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

Pull rdma updates from Jason Gunthorpe:
 "This has been a fairly typical cycle, with the usual sorts of driver
  updates. Several series continue to come through which improve and
  modernize various parts of the core code, and we finally are starting
  to get the uAPI command interface cleaned up.

   - Various driver fixes for bnxt_re, cxgb3/4, hfi1, hns, i40iw, mlx4,
     mlx5, qib, rxe, usnic

   - Rework the entire syscall flow for uverbs to be able to run over
     ioctl(). Finally getting past the historic bad choice to use
     write() for command execution

   - More functional coverage with the mlx5 'devx' user API

   - Start of the HFI1 series for 'TID RDMA'

   - SRQ support in the hns driver

   - Support for new IBTA defined 2x lane widths

   - A big series to consolidate all the driver function pointers into a
     big struct and have drivers provide a 'static const' version of the
     struct instead of open coding initialization

   - New 'advise_mr' uAPI to control device caching/loading of page
     tables

   - Support for inline data in SRPT

   - Modernize how umad uses the driver core and creates cdev's and
     sysfs files

   - First steps toward removing 'uobject' from the view of the drivers"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (193 commits)
  RDMA/srpt: Use kmem_cache_free() instead of kfree()
  RDMA/mlx5: Signedness bug in UVERBS_HANDLER()
  IB/uverbs: Signedness bug in UVERBS_HANDLER()
  IB/mlx5: Allocate the per-port Q counter shared when DEVX is supported
  IB/umad: Start using dev_groups of class
  IB/umad: Use class_groups and let core create class file
  IB/umad: Refactor code to use cdev_device_add()
  IB/umad: Avoid destroying device while it is accessed
  IB/umad: Simplify and avoid dynamic allocation of class
  IB/mlx5: Fix wrong error unwind
  IB/mlx4: Remove set but not used variable 'pd'
  RDMA/iwcm: Don't copy past the end of dev_name() string
  IB/mlx5: Fix long EEH recover time with NVMe offloads
  IB/mlx5: Simplify netdev unbinding
  IB/core: Move query port to ioctl
  RDMA/nldev: Expose port_cap_flags2
  IB/core: uverbs copy to struct or zero helper
  IB/rxe: Reuse code which sets port state
  IB/rxe: Make counters thread safe
  IB/mlx5: Use the correct commands for UMEM and UCTX allocation
  ...

1 files changed, 144 insertions, 22 deletions

diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 87eb4f2cdd7d..47ab34ee1a9d 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -96,7 +96,7 @@ static struct notifier_block ibdev_lsm_nb = {
 
 static int ib_device_check_mandatory(struct ib_device *device)
 {
-#define IB_MANDATORY_FUNC(x) { offsetof(struct ib_device, x), #x }
+#define IB_MANDATORY_FUNC(x) { offsetof(struct ib_device_ops, x), #x }
 	static const struct {
 		size_t offset;
 		char  *name;
@@ -122,7 +122,8 @@ static int ib_device_check_mandatory(struct ib_device *device)
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(mandatory_table); ++i) {
-		if (!*(void **) ((void *) device + mandatory_table[i].offset)) {
+		if (!*(void **) ((void *) &device->ops +
+				 mandatory_table[i].offset)) {
 			dev_warn(&device->dev,
 				 "Device is missing mandatory function %s\n",
 				 mandatory_table[i].name);
@@ -145,7 +146,8 @@ static struct ib_device *__ib_device_get_by_index(u32 index)
 }
 
 /*
- * Caller is responsible to return refrerence count by calling put_device()
+ * Caller must perform ib_device_put() to return the device reference count
+ * when ib_device_get_by_index() returns valid device pointer.
  */
 struct ib_device *ib_device_get_by_index(u32 index)
 {
@@ -153,13 +155,21 @@ struct ib_device *ib_device_get_by_index(u32 index)
 
 	down_read(&lists_rwsem);
 	device = __ib_device_get_by_index(index);
-	if (device)
-		get_device(&device->dev);
-
+	if (device) {
+		/* Do not return a device if unregistration has started. */
+		if (!refcount_inc_not_zero(&device->refcount))
+			device = NULL;
+	}
 	up_read(&lists_rwsem);
 	return device;
 }
 
+void ib_device_put(struct ib_device *device)
+{
+	if (refcount_dec_and_test(&device->refcount))
+		complete(&device->unreg_completion);
+}
+
 static struct ib_device *__ib_device_get_by_name(const char *name)
 {
 	struct ib_device *device;
@@ -293,6 +303,8 @@ struct ib_device *ib_alloc_device(size_t size)
 	rwlock_init(&device->client_data_lock);
 	INIT_LIST_HEAD(&device->client_data_list);
 	INIT_LIST_HEAD(&device->port_list);
+	refcount_set(&device->refcount, 1);
+	init_completion(&device->unreg_completion);
 
 	return device;
 }
@@ -362,8 +374,8 @@ static int read_port_immutable(struct ib_device *device)
 		return -ENOMEM;
 
 	for (port = start_port; port <= end_port; ++port) {
-		ret = device->get_port_immutable(device, port,
-						 &device->port_immutable[port]);
+		ret = device->ops.get_port_immutable(
+			device, port, &device->port_immutable[port]);
 		if (ret)
 			return ret;
 
@@ -375,8 +387,8 @@ static int read_port_immutable(struct ib_device *device)
 
 void ib_get_device_fw_str(struct ib_device *dev, char *str)
 {
-	if (dev->get_dev_fw_str)
-		dev->get_dev_fw_str(dev, str);
+	if (dev->ops.get_dev_fw_str)
+		dev->ops.get_dev_fw_str(dev, str);
 	else
 		str[0] = '\0';
 }
@@ -525,7 +537,7 @@ static int setup_device(struct ib_device *device)
 	}
 
 	memset(&device->attrs, 0, sizeof(device->attrs));
-	ret = device->query_device(device, &device->attrs, &uhw);
+	ret = device->ops.query_device(device, &device->attrs, &uhw);
 	if (ret) {
 		dev_warn(&device->dev,
 			 "Couldn't query the device attributes\n");
@@ -641,6 +653,13 @@ void ib_unregister_device(struct ib_device *device)
 	struct ib_client_data *context, *tmp;
 	unsigned long flags;
 
+	/*
+	 * Wait for all netlink command callers to finish working on the
+	 * device.
+	 */
+	ib_device_put(device);
+	wait_for_completion(&device->unreg_completion);
+
 	mutex_lock(&device_mutex);
 
 	down_write(&lists_rwsem);
@@ -905,14 +924,14 @@ int ib_query_port(struct ib_device *device,
 		return -EINVAL;
 
 	memset(port_attr, 0, sizeof(*port_attr));
-	err = device->query_port(device, port_num, port_attr);
+	err = device->ops.query_port(device, port_num, port_attr);
 	if (err || port_attr->subnet_prefix)
 		return err;
 
 	if (rdma_port_get_link_layer(device, port_num) != IB_LINK_LAYER_INFINIBAND)
 		return 0;
 
-	err = device->query_gid(device, port_num, 0, &gid);
+	err = device->ops.query_gid(device, port_num, 0, &gid);
 	if (err)
 		return err;
 
@@ -946,8 +965,8 @@ void ib_enum_roce_netdev(struct ib_device *ib_dev,
 		if (rdma_protocol_roce(ib_dev, port)) {
 			struct net_device *idev = NULL;
 
-			if (ib_dev->get_netdev)
-				idev = ib_dev->get_netdev(ib_dev, port);
+			if (ib_dev->ops.get_netdev)
+				idev = ib_dev->ops.get_netdev(ib_dev, port);
 
 			if (idev &&
 			    idev->reg_state >= NETREG_UNREGISTERED) {
@@ -1024,7 +1043,10 @@ int ib_enum_all_devs(nldev_callback nldev_cb, struct sk_buff *skb,
 int ib_query_pkey(struct ib_device *device,
 		  u8 port_num, u16 index, u16 *pkey)
 {
-	return device->query_pkey(device, port_num, index, pkey);
+	if (!rdma_is_port_valid(device, port_num))
+		return -EINVAL;
+
+	return device->ops.query_pkey(device, port_num, index, pkey);
 }
 EXPORT_SYMBOL(ib_query_pkey);
 
@@ -1041,11 +1063,11 @@ int ib_modify_device(struct ib_device *device,
 		     int device_modify_mask,
 		     struct ib_device_modify *device_modify)
 {
-	if (!device->modify_device)
+	if (!device->ops.modify_device)
 		return -ENOSYS;
 
-	return device->modify_device(device, device_modify_mask,
-				     device_modify);
+	return device->ops.modify_device(device, device_modify_mask,
+					 device_modify);
 }
 EXPORT_SYMBOL(ib_modify_device);
 
@@ -1069,9 +1091,10 @@ int ib_modify_port(struct ib_device *device,
 	if (!rdma_is_port_valid(device, port_num))
 		return -EINVAL;
 
-	if (device->modify_port)
-		rc = device->modify_port(device, port_num, port_modify_mask,
-					   port_modify);
+	if (device->ops.modify_port)
+		rc = device->ops.modify_port(device, port_num,
+					     port_modify_mask,
+					     port_modify);
 	else
 		rc = rdma_protocol_roce(device, port_num) ? 0 : -ENOSYS;
 	return rc;
@@ -1198,6 +1221,105 @@ struct net_device *ib_get_net_dev_by_params(struct ib_device *dev,
 }
 EXPORT_SYMBOL(ib_get_net_dev_by_params);
 
+void ib_set_device_ops(struct ib_device *dev, const struct ib_device_ops *ops)
+{
+	struct ib_device_ops *dev_ops = &dev->ops;
+#define SET_DEVICE_OP(ptr, name)                                               \
+	do {                                                                   \
+		if (ops->name)                                                 \
+			if (!((ptr)->name))				       \
+				(ptr)->name = ops->name;                       \
+	} while (0)
+
+	SET_DEVICE_OP(dev_ops, add_gid);
+	SET_DEVICE_OP(dev_ops, alloc_dm);
+	SET_DEVICE_OP(dev_ops, alloc_fmr);
+	SET_DEVICE_OP(dev_ops, alloc_hw_stats);
+	SET_DEVICE_OP(dev_ops, alloc_mr);
+	SET_DEVICE_OP(dev_ops, alloc_mw);
+	SET_DEVICE_OP(dev_ops, alloc_pd);
+	SET_DEVICE_OP(dev_ops, alloc_rdma_netdev);
+	SET_DEVICE_OP(dev_ops, alloc_ucontext);
+	SET_DEVICE_OP(dev_ops, alloc_xrcd);
+	SET_DEVICE_OP(dev_ops, attach_mcast);
+	SET_DEVICE_OP(dev_ops, check_mr_status);
+	SET_DEVICE_OP(dev_ops, create_ah);
+	SET_DEVICE_OP(dev_ops, create_counters);
+	SET_DEVICE_OP(dev_ops, create_cq);
+	SET_DEVICE_OP(dev_ops, create_flow);
+	SET_DEVICE_OP(dev_ops, create_flow_action_esp);
+	SET_DEVICE_OP(dev_ops, create_qp);
+	SET_DEVICE_OP(dev_ops, create_rwq_ind_table);
+	SET_DEVICE_OP(dev_ops, create_srq);
+	SET_DEVICE_OP(dev_ops, create_wq);
+	SET_DEVICE_OP(dev_ops, dealloc_dm);
+	SET_DEVICE_OP(dev_ops, dealloc_fmr);
+	SET_DEVICE_OP(dev_ops, dealloc_mw);
+	SET_DEVICE_OP(dev_ops, dealloc_pd);
+	SET_DEVICE_OP(dev_ops, dealloc_ucontext);
+	SET_DEVICE_OP(dev_ops, dealloc_xrcd);
+	SET_DEVICE_OP(dev_ops, del_gid);
+	SET_DEVICE_OP(dev_ops, dereg_mr);
+	SET_DEVICE_OP(dev_ops, destroy_ah);
+	SET_DEVICE_OP(dev_ops, destroy_counters);
+	SET_DEVICE_OP(dev_ops, destroy_cq);
+	SET_DEVICE_OP(dev_ops, destroy_flow);
+	SET_DEVICE_OP(dev_ops, destroy_flow_action);
+	SET_DEVICE_OP(dev_ops, destroy_qp);
+	SET_DEVICE_OP(dev_ops, destroy_rwq_ind_table);
+	SET_DEVICE_OP(dev_ops, destroy_srq);
+	SET_DEVICE_OP(dev_ops, destroy_wq);
+	SET_DEVICE_OP(dev_ops, detach_mcast);
+	SET_DEVICE_OP(dev_ops, disassociate_ucontext);
+	SET_DEVICE_OP(dev_ops, drain_rq);
+	SET_DEVICE_OP(dev_ops, drain_sq);
+	SET_DEVICE_OP(dev_ops, get_dev_fw_str);
+	SET_DEVICE_OP(dev_ops, get_dma_mr);
+	SET_DEVICE_OP(dev_ops, get_hw_stats);
+	SET_DEVICE_OP(dev_ops, get_link_layer);
+	SET_DEVICE_OP(dev_ops, get_netdev);
+	SET_DEVICE_OP(dev_ops, get_port_immutable);
+	SET_DEVICE_OP(dev_ops, get_vector_affinity);
+	SET_DEVICE_OP(dev_ops, get_vf_config);
+	SET_DEVICE_OP(dev_ops, get_vf_stats);
+	SET_DEVICE_OP(dev_ops, map_mr_sg);
+	SET_DEVICE_OP(dev_ops, map_phys_fmr);
+	SET_DEVICE_OP(dev_ops, mmap);
+	SET_DEVICE_OP(dev_ops, modify_ah);
+	SET_DEVICE_OP(dev_ops, modify_cq);
+	SET_DEVICE_OP(dev_ops, modify_device);
+	SET_DEVICE_OP(dev_ops, modify_flow_action_esp);
+	SET_DEVICE_OP(dev_ops, modify_port);
+	SET_DEVICE_OP(dev_ops, modify_qp);
+	SET_DEVICE_OP(dev_ops, modify_srq);
+	SET_DEVICE_OP(dev_ops, modify_wq);
+	SET_DEVICE_OP(dev_ops, peek_cq);
+	SET_DEVICE_OP(dev_ops, poll_cq);
+	SET_DEVICE_OP(dev_ops, post_recv);
+	SET_DEVICE_OP(dev_ops, post_send);
+	SET_DEVICE_OP(dev_ops, post_srq_recv);
+	SET_DEVICE_OP(dev_ops, process_mad);
+	SET_DEVICE_OP(dev_ops, query_ah);
+	SET_DEVICE_OP(dev_ops, query_device);
+	SET_DEVICE_OP(dev_ops, query_gid);
+	SET_DEVICE_OP(dev_ops, query_pkey);
+	SET_DEVICE_OP(dev_ops, query_port);
+	SET_DEVICE_OP(dev_ops, query_qp);
+	SET_DEVICE_OP(dev_ops, query_srq);
+	SET_DEVICE_OP(dev_ops, rdma_netdev_get_params);
+	SET_DEVICE_OP(dev_ops, read_counters);
+	SET_DEVICE_OP(dev_ops, reg_dm_mr);
+	SET_DEVICE_OP(dev_ops, reg_user_mr);
+	SET_DEVICE_OP(dev_ops, req_ncomp_notif);
+	SET_DEVICE_OP(dev_ops, req_notify_cq);
+	SET_DEVICE_OP(dev_ops, rereg_user_mr);
+	SET_DEVICE_OP(dev_ops, resize_cq);
+	SET_DEVICE_OP(dev_ops, set_vf_guid);
+	SET_DEVICE_OP(dev_ops, set_vf_link_state);
+	SET_DEVICE_OP(dev_ops, unmap_fmr);
+}
+EXPORT_SYMBOL(ib_set_device_ops);
+
 static const struct rdma_nl_cbs ibnl_ls_cb_table[RDMA_NL_LS_NUM_OPS] = {
 	[RDMA_NL_LS_OP_RESOLVE] = {
 		.doit = ib_nl_handle_resolve_resp,