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

Pull rdma updates from Jason Gunthorpe:
 "Usual collection of small improvements and fixes, nothing especially
  stands out to me here.

  The new multipath PCI feature is a sign of things to come, I think we
  will see more of this in the next 10 years. Broadcom and HNS continue
  to update their drivers for their new HW generations.

  Summary:

   - Bug fixes and minor improvments in cxgb4, siw, mlx5, rxe, efa, rts,
     hfi, erdma, hns, irdma

   - Code cleanups/typos/etc. Tidy alloc_ordered_workqueue() calls

   - Multipath PCI for mlx5

   - Variable size work queue, SRQ changes, and relaxed ordering for new
     bnxt HW

   - New ODP fault resolution FW protocol in mlx5

   - New 'rdma monitor' netlink mechanism"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (99 commits)
  RDMA/bnxt_re: Remove the unused variable en_dev
  RDMA/nldev: Add missing break in rdma_nl_notify_err_msg()
  RDMA/irdma: fix error message in irdma_modify_qp_roce()
  RDMA/cxgb4: Added NULL check for lookup_atid
  RDMA/hns: Fix ah error counter in sw stat not increasing
  RDMA/bnxt_re: Recover the device when FW error is detected
  RDMA/bnxt_re: Group all operations under add_device and remove_device
  RDMA/bnxt_re: Use the aux device for L2 ULP callbacks
  RDMA/bnxt_re: Change aux driver data to en_info to hold more information
  RDMA/nldev: Expose whether RDMA monitoring is supported
  RDMA/nldev: Add support for RDMA monitoring
  RDMA/mlx5: Use IB set_netdev and get_netdev functions
  RDMA/device: Remove optimization in ib_device_get_netdev()
  RDMA/mlx5: Initialize phys_port_cnt earlier in RDMA device creation
  RDMA/mlx5: Obtain upper net device only when needed
  RDMA/mlx5: Check RoCE LAG status before getting netdev
  RDMA/mlx5: Consider the query_vuid cap for data_direct
  net/mlx5: Handle memory scheme ODP capabilities
  RDMA/mlx5: Add implicit MR handling to ODP memory scheme
  RDMA/mlx5: Add handling for memory scheme page fault events
  ...

Author: torvalds

drivers/infiniband/core/cache.c

@@ -1640,8 +1640,10 @@ int ib_cache_setup_one(struct ib_device *device)
 
 	rdma_for_each_port (device, p) {
 		err = ib_cache_update(device, p, true, true, true);
-		if (err)
+		if (err) {
+			gid_table_cleanup_one(device);
 			return err;
+		}
 	}
 
 	return 0;

drivers/infiniband/core/core_priv.h

@@ -325,9 +325,6 @@ void ib_qp_usecnt_inc(struct ib_qp *qp);
 void ib_qp_usecnt_dec(struct ib_qp *qp);
 
 struct rdma_dev_addr;
-int rdma_resolve_ip_route(struct sockaddr *src_addr,
-			  const struct sockaddr *dst_addr,
-			  struct rdma_dev_addr *addr);
 
 int rdma_addr_find_l2_eth_by_grh(const union ib_gid *sgid,
 				 const union ib_gid *dgid,

drivers/infiniband/core/device.c

@@ -1351,6 +1351,29 @@ static void prevent_dealloc_device(struct ib_device *ib_dev)
 {
 }
 
+static void ib_device_notify_register(struct ib_device *device)
+{
+	struct net_device *netdev;
+	u32 port;
+	int ret;
+
+	ret = rdma_nl_notify_event(device, 0, RDMA_REGISTER_EVENT);
+	if (ret)
+		return;
+
+	rdma_for_each_port(device, port) {
+		netdev = ib_device_get_netdev(device, port);
+		if (!netdev)
+			continue;
+
+		ret = rdma_nl_notify_event(device, port,
+					   RDMA_NETDEV_ATTACH_EVENT);
+		dev_put(netdev);
+		if (ret)
+			return;
+	}
+}
+
 /**
  * ib_register_device - Register an IB device with IB core
  * @device: Device to register
@@ -1449,6 +1472,8 @@ int ib_register_device(struct ib_device *device, const char *name,
 	dev_set_uevent_suppress(&device->dev, false);
 	/* Mark for userspace that device is ready */
 	kobject_uevent(&device->dev.kobj, KOBJ_ADD);
+
+	ib_device_notify_register(device);
 	ib_device_put(device);
 
 	return 0;
@@ -1491,6 +1516,7 @@ static void __ib_unregister_device(struct ib_device *ib_dev)
 		goto out;
 
 	disable_device(ib_dev);
+	rdma_nl_notify_event(ib_dev, 0, RDMA_UNREGISTER_EVENT);
 
 	/* Expedite removing unregistered pointers from the hash table */
 	free_netdevs(ib_dev);
@@ -2159,6 +2185,7 @@ static void add_ndev_hash(struct ib_port_data *pdata)
 int ib_device_set_netdev(struct ib_device *ib_dev, struct net_device *ndev,
 			 u32 port)
 {
+	enum rdma_nl_notify_event_type etype;
 	struct net_device *old_ndev;
 	struct ib_port_data *pdata;
 	unsigned long flags;
@@ -2190,6 +2217,14 @@ int ib_device_set_netdev(struct ib_device *ib_dev, struct net_device *ndev,
 	spin_unlock_irqrestore(&pdata->netdev_lock, flags);
 
 	add_ndev_hash(pdata);
+
+	/* Make sure that the device is registered before we send events */
+	if (xa_load(&devices, ib_dev->index) != ib_dev)
+		return 0;
+
+	etype = ndev ? RDMA_NETDEV_ATTACH_EVENT : RDMA_NETDEV_DETACH_EVENT;
+	rdma_nl_notify_event(ib_dev, port, etype);
+
 	return 0;
 }
 EXPORT_SYMBOL(ib_device_set_netdev);
@@ -2236,6 +2271,9 @@ struct net_device *ib_device_get_netdev(struct ib_device *ib_dev,
 	if (!rdma_is_port_valid(ib_dev, port))
 		return NULL;
 
+	if (!ib_dev->port_data)
+		return NULL;
+
 	pdata = &ib_dev->port_data[port];
 
 	/*
@@ -2252,17 +2290,9 @@ struct net_device *ib_device_get_netdev(struct ib_device *ib_dev,
 		spin_unlock(&pdata->netdev_lock);
 	}
 
-	/*
-	 * If we are starting to unregister expedite things by preventing
-	 * propagation of an unregistering netdev.
-	 */
-	if (res && res->reg_state != NETREG_REGISTERED) {
-		dev_put(res);
-		return NULL;
-	}
-
 	return res;
 }
+EXPORT_SYMBOL(ib_device_get_netdev);
 
 /**
  * ib_device_get_by_netdev - Find an IB device associated with a netdev

drivers/infiniband/core/iwcm.c

@@ -1182,7 +1182,7 @@ static int __init iw_cm_init(void)
 	if (ret)
 		return ret;
 
-	iwcm_wq = alloc_ordered_workqueue("iw_cm_wq", 0);
+	iwcm_wq = alloc_ordered_workqueue("iw_cm_wq", WQ_MEM_RECLAIM);
 	if (!iwcm_wq)
 		goto err_alloc;
 

drivers/infiniband/core/mad.c

@@ -2616,14 +2616,16 @@ static int retry_send(struct ib_mad_send_wr_private *mad_send_wr)
 
 static void timeout_sends(struct work_struct *work)
 {
+	struct ib_mad_send_wr_private *mad_send_wr, *n;
 	struct ib_mad_agent_private *mad_agent_priv;
-	struct ib_mad_send_wr_private *mad_send_wr;
 	struct ib_mad_send_wc mad_send_wc;
+	struct list_head local_list;
 	unsigned long flags, delay;
 
 	mad_agent_priv = container_of(work, struct ib_mad_agent_private,
 				      timed_work.work);
 	mad_send_wc.vendor_err = 0;
+	INIT_LIST_HEAD(&local_list);
 
 	spin_lock_irqsave(&mad_agent_priv->lock, flags);
 	while (!list_empty(&mad_agent_priv->wait_list)) {
@@ -2641,25 +2643,25 @@ static void timeout_sends(struct work_struct *work)
 			break;
 		}
 
-		list_del(&mad_send_wr->agent_list);
+		list_del_init(&mad_send_wr->agent_list);
 		if (mad_send_wr->status == IB_WC_SUCCESS &&
 		    !retry_send(mad_send_wr))
 			continue;
 
-		spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
+		list_add_tail(&mad_send_wr->agent_list, &local_list);
+	}
+	spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
 
+	list_for_each_entry_safe(mad_send_wr, n, &local_list, agent_list) {
 		if (mad_send_wr->status == IB_WC_SUCCESS)
 			mad_send_wc.status = IB_WC_RESP_TIMEOUT_ERR;
 		else
 			mad_send_wc.status = mad_send_wr->status;
 		mad_send_wc.send_buf = &mad_send_wr->send_buf;
 		mad_agent_priv->agent.send_handler(&mad_agent_priv->agent,
 						   &mad_send_wc);
-
 		deref_mad_agent(mad_agent_priv);
-		spin_lock_irqsave(&mad_agent_priv->lock, flags);
 	}
-	spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
 }
 
 /*
@@ -2937,7 +2939,6 @@ static int ib_mad_port_open(struct ib_device *device,
 	int ret, cq_size;
 	struct ib_mad_port_private *port_priv;
 	unsigned long flags;
-	char name[sizeof "ib_mad123"];
 	int has_smi;
 
 	if (WARN_ON(rdma_max_mad_size(device, port_num) < IB_MGMT_MAD_SIZE))
@@ -2990,8 +2991,8 @@ static int ib_mad_port_open(struct ib_device *device,
 			goto error7;
 	}
 
-	snprintf(name, sizeof(name), "ib_mad%u", port_num);
-	port_priv->wq = alloc_ordered_workqueue(name, WQ_MEM_RECLAIM);
+	port_priv->wq = alloc_ordered_workqueue("ib_mad%u", WQ_MEM_RECLAIM,
+						port_num);
 	if (!port_priv->wq) {
 		ret = -ENOMEM;
 		goto error8;

drivers/infiniband/core/netlink.c

@@ -311,6 +311,7 @@ int rdma_nl_net_init(struct rdma_dev_net *rnet)
 	struct net *net = read_pnet(&rnet->net);
 	struct netlink_kernel_cfg cfg = {
 		.input	= rdma_nl_rcv,
+		.flags = NL_CFG_F_NONROOT_RECV,
 	};
 	struct sock *nls;