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

Pull rdma updates from Jason Gunthorpe:
 "Very small update this cycle:

   - Minor code improvements in fi, rxe, ipoib, mana, cxgb4, mlx5,
     irdma, rxe, rtrs, mana

   - Simplify the hns hem mechanism

   - Fix EFA's MSI-X allocation in resource constrained configurations

   - Fix a KASN splat in srpt

   - Narrow hns's congestion control selection to QPs granularity and
     allow userspace to select it

   - Solve a parallel module loading race between the CM module and a
     driver module

   - Flexible array cleanup

   - Dump hns's SCC Conext to 'rdma res' for debugging

   - Make mana build page lists for HW objects that require a 0 offset
     correctly

   - Stuck CM ID debugging"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (29 commits)
  RDMA/cm: add timeout to cm_destroy_id wait
  RDMA/mana_ib: Use virtual address in dma regions for MRs
  RDMA/mana_ib: Fix bug in creation of dma regions
  RDMA/hns: Append SCC context to the raw dump of QPC
  RDMA/uverbs: Avoid -Wflex-array-member-not-at-end warnings
  RDMA/hns: Support userspace configuring congestion control algorithm with QP granularity
  RDMA/rtrs-clt: Check strnlen return len in sysfs mpath_policy_store()
  RDMA/uverbs: Remove flexible arrays from struct *_filter
  RDMA/device: Fix a race between mad_client and cm_client init
  RDMA/hns: Fix mis-modifying default congestion control algorithm
  RDMA/rxe: Remove unused 'iova' parameter from rxe_mr_init_user
  RDMA/srpt: Do not register event handler until srpt device is fully setup
  RDMA/irdma: Remove duplicate assignment
  RDMA/efa: Limit EQs to available MSI-X vectors
  RDMA/mlx5: Delete unused mlx5_ib_copy_pas prototype
  RDMA/cxgb4: Delete unused c4iw_ep_redirect prototype
  RDMA/mana_ib: Introduce mana_ib_install_cq_cb helper function
  RDMA/mana_ib: Introduce mana_ib_get_netdev helper function
  RDMA/mana_ib: Introduce mdev_to_gc helper function
  RDMA/hns: Simplify 'struct hns_roce_hem' allocation
  ...

Author: torvalds

drivers/infiniband/core/cm.c

@@ -34,6 +34,7 @@ MODULE_AUTHOR("Sean Hefty");
 MODULE_DESCRIPTION("InfiniBand CM");
 MODULE_LICENSE("Dual BSD/GPL");
 
+#define CM_DESTROY_ID_WAIT_TIMEOUT 10000 /* msecs */
 static const char * const ibcm_rej_reason_strs[] = {
 	[IB_CM_REJ_NO_QP]			= "no QP",
 	[IB_CM_REJ_NO_EEC]			= "no EEC",
@@ -1025,10 +1026,20 @@ static void cm_reset_to_idle(struct cm_id_private *cm_id_priv)
 	}
 }
 
+static noinline void cm_destroy_id_wait_timeout(struct ib_cm_id *cm_id)
+{
+	struct cm_id_private *cm_id_priv;
+
+	cm_id_priv = container_of(cm_id, struct cm_id_private, id);
+	pr_err("%s: cm_id=%p timed out. state=%d refcnt=%d\n", __func__,
+	       cm_id, cm_id->state, refcount_read(&cm_id_priv->refcount));
+}
+
 static void cm_destroy_id(struct ib_cm_id *cm_id, int err)
 {
 	struct cm_id_private *cm_id_priv;
 	struct cm_work *work;
+	int ret;
 
 	cm_id_priv = container_of(cm_id, struct cm_id_private, id);
 	spin_lock_irq(&cm_id_priv->lock);
@@ -1135,7 +1146,14 @@ static void cm_destroy_id(struct ib_cm_id *cm_id, int err)
 
 	xa_erase(&cm.local_id_table, cm_local_id(cm_id->local_id));
 	cm_deref_id(cm_id_priv);
-	wait_for_completion(&cm_id_priv->comp);
+	do {
+		ret = wait_for_completion_timeout(&cm_id_priv->comp,
+						  msecs_to_jiffies(
+						  CM_DESTROY_ID_WAIT_TIMEOUT));
+		if (!ret) /* timeout happened */
+			cm_destroy_id_wait_timeout(cm_id);
+	} while (!ret);
+
 	while ((work = cm_dequeue_work(cm_id_priv)) != NULL)
 		cm_free_work(work);
 

drivers/infiniband/core/device.c

@@ -1730,7 +1730,7 @@ static int assign_client_id(struct ib_client *client)
 {
 	int ret;
 
-	down_write(&clients_rwsem);
+	lockdep_assert_held(&clients_rwsem);
 	/*
 	 * The add/remove callbacks must be called in FIFO/LIFO order. To
 	 * achieve this we assign client_ids so they are sorted in
@@ -1739,14 +1739,11 @@ static int assign_client_id(struct ib_client *client)
 	client->client_id = highest_client_id;
 	ret = xa_insert(&clients, client->client_id, client, GFP_KERNEL);
 	if (ret)
-		goto out;
+		return ret;
 
 	highest_client_id++;
 	xa_set_mark(&clients, client->client_id, CLIENT_REGISTERED);
-
-out:
-	up_write(&clients_rwsem);
-	return ret;
+	return 0;
 }
 
 static void remove_client_id(struct ib_client *client)
@@ -1776,25 +1773,35 @@ int ib_register_client(struct ib_client *client)
 {
 	struct ib_device *device;
 	unsigned long index;
+	bool need_unreg = false;
 	int ret;
 
 	refcount_set(&client->uses, 1);
 	init_completion(&client->uses_zero);
+
+	/*
+	 * The devices_rwsem is held in write mode to ensure that a racing
+	 * ib_register_device() sees a consisent view of clients and devices.
+	 */
+	down_write(&devices_rwsem);
+	down_write(&clients_rwsem);
 	ret = assign_client_id(client);
 	if (ret)
-		return ret;
+		goto out;
 
-	down_read(&devices_rwsem);
+	need_unreg = true;
 	xa_for_each_marked (&devices, index, device, DEVICE_REGISTERED) {
 		ret = add_client_context(device, client);
-		if (ret) {
-			up_read(&devices_rwsem);
-			ib_unregister_client(client);
-			return ret;
-		}
+		if (ret)
+			goto out;
 	}
-	up_read(&devices_rwsem);
-	return 0;
+	ret = 0;
+out:
+	up_write(&clients_rwsem);
+	up_write(&devices_rwsem);
+	if (need_unreg && ret)
+		ib_unregister_client(client);
+	return ret;
 }
 EXPORT_SYMBOL(ib_register_client);
 

drivers/infiniband/core/uverbs_cmd.c

@@ -2737,7 +2737,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 
 	switch (ib_spec->type & ~IB_FLOW_SPEC_INNER) {
 	case IB_FLOW_SPEC_ETH:
-		ib_filter_sz = offsetof(struct ib_flow_eth_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_eth_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2748,7 +2748,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		memcpy(&ib_spec->eth.mask, kern_spec_mask, actual_filter_sz);
 		break;
 	case IB_FLOW_SPEC_IPV4:
-		ib_filter_sz = offsetof(struct ib_flow_ipv4_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_ipv4_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2759,7 +2759,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		memcpy(&ib_spec->ipv4.mask, kern_spec_mask, actual_filter_sz);
 		break;
 	case IB_FLOW_SPEC_IPV6:
-		ib_filter_sz = offsetof(struct ib_flow_ipv6_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_ipv6_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2775,7 +2775,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		break;
 	case IB_FLOW_SPEC_TCP:
 	case IB_FLOW_SPEC_UDP:
-		ib_filter_sz = offsetof(struct ib_flow_tcp_udp_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_tcp_udp_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2786,7 +2786,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		memcpy(&ib_spec->tcp_udp.mask, kern_spec_mask, actual_filter_sz);
 		break;
 	case IB_FLOW_SPEC_VXLAN_TUNNEL:
-		ib_filter_sz = offsetof(struct ib_flow_tunnel_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_tunnel_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2801,7 +2801,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 			return -EINVAL;
 		break;
 	case IB_FLOW_SPEC_ESP:
-		ib_filter_sz = offsetof(struct ib_flow_esp_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_esp_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2812,7 +2812,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		memcpy(&ib_spec->esp.mask, kern_spec_mask, actual_filter_sz);
 		break;
 	case IB_FLOW_SPEC_GRE:
-		ib_filter_sz = offsetof(struct ib_flow_gre_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_gre_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);
@@ -2823,7 +2823,7 @@ int ib_uverbs_kern_spec_to_ib_spec_filter(enum ib_flow_spec_type type,
 		memcpy(&ib_spec->gre.mask, kern_spec_mask, actual_filter_sz);
 		break;
 	case IB_FLOW_SPEC_MPLS:
-		ib_filter_sz = offsetof(struct ib_flow_mpls_filter, real_sz);
+		ib_filter_sz = sizeof(struct ib_flow_mpls_filter);
 		actual_filter_sz = spec_filter_size(kern_spec_mask,
 						    kern_filter_sz,
 						    ib_filter_sz);