Merge tag 'mlx5e-updates-2018-12-10' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

Saeed Mahameed:

====================
mlx5e-updates-2018-12-10 (gre)

This patch set adds GRE offloading support to Mellanox ethernet driver.

Patches 1-5 replace the existing egdev mechanism with the new TC indirect
block binds mechanism that was introduced by Netronome:
7f76fa36754b ("net: sched: register callbacks for indirect tc block binds")

Patches 6-9 add GRE offloading support along with some required
refactoring work.

Patch 10, Add netif_is_gretap()/netif_is_ip6gretap()
 - Changed the is_gretap_dev and is_ip6gretap_dev logic from structure
   comparison to string comparison of the rtnl_link_ops kind field.

Patch 11, add GRE offloading support to mlx5.

Patch 12 removes the egdev mechanism from TC as it is no longer used by
any of the drivers.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

69 files changed, 3563 insertions, 2924 deletions

diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index 676c1fd1119d..9608681224e6 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -647,8 +647,13 @@ int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 user_virt,
 				flags, local_page_list, NULL, NULL);
 		up_read(&owning_mm->mmap_sem);
 
-		if (npages < 0)
+		if (npages < 0) {
+			if (npages != -EAGAIN)
+				pr_warn("fail to get %zu user pages with error %d\n", gup_num_pages, npages);
+			else
+				pr_debug("fail to get %zu user pages with error %d\n", gup_num_pages, npages);
 			break;
+		}
 
 		bcnt -= min_t(size_t, npages << PAGE_SHIFT, bcnt);
 		mutex_lock(&umem_odp->umem_mutex);
@@ -666,8 +671,13 @@ int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 user_virt,
 			ret = ib_umem_odp_map_dma_single_page(
 					umem_odp, k, local_page_list[j],
 					access_mask, current_seq);
-			if (ret < 0)
+			if (ret < 0) {
+				if (ret != -EAGAIN)
+					pr_warn("ib_umem_odp_map_dma_single_page failed with error %d\n", ret);
+				else
+					pr_debug("ib_umem_odp_map_dma_single_page failed with error %d\n", ret);
 				break;
+			}
 
 			p = page_to_phys(local_page_list[j]);
 			k++;
diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index b8e4b15e2674..33f5adb14e4e 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -1,6 +1,8 @@
 obj-$(CONFIG_MLX5_INFINIBAND)	+= mlx5_ib.o
 
-mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o gsi.o ib_virt.o cmd.o cong.o
+mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq_cmd.o \
+		srq.o mr.o ah.o mad.o gsi.o ib_virt.o cmd.o \
+		cong.o
 mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o
 mlx5_ib-$(CONFIG_MLX5_ESWITCH) += ib_rep.o
 mlx5_ib-$(CONFIG_INFINIBAND_USER_ACCESS) += devx.o
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 7d769b5538b4..26ab9041f94a 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -35,6 +35,7 @@
 #include <rdma/ib_user_verbs.h>
 #include <rdma/ib_cache.h>
 #include "mlx5_ib.h"
+#include "srq.h"
 
 static void mlx5_ib_cq_comp(struct mlx5_core_cq *cq)
 {
@@ -81,7 +82,7 @@ static void *get_sw_cqe(struct mlx5_ib_cq *cq, int n)
 
 	cqe64 = (cq->mcq.cqe_sz == 64) ? cqe : cqe + 64;
 
-	if (likely((cqe64->op_own) >> 4 != MLX5_CQE_INVALID) &&
+	if (likely(get_cqe_opcode(cqe64) != MLX5_CQE_INVALID) &&
 	    !((cqe64->op_own & MLX5_CQE_OWNER_MASK) ^ !!(n & (cq->ibcq.cqe + 1)))) {
 		return cqe;
 	} else {
@@ -177,8 +178,7 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe,
 		struct mlx5_core_srq *msrq = NULL;
 
 		if (qp->ibqp.xrcd) {
-			msrq = mlx5_core_get_srq(dev->mdev,
-						 be32_to_cpu(cqe->srqn));
+			msrq = mlx5_cmd_get_srq(dev, be32_to_cpu(cqe->srqn));
 			srq = to_mibsrq(msrq);
 		} else {
 			srq = to_msrq(qp->ibqp.srq);
@@ -197,7 +197,7 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe,
 	}
 	wc->byte_len = be32_to_cpu(cqe->byte_cnt);
 
-	switch (cqe->op_own >> 4) {
+	switch (get_cqe_opcode(cqe)) {
 	case MLX5_CQE_RESP_WR_IMM:
 		wc->opcode	= IB_WC_RECV_RDMA_WITH_IMM;
 		wc->wc_flags	= IB_WC_WITH_IMM;
@@ -537,7 +537,7 @@ repoll:
 	 */
 	rmb();
 
-	opcode = cqe64->op_own >> 4;
+	opcode = get_cqe_opcode(cqe64);
 	if (unlikely(opcode == MLX5_CQE_RESIZE_CQ)) {
 		if (likely(cq->resize_buf)) {
 			free_cq_buf(dev, &cq->buf);
@@ -1295,7 +1295,7 @@ static int copy_resize_cqes(struct mlx5_ib_cq *cq)
 		return -EINVAL;
 	}
 
-	while ((scqe64->op_own >> 4) != MLX5_CQE_RESIZE_CQ) {
+	while (get_cqe_opcode(scqe64) != MLX5_CQE_RESIZE_CQ) {
 		dcqe = mlx5_frag_buf_get_wqe(&cq->resize_buf->fbc,
 					     (i + 1) & cq->resize_buf->nent);
 		dcqe64 = dsize == 64 ? dcqe : dcqe + 64;
diff --git a/drivers/infiniband/hw/mlx5/ib_rep.c b/drivers/infiniband/hw/mlx5/ib_rep.c
index 584ff2ea7810..8a682d86d634 100644
--- a/drivers/infiniband/hw/mlx5/ib_rep.c
+++ b/drivers/infiniband/hw/mlx5/ib_rep.c
@@ -4,6 +4,7 @@
  */
 
 #include "ib_rep.h"
+#include "srq.h"
 
 static const struct mlx5_ib_profile rep_profile = {
 	STAGE_CREATE(MLX5_IB_STAGE_INIT,
@@ -21,6 +22,9 @@ static const struct mlx5_ib_profile rep_profile = {
 	STAGE_CREATE(MLX5_IB_STAGE_ROCE,
 		     mlx5_ib_stage_rep_roce_init,
 		     mlx5_ib_stage_rep_roce_cleanup),
+	STAGE_CREATE(MLX5_IB_STAGE_SRQ,
+		     mlx5_init_srq_table,
+		     mlx5_cleanup_srq_table),
 	STAGE_CREATE(MLX5_IB_STAGE_DEVICE_RESOURCES,
 		     mlx5_ib_stage_dev_res_init,
 		     mlx5_ib_stage_dev_res_cleanup),
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 3569fda07e07..0eeefff09c1e 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -60,6 +60,7 @@
 #include "mlx5_ib.h"
 #include "ib_rep.h"
 #include "cmd.h"
+#include "srq.h"
 #include <linux/mlx5/fs_helpers.h>
 #include <linux/mlx5/accel.h>
 #include <rdma/uverbs_std_types.h>
@@ -82,10 +83,13 @@ static char mlx5_version[] =
 
 struct mlx5_ib_event_work {
 	struct work_struct	work;
-	struct mlx5_core_dev	*dev;
-	void			*context;
-	enum mlx5_dev_event	event;
-	unsigned long		param;
+	union {
+		struct mlx5_ib_dev	      *dev;
+		struct mlx5_ib_multiport_info *mpi;
+	};
+	bool			is_slave;
+	unsigned int		event;
+	void			*param;
 };
 
 enum {
@@ -2669,11 +2673,11 @@ static int parse_flow_attr(struct mlx5_core_dev *mdev, u32 *match_c,
 			 ntohs(ib_spec->gre.val.protocol));
 
 		memcpy(MLX5_ADDR_OF(fte_match_set_misc, misc_params_c,
-				    gre_key_h),
+				    gre_key.nvgre.hi),
 		       &ib_spec->gre.mask.key,
 		       sizeof(ib_spec->gre.mask.key));
 		memcpy(MLX5_ADDR_OF(fte_match_set_misc, misc_params_v,
-				    gre_key_h),
+				    gre_key.nvgre.hi),
 		       &ib_spec->gre.val.key,
 		       sizeof(ib_spec->gre.val.key));
 		break;
@@ -4226,6 +4230,63 @@ static void delay_drop_handler(struct work_struct *work)
 	mutex_unlock(&delay_drop->lock);
 }
 
+static void handle_general_event(struct mlx5_ib_dev *ibdev, struct mlx5_eqe *eqe,
+				 struct ib_event *ibev)
+{
+	switch (eqe->sub_type) {
+	case MLX5_GENERAL_SUBTYPE_DELAY_DROP_TIMEOUT:
+		schedule_work(&ibdev->delay_drop.delay_drop_work);
+		break;
+	default: /* do nothing */
+		return;
+	}
+}
+
+static int handle_port_change(struct mlx5_ib_dev *ibdev, struct mlx5_eqe *eqe,
+			      struct ib_event *ibev)
+{
+	u8 port = (eqe->data.port.port >> 4) & 0xf;
+
+	ibev->element.port_num = port;
+
+	switch (eqe->sub_type) {
+	case MLX5_PORT_CHANGE_SUBTYPE_ACTIVE:
+	case MLX5_PORT_CHANGE_SUBTYPE_DOWN:
+	case MLX5_PORT_CHANGE_SUBTYPE_INITIALIZED:
+		/* In RoCE, port up/down events are handled in
+		 * mlx5_netdev_event().
+		 */
+		if (mlx5_ib_port_link_layer(&ibdev->ib_dev, port) ==
+					    IB_LINK_LAYER_ETHERNET)
+			return -EINVAL;
+
+		ibev->event = (eqe->sub_type == MLX5_PORT_CHANGE_SUBTYPE_ACTIVE) ?
+				IB_EVENT_PORT_ACTIVE : IB_EVENT_PORT_ERR;
+		break;
+
+	case MLX5_PORT_CHANGE_SUBTYPE_LID:
+		ibev->event = IB_EVENT_LID_CHANGE;
+		break;
+
+	case MLX5_PORT_CHANGE_SUBTYPE_PKEY:
+		ibev->event = IB_EVENT_PKEY_CHANGE;
+		schedule_work(&ibdev->devr.ports[port - 1].pkey_change_work);
+		break;
+
+	case MLX5_PORT_CHANGE_SUBTYPE_GUID:
+		ibev->event = IB_EVENT_GID_CHANGE;
+		break;
+
+	case MLX5_PORT_CHANGE_SUBTYPE_CLIENT_REREG:
+		ibev->event = IB_EVENT_CLIENT_REREGISTER;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static void mlx5_ib_handle_event(struct work_struct *_work)
 {
 	struct mlx5_ib_event_work *work =
@@ -4233,65 +4294,37 @@ static void mlx5_ib_handle_event(struct work_struct *_work)
 	struct mlx5_ib_dev *ibdev;
 	struct ib_event ibev;
 	bool fatal = false;
-	u8 port = (u8)work->param;
 
-	if (mlx5_core_is_mp_slave(work->dev)) {
-		ibdev = mlx5_ib_get_ibdev_from_mpi(work->context);
+	if (work->is_slave) {
+		ibdev = mlx5_ib_get_ibdev_from_mpi(work->mpi);
 		if (!ibdev)
 			goto out;
 	} else {
-		ibdev = work->context;
+		ibdev = work->dev;
 	}
 
 	switch (work->event) {
 	case MLX5_DEV_EVENT_SYS_ERROR:
 		ibev.event = IB_EVENT_DEVICE_FATAL;
 		mlx5_ib_handle_internal_error(ibdev);
+		ibev.element.port_num  = (u8)(unsigned long)work->param;
 		fatal = true;
 		break;
-
-	case MLX5_DEV_EVENT_PORT_UP:
-	case MLX5_DEV_EVENT_PORT_DOWN:
-	case MLX5_DEV_EVENT_PORT_INITIALIZED:
-		/* In RoCE, port up/down events are handled in
-		 * mlx5_netdev_event().
-		 */
-		if (mlx5_ib_port_link_layer(&ibdev->ib_dev, port) ==
-			IB_LINK_LAYER_ETHERNET)
+	case MLX5_EVENT_TYPE_PORT_CHANGE:
+		if (handle_port_change(ibdev, work->param, &ibev))
 			goto out;
-
-		ibev.event = (work->event == MLX5_DEV_EVENT_PORT_UP) ?
-			     IB_EVENT_PORT_ACTIVE : IB_EVENT_PORT_ERR;
 		break;
-
-	case MLX5_DEV_EVENT_LID_CHANGE:
-		ibev.event = IB_EVENT_LID_CHANGE;
-		break;
-
-	case MLX5_DEV_EVENT_PKEY_CHANGE:
-		ibev.event = IB_EVENT_PKEY_CHANGE;
-		schedule_work(&ibdev->devr.ports[port - 1].pkey_change_work);
-		break;
-
-	case MLX5_DEV_EVENT_GUID_CHANGE:
-		ibev.event = IB_EVENT_GID_CHANGE;
-		break;
-
-	case MLX5_DEV_EVENT_CLIENT_REREG:
-		ibev.event = IB_EVENT_CLIENT_REREGISTER;
-		break;
-	case MLX5_DEV_EVENT_DELAY_DROP_TIMEOUT:
-		schedule_work(&ibdev->delay_drop.delay_drop_work);
-		goto out;
+	case MLX5_EVENT_TYPE_GENERAL_EVENT:
+		handle_general_event(ibdev, work->param, &ibev);
+		/* fall through */
 	default:
 		goto out;
 	}
 
-	ibev.device	      = &ibdev->ib_dev;
-	ibev.element.port_num = port;
+	ibev.device = &ibdev->ib_dev;
 
-	if (!rdma_is_port_valid(&ibdev->ib_dev, port)) {
-		mlx5_ib_warn(ibdev, "warning: event on port %d\n", port);
+	if (!rdma_is_port_valid(&ibdev->ib_dev, ibev.element.port_num)) {
+		mlx5_ib_warn(ibdev, "warning: event on port %d\n",  ibev.element.port_num);
 		goto out;
 	}
 
@@ -4304,22 +4337,43 @@ out:
 	kfree(work);
 }
 
-static void mlx5_ib_event(struct mlx5_core_dev *dev, void *context,
-			  enum mlx5_dev_event event, unsigned long param)
+static int mlx5_ib_event(struct notifier_block *nb,
+			 unsigned long event, void *param)
 {
 	struct mlx5_ib_event_work *work;
 
 	work = kmalloc(sizeof(*work), GFP_ATOMIC);
 	if (!work)
-		return;
+		return NOTIFY_DONE;
 
 	INIT_WORK(&work->work, mlx5_ib_handle_event);
-	work->dev = dev;
+	work->dev = container_of(nb, struct mlx5_ib_dev, mdev_events);
+	work->is_slave = false;
 	work->param = param;
-	work->context = context;
 	work->event = event;
 
 	queue_work(mlx5_ib_event_wq, &work->work);
+
+	return NOTIFY_OK;
+}
+
+static int mlx5_ib_event_slave_port(struct notifier_block *nb,
+				    unsigned long event, void *param)
+{
+	struct mlx5_ib_event_work *work;
+
+	work = kmalloc(sizeof(*work), GFP_ATOMIC);
+	if (!work)
+		return NOTIFY_DONE;
+
+	INIT_WORK(&work->work, mlx5_ib_handle_event);
+	work->mpi = container_of(nb, struct mlx5_ib_multiport_info, mdev_events);
+	work->is_slave = true;
+	work->param = param;
+	work->event = event;
+	queue_work(mlx5_ib_event_wq, &work->work);
+
+	return NOTIFY_OK;
 }
 
 static int set_has_smi_cap(struct mlx5_ib_dev *dev)
@@ -5330,7 +5384,7 @@ mlx5_ib_get_vector_affinity(struct ib_device *ibdev, int comp_vector)
 {
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 
-	return mlx5_get_vector_affinity_hint(dev->mdev, comp_vector);
+	return mlx5_comp_irq_get_affinity_mask(dev->mdev, comp_vector);
 }
 
 /* The mlx5_ib_multiport_mutex should be held when calling this function */
@@ -5350,6 +5404,11 @@ static void mlx5_ib_unbind_slave_port(struct mlx5_ib_dev *ibdev,
 		spin_unlock(&port->mp.mpi_lock);
 		return;
 	}
+
+	if (mpi->mdev_events.notifier_call)
+		mlx5_notifier_unregister(mpi->mdev, &mpi->mdev_events);
+	mpi->mdev_events.notifier_call = NULL;
+
 	mpi->ibdev = NULL;
 
 	spin_unlock(&port->mp.mpi_lock);
@@ -5405,6 +5464,7 @@ static bool mlx5_ib_bind_slave_port(struct mlx5_ib_dev *ibdev,
 
 	ibdev->port[port_num].mp.mpi = mpi;
 	mpi->ibdev = ibdev;
+	mpi->mdev_events.notifier_call = NULL;
 	spin_unlock(&ibdev->port[port_num].mp.mpi_lock);
 
 	err = mlx5_nic_vport_affiliate_multiport(ibdev->mdev, mpi->mdev);
@@ -5422,6 +5482,9 @@ static bool mlx5_ib_bind_slave_port(struct mlx5_ib_dev *ibdev,
 		goto unbind;
 	}
 
+	mpi->mdev_events.notifier_call = mlx5_ib_event_slave_port;
+	mlx5_notifier_register(mpi->mdev, &mpi->mdev_events);
+
 	err = mlx5_ib_init_cong_debugfs(ibdev, port_num);
 	if (err)
 		goto unbind;
@@ -5694,8 +5757,7 @@ int mlx5_ib_stage_init_init(struct mlx5_ib_dev *dev)
 	dev->ib_dev.node_type		= RDMA_NODE_IB_CA;
 	dev->ib_dev.local_dma_lkey	= 0 /* not supported for now */;
 	dev->ib_dev.phys_port_cnt	= dev->num_ports;
-	dev->ib_dev.num_comp_vectors    =
-		dev->mdev->priv.eq_table.num_comp_vectors;
+	dev->ib_dev.num_comp_vectors    = mlx5_comp_vectors_count(mdev);
 	dev->ib_dev.dev.parent		= &mdev->pdev->dev;
 
 	mutex_init(&dev->cap_mask_mutex);
@@ -6034,6 +6096,11 @@ static int mlx5_ib_stage_odp_init(struct mlx5_ib_dev *dev)
 	return mlx5_ib_odp_init_one(dev);
 }
 
+void mlx5_ib_stage_odp_cleanup(struct mlx5_ib_dev *dev)
+{
+	mlx5_ib_odp_cleanup_one(dev);
+}
+
 int mlx5_ib_stage_counters_init(struct mlx5_ib_dev *dev)
 {
 	if (MLX5_CAP_GEN(dev->mdev, max_qp_cnt)) {
@@ -6152,6 +6219,34 @@ static void mlx5_ib_stage_rep_reg_cleanup(struct mlx5_ib_dev *dev)
 	mlx5_ib_unregister_vport_reps(dev);
 }
 
+static int mlx5_ib_stage_dev_notifier_init(struct mlx5_ib_dev *dev)
+{
+	dev->mdev_events.notifier_call = mlx5_ib_event;
+	mlx5_notifier_register(dev->mdev, &dev->mdev_events);
+	return 0;
+}
+
+static void mlx5_ib_stage_dev_notifier_cleanup(struct mlx5_ib_dev *dev)
+{
+	mlx5_notifier_unregister(dev->mdev, &dev->mdev_events);
+}
+
+static int mlx5_ib_stage_devx_init(struct mlx5_ib_dev *dev)
+{
+	int uid;
+
+	uid = mlx5_ib_devx_create(dev);
+	if (uid > 0)
+		dev->devx_whitelist_uid = uid;
+
+	return 0;
+}
+static void mlx5_ib_stage_devx_cleanup(struct mlx5_ib_dev *dev)
+{
+	if (dev->devx_whitelist_uid)
+		mlx5_ib_devx_destroy(dev, dev->devx_whitelist_uid);
+}
+
 void __mlx5_ib_remove(struct mlx5_ib_dev *dev,
 		      const struct mlx5_ib_profile *profile,
 		      int stage)
@@ -6163,8 +6258,6 @@ void __mlx5_ib_remove(struct mlx5_ib_dev *dev,
 			profile->stage[stage].cleanup(dev);
 	}
 
-	if (dev->devx_whitelist_uid)
-		mlx5_ib_devx_destroy(dev, dev->devx_whitelist_uid);
 	ib_dealloc_device((struct ib_device *)dev);
 }
 
@@ -6173,7 +6266,6 @@ void *__mlx5_ib_add(struct mlx5_ib_dev *dev,
 {
 	int err;
 	int i;
-	int uid;
 
 	for (i = 0; i < MLX5_IB_STAGE_MAX; i++) {
 		if (profile->stage[i].init) {
@@ -6183,10 +6275,6 @@ void *__mlx