Primary 4.7 merge window changes

- Updates to the new Intel X722 iWARP driver
 - Updates to the hfi1 driver
 - Fixes for the iw_cxgb4 driver
 - Misc core fixes
 - Generic RDMA READ/WRITE API addition
 - SRP updates
 - Misc ipoib updates
 - Minor mlx5 updates
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQIcBAABAgAGBQJXPIAnAAoJELgmozMOVy/dEYUP/0A6NH5ptzUrwPuLrWoz8h+e
 KBfJE7H09mfKBx0Rq8YmnU+pz4lk8vrMLLaqpbGN57mwO0a1lK9bgc3E6KUhQPhc
 dpGEX/NG1+aILomD7M4l1yAKkG17kxFLD75cLCeaxhO76jBRWsunukqk5mT/u0EG
 fUYZs1fRb9t2LDtTNhPfSFR1+dgP5S17xLhpl9ttn87hTmIuiGWR6ig2nTC7azZD
 G0d7RVjohfY2sDD28YgQiUEJ+q+1ymp3XTaCZhPCVl9VCRPweEdtLKcbNWZIvClx
 ewuTCgADXg8tAL/6zu6bEKqZlC17UrmVJee3csKQLT09PJkSEICeFR/ld6ONUKAF
 nDhi3ySa75Xxg9VDcCnuRkXKK+/zi7oDelZuh9mvMG0JJqPK9rTZDD29j2kBf7C0
 bdx4R5cI4KJWQ/GlCyi/nLiuYkmAiCugzcGnRho4ub+EJ0yX1w6n8KVYr37kFsFu
 q6MCnEfArEgDpbq1wo0+9MWtqBYrnOI/XtG81Zd+6X2MW975qU85wUdUSjg6OOb1
 v1osyAmFDy9A0Y80yY+l1HHrSVIvI0IAWZDfxsbCLQY8O03ZNcvxE2RsrzWd5CKL
 iZsX24tjV0WR9+lORHLfAKB3DL9CcfHv/tHo7q+5iAHmIuWZGrEN22ELkwS/4X7x
 d/V0XDjzs6lgQeTJ7R4B
 =e+Zv
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma

Pull rdma updates from Doug Ledford:
 "Primary 4.7 merge window changes

   - Updates to the new Intel X722 iWARP driver
   - Updates to the hfi1 driver
   - Fixes for the iw_cxgb4 driver
   - Misc core fixes
   - Generic RDMA READ/WRITE API addition
   - SRP updates
   - Misc ipoib updates
   - Minor mlx5 updates"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (148 commits)
  IB/mlx5: Fire the CQ completion handler from tasklet
  net/mlx5_core: Use tasklet for user-space CQ completion events
  IB/core: Do not require CAP_NET_ADMIN for packet sniffing
  IB/mlx4: Fix unaligned access in send_reply_to_slave
  IB/mlx5: Report Scatter FCS device capability when supported
  IB/mlx5: Add Scatter FCS support for Raw Packet QP
  IB/core: Add Scatter FCS create flag
  IB/core: Add Raw Scatter FCS device capability
  IB/core: Add extended device capability flags
  i40iw: pass hw_stats by reference rather than by value
  i40iw: Remove unnecessary synchronize_irq() before free_irq()
  i40iw: constify i40iw_vf_cqp_ops structure
  IB/mlx5: Add UARs write-combining and non-cached mapping
  IB/mlx5: Allow mapping the free running counter on PROT_EXEC
  IB/mlx4: Use list_for_each_entry_safe
  IB/SA: Use correct free function
  IB/core: Fix a potential array overrun in CMA and SA agent
  IB/core: Remove unnecessary check in ibnl_rcv_msg
  IB/IWPM: Fix a potential skb leak
  RDMA/nes: replace custom print_hex_dump()
  ...

drivers/net/ethernet/chelsio/cxgb4/t4_msg.h

@@ -1392,6 +1392,10 @@ struct ulp_mem_io {
 #define T5_ULP_MEMIO_ORDER_V(x) ((x) << T5_ULP_MEMIO_ORDER_S)
 #define T5_ULP_MEMIO_ORDER_F    T5_ULP_MEMIO_ORDER_V(1U)
 
+#define T5_ULP_MEMIO_FID_S	4
+#define T5_ULP_MEMIO_FID_M	0x7ff
+#define T5_ULP_MEMIO_FID_V(x)	((x) << T5_ULP_MEMIO_FID_S)
+
 /* ulp_mem_io.lock_addr fields */
 #define ULP_MEMIO_ADDR_S    0
 #define ULP_MEMIO_ADDR_V(x) ((x) << ULP_MEMIO_ADDR_S)

drivers/net/ethernet/mellanox/mlx5/core/cq.c

@@ -39,6 +39,53 @@
 #include <linux/mlx5/cq.h>
 #include "mlx5_core.h"
 
+#define TASKLET_MAX_TIME 2
+#define TASKLET_MAX_TIME_JIFFIES msecs_to_jiffies(TASKLET_MAX_TIME)
+
+void mlx5_cq_tasklet_cb(unsigned long data)
+{
+	unsigned long flags;
+	unsigned long end = jiffies + TASKLET_MAX_TIME_JIFFIES;
+	struct mlx5_eq_tasklet *ctx = (struct mlx5_eq_tasklet *)data;
+	struct mlx5_core_cq *mcq;
+	struct mlx5_core_cq *temp;
+
+	spin_lock_irqsave(&ctx->lock, flags);
+	list_splice_tail_init(&ctx->list, &ctx->process_list);
+	spin_unlock_irqrestore(&ctx->lock, flags);
+
+	list_for_each_entry_safe(mcq, temp, &ctx->process_list,
+				 tasklet_ctx.list) {
+		list_del_init(&mcq->tasklet_ctx.list);
+		mcq->tasklet_ctx.comp(mcq);
+		if (atomic_dec_and_test(&mcq->refcount))
+			complete(&mcq->free);
+		if (time_after(jiffies, end))
+			break;
+	}
+
+	if (!list_empty(&ctx->process_list))
+		tasklet_schedule(&ctx->task);
+}
+
+static void mlx5_add_cq_to_tasklet(struct mlx5_core_cq *cq)
+{
+	unsigned long flags;
+	struct mlx5_eq_tasklet *tasklet_ctx = cq->tasklet_ctx.priv;
+
+	spin_lock_irqsave(&tasklet_ctx->lock, flags);
+	/* When migrating CQs between EQs will be implemented, please note
+	 * that you need to sync this point. It is possible that
+	 * while migrating a CQ, completions on the old EQs could
+	 * still arrive.
+	 */
+	if (list_empty_careful(&cq->tasklet_ctx.list)) {
+		atomic_inc(&cq->refcount);
+		list_add_tail(&cq->tasklet_ctx.list, &tasklet_ctx->list);
+	}
+	spin_unlock_irqrestore(&tasklet_ctx->lock, flags);
+}
+
 void mlx5_cq_completion(struct mlx5_core_dev *dev, u32 cqn)
 {
 	struct mlx5_core_cq *cq;
@@ -96,6 +143,13 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
 	struct mlx5_create_cq_mbox_out out;
 	struct mlx5_destroy_cq_mbox_in din;
 	struct mlx5_destroy_cq_mbox_out dout;
+	int eqn = MLX5_GET(cqc, MLX5_ADDR_OF(create_cq_in, in, cq_context),
+			   c_eqn);
+	struct mlx5_eq *eq;
+
+	eq = mlx5_eqn2eq(dev, eqn);
+	if (IS_ERR(eq))
+		return PTR_ERR(eq);
 
 	in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_CREATE_CQ);
 	memset(&out, 0, sizeof(out));
@@ -111,6 +165,11 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
 	cq->arm_sn     = 0;
 	atomic_set(&cq->refcount, 1);
 	init_completion(&cq->free);
+	if (!cq->comp)
+		cq->comp = mlx5_add_cq_to_tasklet;
+	/* assuming CQ will be deleted before the EQ */
+	cq->tasklet_ctx.priv = &eq->tasklet_ctx;
+	INIT_LIST_HEAD(&cq->tasklet_ctx.list);
 
 	spin_lock_irq(&table->lock);
 	err = radix_tree_insert(&table->tree, cq->cqn, cq);

drivers/net/ethernet/mellanox/mlx5/core/eq.c

@@ -202,7 +202,7 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 	struct mlx5_eqe *eqe;
 	int eqes_found = 0;
 	int set_ci = 0;
-	u32 cqn;
+	u32 cqn = -1;
 	u32 rsn;
 	u8 port;
 
@@ -320,6 +320,9 @@ static int mlx5_eq_int(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 
 	eq_update_ci(eq, 1);
 
+	if (cqn != -1)
+		tasklet_schedule(&eq->tasklet_ctx.task);
+
 	return eqes_found;
 }
 
@@ -403,6 +406,12 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx,
 	if (err)
 		goto err_irq;
 
+	INIT_LIST_HEAD(&eq->tasklet_ctx.list);
+	INIT_LIST_HEAD(&eq->tasklet_ctx.process_list);
+	spin_lock_init(&eq->tasklet_ctx.lock);
+	tasklet_init(&eq->tasklet_ctx.task, mlx5_cq_tasklet_cb,
+		     (unsigned long)&eq->tasklet_ctx);
+
 	/* EQs are created in ARMED state
 	 */
 	eq_update_ci(eq, 1);
@@ -436,6 +445,7 @@ int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 		mlx5_core_warn(dev, "failed to destroy a previously created eq: eqn %d\n",
 			       eq->eqn);
 	synchronize_irq(eq->irqn);
+	tasklet_disable(&eq->tasklet_ctx.task);
 	mlx5_buf_free(dev, &eq->buf);
 
 	return err;

drivers/net/ethernet/mellanox/mlx5/core/main.c

@@ -663,6 +663,23 @@ int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn,
 }
 EXPORT_SYMBOL(mlx5_vector2eqn);
 
+struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn)
+{
+	struct mlx5_eq_table *table = &dev->priv.eq_table;
+	struct mlx5_eq *eq;
+
+	spin_lock(&table->lock);
+	list_for_each_entry(eq, &table->comp_eqs_list, list)
+		if (eq->eqn == eqn) {
+			spin_unlock(&table->lock);
+			return eq;
+		}
+
+	spin_unlock(&table->lock);
+
+	return ERR_PTR(-ENOENT);
+}
+
 static void free_comp_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *table = &dev->priv.eq_table;

drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h

@@ -102,6 +102,8 @@ int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id);
 int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
 cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev);
 u32 mlx5_get_msix_vec(struct mlx5_core_dev *dev, int vecidx);
+struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn);
+void mlx5_cq_tasklet_cb(unsigned long data);
 
 void mlx5e_init(void);
 void mlx5e_cleanup(void);