IB/mlx4: Add IPoIB checksum offload support

ConnectX devices support checksum generation and verification of TCP
and UDP packets for UD IPoIB messages.  This patch checks if the HCA
supports this and sets the IB_DEVICE_UD_IP_CSUM capability flag if it
does.  It implements support for handling the IB_SEND_IP_CSUM send
flag and setting the csum_ok field in receive work completions.

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Ali Ayub <ali@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

4 files changed, 25 insertions, 0 deletions

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 7360bbafbe84..d2e32b03e2f7 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -297,6 +297,20 @@ static void mlx4_ib_handle_error_cqe(struct mlx4_err_cqe *cqe,
         wc->vendor_err = cqe->vendor_err_syndrome;
 }
 
+static int mlx4_ib_ipoib_csum_ok(__be32 status, __be16 checksum)
+{
+        return ((status & cpu_to_be32(MLX4_CQE_IPOIB_STATUS_IPV4        |
+                                      MLX4_CQE_IPOIB_STATUS_IPV4F       |
+                                      MLX4_CQE_IPOIB_STATUS_IPV4OPT     |
+                                      MLX4_CQE_IPOIB_STATUS_IPV6        |
+                                      MLX4_CQE_IPOIB_STATUS_IPOK)) ==
+                cpu_to_be32(MLX4_CQE_IPOIB_STATUS_IPV4  |
+                            MLX4_CQE_IPOIB_STATUS_IPOK))                &&
+                (status & cpu_to_be32(MLX4_CQE_IPOIB_STATUS_UDP |
+                                      MLX4_CQE_IPOIB_STATUS_TCP))       &&
+                checksum == cpu_to_be16(0xffff);
+}
+
 static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
                             struct mlx4_ib_qp **cur_qp,
                             struct ib_wc *wc)
@@ -434,6 +448,8 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq,
                 wc->dlid_path_bits = (g_mlpath_rqpn >> 24) & 0x7f;
                 wc->wc_flags      |= g_mlpath_rqpn & 0x80000000 ? IB_WC_GRH : 0;
                 wc->pkey_index     = be32_to_cpu(cqe->immed_rss_invalid) & 0x7f;
+                wc->csum_ok        = mlx4_ib_ipoib_csum_ok(cqe->ipoib_status,
+                                                           cqe->checksum);
         }
 
         return 0;
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index d5512011999c..6ea4746c2e9b 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -99,6 +99,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
                 props->device_cap_flags |= IB_DEVICE_AUTO_PATH_MIG;
         if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_UD_AV_PORT)
                 props->device_cap_flags |= IB_DEVICE_UD_AV_PORT_ENFORCE;
+        if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
+                props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM;
 
         props->vendor_id           = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
                 0xffffff;
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index ac965ab28845..31b2b5b230bd 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1436,6 +1436,9 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
                          cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE) : 0) |
                         (wr->send_flags & IB_SEND_SOLICITED ?
                          cpu_to_be32(MLX4_WQE_CTRL_SOLICITED) : 0) |
+                        ((wr->send_flags & IB_SEND_IP_CSUM) ?
+                         cpu_to_be32(MLX4_WQE_CTRL_IP_CSUM |
+                                     MLX4_WQE_CTRL_TCP_UDP_CSUM) : 0) |
                         qp->sq_signal_bits;
 
                 if (wr->opcode == IB_WR_SEND_WITH_IMM ||
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 61dc4951d6b0..f494c3e8bce3 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -696,6 +696,10 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
         /* Check port for UD address vector: */
         *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1);
 
+        /* Enable IPoIB checksumming if we can: */
+        if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
+                *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 3);
+
         /* QPC/EEC/CQC/EQC/RDMARC attributes */
 
         MLX4_PUT(inbox, param->qpc_base,      INIT_HCA_QPC_BASE_OFFSET);