

This patch adds hardware offloading support for IPv4, UDP and TCP checksum
calculation, including inner/outer checksums on supported tunnel types.

Signed-off-by: Moti Haimovsky <motih@mellanox.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>

diff --git a/doc/guides/nics/features/mlx4.ini b/doc/guides/nics/features/mlx4.ini
index 19ae688..366e051 100644 (file)
--- a/doc/guides/nics/features/mlx4.ini
+++ b/doc/guides/nics/features/mlx4.ini
@@ -20,6 +20,10 @@ Multicast MAC filter = Y
 RSS hash             = Y
 SR-IOV               = Y
 VLAN filter          = Y
+L3 checksum offload  = Y
+L4 checksum offload  = Y
+Inner L3 checksum    = Y
+Inner L4 checksum    = Y
 Basic stats          = Y
 Stats per queue      = Y
 Other kdrv           = Y

diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 0db9a19..a297b9a 100644 (file)
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/mlx4.c
@@ -566,6 +566,17 @@ mlx4_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
                priv->pd = pd;
                priv->mtu = ETHER_MTU;
                priv->vf = vf;
+               priv->hw_csum = !!(device_attr.device_cap_flags &
+                                  IBV_DEVICE_RAW_IP_CSUM);
+               DEBUG("checksum offloading is %ssupported",
+                     (priv->hw_csum ? "" : "not "));
+               /* Only ConnectX-3 Pro supports tunneling. */
+               priv->hw_csum_l2tun =
+                       priv->hw_csum &&
+                       (device_attr.vendor_part_id ==
+                        PCI_DEVICE_ID_MELLANOX_CONNECTX3PRO);
+               DEBUG("L2 tunnel checksum offloads are %ssupported",
+                     (priv->hw_csum_l2tun ? "" : "not "));
                /* Configure the first MAC address by default. */
                if (mlx4_get_mac(priv, &mac.addr_bytes)) {
                        ERROR("cannot get MAC address, is mlx4_en loaded?"

diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index f4da8c6..e0a9853 100644 (file)
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -113,6 +113,8 @@ struct priv {
        uint32_t vf:1; /**< This is a VF device. */
        uint32_t intr_alarm:1; /**< An interrupt alarm is scheduled. */
        uint32_t isolated:1; /**< Toggle isolated mode. */
+       uint32_t hw_csum:1; /* Checksum offload is supported. */
+       uint32_t hw_csum_l2tun:1; /* Checksum support for L2 tunnels. */
        struct rte_intr_handle intr_handle; /**< Port interrupt handle. */
        struct mlx4_drop *drop; /**< Shared resources for drop flow rules. */
        LIST_HEAD(, mlx4_rss) rss; /**< Shared targets for Rx flow rules. */

diff --git a/drivers/net/mlx4/mlx4_ethdev.c b/drivers/net/mlx4/mlx4_ethdev.c
index 3623909..a8c0ee2 100644 (file)
--- a/drivers/net/mlx4/mlx4_ethdev.c
+++ b/drivers/net/mlx4/mlx4_ethdev.c
@@ -767,6 +767,12 @@ mlx4_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
        info->max_mac_addrs = RTE_DIM(priv->mac);
        info->rx_offload_capa = 0;
        info->tx_offload_capa = 0;
+       if (priv->hw_csum)
+               info->tx_offload_capa |= (DEV_TX_OFFLOAD_IPV4_CKSUM |
+                                         DEV_TX_OFFLOAD_UDP_CKSUM |
+                                         DEV_TX_OFFLOAD_TCP_CKSUM);
+       if (priv->hw_csum_l2tun)
+               info->tx_offload_capa |= DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM;
        if (mlx4_get_ifname(priv, &ifname) == 0)
                info->if_index = if_nametoindex(ifname);
        info->hash_key_size = MLX4_RSS_HASH_KEY_SIZE;

diff --git a/drivers/net/mlx4/mlx4_prm.h b/drivers/net/mlx4/mlx4_prm.h
index 085a595..df5a6b4 100644 (file)
--- a/drivers/net/mlx4/mlx4_prm.h
+++ b/drivers/net/mlx4/mlx4_prm.h
@@ -64,6 +64,8 @@
 
 /* Work queue element (WQE) flags. */
 #define MLX4_BIT_WQE_OWN 0x80000000
+#define MLX4_WQE_CTRL_IIP_HDR_CSUM (1 << 28)
+#define MLX4_WQE_CTRL_IL4_HDR_CSUM (1 << 27)
 
 #define MLX4_SIZE_TO_TXBBS(size) \
        (RTE_ALIGN((size), (MLX4_TXBB_SIZE)) >> (MLX4_TXBB_SHIFT))

diff --git a/drivers/net/mlx4/mlx4_rxtx.c b/drivers/net/mlx4/mlx4_rxtx.c
index cc0baaa..fe7d5d0 100644 (file)
--- a/drivers/net/mlx4/mlx4_rxtx.c
+++ b/drivers/net/mlx4/mlx4_rxtx.c
@@ -431,6 +431,25 @@ mlx4_post_send(struct txq *txq, struct rte_mbuf *pkt)
        } else {
                srcrb_flags = RTE_BE32(MLX4_WQE_CTRL_SOLICIT);
        }
+       /* Enable HW checksum offload if requested */
+       if (txq->csum &&
+           (pkt->ol_flags &
+            (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM))) {
+               const uint64_t is_tunneled = (pkt->ol_flags &
+                                             (PKT_TX_TUNNEL_GRE |
+                                              PKT_TX_TUNNEL_VXLAN));
+
+               if (is_tunneled && txq->csum_l2tun) {
+                       owner_opcode |= MLX4_WQE_CTRL_IIP_HDR_CSUM |
+                                       MLX4_WQE_CTRL_IL4_HDR_CSUM;
+                       if (pkt->ol_flags & PKT_TX_OUTER_IP_CKSUM)
+                               srcrb_flags |=
+                                       RTE_BE32(MLX4_WQE_CTRL_IP_HDR_CSUM);
+               } else {
+                       srcrb_flags |= RTE_BE32(MLX4_WQE_CTRL_IP_HDR_CSUM |
+                                               MLX4_WQE_CTRL_TCP_UDP_CSUM);
+               }
+       }
        ctrl->srcrb_flags = srcrb_flags;
        /*
         * Make sure descriptor is fully written before

diff --git a/drivers/net/mlx4/mlx4_rxtx.h b/drivers/net/mlx4/mlx4_rxtx.h
index fa5738f..6c88efb 100644 (file)
--- a/drivers/net/mlx4/mlx4_rxtx.h
+++ b/drivers/net/mlx4/mlx4_rxtx.h
@@ -124,6 +124,8 @@ struct txq {
        struct txq_elt (*elts)[]; /**< Tx elements. */
        struct mlx4_txq_stats stats; /**< Tx queue counters. */
        uint32_t max_inline; /**< Max inline send size. */
+       uint32_t csum:1; /**< Enable checksum offloading. */
+       uint32_t csum_l2tun:1; /**< Same for L2 tunnels. */
        uint8_t *bounce_buf;
        /**< Memory used for storing the first DWORD of data TXBBs. */
        struct {

diff --git a/drivers/net/mlx4/mlx4_txq.c b/drivers/net/mlx4/mlx4_txq.c
index fbb028a..0e27df2 100644 (file)
--- a/drivers/net/mlx4/mlx4_txq.c
+++ b/drivers/net/mlx4/mlx4_txq.c
@@ -276,6 +276,8 @@ mlx4_tx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t desc,
                        RTE_MIN(MLX4_PMD_TX_PER_COMP_REQ, desc / 4),
                .elts_comp_cd_init =
                        RTE_MIN(MLX4_PMD_TX_PER_COMP_REQ, desc / 4),
+               .csum = priv->hw_csum,
+               .csum_l2tun = priv->hw_csum_l2tun,
                .bounce_buf = bounce_buf,
        };
        txq->cq = ibv_create_cq(priv->ctx, desc, NULL, NULL, 0);