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

Pull rdma updates from Jason Gunthorpe:
"This cycle we got a new RDMA driver "ERDMA" for the Alibaba cloud
environment. Otherwise the changes are dominated by rxe fixes.

There is another RDMA driver on the list that might get merged next
cycle, 'MANA' for the Azure cloud environment.

Summary:

- Bug fixes and small features for irdma, hns, siw, qedr, hfi1, mlx5

- General spelling/grammer fixes

- rdma cm can follow changes in neighbours for control packets

- Significant amounts of rxe fixes and spec compliance changes

- Use the modern NAPI API

- Use the bitmap API instead of open coding

- Performance improvements for rtrs

- Add the ERDMA driver for Alibaba cloud

- Fix a use after free bug in SRP"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (99 commits)
RDMA/ib_srpt: Unify checking rdma_cm_id condition in srpt_cm_req_recv()
RDMA/rxe: Fix error unwind in rxe_create_qp()
RDMA/mlx5: Add missing check for return value in get namespace flow
RDMA/rxe: Split qp state for requester and completer
RDMA/rxe: Generate error completion for error requester QP state
RDMA/rxe: Update wqe_index for each wqe error completion
RDMA/srpt: Fix a use-after-free
RDMA/srpt: Introduce a reference count in struct srpt_device
RDMA/srpt: Duplicate port name members
IB/qib: Fix repeated "in" within comments
RDMA/erdma: Add driver to kernel build environment
RDMA/erdma: Add the ABI definitions
RDMA/erdma: Add the erdma module
RDMA/erdma: Add connection management (CM) support
RDMA/erdma: Add verbs implementation
RDMA/erdma: Add verbs header file
RDMA/erdma: Add event queue implementation
RDMA/erdma: Add cmdq implementation
RDMA/erdma: Add main include file
RDMA/erdma: Add the hardware related definitions
...

MAINTAINERS

···
 F:	Documentation/i2c/busses/i2c-ali1563.rst
 F:	drivers/i2c/busses/i2c-ali1563.c
 
+ALIBABA ELASTIC RDMA DRIVER
+M:	Cheng Xu <[e-mail]>
+M:	Kai Shen <[e-mail]>
+L:	[e-mail]
+S:	Supported
+F:	drivers/infiniband/hw/erdma
+F:	include/uapi/rdma/erdma-abi.h
+
 ALIENWARE WMI DRIVER
 L:	[e-mail]
 S:	Maintained

drivers/infiniband/Kconfig

···
 	def_bool !HIGHMEM
 
 if INFINIBAND_USER_ACCESS || !INFINIBAND_USER_ACCESS
-source "drivers/infiniband/hw/mthca/Kconfig"
-source "drivers/infiniband/hw/qib/Kconfig"
+source "drivers/infiniband/hw/bnxt_re/Kconfig"
 source "drivers/infiniband/hw/cxgb4/Kconfig"
 source "drivers/infiniband/hw/efa/Kconfig"
+source "drivers/infiniband/hw/erdma/Kconfig"
+source "drivers/infiniband/hw/hfi1/Kconfig"
+source "drivers/infiniband/hw/hns/Kconfig"
 source "drivers/infiniband/hw/irdma/Kconfig"
 source "drivers/infiniband/hw/mlx4/Kconfig"
 source "drivers/infiniband/hw/mlx5/Kconfig"
+source "drivers/infiniband/hw/mthca/Kconfig"
 source "drivers/infiniband/hw/ocrdma/Kconfig"
-source "drivers/infiniband/hw/vmw_pvrdma/Kconfig"
-source "drivers/infiniband/hw/usnic/Kconfig"
-source "drivers/infiniband/hw/hns/Kconfig"
-source "drivers/infiniband/hw/bnxt_re/Kconfig"
-source "drivers/infiniband/hw/hfi1/Kconfig"
 source "drivers/infiniband/hw/qedr/Kconfig"
+source "drivers/infiniband/hw/qib/Kconfig"
+source "drivers/infiniband/hw/usnic/Kconfig"
+source "drivers/infiniband/hw/vmw_pvrdma/Kconfig"
 source "drivers/infiniband/sw/rdmavt/Kconfig"
 source "drivers/infiniband/sw/rxe/Kconfig"
 source "drivers/infiniband/sw/siw/Kconfig"

drivers/infiniband/core/cma.c

···
 #include <linux/in6.h>
 #include <linux/mutex.h>
 #include <linux/random.h>
+#include <linux/rbtree.h>
 #include <linux/igmp.h>
 #include <linux/xarray.h>
 #include <linux/inetdevice.h>
···
 
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
+#include <net/netevent.h>
 #include <net/tcp.h>
 #include <net/ipv6.h>
 #include <net/ip_fib.h>
···
 static LIST_HEAD(dev_list);
 static LIST_HEAD(listen_any_list);
 static DEFINE_MUTEX(lock);
+static struct rb_root id_table = RB_ROOT;
+/* Serialize operations of id_table tree */
+static DEFINE_SPINLOCK(id_table_lock);
 static struct workqueue_struct *cma_wq;
 static unsigned int cma_pernet_id;
 
···
 		return NULL;
 	}
 }
+
+struct id_table_entry {
+	struct list_head id_list;
+	struct rb_node rb_node;
+};
 
 struct cma_device {
 	struct list_head	list;
···
 	return hdr->ip_version >> 4;
 }
 
-static inline void cma_set_ip_ver(struct cma_hdr *hdr, u8 ip_ver)
+static void cma_set_ip_ver(struct cma_hdr *hdr, u8 ip_ver)
 {
 	hdr->ip_version = (ip_ver << 4) | (hdr->ip_version & 0xF);
+}
+
+static struct sockaddr *cma_src_addr(struct rdma_id_private *id_priv)
+{
+	return (struct sockaddr *)&id_priv->id.route.addr.src_addr;
+}
+
+static inline struct sockaddr *cma_dst_addr(struct rdma_id_private *id_priv)
+{
+	return (struct sockaddr *)&id_priv->id.route.addr.dst_addr;
 }
 
 static int cma_igmp_send(struct net_device *ndev, union ib_gid *mgid, bool join)
···
 		rtnl_unlock();
 	}
 	return (in_dev) ? 0 : -ENODEV;
+}
+
+static int compare_netdev_and_ip(int ifindex_a, struct sockaddr *sa,
+				 struct id_table_entry *entry_b)
+{
+	struct rdma_id_private *id_priv = list_first_entry(
+		&entry_b->id_list, struct rdma_id_private, id_list_entry);
+	int ifindex_b = id_priv->id.route.addr.dev_addr.bound_dev_if;
+	struct sockaddr *sb = cma_dst_addr(id_priv);
+
+	if (ifindex_a != ifindex_b)
+		return (ifindex_a > ifindex_b) ? 1 : -1;
+
+	if (sa->sa_family != sb->sa_family)
+		return sa->sa_family - sb->sa_family;
+
+	if (sa->sa_family == AF_INET)
+		return memcmp((char *)&((struct sockaddr_in *)sa)->sin_addr,
+			      (char *)&((struct sockaddr_in *)sb)->sin_addr,
+			      sizeof(((struct sockaddr_in *)sa)->sin_addr));
+
+	return ipv6_addr_cmp(&((struct sockaddr_in6 *)sa)->sin6_addr,
+			     &((struct sockaddr_in6 *)sb)->sin6_addr);
+}
+
+static int cma_add_id_to_tree(struct rdma_id_private *node_id_priv)
+{
+	struct rb_node **new, *parent = NULL;
+	struct id_table_entry *this, *node;
+	unsigned long flags;
+	int result;
+
+	node = kzalloc(sizeof(*node), GFP_KERNEL);
+	if (!node)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&id_table_lock, flags);
+	new = &id_table.rb_node;
+	while (*new) {
+		this = container_of(*new, struct id_table_entry, rb_node);
+		result = compare_netdev_and_ip(
+			node_id_priv->id.route.addr.dev_addr.bound_dev_if,
+			cma_dst_addr(node_id_priv), this);
+
+		parent = *new;
+		if (result < 0)
+			new = &((*new)->rb_left);
+		else if (result > 0)
+			new = &((*new)->rb_right);
+		else {
+			list_add_tail(&node_id_priv->id_list_entry,
+				      &this->id_list);
+			kfree(node);
+			goto unlock;
+		}
+	}
+
+	INIT_LIST_HEAD(&node->id_list);
+	list_add_tail(&node_id_priv->id_list_entry, &node->id_list);
+
+	rb_link_node(&node->rb_node, parent, new);
+	rb_insert_color(&node->rb_node, &id_table);
+
+unlock:
+	spin_unlock_irqrestore(&id_table_lock, flags);
+	return 0;
+}
+
+static struct id_table_entry *
+node_from_ndev_ip(struct rb_root *root, int ifindex, struct sockaddr *sa)
+{
+	struct rb_node *node = root->rb_node;
+	struct id_table_entry *data;
+	int result;
+
+	while (node) {
+		data = container_of(node, struct id_table_entry, rb_node);
+		result = compare_netdev_and_ip(ifindex, sa, data);
+		if (result < 0)
+			node = node->rb_left;
+		else if (result > 0)
+			node = node->rb_right;
+		else
+			return data;
+	}
+
+	return NULL;
+}
+
+static void cma_remove_id_from_tree(struct rdma_id_private *id_priv)
+{
+	struct id_table_entry *data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&id_table_lock, flags);
+	if (list_empty(&id_priv->id_list_entry))
+		goto out;
+
+	data = node_from_ndev_ip(&id_table,
+				 id_priv->id.route.addr.dev_addr.bound_dev_if,
+				 cma_dst_addr(id_priv));
+	if (!data)
+		goto out;
+
+	list_del_init(&id_priv->id_list_entry);
+	if (list_empty(&data->id_list)) {
+		rb_erase(&data->rb_node, &id_table);
+		kfree(data);
+	}
+out:
+	spin_unlock_irqrestore(&id_table_lock, flags);
 }
 
 static void _cma_attach_to_dev(struct rdma_id_private *id_priv,
···
 		id_priv->id.route.addr.dev_addr.sgid_attr = NULL;
 	}
 	mutex_unlock(&lock);
-}
-
-static inline struct sockaddr *cma_src_addr(struct rdma_id_private *id_priv)
-{
-	return (struct sockaddr *) &id_priv->id.route.addr.src_addr;
-}
-
-static inline struct sockaddr *cma_dst_addr(struct rdma_id_private *id_priv)
-{
-	return (struct sockaddr *) &id_priv->id.route.addr.dst_addr;
 }
 
 static inline unsigned short cma_family(struct rdma_id_private *id_priv)
···
 	refcount_set(&id_priv->refcount, 1);
 	mutex_init(&id_priv->handler_mutex);
 	INIT_LIST_HEAD(&id_priv->device_item);
+	INIT_LIST_HEAD(&id_priv->id_list_entry);
 	INIT_LIST_HEAD(&id_priv->listen_list);
 	INIT_LIST_HEAD(&id_priv->mc_list);
 	get_random_bytes(&id_priv->seq_num, sizeof id_priv->seq_num);
···
 	cma_cancel_operation(id_priv, state);
 
 	rdma_restrack_del(&id_priv->res);
+	cma_remove_id_from_tree(id_priv);
 	if (id_priv->cma_dev) {
 		if (rdma_cap_ib_cm(id_priv->id.device, 1)) {
 			if (id_priv->cm_id.ib)
···
 	cma_id_get(id_priv);
 	if (rdma_cap_ib_sa(id->device, id->port_num))
 		ret = cma_resolve_ib_route(id_priv, timeout_ms);
-	else if (rdma_protocol_roce(id->device, id->port_num))
+	else if (rdma_protocol_roce(id->device, id->port_num)) {
 		ret = cma_resolve_iboe_route(id_priv);
+		if (!ret)
+			cma_add_id_to_tree(id_priv);
+	}
 	else if (rdma_protocol_iwarp(id->device, id->port_num))
 		ret = cma_resolve_iw_route(id_priv);
 	else
···
 	return ret;
 }
 
+static void cma_netevent_work_handler(struct work_struct *_work)
+{
+	struct rdma_id_private *id_priv =
+		container_of(_work, struct rdma_id_private, id.net_work);
+	struct rdma_cm_event event = {};
+
+	mutex_lock(&id_priv->handler_mutex);
+
+	if (READ_ONCE(id_priv->state) == RDMA_CM_DESTROYING ||
+	    READ_ONCE(id_priv->state) == RDMA_CM_DEVICE_REMOVAL)
+		goto out_unlock;
+
+	event.event = RDMA_CM_EVENT_UNREACHABLE;
+	event.status = -ETIMEDOUT;
+
+	if (cma_cm_event_handler(id_priv, &event)) {
+		__acquire(&id_priv->handler_mutex);
+		id_priv->cm_id.ib = NULL;
+		cma_id_put(id_priv);
+		destroy_id_handler_unlock(id_priv);
+		return;
+	}
+
+out_unlock:
+	mutex_unlock(&id_priv->handler_mutex);
+	cma_id_put(id_priv);
+}
+
+static int cma_netevent_callback(struct notifier_block *self,
+				 unsigned long event, void *ctx)
+{
+	struct id_table_entry *ips_node = NULL;
+	struct rdma_id_private *current_id;
+	struct neighbour *neigh = ctx;
+	unsigned long flags;
+
+	if (event != NETEVENT_NEIGH_UPDATE)
+		return NOTIFY_DONE;
+
+	spin_lock_irqsave(&id_table_lock, flags);
+	if (neigh->tbl->family == AF_INET6) {
+		struct sockaddr_in6 neigh_sock_6;
+
+		neigh_sock_6.sin6_family = AF_INET6;
+		neigh_sock_6.sin6_addr = *(struct in6_addr *)neigh->primary_key;
+		ips_node = node_from_ndev_ip(&id_table, neigh->dev->ifindex,
+					     (struct sockaddr *)&neigh_sock_6);
+	} else if (neigh->tbl->family == AF_INET) {
+		struct sockaddr_in neigh_sock_4;
+
+		neigh_sock_4.sin_family = AF_INET;
+		neigh_sock_4.sin_addr.s_addr = *(__be32 *)(neigh->primary_key);
+		ips_node = node_from_ndev_ip(&id_table, neigh->dev->ifindex,
+					     (struct sockaddr *)&neigh_sock_4);
+	} else
+		goto out;
+
+	if (!ips_node)
+		goto out;
+
+	list_for_each_entry(current_id, &ips_node->id_list, id_list_entry) {
+		if (!memcmp(current_id->id.route.addr.dev_addr.dst_dev_addr,
+			   neigh->ha, ETH_ALEN))
+			continue;
+		INIT_WORK(&current_id->id.net_work, cma_netevent_work_handler);
+		cma_id_get(current_id);
+		queue_work(cma_wq, &current_id->id.net_work);
+	}
+out:
+	spin_unlock_irqrestore(&id_table_lock, flags);
+	return NOTIFY_DONE;
+}
+
 static struct notifier_block cma_nb = {
 	.notifier_call = cma_netdev_callback
+};
+
+static struct notifier_block cma_netevent_cb = {
+	.notifier_call = cma_netevent_callback
 };
 
 static void cma_send_device_removal_put(struct rdma_id_private *id_priv)
···
 
 	ib_sa_register_client(&sa_client);
 	register_netdevice_notifier(&cma_nb);
+	register_netevent_notifier(&cma_netevent_cb);
 
 	ret = ib_register_client(&cma_client);
 	if (ret)
···
 err_ib:
 	ib_unregister_client(&cma_client);
 err:
+	unregister_netevent_notifier(&cma_netevent_cb);
 	unregister_netdevice_notifier(&cma_nb);
 	ib_sa_unregister_client(&sa_client);
 	unregister_pernet_subsys(&cma_pernet_operations);
···
 {
 	cma_configfs_exit();
 	ib_unregister_client(&cma_client);
+	unregister_netevent_notifier(&cma_netevent_cb);
 	unregister_netdevice_notifier(&cma_nb);
 	ib_sa_unregister_client(&sa_client);
 	unregister_pernet_subsys(&cma_pernet_operations);

drivers/infiniband/core/cma_priv.h

···
 		struct list_head listen_item;
 		struct list_head listen_list;
 	};
+	struct list_head        id_list_entry;
 	struct cma_device	*cma_dev;
 	struct list_head	mc_list;
 

drivers/infiniband/core/rdma_core.c

···
 	 * In exclusive access mode, we check that the counter is zero (nobody
 	 * claimed this object) and we set it to -1. Releasing a shared access
 	 * lock is done simply by decreasing the counter. As for exclusive
-	 * access locks, since only a single one of them is is allowed
+	 * access locks, since only a single one of them is allowed
 	 * concurrently, setting the counter to zero is enough for releasing
 	 * this lock.
 	 */

drivers/infiniband/core/roce_gid_mgmt.c

···
 
 /**
  * is_upper_ndev_bond_master_filter - Check if a given netdevice
- * is bond master device of netdevice of the the RDMA device of port.
+ * is bond master device of netdevice of the RDMA device of port.
  * @ib_dev:		IB device to check
  * @port:		Port to consider for adding default GID
  * @rdma_ndev:		Pointer to rdma netdevice

drivers/infiniband/hw/Makefile

···
 obj-$(CONFIG_INFINIBAND_HNS)		+= hns/
 obj-$(CONFIG_INFINIBAND_QEDR)		+= qedr/
 obj-$(CONFIG_INFINIBAND_BNXT_RE)	+= bnxt_re/
+obj-$(CONFIG_INFINIBAND_ERDMA)		+= erdma/

drivers/infiniband/hw/bnxt_re/bnxt_re.h

···
 	/* Max of 2 lossless traffic class supported per port */
 	u16				cosq[2];
 
-	/* QP for for handling QP1 packets */
+	/* QP for handling QP1 packets */
 	struct bnxt_re_gsi_context	gsi_ctx;
 	struct bnxt_re_stats		stats;
 	atomic_t nq_alloc_cnt;

drivers/infiniband/hw/erdma/Kconfig

···
+# SPDX-License-Identifier: GPL-2.0-only
+config INFINIBAND_ERDMA
+	tristate "Alibaba Elastic RDMA Adapter (ERDMA) support"
+	depends on PCI_MSI && 64BIT
+	depends on INFINIBAND_ADDR_TRANS
+	depends on INFINIBAND_USER_ACCESS
+	help
+	  This is a RDMA/iWarp driver for Alibaba Elastic RDMA Adapter(ERDMA),
+	  which supports RDMA features in Alibaba cloud environment.
+
+	  To compile this driver as module, choose M here. The module will be
+	  called erdma.

drivers/infiniband/hw/erdma/Makefile

···
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_INFINIBAND_ERDMA) := erdma.o
+
+erdma-y := erdma_cm.o erdma_main.o erdma_cmdq.o erdma_cq.o erdma_verbs.o erdma_qp.o erdma_eq.o

drivers/infiniband/hw/erdma/erdma.h

···
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#ifndef __ERDMA_H__
+#define __ERDMA_H__
+
+#include <linux/bitfield.h>
+#include <linux/netdevice.h>
+#include <linux/xarray.h>
+#include <rdma/ib_verbs.h>
+
+#include "erdma_hw.h"
+
+#define DRV_MODULE_NAME "erdma"
+#define ERDMA_NODE_DESC "Elastic RDMA(iWARP) stack"
+
+struct erdma_eq {
+	void *qbuf;
+	dma_addr_t qbuf_dma_addr;
+
+	spinlock_t lock;
+
+	u32 depth;
+
+	u16 ci;
+	u16 rsvd;
+
+	atomic64_t event_num;
+	atomic64_t notify_num;
+
+	u64 __iomem *db_addr;
+	u64 *db_record;
+};
+
+struct erdma_cmdq_sq {
+	void *qbuf;
+	dma_addr_t qbuf_dma_addr;
+
+	spinlock_t lock;
+
+	u32 depth;
+	u16 ci;
+	u16 pi;
+
+	u16 wqebb_cnt;
+
+	u64 *db_record;
+};
+
+struct erdma_cmdq_cq {
+	void *qbuf;
+	dma_addr_t qbuf_dma_addr;
+
+	spinlock_t lock;
+
+	u32 depth;
+	u32 ci;
+	u32 cmdsn;
+
+	u64 *db_record;
+
+	atomic64_t armed_num;
+};
+
+enum {
+	ERDMA_CMD_STATUS_INIT,
+	ERDMA_CMD_STATUS_ISSUED,
+	ERDMA_CMD_STATUS_FINISHED,
+	ERDMA_CMD_STATUS_TIMEOUT
+};
+
+struct erdma_comp_wait {
+	struct completion wait_event;
+	u32 cmd_status;
+	u32 ctx_id;
+	u16 sq_pi;
+	u8 comp_status;
+	u8 rsvd;
+	u32 comp_data[4];
+};
+
+enum {
+	ERDMA_CMDQ_STATE_OK_BIT = 0,
+	ERDMA_CMDQ_STATE_TIMEOUT_BIT = 1,
+	ERDMA_CMDQ_STATE_CTX_ERR_BIT = 2,
+};
+
+#define ERDMA_CMDQ_TIMEOUT_MS 15000
+#define ERDMA_REG_ACCESS_WAIT_MS 20
+#define ERDMA_WAIT_DEV_DONE_CNT 500
+
+struct erdma_cmdq {
+	unsigned long *comp_wait_bitmap;
+	struct erdma_comp_wait *wait_pool;
+	spinlock_t lock;
+
+	bool use_event;
+
+	struct erdma_cmdq_sq sq;
+	struct erdma_cmdq_cq cq;
+	struct erdma_eq eq;
+
+	unsigned long state;
+
+	struct semaphore credits;
+	u16 max_outstandings;
+};
+
+#define COMPROMISE_CC ERDMA_CC_CUBIC
+enum erdma_cc_alg {
+	ERDMA_CC_NEWRENO = 0,
+	ERDMA_CC_CUBIC,
+	ERDMA_CC_HPCC_RTT,
+	ERDMA_CC_HPCC_ECN,
+	ERDMA_CC_HPCC_INT,
+	ERDMA_CC_METHODS_NUM
+};
+
+struct erdma_devattr {
+	u32 fw_version;
+
+	unsigned char peer_addr[ETH_ALEN];
+
+	int numa_node;
+	enum erdma_cc_alg cc;
+	u32 grp_num;
+	u32 irq_num;
+
+	bool disable_dwqe;
+	u16 dwqe_pages;
+	u16 dwqe_entries;
+
+	u32 max_qp;
+	u32 max_send_wr;
+	u32 max_recv_wr;
+	u32 max_ord;
+	u32 max_ird;
+
+	u32 max_send_sge;
+	u32 max_recv_sge;
+	u32 max_sge_rd;
+	u32 max_cq;
+	u32 max_cqe;
+	u64 max_mr_size;
+	u32 max_mr;
+	u32 max_pd;
+	u32 max_mw;
+	u32 local_dma_key;
+};
+
+#define ERDMA_IRQNAME_SIZE 50
+
+struct erdma_irq {
+	char name[ERDMA_IRQNAME_SIZE];
+	u32 msix_vector;
+	cpumask_t affinity_hint_mask;
+};
+
+struct erdma_eq_cb {
+	bool ready;
+	void *dev; /* All EQs use this fields to get erdma_dev struct */
+	struct erdma_irq irq;
+	struct erdma_eq eq;
+	struct tasklet_struct tasklet;
+};
+
+struct erdma_resource_cb {
+	unsigned long *bitmap;
+	spinlock_t lock;
+	u32 next_alloc_idx;
+	u32 max_cap;
+};
+
+enum {
+	ERDMA_RES_TYPE_PD = 0,
+	ERDMA_RES_TYPE_STAG_IDX = 1,
+	ERDMA_RES_CNT = 2,
+};
+
+#define ERDMA_EXTRA_BUFFER_SIZE ERDMA_DB_SIZE
+#define WARPPED_BUFSIZE(size) ((size) + ERDMA_EXTRA_BUFFER_SIZE)
+
+struct erdma_dev {
+	struct ib_device ibdev;
+	struct net_device *netdev;
+	struct pci_dev *pdev;
+	struct notifier_block netdev_nb;
+
+	resource_size_t func_bar_addr;
+	resource_size_t func_bar_len;
+	u8 __iomem *func_bar;
+
+	struct erdma_devattr attrs;
+	/* physical port state (only one port per device) */
+	enum ib_port_state state;
+
+	/* cmdq and aeq use the same msix vector */
+	struct erdma_irq comm_irq;
+	struct erdma_cmdq cmdq;
+	struct erdma_eq aeq;
+	struct erdma_eq_cb ceqs[ERDMA_NUM_MSIX_VEC - 1];
+
+	spinlock_t lock;
+	struct erdma_resource_cb res_cb[ERDMA_RES_CNT];
+	struct xarray qp_xa;
+	struct xarray cq_xa;
+
+	u32 next_alloc_qpn;
+	u32 next_alloc_cqn;
+
+	spinlock_t db_bitmap_lock;
+	/* We provide max 64 uContexts that each has one SQ doorbell Page. */
+	DECLARE_BITMAP(sdb_page, ERDMA_DWQE_TYPE0_CNT);
+	/*
+	 * We provide max 496 uContexts that each has one SQ normal Db,
+	 * and one directWQE db。
+	 */
+	DECLARE_BITMAP(sdb_entry, ERDMA_DWQE_TYPE1_CNT);
+
+	atomic_t num_ctx;
+	struct list_head cep_list;
+};
+
+static inline void *get_queue_entry(void *qbuf, u32 idx, u32 depth, u32 shift)
+{
+	idx &= (depth - 1);
+
+	return qbuf + (idx << shift);
+}
+
+static inline struct erdma_dev *to_edev(struct ib_device *ibdev)
+{
+	return container_of(ibdev, struct erdma_dev, ibdev);
+}
+
+static inline u32 erdma_reg_read32(struct erdma_dev *dev, u32 reg)
+{
+	return readl(dev->func_bar + reg);
+}
+
+static inline u64 erdma_reg_read64(struct erdma_dev *dev, u32 reg)
+{
+	return readq(dev->func_bar + reg);
+}
+
+static inline void erdma_reg_write32(struct erdma_dev *dev, u32 reg, u32 value)
+{
+	writel(value, dev->func_bar + reg);
+}
+
+static inline void erdma_reg_write64(struct erdma_dev *dev, u32 reg, u64 value)
+{
+	writeq(value, dev->func_bar + reg);
+}
+
+static inline u32 erdma_reg_read32_filed(struct erdma_dev *dev, u32 reg,
+					 u32 filed_mask)
+{
+	u32 val = erdma_reg_read32(dev, reg);
+
+	return FIELD_GET(filed_mask, val);
+}
+
+int erdma_cmdq_init(struct erdma_dev *dev);
+void erdma_finish_cmdq_init(struct erdma_dev *dev);
+void erdma_cmdq_destroy(struct erdma_dev *dev);
+
+void erdma_cmdq_build_reqhdr(u64 *hdr, u32 mod, u32 op);
+int erdma_post_cmd_wait(struct erdma_cmdq *cmdq, u64 *req, u32 req_size,
+			u64 *resp0, u64 *resp1);
+void erdma_cmdq_completion_handler(struct erdma_cmdq *cmdq);
+
+int erdma_ceqs_init(struct erdma_dev *dev);
+void erdma_ceqs_uninit(struct erdma_dev *dev);
+void notify_eq(struct erdma_eq *eq);
+void *get_next_valid_eqe(struct erdma_eq *eq);
+
+int erdma_aeq_init(struct erdma_dev *dev);
+void erdma_aeq_destroy(struct erdma_dev *dev);
+
+void erdma_aeq_event_handler(struct erdma_dev *dev);
+void erdma_ceq_completion_handler(struct erdma_eq_cb *ceq_cb);
+
+#endif

drivers/infiniband/hw/erdma/erdma_cm.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+/* Authors: Bernard Metzler <[e-mail]> */
+/*          Fredy Neeser */
+/*          Greg Joyce <[e-mail]> */
+/* Copyright (c) 2008-2019, IBM Corporation */
+/* Copyright (c) 2017, Open Grid Computing, Inc. */
+
+#include <linux/errno.h>
+#include <linux/inetdevice.h>
+#include <linux/net.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <net/addrconf.h>
+
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
+
+#include "erdma.h"
+#include "erdma_cm.h"
+#include "erdma_verbs.h"
+
+static struct workqueue_struct *erdma_cm_wq;
+
+static void erdma_cm_llp_state_change(struct sock *sk);
+static void erdma_cm_llp_data_ready(struct sock *sk);
+static void erdma_cm_llp_error_report(struct sock *sk);
+
+static void erdma_sk_assign_cm_upcalls(struct sock *sk)
+{
+	write_lock_bh(&sk->sk_callback_lock);
+	sk->sk_state_change = erdma_cm_llp_state_change;
+	sk->sk_data_ready = erdma_cm_llp_data_ready;
+	sk->sk_error_report = erdma_cm_llp_error_report;
+	write_unlock_bh(&sk->sk_callback_lock);
+}
+
+static void erdma_sk_save_upcalls(struct sock *sk)
+{
+	struct erdma_cep *cep = sk_to_cep(sk);
+
+	write_lock_bh(&sk->sk_callback_lock);
+	cep->sk_state_change = sk->sk_state_change;
+	cep->sk_data_ready = sk->sk_data_ready;
+	cep->sk_error_report = sk->sk_error_report;
+	write_unlock_bh(&sk->sk_callback_lock);
+}
+
+static void erdma_sk_restore_upcalls(struct sock *sk, struct erdma_cep *cep)
+{
+	sk->sk_state_change = cep->sk_state_change;
+	sk->sk_data_ready = cep->sk_data_ready;
+	sk->sk_error_report = cep->sk_error_report;
+	sk->sk_user_data = NULL;
+}
+
+static void erdma_socket_disassoc(struct socket *s)
+{
+	struct sock *sk = s->sk;
+	struct erdma_cep *cep;
+
+	if (sk) {
+		write_lock_bh(&sk->sk_callback_lock);
+		cep = sk_to_cep(sk);
+		if (cep) {
+			erdma_sk_restore_upcalls(sk, cep);
+			erdma_cep_put(cep);
+		} else {
+			WARN_ON_ONCE(1);
+		}
+		write_unlock_bh(&sk->sk_callback_lock);
+	} else {
+		WARN_ON_ONCE(1);
+	}
+}
+
+static void erdma_cep_socket_assoc(struct erdma_cep *cep, struct socket *s)
+{
+	cep->sock = s;
+	erdma_cep_get(cep);
+	s->sk->sk_user_data = cep;
+
+	erdma_sk_save_upcalls(s->sk);
+	erdma_sk_assign_cm_upcalls(s->sk);
+}
+
+static void erdma_disassoc_listen_cep(struct erdma_cep *cep)
+{
+	if (cep->listen_cep) {
+		erdma_cep_put(cep->listen_cep);
+		cep->listen_cep = NULL;
+	}
+}
+
+static struct erdma_cep *erdma_cep_alloc(struct erdma_dev *dev)
+{
+	struct erdma_cep *cep = kzalloc(sizeof(*cep), GFP_KERNEL);
+	unsigned long flags;
+
+	if (!cep)
+		return NULL;
+
+	INIT_LIST_HEAD(&cep->listenq);
+	INIT_LIST_HEAD(&cep->devq);
+	INIT_LIST_HEAD(&cep->work_freelist);
+
+	kref_init(&cep->ref);
+	cep->state = ERDMA_EPSTATE_IDLE;
+	init_waitqueue_head(&cep->waitq);
+	spin_lock_init(&cep->lock);
+	cep->dev = dev;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_add_tail(&cep->devq, &dev->cep_list);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	return cep;
+}
+
+static void erdma_cm_free_work(struct erdma_cep *cep)
+{
+	struct list_head *w, *tmp;
+	struct erdma_cm_work *work;
+
+	list_for_each_safe(w, tmp, &cep->work_freelist) {
+		work = list_entry(w, struct erdma_cm_work, list);
+		list_del(&work->list);
+		kfree(work);
+	}
+}
+
+static void erdma_cancel_mpatimer(struct erdma_cep *cep)
+{
+	spin_lock_bh(&cep->lock);
+	if (cep->mpa_timer) {
+		if (cancel_delayed_work(&cep->mpa_timer->work)) {
+			erdma_cep_put(cep);
+			kfree(cep->mpa_timer);
+		}
+		cep->mpa_timer = NULL;
+	}
+	spin_unlock_bh(&cep->lock);
+}
+
+static void erdma_put_work(struct erdma_cm_work *work)
+{
+	INIT_LIST_HEAD(&work->list);
+	spin_lock_bh(&work->cep->lock);
+	list_add(&work->list, &work->cep->work_freelist);
+	spin_unlock_bh(&work->cep->lock);
+}
+
+static void erdma_cep_set_inuse(struct erdma_cep *cep)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cep->lock, flags);
+	while (cep->in_use) {
+		spin_unlock_irqrestore(&cep->lock, flags);
+		wait_event_interruptible(cep->waitq, !cep->in_use);
+		if (signal_pending(current))
+			flush_signals(current);
+
+		spin_lock_irqsave(&cep->lock, flags);
+	}
+
+	cep->in_use = 1;
+	spin_unlock_irqrestore(&cep->lock, flags);
+}
+
+static void erdma_cep_set_free(struct erdma_cep *cep)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cep->lock, flags);
+	cep->in_use = 0;
+	spin_unlock_irqrestore(&cep->lock, flags);
+
+	wake_up(&cep->waitq);
+}
+
+static void __erdma_cep_dealloc(struct kref *ref)
+{
+	struct erdma_cep *cep = container_of(ref, struct erdma_cep, ref);
+	struct erdma_dev *dev = cep->dev;
+	unsigned long flags;
+
+	WARN_ON(cep->listen_cep);
+
+	kfree(cep->private_data);
+	kfree(cep->mpa.pdata);
+	spin_lock_bh(&cep->lock);
+	if (!list_empty(&cep->work_freelist))
+		erdma_cm_free_work(cep);
+	spin_unlock_bh(&cep->lock);
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_del(&cep->devq);
+	spin_unlock_irqrestore(&dev->lock, flags);
+	kfree(cep);
+}
+
+static struct erdma_cm_work *erdma_get_work(struct erdma_cep *cep)
+{
+	struct erdma_cm_work *work = NULL;
+
+	spin_lock_bh(&cep->lock);
+	if (!list_empty(&cep->work_freelist)) {
+		work = list_entry(cep->work_freelist.next, struct erdma_cm_work,
+				  list);
+		list_del_init(&work->list);
+	}
+
+	spin_unlock_bh(&cep->lock);
+	return work;
+}
+
+static int erdma_cm_alloc_work(struct erdma_cep *cep, int num)
+{
+	struct erdma_cm_work *work;
+
+	while (num--) {
+		work = kmalloc(sizeof(*work), GFP_KERNEL);
+		if (!work) {
+			if (!(list_empty(&cep->work_freelist)))
+				erdma_cm_free_work(cep);
+			return -ENOMEM;
+		}
+		work->cep = cep;
+		INIT_LIST_HEAD(&work->list);
+		list_add(&work->list, &cep->work_freelist);
+	}
+
+	return 0;
+}
+
+static int erdma_cm_upcall(struct erdma_cep *cep, enum iw_cm_event_type reason,
+			   int status)
+{
+	struct iw_cm_event event;
+	struct iw_cm_id *cm_id;
+
+	memset(&event, 0, sizeof(event));
+	event.status = status;
+	event.event = reason;
+
+	if (reason == IW_CM_EVENT_CONNECT_REQUEST) {
+		event.provider_data = cep;
+		cm_id = cep->listen_cep->cm_id;
+
+		event.ird = cep->dev->attrs.max_ird;
+		event.ord = cep->dev->attrs.max_ord;
+	} else {
+		cm_id = cep->cm_id;
+	}
+
+	if (reason == IW_CM_EVENT_CONNECT_REQUEST ||
+	    reason == IW_CM_EVENT_CONNECT_REPLY) {
+		u16 pd_len = be16_to_cpu(cep->mpa.hdr.params.pd_len);
+
+		if (pd_len && cep->mpa.pdata) {
+			event.private_data_len = pd_len;
+			event.private_data = cep->mpa.pdata;
+		}
+
+		getname_local(cep->sock, &event.local_addr);
+		getname_peer(cep->sock, &event.remote_addr);
+	}
+
+	return cm_id->event_handler(cm_id, &event);
+}
+
+void erdma_qp_cm_drop(struct erdma_qp *qp)
+{
+	struct erdma_cep *cep = qp->cep;
+
+	if (!qp->cep)
+		return;
+
+	erdma_cep_set_inuse(cep);
+
+	/* already closed. */
+	if (cep->state == ERDMA_EPSTATE_CLOSED)
+		goto out;
+
+	if (cep->cm_id) {
+		switch (cep->state) {
+		case ERDMA_EPSTATE_AWAIT_MPAREP:
+			erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY,
+					-EINVAL);
+			break;
+		case ERDMA_EPSTATE_RDMA_MODE:
+			erdma_cm_upcall(cep, IW_CM_EVENT_CLOSE, 0);
+			break;
+		case ERDMA_EPSTATE_IDLE:
+		case ERDMA_EPSTATE_LISTENING:
+		case ERDMA_EPSTATE_CONNECTING:
+		case ERDMA_EPSTATE_AWAIT_MPAREQ:
+		case ERDMA_EPSTATE_RECVD_MPAREQ:
+		case ERDMA_EPSTATE_CLOSED:
+		default:
+			break;
+		}
+		cep->cm_id->rem_ref(cep->cm_id);
+		cep->cm_id = NULL;
+		erdma_cep_put(cep);
+	}
+	cep->state = ERDMA_EPSTATE_CLOSED;
+
+	if (cep->sock) {
+		erdma_socket_disassoc(cep->sock);
+		sock_release(cep->sock);
+		cep->sock = NULL;
+	}
+
+	if (cep->qp) {
+		cep->qp = NULL;
+		erdma_qp_put(qp);
+	}
+out:
+	erdma_cep_set_free(cep);
+}
+
+void erdma_cep_put(struct erdma_cep *cep)
+{
+	WARN_ON(kref_read(&cep->ref) < 1);
+	kref_put(&cep->ref, __erdma_cep_dealloc);
+}
+
+void erdma_cep_get(struct erdma_cep *cep)
+{
+	kref_get(&cep->ref);
+}
+
+static int erdma_send_mpareqrep(struct erdma_cep *cep, const void *pdata,
+				u8 pd_len)
+{
+	struct socket *s = cep->sock;
+	struct mpa_rr *rr = &cep->mpa.hdr;
+	struct kvec iov[3];
+	struct msghdr msg;
+	int iovec_num = 0;
+	int ret;
+	int mpa_len;
+
+	memset(&msg, 0, sizeof(msg));
+
+	rr->params.pd_len = cpu_to_be16(pd_len);
+
+	iov[iovec_num].iov_base = rr;
+	iov[iovec_num].iov_len = sizeof(*rr);
+	iovec_num++;
+	mpa_len = sizeof(*rr);
+
+	iov[iovec_num].iov_base = &cep->mpa.ext_data;
+	iov[iovec_num].iov_len = sizeof(cep->mpa.ext_data);
+	iovec_num++;
+	mpa_len += sizeof(cep->mpa.ext_data);
+
+	if (pd_len) {
+		iov[iovec_num].iov_base = (char *)pdata;
+		iov[iovec_num].iov_len = pd_len;
+		mpa_len += pd_len;
+		iovec_num++;
+	}
+
+	ret = kernel_sendmsg(s, &msg, iov, iovec_num, mpa_len);
+
+	return ret < 0 ? ret : 0;
+}
+
+static inline int ksock_recv(struct socket *sock, char *buf, size_t size,
+			     int flags)
+{
+	struct kvec iov = { buf, size };
+	struct msghdr msg = { .msg_name = NULL, .msg_flags = flags };
+
+	return kernel_recvmsg(sock, &msg, &iov, 1, size, flags);
+}
+
+static int __recv_mpa_hdr(struct erdma_cep *cep, int hdr_rcvd, char *hdr,
+			  int hdr_size, int *rcvd_out)
+{
+	struct socket *s = cep->sock;
+	int rcvd;
+
+	*rcvd_out = 0;
+	if (hdr_rcvd < hdr_size) {
+		rcvd = ksock_recv(s, hdr + hdr_rcvd, hdr_size - hdr_rcvd,
+				  MSG_DONTWAIT);
+		if (rcvd == -EAGAIN)
+			return -EAGAIN;
+
+		if (rcvd <= 0)
+			return -ECONNABORTED;
+
+		hdr_rcvd += rcvd;
+		*rcvd_out = rcvd;
+
+		if (hdr_rcvd < hdr_size)
+			return -EAGAIN;
+	}
+
+	return 0;
+}
+
+static void __mpa_rr_set_revision(__be16 *bits, u8 rev)
+{
+	*bits = (*bits & ~MPA_RR_MASK_REVISION) |
+		(cpu_to_be16(rev) & MPA_RR_MASK_REVISION);
+}
+
+static u8 __mpa_rr_revision(__be16 mpa_rr_bits)
+{
+	__be16 rev = mpa_rr_bits & MPA_RR_MASK_REVISION;
+
+	return (u8)be16_to_cpu(rev);
+}
+
+static void __mpa_ext_set_cc(__be32 *bits, u32 cc)
+{
+	*bits = (*bits & ~MPA_EXT_FLAG_CC) |
+		(cpu_to_be32(cc) & MPA_EXT_FLAG_CC);
+}
+
+static u8 __mpa_ext_cc(__be32 mpa_ext_bits)
+{
+	__be32 cc = mpa_ext_bits & MPA_EXT_FLAG_CC;
+
+	return (u8)be32_to_cpu(cc);
+}
+
+/*
+ * Receive MPA Request/Reply header.
+ *
+ * Returns 0 if complete MPA Request/Reply haeder including
+ * eventual private data was received. Returns -EAGAIN if
+ * header was partially received or negative error code otherwise.
+ *
+ * Context: May be called in process context only
+ */
+static int erdma_recv_mpa_rr(struct erdma_cep *cep)
+{
+	struct mpa_rr *hdr = &cep->mpa.hdr;
+	struct socket *s = cep->sock;
+	u16 pd_len;
+	int rcvd, to_rcv, ret, pd_rcvd;
+
+	if (cep->mpa.bytes_rcvd < sizeof(struct mpa_rr)) {
+		ret = __recv_mpa_hdr(cep, cep->mpa.bytes_rcvd,
+				     (char *)&cep->mpa.hdr,
+				     sizeof(struct mpa_rr), &rcvd);
+		cep->mpa.bytes_rcvd += rcvd;
+		if (ret)
+			return ret;
+	}
+
+	if (be16_to_cpu(hdr->params.pd_len) > MPA_MAX_PRIVDATA ||
+	    __mpa_rr_revision(hdr->params.bits) != MPA_REVISION_EXT_1)
+		return -EPROTO;
+
+	if (cep->mpa.bytes_rcvd - sizeof(struct mpa_rr) <
+	    sizeof(struct erdma_mpa_ext)) {
+		ret = __recv_mpa_hdr(
+			cep, cep->mpa.bytes_rcvd - sizeof(struct mpa_rr),
+			(char *)&cep->mpa.ext_data,
+			sizeof(struct erdma_mpa_ext), &rcvd);
+		cep->mpa.bytes_rcvd += rcvd;
+		if (ret)
+			return ret;
+	}
+
+	pd_len = be16_to_cpu(hdr->params.pd_len);
+	pd_rcvd = cep->mpa.bytes_rcvd - sizeof(struct mpa_rr) -
+		  sizeof(struct erdma_mpa_ext);
+	to_rcv = pd_len - pd_rcvd;
+
+	if (!to_rcv) {
+		/*
+		 * We have received the whole MPA Request/Reply message.
+		 * Check against peer protocol violation.
+		 */
+		u32 word;
+
+		ret = __recv_mpa_hdr(cep, 0, (char *)&word, sizeof(word),
+				     &rcvd);
+		if (ret == -EAGAIN && rcvd == 0)
+			return 0;
+
+		if (ret)
+			return ret;
+
+		return -EPROTO;
+	}
+
+	/*
+	 * At this point, MPA header has been fully received, and pd_len != 0.
+	 * So, begin to receive private data.
+	 */
+	if (!cep->mpa.pdata) {
+		cep->mpa.pdata = kmalloc(pd_len + 4, GFP_KERNEL);
+		if (!cep->mpa.pdata)
+			return -ENOMEM;
+	}
+
+	rcvd = ksock_recv(s, cep->mpa.pdata + pd_rcvd, to_rcv + 4,
+			  MSG_DONTWAIT);
+	if (rcvd < 0)
+		return rcvd;
+
+	if (rcvd > to_rcv)
+		return -EPROTO;
+
+	cep->mpa.bytes_rcvd += rcvd;
+
+	if (to_rcv == rcvd)
+		return 0;
+
+	return -EAGAIN;
+}
+
+/*
+ * erdma_proc_mpareq()
+ *
+ * Read MPA Request from socket and signal new connection to IWCM
+ * if success. Caller must hold lock on corresponding listening CEP.
+ */
+static int erdma_proc_mpareq(struct erdma_cep *cep)
+{
+	struct mpa_rr *req;
+	int ret;
+
+	ret = erdma_recv_mpa_rr(cep);
+	if (ret)
+		return ret;
+
+	req = &cep->mpa.hdr;
+
+	if (memcmp(req->key, MPA_KEY_REQ, MPA_KEY_SIZE))
+		return -EPROTO;
+
+	memcpy(req->key, MPA_KEY_REP, MPA_KEY_SIZE);
+
+	/* Currently does not support marker and crc. */
+	if (req->params.bits & MPA_RR_FLAG_MARKERS ||
+	    req->params.bits & MPA_RR_FLAG_CRC)
+		goto reject_conn;
+
+	cep->state = ERDMA_EPSTATE_RECVD_MPAREQ;
+
+	/* Keep reference until IWCM accepts/rejects */
+	erdma_cep_get(cep);
+	ret = erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REQUEST, 0);
+	if (ret)
+		erdma_cep_put(cep);
+
+	return ret;
+
+reject_conn:
+	req->params.bits &= ~MPA_RR_FLAG_MARKERS;
+	req->params.bits |= MPA_RR_FLAG_REJECT;
+	req->params.bits &= ~MPA_RR_FLAG_CRC;
+
+	kfree(cep->mpa.pdata);
+	cep->mpa.pdata = NULL;
+	erdma_send_mpareqrep(cep, NULL, 0);
+
+	return -EOPNOTSUPP;
+}
+
+static int erdma_proc_mpareply(struct erdma_cep *cep)
+{
+	struct erdma_qp_attrs qp_attrs;
+	struct erdma_qp *qp = cep->qp;
+	struct mpa_rr *rep;
+	int ret;
+
+	ret = erdma_recv_mpa_rr(cep);
+	if (ret)
+		goto out_err;
+
+	erdma_cancel_mpatimer(cep);
+
+	rep = &cep->mpa.hdr;
+
+	if (memcmp(rep->key, MPA_KEY_REP, MPA_KEY_SIZE)) {
+		ret = -EPROTO;
+		goto out_err;
+	}
+
+	if (rep->params.bits & MPA_RR_FLAG_REJECT) {
+		erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, -ECONNRESET);
+		return -ECONNRESET;
+	}
+
+	/* Currently does not support marker and crc. */
+	if ((rep->params.bits & MPA_RR_FLAG_MARKERS) ||
+	    (rep->params.bits & MPA_RR_FLAG_CRC)) {
+		erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, -ECONNREFUSED);
+		return -EINVAL;
+	}
+
+	memset(&qp_attrs, 0, sizeof(qp_attrs));
+	qp_attrs.irq_size = cep->ird;
+	qp_attrs.orq_size = cep->ord;
+	qp_attrs.state = ERDMA_QP_STATE_RTS;
+
+	down_write(&qp->state_lock);
+	if (qp->attrs.state > ERDMA_QP_STATE_RTR) {
+		ret = -EINVAL;
+		up_write(&qp->state_lock);
+		goto out_err;
+	}
+
+	qp->attrs.qp_type = ERDMA_QP_ACTIVE;
+	if (__mpa_ext_cc(cep->mpa.ext_data.bits) != qp->attrs.cc)
+		qp->attrs.cc = COMPROMISE_CC;
+
+	ret = erdma_modify_qp_internal(qp, &qp_attrs,
+				       ERDMA_QP_ATTR_STATE |
+				       ERDMA_QP_ATTR_LLP_HANDLE |
+				       ERDMA_QP_ATTR_MPA);
+
+	up_write(&qp->state_lock);
+
+	if (!ret) {
+		ret = erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, 0);
+		if (!ret)
+			cep->state = ERDMA_EPSTATE_RDMA_MODE;
+
+		return 0;
+	}
+
+out_err:
+	if (ret != -EAGAIN)
+		erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, -EINVAL);
+
+	return ret;
+}
+
+static void erdma_accept_newconn(struct erdma_cep *cep)
+{
+	struct socket *s = cep->sock;
+	struct socket *new_s = NULL;
+	struct erdma_cep *new_cep = NULL;
+	int ret = 0;
+
+	if (cep->state != ERDMA_EPSTATE_LISTENING)
+		goto error;
+
+	new_cep = erdma_cep_alloc(cep->dev);
+	if (!new_cep)
+		goto error;
+
+	/*
+	 * 4: Allocate a sufficient number of work elements
+	 * to allow concurrent handling of local + peer close
+	 * events, MPA header processing + MPA timeout.
+	 */
+	if (erdma_cm_alloc_work(new_cep, 4) != 0)
+		goto error;
+
+	/*
+	 * Copy saved socket callbacks from listening CEP
+	 * and assign new socket with new CEP
+	 */
+	new_cep->sk_state_change = cep->sk_state_change;
+	new_cep->sk_data_ready = cep->sk_data_ready;
+	new_cep->sk_error_report = cep->sk_error_report;
+
+	ret = kernel_accept(s, &new_s, O_NONBLOCK);
+	if (ret != 0)
+		goto error;
+
+	new_cep->sock = new_s;
+	erdma_cep_get(new_cep);
+	new_s->sk->sk_user_data = new_cep;
+
+	tcp_sock_set_nodelay(new_s->sk);
+	new_cep->state = ERDMA_EPSTATE_AWAIT_MPAREQ;
+
+	ret = erdma_cm_queue_work(new_cep, ERDMA_CM_WORK_MPATIMEOUT);
+	if (ret)
+		goto error;
+
+	new_cep->listen_cep = cep;
+	erdma_cep_get(cep);
+
+	if (atomic_read(&new_s->sk->sk_rmem_alloc)) {
+		/* MPA REQ already queued */
+		erdma_cep_set_inuse(new_cep);
+		ret = erdma_proc_mpareq(new_cep);
+		if (ret != -EAGAIN) {
+			erdma_cep_put(cep);
+			new_cep->listen_cep = NULL;
+			if (ret) {
+				erdma_cep_set_free(new_cep);
+				goto error;
+			}
+		}
+		erdma_cep_set_free(new_cep);
+	}
+	return;
+
+error:
+	if (new_cep) {
+		new_cep->state = ERDMA_EPSTATE_CLOSED;
+		erdma_cancel_mpatimer(new_cep);
+
+		erdma_cep_put(new_cep);
+		new_cep->sock = NULL;
+	}
+
+	if (new_s) {
+		erdma_socket_disassoc(new_s);
+		sock_release(new_s);
+	}
+}
+
+static int erdma_newconn_connected(struct erdma_cep *cep)
+{
+	int ret = 0;
+
+	cep->mpa.hdr.params.bits = 0;
+	__mpa_rr_set_revision(&cep->mpa.hdr.params.bits, MPA_REVISION_EXT_1);
+
+	memcpy(cep->mpa.hdr.key, MPA_KEY_REQ, MPA_KEY_SIZE);
+	cep->mpa.ext_data.cookie = cpu_to_be32(cep->qp->attrs.cookie);
+	__mpa_ext_set_cc(&cep->mpa.ext_data.bits, cep->qp->attrs.cc);
+
+	ret = erdma_send_mpareqrep(cep, cep->private_data, cep->pd_len);
+	cep->state = ERDMA_EPSTATE_AWAIT_MPAREP;
+	cep->mpa.hdr.params.pd_len = 0;
+
+	if (ret >= 0)
+		ret = erdma_cm_queue_work(cep, ERDMA_CM_WORK_MPATIMEOUT);
+
+	return ret;
+}
+
+static void erdma_cm_work_handler(struct work_struct *w)
+{
+	struct erdma_cm_work *work;
+	struct erdma_cep *cep;
+	int release_cep = 0, ret = 0;
+
+	work = container_of(w, struct erdma_cm_work, work.work);
+	cep = work->cep;
+
+	erdma_cep_set_inuse(cep);
+
+	switch (work->type) {
+	case ERDMA_CM_WORK_CONNECTED:
+		erdma_cancel_mpatimer(cep);
+		if (cep->state == ERDMA_EPSTATE_CONNECTING) {
+			ret = erdma_newconn_connected(cep);
+			if (ret) {
+				erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY,
+						-EIO);
+				release_cep = 1;
+			}
+		}
+		break;
+	case ERDMA_CM_WORK_CONNECTTIMEOUT:
+		if (cep->state == ERDMA_EPSTATE_CONNECTING) {
+			cep->mpa_timer = NULL;
+			erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY,
+					-ETIMEDOUT);
+			release_cep = 1;
+		}
+		break;
+	case ERDMA_CM_WORK_ACCEPT:
+		erdma_accept_newconn(cep);
+		break;
+	case ERDMA_CM_WORK_READ_MPAHDR:
+		if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREQ) {
+			if (cep->listen_cep) {
+				erdma_cep_set_inuse(cep->listen_cep);
+
+				if (cep->listen_cep->state ==
+				    ERDMA_EPSTATE_LISTENING)
+					ret = erdma_proc_mpareq(cep);
+				else
+					ret = -EFAULT;
+
+				erdma_cep_set_free(cep->listen_cep);
+
+				if (ret != -EAGAIN) {
+					erdma_cep_put(cep->listen_cep);
+					cep->listen_cep = NULL;
+					if (ret)
+						erdma_cep_put(cep);
+				}
+			}
+		} else if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREP) {
+			ret = erdma_proc_mpareply(cep);
+		}
+
+		if (ret && ret != -EAGAIN)
+			release_cep = 1;
+		break;
+	case ERDMA_CM_WORK_CLOSE_LLP:
+		if (cep->cm_id)
+			erdma_cm_upcall(cep, IW_CM_EVENT_CLOSE, 0);
+		release_cep = 1;
+		break;
+	case ERDMA_CM_WORK_PEER_CLOSE:
+		if (cep->cm_id) {
+			if (cep->state == ERDMA_EPSTATE_CONNECTING ||
+			    cep->state == ERDMA_EPSTATE_AWAIT_MPAREP) {
+				/*
+				 * MPA reply not received, but connection drop
+				 */
+				erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY,
+						-ECONNRESET);
+			} else if (cep->state == ERDMA_EPSTATE_RDMA_MODE) {
+				/*
+				 * NOTE: IW_CM_EVENT_DISCONNECT is given just
+				 *       to transition IWCM into CLOSING.
+				 */
+				erdma_cm_upcall(cep, IW_CM_EVENT_DISCONNECT, 0);
+				erdma_cm_upcall(cep, IW_CM_EVENT_CLOSE, 0);
+			}
+		} else if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREQ) {
+			/* Socket close before MPA request received. */
+			erdma_disassoc_listen_cep(cep);
+			erdma_cep_put(cep);
+		}
+		release_cep = 1;
+		break;
+	case ERDMA_CM_WORK_MPATIMEOUT:
+		cep->mpa_timer = NULL;
+		if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREP) {
+			/*
+			 * MPA request timed out:
+			 * Hide any partially received private data and signal
+			 * timeout
+			 */
+			cep->mpa.hdr.params.pd_len = 0;
+
+			if (cep->cm_id)
+				erdma_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY,
+						-ETIMEDOUT);
+			release_cep = 1;
+		} else if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREQ) {
+			/* No MPA req received after peer TCP stream setup. */
+			erdma_disassoc_listen_cep(cep);
+
+			erdma_cep_put(cep);
+			release_cep = 1;
+		}
+		break;
+	default:
+		WARN(1, "Undefined CM work type: %d\n", work->type);
+	}
+
+	if (release_cep) {
+		erdma_cancel_mpatimer(cep);
+		cep->state = ERDMA_EPSTATE_CLOSED;
+		if (cep->qp) {
+			struct erdma_qp *qp = cep->qp;
+			/*
+			 * Serialize a potential race with application
+			 * closing the QP and calling erdma_qp_cm_drop()
+			 */
+			erdma_qp_get(qp);
+			erdma_cep_set_free(cep);
+
+			erdma_qp_llp_close(qp);
+			erdma_qp_put(qp);
+
+			erdma_cep_set_inuse(cep);
+			cep->qp = NULL;
+			erdma_qp_put(qp);
+		}
+
+		if (cep->sock) {
+			erdma_socket_disassoc(cep->sock);
+			sock_release(cep->sock);
+			cep->sock = NULL;
+		}
+
+		if (cep->cm_id) {
+			cep->cm_id->rem_ref(cep->cm_id);
+			cep->cm_id = NULL;
+			if (cep->state != ERDMA_EPSTATE_LISTENING)
+				erdma_cep_put(cep);
+		}
+	}
+	erdma_cep_set_free(cep);
+	erdma_put_work(work);
+	erdma_cep_put(cep);
+}
+
+int erdma_cm_queue_work(struct erdma_cep *cep, enum erdma_work_type type)
+{
+	struct erdma_cm_work *work = erdma_get_work(cep);
+	unsigned long delay = 0;
+
+	if (!work)
+		return -ENOMEM;
+
+	work->type = type;
+	work->cep = cep;
+
+	erdma_cep_get(cep);
+
+	INIT_DELAYED_WORK(&work->work, erdma_cm_work_handler);
+
+	if (type == ERDMA_CM_WORK_MPATIMEOUT) {
+		cep->mpa_timer = work;
+
+		if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREP)
+			delay = MPAREP_TIMEOUT;
+		else
+			delay = MPAREQ_TIMEOUT;
+	} else if (type == ERDMA_CM_WORK_CONNECTTIMEOUT) {
+		cep->mpa_timer = work;
+
+		delay = CONNECT_TIMEOUT;
+	}
+
+	queue_delayed_work(erdma_cm_wq, &work->work, delay);
+
+	return 0;
+}
+
+static void erdma_cm_llp_data_ready(struct sock *sk)
+{
+	struct erdma_cep *cep;
+
+	read_lock(&sk->sk_callback_lock);
+
+	cep = sk_to_cep(sk);
+	if (!cep)
+		goto out;
+
+	if (cep->state == ERDMA_EPSTATE_AWAIT_MPAREQ ||
+	    cep->state == ERDMA_EPSTATE_AWAIT_MPAREP)
+		erdma_cm_queue_work(cep, ERDMA_CM_WORK_READ_MPAHDR);
+
+out:
+	read_unlock(&sk->sk_callback_lock);
+}
+
+static void erdma_cm_llp_error_report(struct sock *sk)
+{
+	struct erdma_cep *cep = sk_to_cep(sk);
+
+	if (cep)
+		cep->sk_error_report(sk);
+}
+
+static void erdma_cm_llp_state_change(struct sock *sk)
+{
+	struct erdma_cep *cep;
+	void (*orig_state_change)(struct sock *sk);
+
+	read_lock(&sk->sk_callback_lock);
+
+	cep = sk_to_cep(sk);
+	if (!cep) {
+		read_unlock(&sk->sk_callback_lock);
+		return;
+	}
+	orig_state_change = cep->sk_state_change;
+
+	switch (sk->sk_state) {
+	case TCP_ESTABLISHED:
+		if (cep->state == ERDMA_EPSTATE_CONNECTING)
+			erdma_cm_queue_work(cep, ERDMA_CM_WORK_CONNECTED);
+		else
+			erdma_cm_queue_work(cep, ERDMA_CM_WORK_ACCEPT);
+		break;
+	case TCP_CLOSE:
+	case TCP_CLOSE_WAIT:
+		if (cep->state != ERDMA_EPSTATE_LISTENING)
+			erdma_cm_queue_work(cep, ERDMA_CM_WORK_PEER_CLOSE);
+		break;
+	default:
+		break;
+	}
+	read_unlock(&sk->sk_callback_lock);
+	orig_state_change(sk);
+}
+
+static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr,
+			      int laddrlen, struct sockaddr *raddr,
+			      int raddrlen, int flags)
+{
+	int ret;
+
+	sock_set_reuseaddr(s->sk);
+	ret = s->ops->bind(s, laddr, laddrlen);
+	if (ret)
+		return ret;
+	ret = s->ops->connect(s, raddr, raddrlen, flags);
+	return ret < 0 ? ret : 0;
+}
+
+int erdma_connect(struct iw_cm_id *id, struct iw_cm_conn_param *params)
+{
+	struct erdma_dev *dev = to_edev(id->device);
+	struct erdma_qp *qp;
+	struct erdma_cep *cep = NULL;
+	struct socket *s = NULL;
+	struct sockaddr *laddr = (struct sockaddr *)&id->m_local_addr;
+	struct sockaddr *raddr = (struct sockaddr *)&id->m_remote_addr;
+	u16 pd_len = params->private_data_len;
+	int ret;
+
+	if (pd_len > MPA_MAX_PRIVDATA)
+		return -EINVAL;
+
+	if (params->ird > dev->attrs.max_ird ||
+	    params->ord > dev->attrs.max_ord)
+		return -EINVAL;
+
+	if (laddr->sa_family != AF_INET || raddr->sa_family != AF_INET)
+		return -EAFNOSUPPORT;
+
+	qp = find_qp_by_qpn(dev, params->qpn);
+	if (!qp)
+		return -ENOENT;
+	erdma_qp_get(qp);
+
+	ret = sock_create(AF_INET, SOCK_STREAM, IPPROTO_TCP, &s);
+	if (ret < 0)
+		goto error_put_qp;
+
+	cep = erdma_cep_alloc(dev);
+	if (!cep) {
+		ret = -ENOMEM;
+		goto error_release_sock;
+	}
+
+	erdma_cep_set_inuse(cep);
+
+	/* Associate QP with CEP */
+	erdma_cep_get(cep);
+	qp->cep = cep;
+	cep->qp = qp;
+
+	/* Associate cm_id with CEP */
+	id->add_ref(id);
+	cep->cm_id = id;
+
+	/*
+	 * 6: Allocate a sufficient number of work elements
+	 * to allow concurrent handling of local + peer close
+	 * events, MPA header processing + MPA timeout, connected event
+	 * and connect timeout.
+	 */
+	ret = erdma_cm_alloc_work(cep, 6);
+	if (ret != 0) {
+		ret = -ENOMEM;
+		goto error_release_cep;
+	}
+
+	cep->ird = params->ird;
+	cep->ord = params->ord;
+	cep->state = ERDMA_EPSTATE_CONNECTING;
+
+	erdma_cep_socket_assoc(cep, s);
+
+	if (pd_len) {
+		cep->pd_len = pd_len;
+		cep->private_data = kmalloc(pd_len, GFP_KERNEL);
+		if (!cep->private_data) {
+			ret = -ENOMEM;
+			goto error_disassoc;
+		}
+
+		memcpy(cep->private_data, params->private_data,
+		       params->private_data_len);
+	}
+
+	ret = kernel_bindconnect(s, laddr, sizeof(*laddr), raddr,
+				 sizeof(*raddr), O_NONBLOCK);
+	if (ret != -EINPROGRESS && ret != 0) {
+		goto error_disassoc;
+	} else if (ret == 0) {
+		ret = erdma_cm_queue_work(cep, ERDMA_CM_WORK_CONNECTED);
+		if (ret)
+			goto error_disassoc;
+	} else {
+		ret = erdma_cm_queue_work(cep, ERDMA_CM_WORK_CONNECTTIMEOUT);
+		if (ret)
+			goto error_disassoc;
+	}
+
+	erdma_cep_set_free(cep);
+	return 0;
+
+error_disassoc:
+	kfree(cep->private_data);
+	cep->private_data = NULL;
+	cep->pd_len = 0;
+
+	erdma_socket_disassoc(s);
+
+error_release_cep:
+	/* disassoc with cm_id */
+	cep->cm_id = NULL;
+	id->rem_ref(id);
+
+	/* disassoc with qp */
+	qp->cep = NULL;
+	erdma_cep_put(cep);
+	cep->qp = NULL;
+
+	cep->state = ERDMA_EPSTATE_CLOSED;
+
+	erdma_cep_set_free(cep);
+
+	/* release the cep. */
+	erdma_cep_put(cep);
+
+error_release_sock:
+	if (s)
+		sock_release(s);
+error_put_qp:
+	erdma_qp_put(qp);
+
+	return ret;
+}
+
+int erdma_accept(struct iw_cm_id *id, struct iw_cm_conn_param *params)
+{
+	struct erdma_dev *dev = to_edev(id->device);
+	struct erdma_cep *cep = (struct erdma_cep *)id->provider_data;
+	struct erdma_qp *qp;
+	struct erdma_qp_attrs qp_attrs;
+	int ret;
+
+	erdma_cep_set_inuse(cep);
+	erdma_cep_put(cep);
+
+	/* Free lingering inbound private data */
+	if (cep->mpa.hdr.params.pd_len) {
+		cep->mpa.hdr.params.pd_len = 0;
+		kfree(cep->mpa.pdata);
+		cep->mpa.pdata = NULL;
+	}
+	erdma_cancel_mpatimer(cep);
+
+	if (cep->state != ERDMA_EPSTATE_RECVD_MPAREQ) {
+		erdma_cep_set_free(cep);
+		erdma_cep_put(cep);
+
+		return -ECONNRESET;
+	}
+
+	qp = find_qp_by_qpn(dev, params->qpn);
+	if (!qp)
+		return -ENOENT;
+	erdma_qp_get(qp);
+
+	down_write(&qp->state_lock);
+	if (qp->attrs.state > ERDMA_QP_STATE_RTR) {
+		ret = -EINVAL;
+		up_write(&qp->state_lock);
+		goto error;
+	}
+
+	if (params->ord > dev->attrs.max_ord ||
+	    params->ird > dev->attrs.max_ord) {
+		ret = -EINVAL;
+		up_write(&qp->state_lock);
+		goto error;
+	}
+
+	if (params->private_data_len > MPA_MAX_PRIVDATA) {
+		ret = -EINVAL;
+		up_write(&qp->state_lock);
+		goto error;
+	}
+
+	cep->ird = params->ird;
+	cep->ord = params->ord;
+
+	cep->cm_id = id;
+	id->add_ref(id);
+
+	memset(&qp_attrs, 0, sizeof(qp_attrs));
+	qp_attrs.orq_size = params->ord;
+	qp_attrs.irq_size = params->ird;
+
+	qp_attrs.state = ERDMA_QP_STATE_RTS;
+
+	/* Associate QP with CEP */
+	erdma_cep_get(cep);
+	qp->cep = cep;
+	cep->qp = qp;
+
+	cep->state = ERDMA_EPSTATE_RDMA_MODE;
+
+	qp->attrs.qp_type = ERDMA_QP_PASSIVE;
+	qp->attrs.pd_len = params->private_data_len;
+
+	if (qp->attrs.cc != __mpa_ext_cc(cep->mpa.ext_data.bits))
+		qp->attrs.cc = COMPROMISE_CC;
+
+	/* move to rts */
+	ret = erdma_modify_qp_internal(qp, &qp_attrs,
+				       ERDMA_QP_ATTR_STATE |
+				       ERDMA_QP_ATTR_ORD |
+				       ERDMA_QP_ATTR_LLP_HANDLE |
+				       ERDMA_QP_ATTR_IRD |
+				       ERDMA_QP_ATTR_MPA);
+	up_write(&qp->state_lock);
+
+	if (ret)
+		goto error;
+
+	cep->mpa.ext_data.bits = 0;
+	__mpa_ext_set_cc(&cep->mpa.ext_data.bits, qp->attrs.cc);
+	cep->mpa.ext_data.cookie = cpu_to_be32(cep->qp->attrs.cookie);
+
+	ret = erdma_send_mpareqrep(cep, params->private_data,
+				   params->private_data_len);
+	if (!ret) {
+		ret = erdma_cm_upcall(cep, IW_CM_EVENT_ESTABLISHED, 0);
+		if (ret)
+			goto error;
+
+		erdma_cep_set_free(cep);
+
+		return 0;
+	}
+
+error:
+	erdma_socket_disassoc(cep->sock);
+	sock_release(cep->sock);
+	cep->sock = NULL;
+
+	cep->state = ERDMA_EPSTATE_CLOSED;
+
+	if (cep->cm_id) {
+		cep->cm_id->rem_ref(id);
+		cep->cm_id = NULL;
+	}
+
+	if (qp->cep) {
+		erdma_cep_put(cep);
+		qp->cep = NULL;
+	}
+
+	cep->qp = NULL;
+	erdma_qp_put(qp);
+
+	erdma_cep_set_free(cep);
+	erdma_cep_put(cep);
+
+	return ret;
+}
+
+int erdma_reject(struct iw_cm_id *id, const void *pdata, u8 plen)
+{
+	struct erdma_cep *cep = (struct erdma_cep *)id->provider_data;
+
+	erdma_cep_set_inuse(cep);
+	erdma_cep_put(cep);
+
+	erdma_cancel_mpatimer(cep);
+
+	if (cep->state != ERDMA_EPSTATE_RECVD_MPAREQ) {
+		erdma_cep_set_free(cep);
+		erdma_cep_put(cep);
+
+		return -ECONNRESET;
+	}
+
+	if (__mpa_rr_revision(cep->mpa.hdr.params.bits) == MPA_REVISION_EXT_1) {
+		cep->mpa.hdr.params.bits |= MPA_RR_FLAG_REJECT; /* reject */
+		erdma_send_mpareqrep(cep, pdata, plen);
+	}
+
+	erdma_socket_disassoc(cep->sock);
+	sock_release(cep->sock);
+	cep->sock = NULL;
+
+	cep->state = ERDMA_EPSTATE_CLOSED;
+
+	erdma_cep_set_free(cep);
+	erdma_cep_put(cep);
+
+	return 0;
+}
+
+int erdma_create_listen(struct iw_cm_id *id, int backlog)
+{
+	struct socket *s;
+	struct erdma_cep *cep = NULL;
+	int ret = 0;
+	struct erdma_dev *dev = to_edev(id->device);
+	int addr_family = id->local_addr.ss_family;
+	struct sockaddr_in *laddr = &to_sockaddr_in(id->local_addr);
+
+	if (addr_family != AF_INET)
+		return -EAFNOSUPPORT;
+
+	ret = sock_create(addr_family, SOCK_STREAM, IPPROTO_TCP, &s);
+	if (ret < 0)
+		return ret;
+
+	sock_set_reuseaddr(s->sk);
+
+	/* For wildcard addr, limit binding to current device only */
+	if (ipv4_is_zeronet(laddr->sin_addr.s_addr))
+		s->sk->sk_bound_dev_if = dev->netdev->ifindex;
+
+	ret = s->ops->bind(s, (struct sockaddr *)laddr,
+			   sizeof(struct sockaddr_in));
+	if (ret)
+		goto error;
+
+	cep = erdma_cep_alloc(dev);
+	if (!cep) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	erdma_cep_socket_assoc(cep, s);
+
+	ret = erdma_cm_alloc_work(cep, backlog);
+	if (ret)
+		goto error;
+
+	ret = s->ops->listen(s, backlog);
+	if (ret)
+		goto error;
+
+	cep->cm_id = id;
+	id->add_ref(id);
+
+	if (!id->provider_data) {
+		id->provider_data =
+			kmalloc(sizeof(struct list_head), GFP_KERNEL);
+		if (!id->provider_data) {
+			ret = -ENOMEM;
+			goto error;
+		}
+		INIT_LIST_HEAD((struct list_head *)id->provider_data);
+	}
+
+	list_add_tail(&cep->listenq, (struct list_head *)id->provider_data);
+	cep->state = ERDMA_EPSTATE_LISTENING;
+
+	return 0;
+
+error:
+	if (cep) {
+		erdma_cep_set_inuse(cep);
+
+		if (cep->cm_id) {
+			cep->cm_id->rem_ref(cep->cm_id);
+			cep->cm_id = NULL;
+		}
+		cep->sock = NULL;
+		erdma_socket_disassoc(s);
+		cep->state = ERDMA_EPSTATE_CLOSED;
+
+		erdma_cep_set_free(cep);
+		erdma_cep_put(cep);
+	}
+	sock_release(s);
+
+	return ret;
+}
+
+static void erdma_drop_listeners(struct iw_cm_id *id)
+{
+	struct list_head *p, *tmp;
+	/*
+	 * In case of a wildcard rdma_listen on a multi-homed device,
+	 * a listener's IWCM id is associated with more than one listening CEP.
+	 */
+	list_for_each_safe(p, tmp, (struct list_head *)id->provider_data) {
+		struct erdma_cep *cep =
+			list_entry(p, struct erdma_cep, listenq);
+
+		list_del(p);
+
+		erdma_cep_set_inuse(cep);
+
+		if (cep->cm_id) {
+			cep->cm_id->rem_ref(cep->cm_id);
+			cep->cm_id = NULL;
+		}
+		if (cep->sock) {
+			erdma_socket_disassoc(cep->sock);
+			sock_release(cep->sock);
+			cep->sock = NULL;
+		}
+		cep->state = ERDMA_EPSTATE_CLOSED;
+		erdma_cep_set_free(cep);
+		erdma_cep_put(cep);
+	}
+}
+
+int erdma_destroy_listen(struct iw_cm_id *id)
+{
+	if (!id->provider_data)
+		return 0;
+
+	erdma_drop_listeners(id);
+	kfree(id->provider_data);
+	id->provider_data = NULL;
+
+	return 0;
+}
+
+int erdma_cm_init(void)
+{
+	erdma_cm_wq = create_singlethread_workqueue("erdma_cm_wq");
+	if (!erdma_cm_wq)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void erdma_cm_exit(void)
+{
+	if (erdma_cm_wq)
+		destroy_workqueue(erdma_cm_wq);
+}

drivers/infiniband/hw/erdma/erdma_cm.h

···
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+/* Authors: Bernard Metzler <[e-mail]> */
+/*          Greg Joyce <[e-mail]> */
+/* Copyright (c) 2008-2019, IBM Corporation */
+/* Copyright (c) 2017, Open Grid Computing, Inc. */
+
+#ifndef __ERDMA_CM_H__
+#define __ERDMA_CM_H__
+
+#include <linux/tcp.h>
+#include <net/sock.h>
+#include <rdma/iw_cm.h>
+
+/* iWarp MPA protocol defs */
+#define MPA_REVISION_EXT_1 129
+#define MPA_MAX_PRIVDATA RDMA_MAX_PRIVATE_DATA
+#define MPA_KEY_REQ "MPA ID Req Frame"
+#define MPA_KEY_REP "MPA ID Rep Frame"
+#define MPA_KEY_SIZE 16
+#define MPA_DEFAULT_HDR_LEN 28
+
+struct mpa_rr_params {
+	__be16 bits;
+	__be16 pd_len;
+};
+
+/*
+ * MPA request/response Hdr bits & fields
+ */
+enum {
+	MPA_RR_FLAG_MARKERS = __cpu_to_be16(0x8000),
+	MPA_RR_FLAG_CRC = __cpu_to_be16(0x4000),
+	MPA_RR_FLAG_REJECT = __cpu_to_be16(0x2000),
+	MPA_RR_RESERVED = __cpu_to_be16(0x1f00),
+	MPA_RR_MASK_REVISION = __cpu_to_be16(0x00ff)
+};
+
+/*
+ * MPA request/reply header
+ */
+struct mpa_rr {
+	u8 key[16];
+	struct mpa_rr_params params;
+};
+
+struct erdma_mpa_ext {
+	__be32 cookie;
+	__be32 bits;
+};
+
+enum {
+	MPA_EXT_FLAG_CC = cpu_to_be32(0x0000000f),
+};
+
+struct erdma_mpa_info {
+	struct mpa_rr hdr; /* peer mpa hdr in host byte order */
+	struct erdma_mpa_ext ext_data;
+	char *pdata;
+	int bytes_rcvd;
+};
+
+struct erdma_sk_upcalls {
+	void (*sk_state_change)(struct sock *sk);
+	void (*sk_data_ready)(struct sock *sk, int bytes);
+	void (*sk_error_report)(struct sock *sk);
+};
+
+struct erdma_dev;
+
+enum erdma_cep_state {
+	ERDMA_EPSTATE_IDLE = 1,
+	ERDMA_EPSTATE_LISTENING,
+	ERDMA_EPSTATE_CONNECTING,
+	ERDMA_EPSTATE_AWAIT_MPAREQ,
+	ERDMA_EPSTATE_RECVD_MPAREQ,
+	ERDMA_EPSTATE_AWAIT_MPAREP,
+	ERDMA_EPSTATE_RDMA_MODE,
+	ERDMA_EPSTATE_CLOSED
+};
+
+struct erdma_cep {
+	struct iw_cm_id *cm_id;
+	struct erdma_dev *dev;
+	struct list_head devq;
+	spinlock_t lock;
+	struct kref ref;
+	int in_use;
+	wait_queue_head_t waitq;
+	enum erdma_cep_state state;
+
+	struct list_head listenq;
+	struct erdma_cep *listen_cep;
+
+	struct erdma_qp *qp;
+	struct socket *sock;
+
+	struct erdma_cm_work *mpa_timer;
+	struct list_head work_freelist;
+
+	struct erdma_mpa_info mpa;
+	int ord;
+	int ird;
+
+	int pd_len;
+	/* hold user's private data. */
+	void *private_data;
+
+	/* Saved upcalls of socket llp.sock */
+	void (*sk_state_change)(struct sock *sk);
+	void (*sk_data_ready)(struct sock *sk);
+	void (*sk_error_report)(struct sock *sk);
+};
+
+#define MPAREQ_TIMEOUT (HZ * 20)
+#define MPAREP_TIMEOUT (HZ * 10)
+#define CONNECT_TIMEOUT (HZ * 10)
+
+enum erdma_work_type {
+	ERDMA_CM_WORK_ACCEPT = 1,
+	ERDMA_CM_WORK_READ_MPAHDR,
+	ERDMA_CM_WORK_CLOSE_LLP, /* close socket */
+	ERDMA_CM_WORK_PEER_CLOSE, /* socket indicated peer close */
+	ERDMA_CM_WORK_MPATIMEOUT,
+	ERDMA_CM_WORK_CONNECTED,
+	ERDMA_CM_WORK_CONNECTTIMEOUT
+};
+
+struct erdma_cm_work {
+	struct delayed_work work;
+	struct list_head list;
+	enum erdma_work_type type;
+	struct erdma_cep *cep;
+};
+
+#define to_sockaddr_in(a) (*(struct sockaddr_in *)(&(a)))
+
+static inline int getname_peer(struct socket *s, struct sockaddr_storage *a)
+{
+	return s->ops->getname(s, (struct sockaddr *)a, 1);
+}
+
+static inline int getname_local(struct socket *s, struct sockaddr_storage *a)
+{
+	return s->ops->getname(s, (struct sockaddr *)a, 0);
+}
+
+int erdma_connect(struct iw_cm_id *id, struct iw_cm_conn_param *param);
+int erdma_accept(struct iw_cm_id *id, struct iw_cm_conn_param *param);
+int erdma_reject(struct iw_cm_id *id, const void *pdata, u8 plen);
+int erdma_create_listen(struct iw_cm_id *id, int backlog);
+int erdma_destroy_listen(struct iw_cm_id *id);
+
+void erdma_cep_get(struct erdma_cep *ceq);
+void erdma_cep_put(struct erdma_cep *ceq);
+int erdma_cm_queue_work(struct erdma_cep *ceq, enum erdma_work_type type);
+
+int erdma_cm_init(void);
+void erdma_cm_exit(void);
+
+#define sk_to_cep(sk) ((struct erdma_cep *)((sk)->sk_user_data))
+
+#endif

drivers/infiniband/hw/erdma/erdma_cmdq.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+
+#include "erdma.h"
+#include "erdma_hw.h"
+#include "erdma_verbs.h"
+
+static void arm_cmdq_cq(struct erdma_cmdq *cmdq)
+{
+	struct erdma_dev *dev = container_of(cmdq, struct erdma_dev, cmdq);
+	u64 db_data = FIELD_PREP(ERDMA_CQDB_CI_MASK, cmdq->cq.ci) |
+		      FIELD_PREP(ERDMA_CQDB_ARM_MASK, 1) |
+		      FIELD_PREP(ERDMA_CQDB_CMDSN_MASK, cmdq->cq.cmdsn) |
+		      FIELD_PREP(ERDMA_CQDB_IDX_MASK, cmdq->cq.cmdsn);
+
+	*cmdq->cq.db_record = db_data;
+	writeq(db_data, dev->func_bar + ERDMA_CMDQ_CQDB_REG);
+
+	atomic64_inc(&cmdq->cq.armed_num);
+}
+
+static void kick_cmdq_db(struct erdma_cmdq *cmdq)
+{
+	struct erdma_dev *dev = container_of(cmdq, struct erdma_dev, cmdq);
+	u64 db_data = FIELD_PREP(ERDMA_CMD_HDR_WQEBB_INDEX_MASK, cmdq->sq.pi);
+
+	*cmdq->sq.db_record = db_data;
+	writeq(db_data, dev->func_bar + ERDMA_CMDQ_SQDB_REG);
+}
+
+static struct erdma_comp_wait *get_comp_wait(struct erdma_cmdq *cmdq)
+{
+	int comp_idx;
+
+	spin_lock(&cmdq->lock);
+	comp_idx = find_first_zero_bit(cmdq->comp_wait_bitmap,
+				       cmdq->max_outstandings);
+	if (comp_idx == cmdq->max_outstandings) {
+		spin_unlock(&cmdq->lock);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	__set_bit(comp_idx, cmdq->comp_wait_bitmap);
+	spin_unlock(&cmdq->lock);
+
+	return &cmdq->wait_pool[comp_idx];
+}
+
+static void put_comp_wait(struct erdma_cmdq *cmdq,
+			  struct erdma_comp_wait *comp_wait)
+{
+	int used;
+
+	cmdq->wait_pool[comp_wait->ctx_id].cmd_status = ERDMA_CMD_STATUS_INIT;
+	spin_lock(&cmdq->lock);
+	used = __test_and_clear_bit(comp_wait->ctx_id, cmdq->comp_wait_bitmap);
+	spin_unlock(&cmdq->lock);
+
+	WARN_ON(!used);
+}
+
+static int erdma_cmdq_wait_res_init(struct erdma_dev *dev,
+				    struct erdma_cmdq *cmdq)
+{
+	int i;
+
+	cmdq->wait_pool =
+		devm_kcalloc(&dev->pdev->dev, cmdq->max_outstandings,
+			     sizeof(struct erdma_comp_wait), GFP_KERNEL);
+	if (!cmdq->wait_pool)
+		return -ENOMEM;
+
+	spin_lock_init(&cmdq->lock);
+	cmdq->comp_wait_bitmap = devm_bitmap_zalloc(
+		&dev->pdev->dev, cmdq->max_outstandings, GFP_KERNEL);
+	if (!cmdq->comp_wait_bitmap)
+		return -ENOMEM;
+
+	for (i = 0; i < cmdq->max_outstandings; i++) {
+		init_completion(&cmdq->wait_pool[i].wait_event);
+		cmdq->wait_pool[i].ctx_id = i;
+	}
+
+	return 0;
+}
+
+static int erdma_cmdq_sq_init(struct erdma_dev *dev)
+{
+	struct erdma_cmdq *cmdq = &dev->cmdq;
+	struct erdma_cmdq_sq *sq = &cmdq->sq;
+	u32 buf_size;
+
+	sq->wqebb_cnt = SQEBB_COUNT(ERDMA_CMDQ_SQE_SIZE);
+	sq->depth = cmdq->max_outstandings * sq->wqebb_cnt;
+
+	buf_size = sq->depth << SQEBB_SHIFT;
+
+	sq->qbuf =
+		dma_alloc_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size),
+				   &sq->qbuf_dma_addr, GFP_KERNEL);
+	if (!sq->qbuf)
+		return -ENOMEM;
+
+	sq->db_record = (u64 *)(sq->qbuf + buf_size);
+
+	spin_lock_init(&sq->lock);
+
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_SQ_ADDR_H_REG,
+			  upper_32_bits(sq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_SQ_ADDR_L_REG,
+			  lower_32_bits(sq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_DEPTH_REG, sq->depth);
+	erdma_reg_write64(dev, ERDMA_CMDQ_SQ_DB_HOST_ADDR_REG,
+			  sq->qbuf_dma_addr + buf_size);
+
+	return 0;
+}
+
+static int erdma_cmdq_cq_init(struct erdma_dev *dev)
+{
+	struct erdma_cmdq *cmdq = &dev->cmdq;
+	struct erdma_cmdq_cq *cq = &cmdq->cq;
+	u32 buf_size;
+
+	cq->depth = cmdq->sq.depth;
+	buf_size = cq->depth << CQE_SHIFT;
+
+	cq->qbuf =
+		dma_alloc_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size),
+				   &cq->qbuf_dma_addr, GFP_KERNEL | __GFP_ZERO);
+	if (!cq->qbuf)
+		return -ENOMEM;
+
+	spin_lock_init(&cq->lock);
+
+	cq->db_record = (u64 *)(cq->qbuf + buf_size);
+
+	atomic64_set(&cq->armed_num, 0);
+
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_CQ_ADDR_H_REG,
+			  upper_32_bits(cq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_CQ_ADDR_L_REG,
+			  lower_32_bits(cq->qbuf_dma_addr));
+	erdma_reg_write64(dev, ERDMA_CMDQ_CQ_DB_HOST_ADDR_REG,
+			  cq->qbuf_dma_addr + buf_size);
+
+	return 0;
+}
+
+static int erdma_cmdq_eq_init(struct erdma_dev *dev)
+{
+	struct erdma_cmdq *cmdq = &dev->cmdq;
+	struct erdma_eq *eq = &cmdq->eq;
+	u32 buf_size;
+
+	eq->depth = cmdq->max_outstandings;
+	buf_size = eq->depth << EQE_SHIFT;
+
+	eq->qbuf =
+		dma_alloc_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size),
+				   &eq->qbuf_dma_addr, GFP_KERNEL | __GFP_ZERO);
+	if (!eq->qbuf)
+		return -ENOMEM;
+
+	spin_lock_init(&eq->lock);
+	atomic64_set(&eq->event_num, 0);
+
+	eq->db_addr =
+		(u64 __iomem *)(dev->func_bar + ERDMA_REGS_CEQ_DB_BASE_REG);
+	eq->db_record = (u64 *)(eq->qbuf + buf_size);
+
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_EQ_ADDR_H_REG,
+			  upper_32_bits(eq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_EQ_ADDR_L_REG,
+			  lower_32_bits(eq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_CMDQ_EQ_DEPTH_REG, eq->depth);
+	erdma_reg_write64(dev, ERDMA_CMDQ_EQ_DB_HOST_ADDR_REG,
+			  eq->qbuf_dma_addr + buf_size);
+
+	return 0;
+}
+
+int erdma_cmdq_init(struct erdma_dev *dev)
+{
+	int err, i;
+	struct erdma_cmdq *cmdq = &dev->cmdq;
+	u32 sts, ctrl;
+
+	cmdq->max_outstandings = ERDMA_CMDQ_MAX_OUTSTANDING;
+	cmdq->use_event = false;
+
+	sema_init(&cmdq->credits, cmdq->max_outstandings);
+
+	err = erdma_cmdq_wait_res_init(dev, cmdq);
+	if (err)
+		return err;
+
+	err = erdma_cmdq_sq_init(dev);
+	if (err)
+		return err;
+
+	err = erdma_cmdq_cq_init(dev);
+	if (err)
+		goto err_destroy_sq;
+
+	err = erdma_cmdq_eq_init(dev);
+	if (err)
+		goto err_destroy_cq;
+
+	ctrl = FIELD_PREP(ERDMA_REG_DEV_CTRL_INIT_MASK, 1);
+	erdma_reg_write32(dev, ERDMA_REGS_DEV_CTRL_REG, ctrl);
+
+	for (i = 0; i < ERDMA_WAIT_DEV_DONE_CNT; i++) {
+		sts = erdma_reg_read32_filed(dev, ERDMA_REGS_DEV_ST_REG,
+					     ERDMA_REG_DEV_ST_INIT_DONE_MASK);
+		if (sts)
+			break;
+
+		msleep(ERDMA_REG_ACCESS_WAIT_MS);
+	}
+
+	if (i == ERDMA_WAIT_DEV_DONE_CNT) {
+		dev_err(&dev->pdev->dev, "wait init done failed.\n");
+		err = -ETIMEDOUT;
+		goto err_destroy_eq;
+	}
+
+	set_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state);
+
+	return 0;
+
+err_destroy_eq:
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->eq.depth << EQE_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->eq.qbuf, cmdq->eq.qbuf_dma_addr);
+
+err_destroy_cq:
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->cq.depth << CQE_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->cq.qbuf, cmdq->cq.qbuf_dma_addr);
+
+err_destroy_sq:
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->sq.depth << SQEBB_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->sq.qbuf, cmdq->sq.qbuf_dma_addr);
+
+	return err;
+}
+
+void erdma_finish_cmdq_init(struct erdma_dev *dev)
+{
+	/* after device init successfully, change cmdq to event mode. */
+	dev->cmdq.use_event = true;
+	arm_cmdq_cq(&dev->cmdq);
+}
+
+void erdma_cmdq_destroy(struct erdma_dev *dev)
+{
+	struct erdma_cmdq *cmdq = &dev->cmdq;
+
+	clear_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state);
+
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->eq.depth << EQE_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->eq.qbuf, cmdq->eq.qbuf_dma_addr);
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->sq.depth << SQEBB_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->sq.qbuf, cmdq->sq.qbuf_dma_addr);
+	dma_free_coherent(&dev->pdev->dev,
+			  (cmdq->cq.depth << CQE_SHIFT) +
+				  ERDMA_EXTRA_BUFFER_SIZE,
+			  cmdq->cq.qbuf, cmdq->cq.qbuf_dma_addr);
+}
+
+static void *get_next_valid_cmdq_cqe(struct erdma_cmdq *cmdq)
+{
+	__be32 *cqe = get_queue_entry(cmdq->cq.qbuf, cmdq->cq.ci,
+				      cmdq->cq.depth, CQE_SHIFT);
+	u32 owner = FIELD_GET(ERDMA_CQE_HDR_OWNER_MASK,
+			      __be32_to_cpu(READ_ONCE(*cqe)));
+
+	return owner ^ !!(cmdq->cq.ci & cmdq->cq.depth) ? cqe : NULL;
+}
+
+static void push_cmdq_sqe(struct erdma_cmdq *cmdq, u64 *req, size_t req_len,
+			  struct erdma_comp_wait *comp_wait)
+{
+	__le64 *wqe;
+	u64 hdr = *req;
+
+	comp_wait->cmd_status = ERDMA_CMD_STATUS_ISSUED;
+	reinit_completion(&comp_wait->wait_event);
+	comp_wait->sq_pi = cmdq->sq.pi;
+
+	wqe = get_queue_entry(cmdq->sq.qbuf, cmdq->sq.pi, cmdq->sq.depth,
+			      SQEBB_SHIFT);
+	memcpy(wqe, req, req_len);
+
+	cmdq->sq.pi += cmdq->sq.wqebb_cnt;
+	hdr |= FIELD_PREP(ERDMA_CMD_HDR_WQEBB_INDEX_MASK, cmdq->sq.pi) |
+	       FIELD_PREP(ERDMA_CMD_HDR_CONTEXT_COOKIE_MASK,
+			  comp_wait->ctx_id) |
+	       FIELD_PREP(ERDMA_CMD_HDR_WQEBB_CNT_MASK, cmdq->sq.wqebb_cnt - 1);
+	*wqe = cpu_to_le64(hdr);
+
+	kick_cmdq_db(cmdq);
+}
+
+static int erdma_poll_single_cmd_completion(struct erdma_cmdq *cmdq)
+{
+	struct erdma_comp_wait *comp_wait;
+	u32 hdr0, sqe_idx;
+	__be32 *cqe;
+	u16 ctx_id;
+	u64 *sqe;
+	int i;
+
+	cqe = get_next_valid_cmdq_cqe(cmdq);
+	if (!cqe)
+		return -EAGAIN;
+
+	cmdq->cq.ci++;
+
+	dma_rmb();
+	hdr0 = __be32_to_cpu(*cqe);
+	sqe_idx = __be32_to_cpu(*(cqe + 1));
+
+	sqe = get_queue_entry(cmdq->sq.qbuf, sqe_idx, cmdq->sq.depth,
+			      SQEBB_SHIFT);
+	ctx_id = FIELD_GET(ERDMA_CMD_HDR_CONTEXT_COOKIE_MASK, *sqe);
+	comp_wait = &cmdq->wait_pool[ctx_id];
+	if (comp_wait->cmd_status != ERDMA_CMD_STATUS_ISSUED)
+		return -EIO;
+
+	comp_wait->cmd_status = ERDMA_CMD_STATUS_FINISHED;
+	comp_wait->comp_status = FIELD_GET(ERDMA_CQE_HDR_SYNDROME_MASK, hdr0);
+	cmdq->sq.ci += cmdq->sq.wqebb_cnt;
+
+	for (i = 0; i < 4; i++)
+		comp_wait->comp_data[i] = __be32_to_cpu(*(cqe + 2 + i));
+
+	if (cmdq->use_event)
+		complete(&comp_wait->wait_event);
+
+	return 0;
+}
+
+static void erdma_polling_cmd_completions(struct erdma_cmdq *cmdq)
+{
+	unsigned long flags;
+	u16 comp_num;
+
+	spin_lock_irqsave(&cmdq->cq.lock, flags);
+
+	/* We must have less than # of max_outstandings
+	 * completions at one time.
+	 */
+	for (comp_num = 0; comp_num < cmdq->max_outstandings; comp_num++)
+		if (erdma_poll_single_cmd_completion(cmdq))
+			break;
+
+	if (comp_num && cmdq->use_event)
+		arm_cmdq_cq(cmdq);
+
+	spin_unlock_irqrestore(&cmdq->cq.lock, flags);
+}
+
+void erdma_cmdq_completion_handler(struct erdma_cmdq *cmdq)
+{
+	int got_event = 0;
+
+	if (!test_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state) ||
+	    !cmdq->use_event)
+		return;
+
+	while (get_next_valid_eqe(&cmdq->eq)) {
+		cmdq->eq.ci++;
+		got_event++;
+	}
+
+	if (got_event) {
+		cmdq->cq.cmdsn++;
+		erdma_polling_cmd_completions(cmdq);
+	}
+
+	notify_eq(&cmdq->eq);
+}
+
+static int erdma_poll_cmd_completion(struct erdma_comp_wait *comp_ctx,
+				     struct erdma_cmdq *cmdq, u32 timeout)
+{
+	unsigned long comp_timeout = jiffies + msecs_to_jiffies(timeout);
+
+	while (1) {
+		erdma_polling_cmd_completions(cmdq);
+		if (comp_ctx->cmd_status != ERDMA_CMD_STATUS_ISSUED)
+			break;
+
+		if (time_is_before_jiffies(comp_timeout))
+			return -ETIME;
+
+		msleep(20);
+	}
+
+	return 0;
+}
+
+static int erdma_wait_cmd_completion(struct erdma_comp_wait *comp_ctx,
+				     struct erdma_cmdq *cmdq, u32 timeout)
+{
+	unsigned long flags = 0;
+
+	wait_for_completion_timeout(&comp_ctx->wait_event,
+				    msecs_to_jiffies(timeout));
+
+	if (unlikely(comp_ctx->cmd_status != ERDMA_CMD_STATUS_FINISHED)) {
+		spin_lock_irqsave(&cmdq->cq.lock, flags);
+		comp_ctx->cmd_status = ERDMA_CMD_STATUS_TIMEOUT;
+		spin_unlock_irqrestore(&cmdq->cq.lock, flags);
+		return -ETIME;
+	}
+
+	return 0;
+}
+
+void erdma_cmdq_build_reqhdr(u64 *hdr, u32 mod, u32 op)
+{
+	*hdr = FIELD_PREP(ERDMA_CMD_HDR_SUB_MOD_MASK, mod) |
+	       FIELD_PREP(ERDMA_CMD_HDR_OPCODE_MASK, op);
+}
+
+int erdma_post_cmd_wait(struct erdma_cmdq *cmdq, u64 *req, u32 req_size,
+			u64 *resp0, u64 *resp1)
+{
+	struct erdma_comp_wait *comp_wait;
+	int ret;
+
+	if (!test_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state))
+		return -ENODEV;
+
+	down(&cmdq->credits);
+
+	comp_wait = get_comp_wait(cmdq);
+	if (IS_ERR(comp_wait)) {
+		clear_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state);
+		set_bit(ERDMA_CMDQ_STATE_CTX_ERR_BIT, &cmdq->state);
+		up(&cmdq->credits);
+		return PTR_ERR(comp_wait);
+	}
+
+	spin_lock(&cmdq->sq.lock);
+	push_cmdq_sqe(cmdq, req, req_size, comp_wait);
+	spin_unlock(&cmdq->sq.lock);
+
+	if (cmdq->use_event)
+		ret = erdma_wait_cmd_completion(comp_wait, cmdq,
+						ERDMA_CMDQ_TIMEOUT_MS);
+	else
+		ret = erdma_poll_cmd_completion(comp_wait, cmdq,
+						ERDMA_CMDQ_TIMEOUT_MS);
+
+	if (ret) {
+		set_bit(ERDMA_CMDQ_STATE_TIMEOUT_BIT, &cmdq->state);
+		clear_bit(ERDMA_CMDQ_STATE_OK_BIT, &cmdq->state);
+		goto out;
+	}
+
+	if (comp_wait->comp_status)
+		ret = -EIO;
+
+	if (resp0 && resp1) {
+		*resp0 = *((u64 *)&comp_wait->comp_data[0]);
+		*resp1 = *((u64 *)&comp_wait->comp_data[2]);
+	}
+	put_comp_wait(cmdq, comp_wait);
+
+out:
+	up(&cmdq->credits);
+
+	return ret;
+}

drivers/infiniband/hw/erdma/erdma_cq.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#include <rdma/ib_verbs.h>
+
+#include "erdma_hw.h"
+#include "erdma_verbs.h"
+
+static void *get_next_valid_cqe(struct erdma_cq *cq)
+{
+	__be32 *cqe = get_queue_entry(cq->kern_cq.qbuf, cq->kern_cq.ci,
+				      cq->depth, CQE_SHIFT);
+	u32 owner = FIELD_GET(ERDMA_CQE_HDR_OWNER_MASK,
+			      __be32_to_cpu(READ_ONCE(*cqe)));
+
+	return owner ^ !!(cq->kern_cq.ci & cq->depth) ? cqe : NULL;
+}
+
+static void notify_cq(struct erdma_cq *cq, u8 solcitied)
+{
+	u64 db_data =
+		FIELD_PREP(ERDMA_CQDB_IDX_MASK, (cq->kern_cq.notify_cnt)) |
+		FIELD_PREP(ERDMA_CQDB_CQN_MASK, cq->cqn) |
+		FIELD_PREP(ERDMA_CQDB_ARM_MASK, 1) |
+		FIELD_PREP(ERDMA_CQDB_SOL_MASK, solcitied) |
+		FIELD_PREP(ERDMA_CQDB_CMDSN_MASK, cq->kern_cq.cmdsn) |
+		FIELD_PREP(ERDMA_CQDB_CI_MASK, cq->kern_cq.ci);
+
+	*cq->kern_cq.db_record = db_data;
+	writeq(db_data, cq->kern_cq.db);
+}
+
+int erdma_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags)
+{
+	struct erdma_cq *cq = to_ecq(ibcq);
+	unsigned long irq_flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&cq->kern_cq.lock, irq_flags);
+
+	notify_cq(cq, (flags & IB_CQ_SOLICITED_MASK) == IB_CQ_SOLICITED);
+
+	if ((flags & IB_CQ_REPORT_MISSED_EVENTS) && get_next_valid_cqe(cq))
+		ret = 1;
+
+	cq->kern_cq.notify_cnt++;
+
+	spin_unlock_irqrestore(&cq->kern_cq.lock, irq_flags);
+
+	return ret;
+}
+
+static const enum ib_wc_opcode wc_mapping_table[ERDMA_NUM_OPCODES] = {
+	[ERDMA_OP_WRITE] = IB_WC_RDMA_WRITE,
+	[ERDMA_OP_READ] = IB_WC_RDMA_READ,
+	[ERDMA_OP_SEND] = IB_WC_SEND,
+	[ERDMA_OP_SEND_WITH_IMM] = IB_WC_SEND,
+	[ERDMA_OP_RECEIVE] = IB_WC_RECV,
+	[ERDMA_OP_RECV_IMM] = IB_WC_RECV_RDMA_WITH_IMM,
+	[ERDMA_OP_RECV_INV] = IB_WC_RECV,
+	[ERDMA_OP_WRITE_WITH_IMM] = IB_WC_RDMA_WRITE,
+	[ERDMA_OP_INVALIDATE] = IB_WC_LOCAL_INV,
+	[ERDMA_OP_RSP_SEND_IMM] = IB_WC_RECV,
+	[ERDMA_OP_SEND_WITH_INV] = IB_WC_SEND,
+	[ERDMA_OP_REG_MR] = IB_WC_REG_MR,
+	[ERDMA_OP_LOCAL_INV] = IB_WC_LOCAL_INV,
+	[ERDMA_OP_READ_WITH_INV] = IB_WC_RDMA_READ,
+};
+
+static const struct {
+	enum erdma_wc_status erdma;
+	enum ib_wc_status base;
+	enum erdma_vendor_err vendor;
+} map_cqe_status[ERDMA_NUM_WC_STATUS] = {
+	{ ERDMA_WC_SUCCESS, IB_WC_SUCCESS, ERDMA_WC_VENDOR_NO_ERR },
+	{ ERDMA_WC_GENERAL_ERR, IB_WC_GENERAL_ERR, ERDMA_WC_VENDOR_NO_ERR },
+	{ ERDMA_WC_RECV_WQE_FORMAT_ERR, IB_WC_GENERAL_ERR,
+	  ERDMA_WC_VENDOR_INVALID_RQE },
+	{ ERDMA_WC_RECV_STAG_INVALID_ERR, IB_WC_REM_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_RQE_INVALID_STAG },
+	{ ERDMA_WC_RECV_ADDR_VIOLATION_ERR, IB_WC_REM_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_RQE_ADDR_VIOLATION },
+	{ ERDMA_WC_RECV_RIGHT_VIOLATION_ERR, IB_WC_REM_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_RQE_ACCESS_RIGHT_ERR },
+	{ ERDMA_WC_RECV_PDID_ERR, IB_WC_REM_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_RQE_INVALID_PD },
+	{ ERDMA_WC_RECV_WARRPING_ERR, IB_WC_REM_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_RQE_WRAP_ERR },
+	{ ERDMA_WC_SEND_WQE_FORMAT_ERR, IB_WC_LOC_QP_OP_ERR,
+	  ERDMA_WC_VENDOR_INVALID_SQE },
+	{ ERDMA_WC_SEND_WQE_ORD_EXCEED, IB_WC_GENERAL_ERR,
+	  ERDMA_WC_VENDOR_ZERO_ORD },
+	{ ERDMA_WC_SEND_STAG_INVALID_ERR, IB_WC_LOC_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_SQE_INVALID_STAG },
+	{ ERDMA_WC_SEND_ADDR_VIOLATION_ERR, IB_WC_LOC_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_SQE_ADDR_VIOLATION },
+	{ ERDMA_WC_SEND_RIGHT_VIOLATION_ERR, IB_WC_LOC_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_SQE_ACCESS_ERR },
+	{ ERDMA_WC_SEND_PDID_ERR, IB_WC_LOC_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_SQE_INVALID_PD },
+	{ ERDMA_WC_SEND_WARRPING_ERR, IB_WC_LOC_ACCESS_ERR,
+	  ERDMA_WC_VENDOR_SQE_WARP_ERR },
+	{ ERDMA_WC_FLUSH_ERR, IB_WC_WR_FLUSH_ERR, ERDMA_WC_VENDOR_NO_ERR },
+	{ ERDMA_WC_RETRY_EXC_ERR, IB_WC_RETRY_EXC_ERR, ERDMA_WC_VENDOR_NO_ERR },
+};
+
+#define ERDMA_POLLCQ_NO_QP 1
+
+static int erdma_poll_one_cqe(struct erdma_cq *cq, struct ib_wc *wc)
+{
+	struct erdma_dev *dev = to_edev(cq->ibcq.device);
+	u8 opcode, syndrome, qtype;
+	struct erdma_kqp *kern_qp;
+	struct erdma_cqe *cqe;
+	struct erdma_qp *qp;
+	u16 wqe_idx, depth;
+	u32 qpn, cqe_hdr;
+	u64 *id_table;
+	u64 *wqe_hdr;
+
+	cqe = get_next_valid_cqe(cq);
+	if (!cqe)
+		return -EAGAIN;
+
+	cq->kern_cq.ci++;
+
+	/* cqbuf should be ready when we poll */
+	dma_rmb();
+
+	qpn = be32_to_cpu(cqe->qpn);
+	wqe_idx = be32_to_cpu(cqe->qe_idx);
+	cqe_hdr = be32_to_cpu(cqe->hdr);
+
+	qp = find_qp_by_qpn(dev, qpn);
+	if (!qp)
+		return ERDMA_POLLCQ_NO_QP;
+
+	kern_qp = &qp->kern_qp;
+
+	qtype = FIELD_GET(ERDMA_CQE_HDR_QTYPE_MASK, cqe_hdr);
+	syndrome = FIELD_GET(ERDMA_CQE_HDR_SYNDROME_MASK, cqe_hdr);
+	opcode = FIELD_GET(ERDMA_CQE_HDR_OPCODE_MASK, cqe_hdr);
+
+	if (qtype == ERDMA_CQE_QTYPE_SQ) {
+		id_table = kern_qp->swr_tbl;
+		depth = qp->attrs.sq_size;
+		wqe_hdr = get_queue_entry(qp->kern_qp.sq_buf, wqe_idx,
+					  qp->attrs.sq_size, SQEBB_SHIFT);
+		kern_qp->sq_ci =
+			FIELD_GET(ERDMA_SQE_HDR_WQEBB_CNT_MASK, *wqe_hdr) +
+			wqe_idx + 1;
+	} else {
+		id_table = kern_qp->rwr_tbl;
+		depth = qp->attrs.rq_size;
+	}
+	wc->wr_id = id_table[wqe_idx & (depth - 1)];
+	wc->byte_len = be32_to_cpu(cqe->size);
+
+	wc->wc_flags = 0;
+
+	wc->opcode = wc_mapping_table[opcode];
+	if (opcode == ERDMA_OP_RECV_IMM || opcode == ERDMA_OP_RSP_SEND_IMM) {
+		wc->ex.imm_data = cpu_to_be32(le32_to_cpu(cqe->imm_data));
+		wc->wc_flags |= IB_WC_WITH_IMM;
+	} else if (opcode == ERDMA_OP_RECV_INV) {
+		wc->ex.invalidate_rkey = be32_to_cpu(cqe->inv_rkey);
+		wc->wc_flags |= IB_WC_WITH_INVALIDATE;
+	}
+
+	if (syndrome >= ERDMA_NUM_WC_STATUS)
+		syndrome = ERDMA_WC_GENERAL_ERR;
+
+	wc->status = map_cqe_status[syndrome].base;
+	wc->vendor_err = map_cqe_status[syndrome].vendor;
+	wc->qp = &qp->ibqp;
+
+	return 0;
+}
+
+int erdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc)
+{
+	struct erdma_cq *cq = to_ecq(ibcq);
+	unsigned long flags;
+	int npolled, ret;
+
+	spin_lock_irqsave(&cq->kern_cq.lock, flags);
+
+	for (npolled = 0; npolled < num_entries;) {
+		ret = erdma_poll_one_cqe(cq, wc + npolled);
+
+		if (ret == -EAGAIN) /* no received new CQEs. */
+			break;
+		else if (ret) /* ignore invalid CQEs. */
+			continue;
+
+		npolled++;
+	}
+
+	spin_unlock_irqrestore(&cq->kern_cq.lock, flags);
+
+	return npolled;
+}

drivers/infiniband/hw/erdma/erdma_eq.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+
+#include "erdma.h"
+#include "erdma_hw.h"
+#include "erdma_verbs.h"
+
+#define MAX_POLL_CHUNK_SIZE 16
+
+void notify_eq(struct erdma_eq *eq)
+{
+	u64 db_data = FIELD_PREP(ERDMA_EQDB_CI_MASK, eq->ci) |
+		      FIELD_PREP(ERDMA_EQDB_ARM_MASK, 1);
+
+	*eq->db_record = db_data;
+	writeq(db_data, eq->db_addr);
+
+	atomic64_inc(&eq->notify_num);
+}
+
+void *get_next_valid_eqe(struct erdma_eq *eq)
+{
+	u64 *eqe = get_queue_entry(eq->qbuf, eq->ci, eq->depth, EQE_SHIFT);
+	u32 owner = FIELD_GET(ERDMA_CEQE_HDR_O_MASK, READ_ONCE(*eqe));
+
+	return owner ^ !!(eq->ci & eq->depth) ? eqe : NULL;
+}
+
+void erdma_aeq_event_handler(struct erdma_dev *dev)
+{
+	struct erdma_aeqe *aeqe;
+	u32 cqn, qpn;
+	struct erdma_qp *qp;
+	struct erdma_cq *cq;
+	struct ib_event event;
+	u32 poll_cnt = 0;
+
+	memset(&event, 0, sizeof(event));
+
+	while (poll_cnt < MAX_POLL_CHUNK_SIZE) {
+		aeqe = get_next_valid_eqe(&dev->aeq);
+		if (!aeqe)
+			break;
+
+		dma_rmb();
+
+		dev->aeq.ci++;
+		atomic64_inc(&dev->aeq.event_num);
+		poll_cnt++;
+
+		if (FIELD_GET(ERDMA_AEQE_HDR_TYPE_MASK,
+			      le32_to_cpu(aeqe->hdr)) == ERDMA_AE_TYPE_CQ_ERR) {
+			cqn = le32_to_cpu(aeqe->event_data0);
+			cq = find_cq_by_cqn(dev, cqn);
+			if (!cq)
+				continue;
+
+			event.device = cq->ibcq.device;
+			event.element.cq = &cq->ibcq;
+			event.event = IB_EVENT_CQ_ERR;
+			if (cq->ibcq.event_handler)
+				cq->ibcq.event_handler(&event,
+						       cq->ibcq.cq_context);
+		} else {
+			qpn = le32_to_cpu(aeqe->event_data0);
+			qp = find_qp_by_qpn(dev, qpn);
+			if (!qp)
+				continue;
+
+			event.device = qp->ibqp.device;
+			event.element.qp = &qp->ibqp;
+			event.event = IB_EVENT_QP_FATAL;
+			if (qp->ibqp.event_handler)
+				qp->ibqp.event_handler(&event,
+						       qp->ibqp.qp_context);
+		}
+	}
+
+	notify_eq(&dev->aeq);
+}
+
+int erdma_aeq_init(struct erdma_dev *dev)
+{
+	struct erdma_eq *eq = &dev->aeq;
+	u32 buf_size;
+
+	eq->depth = ERDMA_DEFAULT_EQ_DEPTH;
+	buf_size = eq->depth << EQE_SHIFT;
+
+	eq->qbuf =
+		dma_alloc_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size),
+				   &eq->qbuf_dma_addr, GFP_KERNEL | __GFP_ZERO);
+	if (!eq->qbuf)
+		return -ENOMEM;
+
+	spin_lock_init(&eq->lock);
+	atomic64_set(&eq->event_num, 0);
+	atomic64_set(&eq->notify_num, 0);
+
+	eq->db_addr = (u64 __iomem *)(dev->func_bar + ERDMA_REGS_AEQ_DB_REG);
+	eq->db_record = (u64 *)(eq->qbuf + buf_size);
+
+	erdma_reg_write32(dev, ERDMA_REGS_AEQ_ADDR_H_REG,
+			  upper_32_bits(eq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_AEQ_ADDR_L_REG,
+			  lower_32_bits(eq->qbuf_dma_addr));
+	erdma_reg_write32(dev, ERDMA_REGS_AEQ_DEPTH_REG, eq->depth);
+	erdma_reg_write64(dev, ERDMA_AEQ_DB_HOST_ADDR_REG,
+			  eq->qbuf_dma_addr + buf_size);
+
+	return 0;
+}
+
+void erdma_aeq_destroy(struct erdma_dev *dev)
+{
+	struct erdma_eq *eq = &dev->aeq;
+
+	dma_free_coherent(&dev->pdev->dev,
+			  WARPPED_BUFSIZE(eq->depth << EQE_SHIFT), eq->qbuf,
+			  eq->qbuf_dma_addr);
+}
+
+void erdma_ceq_completion_handler(struct erdma_eq_cb *ceq_cb)
+{
+	struct erdma_dev *dev = ceq_cb->dev;
+	struct erdma_cq *cq;
+	u32 poll_cnt = 0;
+	u64 *ceqe;
+	int cqn;
+
+	if (!ceq_cb->ready)
+		return;
+
+	while (poll_cnt < MAX_POLL_CHUNK_SIZE) {
+		ceqe = get_next_valid_eqe(&ceq_cb->eq);
+		if (!ceqe)
+			break;
+
+		dma_rmb();
+		ceq_cb->eq.ci++;
+		poll_cnt++;
+		cqn = FIELD_GET(ERDMA_CEQE_HDR_CQN_MASK, READ_ONCE(*ceqe));
+
+		cq = find_cq_by_cqn(dev, cqn);
+		if (!cq)
+			continue;
+
+		if (rdma_is_kernel_res(&cq->ibcq.res))
+			cq->kern_cq.cmdsn++;
+
+		if (cq->ibcq.comp_handler)
+			cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context);
+	}
+
+	notify_eq(&ceq_cb->eq);
+}
+
+static irqreturn_t erdma_intr_ceq_handler(int irq, void *data)
+{
+	struct erdma_eq_cb *ceq_cb = data;
+
+	tasklet_schedule(&ceq_cb->tasklet);
+
+	return IRQ_HANDLED;
+}
+
+static void erdma_intr_ceq_task(unsigned long data)
+{
+	erdma_ceq_completion_handler((struct erdma_eq_cb *)data);
+}
+
+static int erdma_set_ceq_irq(struct erdma_dev *dev, u16 ceqn)
+{
+	struct erdma_eq_cb *eqc = &dev->ceqs[ceqn];
+	int err;
+
+	snprintf(eqc->irq.name, ERDMA_IRQNAME_SIZE, "erdma-ceq%u@pci:%s", ceqn,
+		 pci_name(dev->pdev));
+	eqc->irq.msix_vector = pci_irq_vector(dev->pdev, ceqn + 1);
+
+	tasklet_init(&dev->ceqs[ceqn].tasklet, erdma_intr_ceq_task,
+		     (unsigned long)&dev->ceqs[ceqn]);
+
+	cpumask_set_cpu(cpumask_local_spread(ceqn + 1, dev->attrs.numa_node),
+			&eqc->irq.affinity_hint_mask);
+
+	err = request_irq(eqc->irq.msix_vector, erdma_intr_ceq_handler, 0,
+			  eqc->irq.name, eqc);
+	if (err) {
+		dev_err(&dev->pdev->dev, "failed to request_irq(%d)\n", err);
+		return err;
+	}
+
+	irq_set_affinity_hint(eqc->irq.msix_vector,
+			      &eqc->irq.affinity_hint_mask);
+
+	return 0;
+}
+
+static void erdma_free_ceq_irq(struct erdma_dev *dev, u16 ceqn)
+{
+	struct erdma_eq_cb *eqc = &dev->ceqs[ceqn];
+
+	irq_set_affinity_hint(eqc->irq.msix_vector, NULL);
+	free_irq(eqc->irq.msix_vector, eqc);
+}
+
+static int create_eq_cmd(struct erdma_dev *dev, u32 eqn, struct erdma_eq *eq)
+{
+	struct erdma_cmdq_create_eq_req req;
+	dma_addr_t db_info_dma_addr;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_COMMON,
+				CMDQ_OPCODE_CREATE_EQ);
+	req.eqn = eqn;
+	req.depth = ilog2(eq->depth);
+	req.qbuf_addr = eq->qbuf_dma_addr;
+	req.qtype = ERDMA_EQ_TYPE_CEQ;
+	/* Vector index is the same as EQN. */
+	req.vector_idx = eqn;
+	db_info_dma_addr = eq->qbuf_dma_addr + (eq->depth << EQE_SHIFT);
+	req.db_dma_addr_l = lower_32_bits(db_info_dma_addr);
+	req.db_dma_addr_h = upper_32_bits(db_info_dma_addr);
+
+	return erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req,
+				   sizeof(struct erdma_cmdq_create_eq_req),
+				   NULL, NULL);
+}
+
+static int erdma_ceq_init_one(struct erdma_dev *dev, u16 ceqn)
+{
+	struct erdma_eq *eq = &dev->ceqs[ceqn].eq;
+	u32 buf_size = ERDMA_DEFAULT_EQ_DEPTH << EQE_SHIFT;
+	int ret;
+
+	eq->qbuf =
+		dma_alloc_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size),
+				   &eq->qbuf_dma_addr, GFP_KERNEL | __GFP_ZERO);
+	if (!eq->qbuf)
+		return -ENOMEM;
+
+	spin_lock_init(&eq->lock);
+	atomic64_set(&eq->event_num, 0);
+	atomic64_set(&eq->notify_num, 0);
+
+	eq->depth = ERDMA_DEFAULT_EQ_DEPTH;
+	eq->db_addr =
+		(u64 __iomem *)(dev->func_bar + ERDMA_REGS_CEQ_DB_BASE_REG +
+				(ceqn + 1) * ERDMA_DB_SIZE);
+	eq->db_record = (u64 *)(eq->qbuf + buf_size);
+	eq->ci = 0;
+	dev->ceqs[ceqn].dev = dev;
+
+	/* CEQ indexed from 1, 0 rsvd for CMDQ-EQ. */
+	ret = create_eq_cmd(dev, ceqn + 1, eq);
+	dev->ceqs[ceqn].ready = ret ? false : true;
+
+	return ret;
+}
+
+static void erdma_ceq_uninit_one(struct erdma_dev *dev, u16 ceqn)
+{
+	struct erdma_eq *eq = &dev->ceqs[ceqn].eq;
+	u32 buf_size = ERDMA_DEFAULT_EQ_DEPTH << EQE_SHIFT;
+	struct erdma_cmdq_destroy_eq_req req;
+	int err;
+
+	dev->ceqs[ceqn].ready = 0;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_COMMON,
+				CMDQ_OPCODE_DESTROY_EQ);
+	/* CEQ indexed from 1, 0 rsvd for CMDQ-EQ. */
+	req.eqn = ceqn + 1;
+	req.qtype = ERDMA_EQ_TYPE_CEQ;
+	req.vector_idx = ceqn + 1;
+
+	err = erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				  NULL);
+	if (err)
+		return;
+
+	dma_free_coherent(&dev->pdev->dev, WARPPED_BUFSIZE(buf_size), eq->qbuf,
+			  eq->qbuf_dma_addr);
+}
+
+int erdma_ceqs_init(struct erdma_dev *dev)
+{
+	u32 i, j;
+	int err;
+
+	for (i = 0; i < dev->attrs.irq_num - 1; i++) {
+		err = erdma_ceq_init_one(dev, i);
+		if (err)
+			goto out_err;
+
+		err = erdma_set_ceq_irq(dev, i);
+		if (err) {
+			erdma_ceq_uninit_one(dev, i);
+			goto out_err;
+		}
+	}
+
+	return 0;
+
+out_err:
+	for (j = 0; j < i; j++) {
+		erdma_free_ceq_irq(dev, j);
+		erdma_ceq_uninit_one(dev, j);
+	}
+
+	return err;
+}
+
+void erdma_ceqs_uninit(struct erdma_dev *dev)
+{
+	u32 i;
+
+	for (i = 0; i < dev->attrs.irq_num - 1; i++) {
+		erdma_free_ceq_irq(dev, i);
+		erdma_ceq_uninit_one(dev, i);
+	}
+}

drivers/infiniband/hw/erdma/erdma_hw.h

···
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#ifndef __ERDMA_HW_H__
+#define __ERDMA_HW_H__
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+
+/* PCIe device related definition. */
+#define PCI_VENDOR_ID_ALIBABA 0x1ded
+
+#define ERDMA_PCI_WIDTH 64
+#define ERDMA_FUNC_BAR 0
+#define ERDMA_MISX_BAR 2
+
+#define ERDMA_BAR_MASK (BIT(ERDMA_FUNC_BAR) | BIT(ERDMA_MISX_BAR))
+
+/* MSI-X related. */
+#define ERDMA_NUM_MSIX_VEC 32U
+#define ERDMA_MSIX_VECTOR_CMDQ 0
+
+/* PCIe Bar0 Registers. */
+#define ERDMA_REGS_VERSION_REG 0x0
+#define ERDMA_REGS_DEV_CTRL_REG 0x10
+#define ERDMA_REGS_DEV_ST_REG 0x14
+#define ERDMA_REGS_NETDEV_MAC_L_REG 0x18
+#define ERDMA_REGS_NETDEV_MAC_H_REG 0x1C
+#define ERDMA_REGS_CMDQ_SQ_ADDR_L_REG 0x20
+#define ERDMA_REGS_CMDQ_SQ_ADDR_H_REG 0x24
+#define ERDMA_REGS_CMDQ_CQ_ADDR_L_REG 0x28
+#define ERDMA_REGS_CMDQ_CQ_ADDR_H_REG 0x2C
+#define ERDMA_REGS_CMDQ_DEPTH_REG 0x30
+#define ERDMA_REGS_CMDQ_EQ_DEPTH_REG 0x34
+#define ERDMA_REGS_CMDQ_EQ_ADDR_L_REG 0x38
+#define ERDMA_REGS_CMDQ_EQ_ADDR_H_REG 0x3C
+#define ERDMA_REGS_AEQ_ADDR_L_REG 0x40
+#define ERDMA_REGS_AEQ_ADDR_H_REG 0x44
+#define ERDMA_REGS_AEQ_DEPTH_REG 0x48
+#define ERDMA_REGS_GRP_NUM_REG 0x4c
+#define ERDMA_REGS_AEQ_DB_REG 0x50
+#define ERDMA_CMDQ_SQ_DB_HOST_ADDR_REG 0x60
+#define ERDMA_CMDQ_CQ_DB_HOST_ADDR_REG 0x68
+#define ERDMA_CMDQ_EQ_DB_HOST_ADDR_REG 0x70
+#define ERDMA_AEQ_DB_HOST_ADDR_REG 0x78
+#define ERDMA_REGS_STATS_TSO_IN_PKTS_REG 0x80
+#define ERDMA_REGS_STATS_TSO_OUT_PKTS_REG 0x88
+#define ERDMA_REGS_STATS_TSO_OUT_BYTES_REG 0x90
+#define ERDMA_REGS_STATS_TX_DROP_PKTS_REG 0x98
+#define ERDMA_REGS_STATS_TX_BPS_METER_DROP_PKTS_REG 0xa0
+#define ERDMA_REGS_STATS_TX_PPS_METER_DROP_PKTS_REG 0xa8
+#define ERDMA_REGS_STATS_RX_PKTS_REG 0xc0
+#define ERDMA_REGS_STATS_RX_BYTES_REG 0xc8
+#define ERDMA_REGS_STATS_RX_DROP_PKTS_REG 0xd0
+#define ERDMA_REGS_STATS_RX_BPS_METER_DROP_PKTS_REG 0xd8
+#define ERDMA_REGS_STATS_RX_PPS_METER_DROP_PKTS_REG 0xe0
+#define ERDMA_REGS_CEQ_DB_BASE_REG 0x100
+#define ERDMA_CMDQ_SQDB_REG 0x200
+#define ERDMA_CMDQ_CQDB_REG 0x300
+
+/* DEV_CTRL_REG details. */
+#define ERDMA_REG_DEV_CTRL_RESET_MASK 0x00000001
+#define ERDMA_REG_DEV_CTRL_INIT_MASK 0x00000002
+
+/* DEV_ST_REG details. */
+#define ERDMA_REG_DEV_ST_RESET_DONE_MASK 0x00000001U
+#define ERDMA_REG_DEV_ST_INIT_DONE_MASK 0x00000002U
+
+/* eRDMA PCIe DBs definition. */
+#define ERDMA_BAR_DB_SPACE_BASE 4096
+
+#define ERDMA_BAR_SQDB_SPACE_OFFSET ERDMA_BAR_DB_SPACE_BASE
+#define ERDMA_BAR_SQDB_SPACE_SIZE (384 * 1024)
+
+#define ERDMA_BAR_RQDB_SPACE_OFFSET \
+	(ERDMA_BAR_SQDB_SPACE_OFFSET + ERDMA_BAR_SQDB_SPACE_SIZE)
+#define ERDMA_BAR_RQDB_SPACE_SIZE (96 * 1024)
+
+#define ERDMA_BAR_CQDB_SPACE_OFFSET \
+	(ERDMA_BAR_RQDB_SPACE_OFFSET + ERDMA_BAR_RQDB_SPACE_SIZE)
+
+/* Doorbell page resources related. */
+/*
+ * Max # of parallelly issued directSQE is 3072 per device,
+ * hardware organizes this into 24 group, per group has 128 credits.
+ */
+#define ERDMA_DWQE_MAX_GRP_CNT 24
+#define ERDMA_DWQE_NUM_PER_GRP 128
+
+#define ERDMA_DWQE_TYPE0_CNT 64
+#define ERDMA_DWQE_TYPE1_CNT 496
+/* type1 DB contains 2 DBs, takes 256Byte. */
+#define ERDMA_DWQE_TYPE1_CNT_PER_PAGE 16
+
+#define ERDMA_SDB_SHARED_PAGE_INDEX 95
+
+/* Doorbell related. */
+#define ERDMA_DB_SIZE 8
+
+#define ERDMA_CQDB_IDX_MASK GENMASK_ULL(63, 56)
+#define ERDMA_CQDB_CQN_MASK GENMASK_ULL(55, 32)
+#define ERDMA_CQDB_ARM_MASK BIT_ULL(31)
+#define ERDMA_CQDB_SOL_MASK BIT_ULL(30)
+#define ERDMA_CQDB_CMDSN_MASK GENMASK_ULL(29, 28)
+#define ERDMA_CQDB_CI_MASK GENMASK_ULL(23, 0)
+
+#define ERDMA_EQDB_ARM_MASK BIT(31)
+#define ERDMA_EQDB_CI_MASK GENMASK_ULL(23, 0)
+
+#define ERDMA_PAGE_SIZE_SUPPORT 0x7FFFF000
+
+/* WQE related. */
+#define EQE_SIZE 16
+#define EQE_SHIFT 4
+#define RQE_SIZE 32
+#define RQE_SHIFT 5
+#define CQE_SIZE 32
+#define CQE_SHIFT 5
+#define SQEBB_SIZE 32
+#define SQEBB_SHIFT 5
+#define SQEBB_MASK (~(SQEBB_SIZE - 1))
+#define SQEBB_ALIGN(size) ((size + SQEBB_SIZE - 1) & SQEBB_MASK)
+#define SQEBB_COUNT(size) (SQEBB_ALIGN(size) >> SQEBB_SHIFT)
+
+#define ERDMA_MAX_SQE_SIZE 128
+#define ERDMA_MAX_WQEBB_PER_SQE 4
+
+/* CMDQ related. */
+#define ERDMA_CMDQ_MAX_OUTSTANDING 128
+#define ERDMA_CMDQ_SQE_SIZE 64
+
+/* cmdq sub module definition. */
+enum CMDQ_WQE_SUB_MOD {
+	CMDQ_SUBMOD_RDMA = 0,
+	CMDQ_SUBMOD_COMMON = 1
+};
+
+enum CMDQ_RDMA_OPCODE {
+	CMDQ_OPCODE_QUERY_DEVICE = 0,
+	CMDQ_OPCODE_CREATE_QP = 1,
+	CMDQ_OPCODE_DESTROY_QP = 2,
+	CMDQ_OPCODE_MODIFY_QP = 3,
+	CMDQ_OPCODE_CREATE_CQ = 4,
+	CMDQ_OPCODE_DESTROY_CQ = 5,
+	CMDQ_OPCODE_REG_MR = 8,
+	CMDQ_OPCODE_DEREG_MR = 9
+};
+
+enum CMDQ_COMMON_OPCODE {
+	CMDQ_OPCODE_CREATE_EQ = 0,
+	CMDQ_OPCODE_DESTROY_EQ = 1,
+	CMDQ_OPCODE_QUERY_FW_INFO = 2,
+};
+
+/* cmdq-SQE HDR */
+#define ERDMA_CMD_HDR_WQEBB_CNT_MASK GENMASK_ULL(54, 52)
+#define ERDMA_CMD_HDR_CONTEXT_COOKIE_MASK GENMASK_ULL(47, 32)
+#define ERDMA_CMD_HDR_SUB_MOD_MASK GENMASK_ULL(25, 24)
+#define ERDMA_CMD_HDR_OPCODE_MASK GENMASK_ULL(23, 16)
+#define ERDMA_CMD_HDR_WQEBB_INDEX_MASK GENMASK_ULL(15, 0)
+
+struct erdma_cmdq_destroy_cq_req {
+	u64 hdr;
+	u32 cqn;
+};
+
+#define ERDMA_EQ_TYPE_AEQ 0
+#define ERDMA_EQ_TYPE_CEQ 1
+
+struct erdma_cmdq_create_eq_req {
+	u64 hdr;
+	u64 qbuf_addr;
+	u8 vector_idx;
+	u8 eqn;
+	u8 depth;
+	u8 qtype;
+	u32 db_dma_addr_l;
+	u32 db_dma_addr_h;
+};
+
+struct erdma_cmdq_destroy_eq_req {
+	u64 hdr;
+	u64 rsvd0;
+	u8 vector_idx;
+	u8 eqn;
+	u8 rsvd1;
+	u8 qtype;
+};
+
+/* create_cq cfg0 */
+#define ERDMA_CMD_CREATE_CQ_DEPTH_MASK GENMASK(31, 24)
+#define ERDMA_CMD_CREATE_CQ_PAGESIZE_MASK GENMASK(23, 20)
+#define ERDMA_CMD_CREATE_CQ_CQN_MASK GENMASK(19, 0)
+
+/* create_cq cfg1 */
+#define ERDMA_CMD_CREATE_CQ_MTT_CNT_MASK GENMASK(31, 16)
+#define ERDMA_CMD_CREATE_CQ_MTT_TYPE_MASK BIT(15)
+#define ERDMA_CMD_CREATE_CQ_EQN_MASK GENMASK(9, 0)
+
+struct erdma_cmdq_create_cq_req {
+	u64 hdr;
+	u32 cfg0;
+	u32 qbuf_addr_l;
+	u32 qbuf_addr_h;
+	u32 cfg1;
+	u64 cq_db_info_addr;
+	u32 first_page_offset;
+};
+
+/* regmr/deregmr cfg0 */
+#define ERDMA_CMD_MR_VALID_MASK BIT(31)
+#define ERDMA_CMD_MR_KEY_MASK GENMASK(27, 20)
+#define ERDMA_CMD_MR_MPT_IDX_MASK GENMASK(19, 0)
+
+/* regmr cfg1 */
+#define ERDMA_CMD_REGMR_PD_MASK GENMASK(31, 12)
+#define ERDMA_CMD_REGMR_TYPE_MASK GENMASK(7, 6)
+#define ERDMA_CMD_REGMR_RIGHT_MASK GENMASK(5, 2)
+#define ERDMA_CMD_REGMR_ACC_MODE_MASK GENMASK(1, 0)
+
+/* regmr cfg2 */
+#define ERDMA_CMD_REGMR_PAGESIZE_MASK GENMASK(31, 27)
+#define ERDMA_CMD_REGMR_MTT_TYPE_MASK GENMASK(21, 20)
+#define ERDMA_CMD_REGMR_MTT_CNT_MASK GENMASK(19, 0)
+
+struct erdma_cmdq_reg_mr_req {
+	u64 hdr;
+	u32 cfg0;
+	u32 cfg1;
+	u64 start_va;
+	u32 size;
+	u32 cfg2;
+	u64 phy_addr[4];
+};
+
+struct erdma_cmdq_dereg_mr_req {
+	u64 hdr;
+	u32 cfg;
+};
+
+/* modify qp cfg */
+#define ERDMA_CMD_MODIFY_QP_STATE_MASK GENMASK(31, 24)
+#define ERDMA_CMD_MODIFY_QP_CC_MASK GENMASK(23, 20)
+#define ERDMA_CMD_MODIFY_QP_QPN_MASK GENMASK(19, 0)
+
+struct erdma_cmdq_modify_qp_req {
+	u64 hdr;
+	u32 cfg;
+	u32 cookie;
+	__be32 dip;
+	__be32 sip;
+	__be16 sport;
+	__be16 dport;
+	u32 send_nxt;
+	u32 recv_nxt;
+};
+
+/* create qp cfg0 */
+#define ERDMA_CMD_CREATE_QP_SQ_DEPTH_MASK GENMASK(31, 20)
+#define ERDMA_CMD_CREATE_QP_QPN_MASK GENMASK(19, 0)
+
+/* create qp cfg1 */
+#define ERDMA_CMD_CREATE_QP_RQ_DEPTH_MASK GENMASK(31, 20)
+#define ERDMA_CMD_CREATE_QP_PD_MASK GENMASK(19, 0)
+
+/* create qp cqn_mtt_cfg */
+#define ERDMA_CMD_CREATE_QP_PAGE_SIZE_MASK GENMASK(31, 28)
+#define ERDMA_CMD_CREATE_QP_CQN_MASK GENMASK(23, 0)
+
+/* create qp mtt_cfg */
+#define ERDMA_CMD_CREATE_QP_PAGE_OFFSET_MASK GENMASK(31, 12)
+#define ERDMA_CMD_CREATE_QP_MTT_CNT_MASK GENMASK(11, 1)
+#define ERDMA_CMD_CREATE_QP_MTT_TYPE_MASK BIT(0)
+
+#define ERDMA_CMDQ_CREATE_QP_RESP_COOKIE_MASK GENMASK_ULL(31, 0)
+
+struct erdma_cmdq_create_qp_req {
+	u64 hdr;
+	u32 cfg0;
+	u32 cfg1;
+	u32 sq_cqn_mtt_cfg;
+	u32 rq_cqn_mtt_cfg;
+	u64 sq_buf_addr;
+	u64 rq_buf_addr;
+	u32 sq_mtt_cfg;
+	u32 rq_mtt_cfg;
+	u64 sq_db_info_dma_addr;
+	u64 rq_db_info_dma_addr;
+};
+
+struct erdma_cmdq_destroy_qp_req {
+	u64 hdr;
+	u32 qpn;
+};
+
+/* cap qword 0 definition */
+#define ERDMA_CMD_DEV_CAP_MAX_CQE_MASK GENMASK_ULL(47, 40)
+#define ERDMA_CMD_DEV_CAP_MAX_RECV_WR_MASK GENMASK_ULL(23, 16)
+#define ERDMA_CMD_DEV_CAP_MAX_MR_SIZE_MASK GENMASK_ULL(7, 0)
+
+/* cap qword 1 definition */
+#define ERDMA_CMD_DEV_CAP_DMA_LOCAL_KEY_MASK GENMASK_ULL(63, 32)
+#define ERDMA_CMD_DEV_CAP_DEFAULT_CC_MASK GENMASK_ULL(31, 28)
+#define ERDMA_CMD_DEV_CAP_QBLOCK_MASK GENMASK_ULL(27, 16)
+#define ERDMA_CMD_DEV_CAP_MAX_MW_MASK GENMASK_ULL(7, 0)
+
+#define ERDMA_NQP_PER_QBLOCK 1024
+
+#define ERDMA_CMD_INFO0_FW_VER_MASK GENMASK_ULL(31, 0)
+
+/* CQE hdr */
+#define ERDMA_CQE_HDR_OWNER_MASK BIT(31)
+#define ERDMA_CQE_HDR_OPCODE_MASK GENMASK(23, 16)
+#define ERDMA_CQE_HDR_QTYPE_MASK GENMASK(15, 8)
+#define ERDMA_CQE_HDR_SYNDROME_MASK GENMASK(7, 0)
+
+#define ERDMA_CQE_QTYPE_SQ 0
+#define ERDMA_CQE_QTYPE_RQ 1
+#define ERDMA_CQE_QTYPE_CMDQ 2
+
+struct erdma_cqe {
+	__be32 hdr;
+	__be32 qe_idx;
+	__be32 qpn;
+	union {
+		__le32 imm_data;
+		__be32 inv_rkey;
+	};
+	__be32 size;
+	__be32 rsvd[3];
+};
+
+struct erdma_sge {
+	__aligned_le64 laddr;
+	__le32 length;
+	__le32 lkey;
+};
+
+/* Receive Queue Element */
+struct erdma_rqe {
+	__le16 qe_idx;
+	__le16 rsvd0;
+	__le32 qpn;
+	__le32 rsvd1;
+	__le32 rsvd2;
+	__le64 to;
+	__le32 length;
+	__le32 stag;
+};
+
+/* SQE */
+#define ERDMA_SQE_HDR_SGL_LEN_MASK GENMASK_ULL(63, 56)
+#define ERDMA_SQE_HDR_WQEBB_CNT_MASK GENMASK_ULL(54, 52)
+#define ERDMA_SQE_HDR_QPN_MASK GENMASK_ULL(51, 32)
+#define ERDMA_SQE_HDR_OPCODE_MASK GENMASK_ULL(31, 27)
+#define ERDMA_SQE_HDR_DWQE_MASK BIT_ULL(26)
+#define ERDMA_SQE_HDR_INLINE_MASK BIT_ULL(25)
+#define ERDMA_SQE_HDR_FENCE_MASK BIT_ULL(24)
+#define ERDMA_SQE_HDR_SE_MASK BIT_ULL(23)
+#define ERDMA_SQE_HDR_CE_MASK BIT_ULL(22)
+#define ERDMA_SQE_HDR_WQEBB_INDEX_MASK GENMASK_ULL(15, 0)
+
+/* REG MR attrs */
+#define ERDMA_SQE_MR_MODE_MASK GENMASK(1, 0)
+#define ERDMA_SQE_MR_ACCESS_MASK GENMASK(5, 2)
+#define ERDMA_SQE_MR_MTT_TYPE_MASK GENMASK(7, 6)
+#define ERDMA_SQE_MR_MTT_CNT_MASK GENMASK(31, 12)
+
+struct erdma_write_sqe {
+	__le64 hdr;
+	__be32 imm_data;
+	__le32 length;
+
+	__le32 sink_stag;
+	__le32 sink_to_l;
+	__le32 sink_to_h;
+
+	__le32 rsvd;
+
+	struct erdma_sge sgl[0];
+};
+
+struct erdma_send_sqe {
+	__le64 hdr;
+	union {
+		__be32 imm_data;
+		__le32 invalid_stag;
+	};
+
+	__le32 length;
+	struct erdma_sge sgl[0];
+};
+
+struct erdma_readreq_sqe {
+	__le64 hdr;
+	__le32 invalid_stag;
+	__le32 length;
+	__le32 sink_stag;
+	__le32 sink_to_l;
+	__le32 sink_to_h;
+	__le32 rsvd;
+};
+
+struct erdma_reg_mr_sqe {
+	__le64 hdr;
+	__le64 addr;
+	__le32 length;
+	__le32 stag;
+	__le32 attrs;
+	__le32 rsvd;
+};
+
+/* EQ related. */
+#define ERDMA_DEFAULT_EQ_DEPTH 256
+
+/* ceqe */
+#define ERDMA_CEQE_HDR_DB_MASK BIT_ULL(63)
+#define ERDMA_CEQE_HDR_PI_MASK GENMASK_ULL(55, 32)
+#define ERDMA_CEQE_HDR_O_MASK BIT_ULL(31)
+#define ERDMA_CEQE_HDR_CQN_MASK GENMASK_ULL(19, 0)
+
+/* aeqe */
+#define ERDMA_AEQE_HDR_O_MASK BIT(31)
+#define ERDMA_AEQE_HDR_TYPE_MASK GENMASK(23, 16)
+#define ERDMA_AEQE_HDR_SUBTYPE_MASK GENMASK(7, 0)
+
+#define ERDMA_AE_TYPE_QP_FATAL_EVENT 0
+#define ERDMA_AE_TYPE_QP_ERQ_ERR_EVENT 1
+#define ERDMA_AE_TYPE_ACC_ERR_EVENT 2
+#define ERDMA_AE_TYPE_CQ_ERR 3
+#define ERDMA_AE_TYPE_OTHER_ERROR 4
+
+struct erdma_aeqe {
+	__le32 hdr;
+	__le32 event_data0;
+	__le32 event_data1;
+	__le32 rsvd;
+};
+
+enum erdma_opcode {
+	ERDMA_OP_WRITE = 0,
+	ERDMA_OP_READ = 1,
+	ERDMA_OP_SEND = 2,
+	ERDMA_OP_SEND_WITH_IMM = 3,
+
+	ERDMA_OP_RECEIVE = 4,
+	ERDMA_OP_RECV_IMM = 5,
+	ERDMA_OP_RECV_INV = 6,
+
+	ERDMA_OP_REQ_ERR = 7,
+	ERDMA_OP_READ_RESPONSE = 8,
+	ERDMA_OP_WRITE_WITH_IMM = 9,
+
+	ERDMA_OP_RECV_ERR = 10,
+
+	ERDMA_OP_INVALIDATE = 11,
+	ERDMA_OP_RSP_SEND_IMM = 12,
+	ERDMA_OP_SEND_WITH_INV = 13,
+
+	ERDMA_OP_REG_MR = 14,
+	ERDMA_OP_LOCAL_INV = 15,
+	ERDMA_OP_READ_WITH_INV = 16,
+	ERDMA_NUM_OPCODES = 17,
+	ERDMA_OP_INVALID = ERDMA_NUM_OPCODES + 1
+};
+
+enum erdma_wc_status {
+	ERDMA_WC_SUCCESS = 0,
+	ERDMA_WC_GENERAL_ERR = 1,
+	ERDMA_WC_RECV_WQE_FORMAT_ERR = 2,
+	ERDMA_WC_RECV_STAG_INVALID_ERR = 3,
+	ERDMA_WC_RECV_ADDR_VIOLATION_ERR = 4,
+	ERDMA_WC_RECV_RIGHT_VIOLATION_ERR = 5,
+	ERDMA_WC_RECV_PDID_ERR = 6,
+	ERDMA_WC_RECV_WARRPING_ERR = 7,
+	ERDMA_WC_SEND_WQE_FORMAT_ERR = 8,
+	ERDMA_WC_SEND_WQE_ORD_EXCEED = 9,
+	ERDMA_WC_SEND_STAG_INVALID_ERR = 10,
+	ERDMA_WC_SEND_ADDR_VIOLATION_ERR = 11,
+	ERDMA_WC_SEND_RIGHT_VIOLATION_ERR = 12,
+	ERDMA_WC_SEND_PDID_ERR = 13,
+	ERDMA_WC_SEND_WARRPING_ERR = 14,
+	ERDMA_WC_FLUSH_ERR = 15,
+	ERDMA_WC_RETRY_EXC_ERR = 16,
+	ERDMA_NUM_WC_STATUS
+};
+
+enum erdma_vendor_err {
+	ERDMA_WC_VENDOR_NO_ERR = 0,
+	ERDMA_WC_VENDOR_INVALID_RQE = 1,
+	ERDMA_WC_VENDOR_RQE_INVALID_STAG = 2,
+	ERDMA_WC_VENDOR_RQE_ADDR_VIOLATION = 3,
+	ERDMA_WC_VENDOR_RQE_ACCESS_RIGHT_ERR = 4,
+	ERDMA_WC_VENDOR_RQE_INVALID_PD = 5,
+	ERDMA_WC_VENDOR_RQE_WRAP_ERR = 6,
+	ERDMA_WC_VENDOR_INVALID_SQE = 0x20,
+	ERDMA_WC_VENDOR_ZERO_ORD = 0x21,
+	ERDMA_WC_VENDOR_SQE_INVALID_STAG = 0x30,
+	ERDMA_WC_VENDOR_SQE_ADDR_VIOLATION = 0x31,
+	ERDMA_WC_VENDOR_SQE_ACCESS_ERR = 0x32,
+	ERDMA_WC_VENDOR_SQE_INVALID_PD = 0x33,
+	ERDMA_WC_VENDOR_SQE_WARP_ERR = 0x34
+};
+
+#endif

drivers/infiniband/hw/erdma/erdma_main.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <net/addrconf.h>
+#include <rdma/erdma-abi.h>
+#include <rdma/ib_verbs.h>
+#include <rdma/ib_user_verbs.h>
+
+#include "erdma.h"
+#include "erdma_cm.h"
+#include "erdma_hw.h"
+#include "erdma_verbs.h"
+
+MODULE_AUTHOR("Cheng Xu <[e-mail]>");
+MODULE_DESCRIPTION("Alibaba elasticRDMA adapter driver");
+MODULE_LICENSE("Dual BSD/GPL");
+
+static int erdma_netdev_event(struct notifier_block *nb, unsigned long event,
+			      void *arg)
+{
+	struct net_device *netdev = netdev_notifier_info_to_dev(arg);
+	struct erdma_dev *dev = container_of(nb, struct erdma_dev, netdev_nb);
+
+	if (dev->netdev == NULL || dev->netdev != netdev)
+		goto done;
+
+	switch (event) {
+	case NETDEV_UP:
+		dev->state = IB_PORT_ACTIVE;
+		erdma_port_event(dev, IB_EVENT_PORT_ACTIVE);
+		break;
+	case NETDEV_DOWN:
+		dev->state = IB_PORT_DOWN;
+		erdma_port_event(dev, IB_EVENT_PORT_ERR);
+		break;
+	case NETDEV_REGISTER:
+	case NETDEV_UNREGISTER:
+	case NETDEV_CHANGEADDR:
+	case NETDEV_CHANGEMTU:
+	case NETDEV_GOING_DOWN:
+	case NETDEV_CHANGE:
+	default:
+		break;
+	}
+
+done:
+	return NOTIFY_OK;
+}
+
+static int erdma_enum_and_get_netdev(struct erdma_dev *dev)
+{
+	struct net_device *netdev;
+	int ret = -ENODEV;
+
+	/* Already binded to a net_device, so we skip. */
+	if (dev->netdev)
+		return 0;
+
+	rtnl_lock();
+	for_each_netdev(&init_net, netdev) {
+		/*
+		 * In erdma, the paired netdev and ibdev should have the same
+		 * MAC address. erdma can get the value from its PCIe bar
+		 * registers. Since erdma can not get the paired netdev
+		 * reference directly, we do a traverse here to get the paired
+		 * netdev.
+		 */
+		if (ether_addr_equal_unaligned(netdev->perm_addr,
+					       dev->attrs.peer_addr)) {
+			ret = ib_device_set_netdev(&dev->ibdev, netdev, 1);
+			if (ret) {
+				rtnl_unlock();
+				ibdev_warn(&dev->ibdev,
+					   "failed (%d) to link netdev", ret);
+				return ret;
+			}
+
+			dev->netdev = netdev;
+			break;
+		}
+	}
+
+	rtnl_unlock();
+
+	return ret;
+}
+
+static int erdma_device_register(struct erdma_dev *dev)
+{
+	struct ib_device *ibdev = &dev->ibdev;
+	int ret;
+
+	ret = erdma_enum_and_get_netdev(dev);
+	if (ret)
+		return ret;
+
+	addrconf_addr_eui48((u8 *)&ibdev->node_guid, dev->netdev->dev_addr);
+
+	ret = ib_register_device(ibdev, "erdma_%d", &dev->pdev->dev);
+	if (ret) {
+		dev_err(&dev->pdev->dev,
+			"ib_register_device failed: ret = %d\n", ret);
+		return ret;
+	}
+
+	dev->netdev_nb.notifier_call = erdma_netdev_event;
+	ret = register_netdevice_notifier(&dev->netdev_nb);
+	if (ret) {
+		ibdev_err(&dev->ibdev, "failed to register notifier.\n");
+		ib_unregister_device(ibdev);
+	}
+
+	return ret;
+}
+
+static irqreturn_t erdma_comm_irq_handler(int irq, void *data)
+{
+	struct erdma_dev *dev = data;
+
+	erdma_cmdq_completion_handler(&dev->cmdq);
+	erdma_aeq_event_handler(dev);
+
+	return IRQ_HANDLED;
+}
+
+static void erdma_dwqe_resource_init(struct erdma_dev *dev)
+{
+	int total_pages, type0, type1;
+
+	dev->attrs.grp_num = erdma_reg_read32(dev, ERDMA_REGS_GRP_NUM_REG);
+
+	if (dev->attrs.grp_num < 4)
+		dev->attrs.disable_dwqe = true;
+	else
+		dev->attrs.disable_dwqe = false;
+
+	/* One page contains 4 goups. */
+	total_pages = dev->attrs.grp_num * 4;
+
+	if (dev->attrs.grp_num >= ERDMA_DWQE_MAX_GRP_CNT) {
+		dev->attrs.grp_num = ERDMA_DWQE_MAX_GRP_CNT;
+		type0 = ERDMA_DWQE_TYPE0_CNT;
+		type1 = ERDMA_DWQE_TYPE1_CNT / ERDMA_DWQE_TYPE1_CNT_PER_PAGE;
+	} else {
+		type1 = total_pages / 3;
+		type0 = total_pages - type1 - 1;
+	}
+
+	dev->attrs.dwqe_pages = type0;
+	dev->attrs.dwqe_entries = type1 * ERDMA_DWQE_TYPE1_CNT_PER_PAGE;
+}
+
+static int erdma_request_vectors(struct erdma_dev *dev)
+{
+	int expect_irq_num = min(num_possible_cpus() + 1, ERDMA_NUM_MSIX_VEC);
+	int ret;
+
+	ret = pci_alloc_irq_vectors(dev->pdev, 1, expect_irq_num, PCI_IRQ_MSIX);
+	if (ret < 0) {
+		dev_err(&dev->pdev->dev, "request irq vectors failed(%d)\n",
+			ret);
+		return ret;
+	}
+	dev->attrs.irq_num = ret;
+
+	return 0;
+}
+
+static int erdma_comm_irq_init(struct erdma_dev *dev)
+{
+	snprintf(dev->comm_irq.name, ERDMA_IRQNAME_SIZE, "erdma-common@pci:%s",
+		 pci_name(dev->pdev));
+	dev->comm_irq.msix_vector =
+		pci_irq_vector(dev->pdev, ERDMA_MSIX_VECTOR_CMDQ);
+
+	cpumask_set_cpu(cpumask_first(cpumask_of_pcibus(dev->pdev->bus)),
+			&dev->comm_irq.affinity_hint_mask);
+	irq_set_affinity_hint(dev->comm_irq.msix_vector,
+			      &dev->comm_irq.affinity_hint_mask);
+
+	return request_irq(dev->comm_irq.msix_vector, erdma_comm_irq_handler, 0,
+			   dev->comm_irq.name, dev);
+}
+
+static void erdma_comm_irq_uninit(struct erdma_dev *dev)
+{
+	irq_set_affinity_hint(dev->comm_irq.msix_vector, NULL);
+	free_irq(dev->comm_irq.msix_vector, dev);
+}
+
+static int erdma_device_init(struct erdma_dev *dev, struct pci_dev *pdev)
+{
+	int ret;
+
+	erdma_dwqe_resource_init(dev);
+
+	ret = dma_set_mask_and_coherent(&pdev->dev,
+					DMA_BIT_MASK(ERDMA_PCI_WIDTH));
+	if (ret)
+		return ret;
+
+	dma_set_max_seg_size(&pdev->dev, UINT_MAX);
+
+	return 0;
+}
+
+static void erdma_device_uninit(struct erdma_dev *dev)
+{
+	u32 ctrl = FIELD_PREP(ERDMA_REG_DEV_CTRL_RESET_MASK, 1);
+
+	erdma_reg_write32(dev, ERDMA_REGS_DEV_CTRL_REG, ctrl);
+}
+
+static const struct pci_device_id erdma_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_ALIBABA, 0x107f) },
+	{}
+};
+
+static int erdma_probe_dev(struct pci_dev *pdev)
+{
+	struct erdma_dev *dev;
+	int bars, err;
+	u32 version;
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "pci_enable_device failed(%d)\n", err);
+		return err;
+	}
+
+	pci_set_master(pdev);
+
+	dev = ib_alloc_device(erdma_dev, ibdev);
+	if (!dev) {
+		dev_err(&pdev->dev, "ib_alloc_device failed\n");
+		err = -ENOMEM;
+		goto err_disable_device;
+	}
+
+	pci_set_drvdata(pdev, dev);
+	dev->pdev = pdev;
+	dev->attrs.numa_node = dev_to_node(&pdev->dev);
+
+	bars = pci_select_bars(pdev, IORESOURCE_MEM);
+	err = pci_request_selected_regions(pdev, bars, DRV_MODULE_NAME);
+	if (bars != ERDMA_BAR_MASK || err) {
+		err = err ? err : -EINVAL;
+		goto err_ib_device_release;
+	}
+
+	dev->func_bar_addr = pci_resource_start(pdev, ERDMA_FUNC_BAR);
+	dev->func_bar_len = pci_resource_len(pdev, ERDMA_FUNC_BAR);
+
+	dev->func_bar =
+		devm_ioremap(&pdev->dev, dev->func_bar_addr, dev->func_bar_len);
+	if (!dev->func_bar) {
+		dev_err(&pdev->dev, "devm_ioremap failed.\n");
+		err = -EFAULT;
+		goto err_release_bars;
+	}
+
+	version = erdma_reg_read32(dev, ERDMA_REGS_VERSION_REG);
+	if (version == 0) {
+		/* we knows that it is a non-functional function. */
+		err = -ENODEV;
+		goto err_iounmap_func_bar;
+	}
+
+	err = erdma_device_init(dev, pdev);
+	if (err)
+		goto err_iounmap_func_bar;
+
+	err = erdma_request_vectors(dev);
+	if (err)
+		goto err_iounmap_func_bar;
+
+	err = erdma_comm_irq_init(dev);
+	if (err)
+		goto err_free_vectors;
+
+	err = erdma_aeq_init(dev);
+	if (err)
+		goto err_uninit_comm_irq;
+
+	err = erdma_cmdq_init(dev);
+	if (err)
+		goto err_uninit_aeq;
+
+	err = erdma_ceqs_init(dev);
+	if (err)
+		goto err_uninit_cmdq;
+
+	erdma_finish_cmdq_init(dev);
+
+	return 0;
+
+err_uninit_cmdq:
+	erdma_device_uninit(dev);
+	erdma_cmdq_destroy(dev);
+
+err_uninit_aeq:
+	erdma_aeq_destroy(dev);
+
+err_uninit_comm_irq:
+	erdma_comm_irq_uninit(dev);
+
+err_free_vectors:
+	pci_free_irq_vectors(dev->pdev);
+
+err_iounmap_func_bar:
+	devm_iounmap(&pdev->dev, dev->func_bar);
+
+err_release_bars:
+	pci_release_selected_regions(pdev, bars);
+
+err_ib_device_release:
+	ib_dealloc_device(&dev->ibdev);
+
+err_disable_device:
+	pci_disable_device(pdev);
+
+	return err;
+}
+
+static void erdma_remove_dev(struct pci_dev *pdev)
+{
+	struct erdma_dev *dev = pci_get_drvdata(pdev);
+
+	erdma_ceqs_uninit(dev);
+
+	erdma_device_uninit(dev);
+
+	erdma_cmdq_destroy(dev);
+	erdma_aeq_destroy(dev);
+	erdma_comm_irq_uninit(dev);
+	pci_free_irq_vectors(dev->pdev);
+
+	devm_iounmap(&pdev->dev, dev->func_bar);
+	pci_release_selected_regions(pdev, ERDMA_BAR_MASK);
+
+	ib_dealloc_device(&dev->ibdev);
+
+	pci_disable_device(pdev);
+}
+
+#define ERDMA_GET_CAP(name, cap) FIELD_GET(ERDMA_CMD_DEV_CAP_##name##_MASK, cap)
+
+static int erdma_dev_attrs_init(struct erdma_dev *dev)
+{
+	int err;
+	u64 req_hdr, cap0, cap1;
+
+	erdma_cmdq_build_reqhdr(&req_hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_QUERY_DEVICE);
+
+	err = erdma_post_cmd_wait(&dev->cmdq, &req_hdr, sizeof(req_hdr), &cap0,
+				  &cap1);
+	if (err)
+		return err;
+
+	dev->attrs.max_cqe = 1 << ERDMA_GET_CAP(MAX_CQE, cap0);
+	dev->attrs.max_mr_size = 1ULL << ERDMA_GET_CAP(MAX_MR_SIZE, cap0);
+	dev->attrs.max_mw = 1 << ERDMA_GET_CAP(MAX_MW, cap1);
+	dev->attrs.max_recv_wr = 1 << ERDMA_GET_CAP(MAX_RECV_WR, cap0);
+	dev->attrs.local_dma_key = ERDMA_GET_CAP(DMA_LOCAL_KEY, cap1);
+	dev->attrs.cc = ERDMA_GET_CAP(DEFAULT_CC, cap1);
+	dev->attrs.max_qp = ERDMA_NQP_PER_QBLOCK * ERDMA_GET_CAP(QBLOCK, cap1);
+	dev->attrs.max_mr = dev->attrs.max_qp << 1;
+	dev->attrs.max_cq = dev->attrs.max_qp << 1;
+
+	dev->attrs.max_send_wr = ERDMA_MAX_SEND_WR;
+	dev->attrs.max_ord = ERDMA_MAX_ORD;
+	dev->attrs.max_ird = ERDMA_MAX_IRD;
+	dev->attrs.max_send_sge = ERDMA_MAX_SEND_SGE;
+	dev->attrs.max_recv_sge = ERDMA_MAX_RECV_SGE;
+	dev->attrs.max_sge_rd = ERDMA_MAX_SGE_RD;
+	dev->attrs.max_pd = ERDMA_MAX_PD;
+
+	dev->res_cb[ERDMA_RES_TYPE_PD].max_cap = ERDMA_MAX_PD;
+	dev->res_cb[ERDMA_RES_TYPE_STAG_IDX].max_cap = dev->attrs.max_mr;
+
+	erdma_cmdq_build_reqhdr(&req_hdr, CMDQ_SUBMOD_COMMON,
+				CMDQ_OPCODE_QUERY_FW_INFO);
+
+	err = erdma_post_cmd_wait(&dev->cmdq, &req_hdr, sizeof(req_hdr), &cap0,
+				  &cap1);
+	if (!err)
+		dev->attrs.fw_version =
+			FIELD_GET(ERDMA_CMD_INFO0_FW_VER_MASK, cap0);
+
+	return err;
+}
+
+static int erdma_res_cb_init(struct erdma_dev *dev)
+{
+	int i, j;
+
+	for (i = 0; i < ERDMA_RES_CNT; i++) {
+		dev->res_cb[i].next_alloc_idx = 1;
+		spin_lock_init(&dev->res_cb[i].lock);
+		dev->res_cb[i].bitmap =
+			bitmap_zalloc(dev->res_cb[i].max_cap, GFP_KERNEL);
+		if (!dev->res_cb[i].bitmap)
+			goto err;
+	}
+
+	return 0;
+
+err:
+	for (j = 0; j < i; j++)
+		bitmap_free(dev->res_cb[j].bitmap);
+
+	return -ENOMEM;
+}
+
+static void erdma_res_cb_free(struct erdma_dev *dev)
+{
+	int i;
+
+	for (i = 0; i < ERDMA_RES_CNT; i++)
+		bitmap_free(dev->res_cb[i].bitmap);
+}
+
+static const struct ib_device_ops erdma_device_ops = {
+	.owner = THIS_MODULE,
+	.driver_id = RDMA_DRIVER_ERDMA,
+	.uverbs_abi_ver = ERDMA_ABI_VERSION,
+
+	.alloc_mr = erdma_ib_alloc_mr,
+	.alloc_pd = erdma_alloc_pd,
+	.alloc_ucontext = erdma_alloc_ucontext,
+	.create_cq = erdma_create_cq,
+	.create_qp = erdma_create_qp,
+	.dealloc_pd = erdma_dealloc_pd,
+	.dealloc_ucontext = erdma_dealloc_ucontext,
+	.dereg_mr = erdma_dereg_mr,
+	.destroy_cq = erdma_destroy_cq,
+	.destroy_qp = erdma_destroy_qp,
+	.get_dma_mr = erdma_get_dma_mr,
+	.get_port_immutable = erdma_get_port_immutable,
+	.iw_accept = erdma_accept,
+	.iw_add_ref = erdma_qp_get_ref,
+	.iw_connect = erdma_connect,
+	.iw_create_listen = erdma_create_listen,
+	.iw_destroy_listen = erdma_destroy_listen,
+	.iw_get_qp = erdma_get_ibqp,
+	.iw_reject = erdma_reject,
+	.iw_rem_ref = erdma_qp_put_ref,
+	.map_mr_sg = erdma_map_mr_sg,
+	.mmap = erdma_mmap,
+	.mmap_free = erdma_mmap_free,
+	.modify_qp = erdma_modify_qp,
+	.post_recv = erdma_post_recv,
+	.post_send = erdma_post_send,
+	.poll_cq = erdma_poll_cq,
+	.query_device = erdma_query_device,
+	.query_gid = erdma_query_gid,
+	.query_port = erdma_query_port,
+	.query_qp = erdma_query_qp,
+	.req_notify_cq = erdma_req_notify_cq,
+	.reg_user_mr = erdma_reg_user_mr,
+
+	INIT_RDMA_OBJ_SIZE(ib_cq, erdma_cq, ibcq),
+	INIT_RDMA_OBJ_SIZE(ib_pd, erdma_pd, ibpd),
+	INIT_RDMA_OBJ_SIZE(ib_ucontext, erdma_ucontext, ibucontext),
+	INIT_RDMA_OBJ_SIZE(ib_qp, erdma_qp, ibqp),
+};
+
+static int erdma_ib_device_add(struct pci_dev *pdev)
+{
+	struct erdma_dev *dev = pci_get_drvdata(pdev);
+	struct ib_device *ibdev = &dev->ibdev;
+	u64 mac;
+	int ret;
+
+	ret = erdma_dev_attrs_init(dev);
+	if (ret)
+		return ret;
+
+	ibdev->node_type = RDMA_NODE_RNIC;
+	memcpy(ibdev->node_desc, ERDMA_NODE_DESC, sizeof(ERDMA_NODE_DESC));
+
+	/*
+	 * Current model (one-to-one device association):
+	 * One ERDMA device per net_device or, equivalently,
+	 * per physical port.
+	 */
+	ibdev->phys_port_cnt = 1;
+	ibdev->num_comp_vectors = dev->attrs.irq_num - 1;
+
+	ib_set_device_ops(ibdev, &erdma_device_ops);
+
+	INIT_LIST_HEAD(&dev->cep_list);
+
+	spin_lock_init(&dev->lock);
+	xa_init_flags(&dev->qp_xa, XA_FLAGS_ALLOC1);
+	xa_init_flags(&dev->cq_xa, XA_FLAGS_ALLOC1);
+	dev->next_alloc_cqn = 1;
+	dev->next_alloc_qpn = 1;
+
+	ret = erdma_res_cb_init(dev);
+	if (ret)
+		return ret;
+
+	spin_lock_init(&dev->db_bitmap_lock);
+	bitmap_zero(dev->sdb_page, ERDMA_DWQE_TYPE0_CNT);
+	bitmap_zero(dev->sdb_entry, ERDMA_DWQE_TYPE1_CNT);
+
+	atomic_set(&dev->num_ctx, 0);
+
+	mac = erdma_reg_read32(dev, ERDMA_REGS_NETDEV_MAC_L_REG);
+	mac |= (u64)erdma_reg_read32(dev, ERDMA_REGS_NETDEV_MAC_H_REG) << 32;
+
+	u64_to_ether_addr(mac, dev->attrs.peer_addr);
+
+	ret = erdma_device_register(dev);
+	if (ret)
+		goto err_out;
+
+	return 0;
+
+err_out:
+	xa_destroy(&dev->qp_xa);
+	xa_destroy(&dev->cq_xa);
+
+	erdma_res_cb_free(dev);
+
+	return ret;
+}
+
+static void erdma_ib_device_remove(struct pci_dev *pdev)
+{
+	struct erdma_dev *dev = pci_get_drvdata(pdev);
+
+	unregister_netdevice_notifier(&dev->netdev_nb);
+	ib_unregister_device(&dev->ibdev);
+
+	erdma_res_cb_free(dev);
+	xa_destroy(&dev->qp_xa);
+	xa_destroy(&dev->cq_xa);
+}
+
+static int erdma_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	int ret;
+
+	ret = erdma_probe_dev(pdev);
+	if (ret)
+		return ret;
+
+	ret = erdma_ib_device_add(pdev);
+	if (ret) {
+		erdma_remove_dev(pdev);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void erdma_remove(struct pci_dev *pdev)
+{
+	erdma_ib_device_remove(pdev);
+	erdma_remove_dev(pdev);
+}
+
+static struct pci_driver erdma_pci_driver = {
+	.name = DRV_MODULE_NAME,
+	.id_table = erdma_pci_tbl,
+	.probe = erdma_probe,
+	.remove = erdma_remove
+};
+
+MODULE_DEVICE_TABLE(pci, erdma_pci_tbl);
+
+static __init int erdma_init_module(void)
+{
+	int ret;
+
+	ret = erdma_cm_init();
+	if (ret)
+		return ret;
+
+	ret = pci_register_driver(&erdma_pci_driver);
+	if (ret)
+		erdma_cm_exit();
+
+	return ret;
+}
+
+static void __exit erdma_exit_module(void)
+{
+	pci_unregister_driver(&erdma_pci_driver);
+
+	erdma_cm_exit();
+}
+
+module_init(erdma_init_module);
+module_exit(erdma_exit_module);

drivers/infiniband/hw/erdma/erdma_qp.c

···
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2021, Alibaba Group */
+/* Authors: Bernard Metzler <[e-mail]> */
+/* Copyright (c) 2008-2019, IBM Corporation */
+
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/scatterlist.h>
+#include <linux/types.h>
+
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
+
+#include "erdma.h"
+#include "erdma_cm.h"
+#include "erdma_verbs.h"
+
+void erdma_qp_llp_close(struct erdma_qp *qp)
+{
+	struct erdma_qp_attrs qp_attrs;
+
+	down_write(&qp->state_lock);
+
+	switch (qp->attrs.state) {
+	case ERDMA_QP_STATE_RTS:
+	case ERDMA_QP_STATE_RTR:
+	case ERDMA_QP_STATE_IDLE:
+	case ERDMA_QP_STATE_TERMINATE:
+		qp_attrs.state = ERDMA_QP_STATE_CLOSING;
+		erdma_modify_qp_internal(qp, &qp_attrs, ERDMA_QP_ATTR_STATE);
+		break;
+	case ERDMA_QP_STATE_CLOSING:
+		qp->attrs.state = ERDMA_QP_STATE_IDLE;
+		break;
+	default:
+		break;
+	}
+
+	if (qp->cep) {
+		erdma_cep_put(qp->cep);
+		qp->cep = NULL;
+	}
+
+	up_write(&qp->state_lock);
+}
+
+struct ib_qp *erdma_get_ibqp(struct ib_device *ibdev, int id)
+{
+	struct erdma_qp *qp = find_qp_by_qpn(to_edev(ibdev), id);
+
+	if (qp)
+		return &qp->ibqp;
+
+	return NULL;
+}
+
+static int erdma_modify_qp_state_to_rts(struct erdma_qp *qp,
+					struct erdma_qp_attrs *attrs,
+					enum erdma_qp_attr_mask mask)
+{
+	int ret;
+	struct erdma_dev *dev = qp->dev;
+	struct erdma_cmdq_modify_qp_req req;
+	struct tcp_sock *tp;
+	struct erdma_cep *cep = qp->cep;
+	struct sockaddr_storage local_addr, remote_addr;
+
+	if (!(mask & ERDMA_QP_ATTR_LLP_HANDLE))
+		return -EINVAL;
+
+	if (!(mask & ERDMA_QP_ATTR_MPA))
+		return -EINVAL;
+
+	ret = getname_local(cep->sock, &local_addr);
+	if (ret < 0)
+		return ret;
+
+	ret = getname_peer(cep->sock, &remote_addr);
+	if (ret < 0)
+		return ret;
+
+	qp->attrs.state = ERDMA_QP_STATE_RTS;
+
+	tp = tcp_sk(qp->cep->sock->sk);
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_MODIFY_QP);
+
+	req.cfg = FIELD_PREP(ERDMA_CMD_MODIFY_QP_STATE_MASK, qp->attrs.state) |
+		  FIELD_PREP(ERDMA_CMD_MODIFY_QP_CC_MASK, qp->attrs.cc) |
+		  FIELD_PREP(ERDMA_CMD_MODIFY_QP_QPN_MASK, QP_ID(qp));
+
+	req.cookie = be32_to_cpu(qp->cep->mpa.ext_data.cookie);
+	req.dip = to_sockaddr_in(remote_addr).sin_addr.s_addr;
+	req.sip = to_sockaddr_in(local_addr).sin_addr.s_addr;
+	req.dport = to_sockaddr_in(remote_addr).sin_port;
+	req.sport = to_sockaddr_in(local_addr).sin_port;
+
+	req.send_nxt = tp->snd_nxt;
+	/* rsvd tcp seq for mpa-rsp in server. */
+	if (qp->attrs.qp_type == ERDMA_QP_PASSIVE)
+		req.send_nxt += MPA_DEFAULT_HDR_LEN + qp->attrs.pd_len;
+	req.recv_nxt = tp->rcv_nxt;
+
+	return erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				   NULL);
+}
+
+static int erdma_modify_qp_state_to_stop(struct erdma_qp *qp,
+					 struct erdma_qp_attrs *attrs,
+					 enum erdma_qp_attr_mask mask)
+{
+	struct erdma_dev *dev = qp->dev;
+	struct erdma_cmdq_modify_qp_req req;
+
+	qp->attrs.state = attrs->state;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_MODIFY_QP);
+
+	req.cfg = FIELD_PREP(ERDMA_CMD_MODIFY_QP_STATE_MASK, attrs->state) |
+		  FIELD_PREP(ERDMA_CMD_MODIFY_QP_QPN_MASK, QP_ID(qp));
+
+	return erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				   NULL);
+}
+
+int erdma_modify_qp_internal(struct erdma_qp *qp, struct erdma_qp_attrs *attrs,
+			     enum erdma_qp_attr_mask mask)
+{
+	int drop_conn, ret = 0;
+
+	if (!mask)
+		return 0;
+
+	if (!(mask & ERDMA_QP_ATTR_STATE))
+		return 0;
+
+	switch (qp->attrs.state) {
+	case ERDMA_QP_STATE_IDLE:
+	case ERDMA_QP_STATE_RTR:
+		if (attrs->state == ERDMA_QP_STATE_RTS) {
+			ret = erdma_modify_qp_state_to_rts(qp, attrs, mask);
+		} else if (attrs->state == ERDMA_QP_STATE_ERROR) {
+			qp->attrs.state = ERDMA_QP_STATE_ERROR;
+			if (qp->cep) {
+				erdma_cep_put(qp->cep);
+				qp->cep = NULL;
+			}
+			ret = erdma_modify_qp_state_to_stop(qp, attrs, mask);
+		}
+		break;
+	case ERDMA_QP_STATE_RTS:
+		drop_conn = 0;
+
+		if (attrs->state == ERDMA_QP_STATE_CLOSING) {
+			ret = erdma_modify_qp_state_to_stop(qp, attrs, mask);
+			drop_conn = 1;
+		} else if (attrs->state == ERDMA_QP_STATE_TERMINATE) {
+			qp->attrs.state = ERDMA_QP_STATE_TERMINATE;
+			ret = erdma_modify_qp_state_to_stop(qp, attrs, mask);
+			drop_conn = 1;
+		} else if (attrs->state == ERDMA_QP_STATE_ERROR) {
+			ret = erdma_modify_qp_state_to_stop(qp, attrs, mask);
+			qp->attrs.state = ERDMA_QP_STATE_ERROR;
+			drop_conn = 1;
+		}
+
+		if (drop_conn)
+			erdma_qp_cm_drop(qp);
+
+		break;
+	case ERDMA_QP_STATE_TERMINATE:
+		if (attrs->state == ERDMA_QP_STATE_ERROR)
+			qp->attrs.state = ERDMA_QP_STATE_ERROR;
+		break;
+	case ERDMA_QP_STATE_CLOSING:
+		if (attrs->state == ERDMA_QP_STATE_IDLE) {
+			qp->attrs.state = ERDMA_QP_STATE_IDLE;
+		} else if (attrs->state == ERDMA_QP_STATE_ERROR) {
+			ret = erdma_modify_qp_state_to_stop(qp, attrs, mask);
+			qp->attrs.state = ERDMA_QP_STATE_ERROR;
+		} else if (attrs->state != ERDMA_QP_STATE_CLOSING) {
+			return -ECONNABORTED;
+		}
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static void erdma_qp_safe_free(struct kref *ref)
+{
+	struct erdma_qp *qp = container_of(ref, struct erdma_qp, ref);
+
+	complete(&qp->safe_free);
+}
+
+void erdma_qp_put(struct erdma_qp *qp)
+{
+	WARN_ON(kref_read(&qp->ref) < 1);
+	kref_put(&qp->ref, erdma_qp_safe_free);
+}
+
+void erdma_qp_get(struct erdma_qp *qp)
+{
+	kref_get(&qp->ref);
+}
+
+static int fill_inline_data(struct erdma_qp *qp,
+			    const struct ib_send_wr *send_wr, u16 wqe_idx,
+			    u32 sgl_offset, __le32 *length_field)
+{
+	u32 remain_size, copy_size, data_off, bytes = 0;
+	char *data;
+	int i = 0;
+
+	wqe_idx += (sgl_offset >> SQEBB_SHIFT);
+	sgl_offset &= (SQEBB_SIZE - 1);
+	data = get_queue_entry(qp->kern_qp.sq_buf, wqe_idx, qp->attrs.sq_size,
+			       SQEBB_SHIFT);
+
+	while (i < send_wr->num_sge) {
+		bytes += send_wr->sg_list[i].length;
+		if (bytes > (int)ERDMA_MAX_INLINE)
+			return -EINVAL;
+
+		remain_size = send_wr->sg_list[i].length;
+		data_off = 0;
+
+		while (1) {
+			copy_size = min(remain_size, SQEBB_SIZE - sgl_offset);
+
+			memcpy(data + sgl_offset,
+			       (void *)(uintptr_t)send_wr->sg_list[i].addr +
+				       data_off,
+			       copy_size);
+			remain_size -= copy_size;
+			data_off += copy_size;
+			sgl_offset += copy_size;
+			wqe_idx += (sgl_offset >> SQEBB_SHIFT);
+			sgl_offset &= (SQEBB_SIZE - 1);
+
+			data = get_queue_entry(qp->kern_qp.sq_buf, wqe_idx,
+					       qp->attrs.sq_size, SQEBB_SHIFT);
+			if (!remain_size)
+				break;
+		}
+
+		i++;
+	}
+	*length_field = cpu_to_le32(bytes);
+
+	return bytes;
+}
+
+static int fill_sgl(struct erdma_qp *qp, const struct ib_send_wr *send_wr,
+		    u16 wqe_idx, u32 sgl_offset, __le32 *length_field)
+{
+	int i = 0;
+	u32 bytes = 0;
+	char *sgl;
+
+	if (send_wr->num_sge > qp->dev->attrs.max_send_sge)
+		return -EINVAL;
+
+	if (sgl_offset & 0xF)
+		return -EINVAL;
+
+	while (i < send_wr->num_sge) {
+		wqe_idx += (sgl_offset >> SQEBB_SHIFT);
+		sgl_offset &= (SQEBB_SIZE - 1);
+		sgl = get_queue_entry(qp->kern_qp.sq_buf, wqe_idx,
+				      qp->attrs.sq_size, SQEBB_SHIFT);
+
+		bytes += send_wr->sg_list[i].length;
+		memcpy(sgl + sgl_offset, &send_wr->sg_list[i],
+		       sizeof(struct ib_sge));
+
+		sgl_offset += sizeof(struct ib_sge);
+		i++;
+	}
+
+	*length_field = cpu_to_le32(bytes);
+	return 0;
+}
+
+static int erdma_push_one_sqe(struct erdma_qp *qp, u16 *pi,
+			      const struct ib_send_wr *send_wr)
+{
+	u32 wqe_size, wqebb_cnt, hw_op, flags, sgl_offset;
+	u32 idx = *pi & (qp->attrs.sq_size - 1);
+	enum ib_wr_opcode op = send_wr->opcode;
+	struct erdma_readreq_sqe *read_sqe;
+	struct erdma_reg_mr_sqe *regmr_sge;
+	struct erdma_write_sqe *write_sqe;
+	struct erdma_send_sqe *send_sqe;
+	struct ib_rdma_wr *rdma_wr;
+	struct erdma_mr *mr;
+	__le32 *length_field;
+	u64 wqe_hdr, *entry;
+	struct ib_sge *sge;
+	u32 attrs;
+	int ret;
+
+	entry = get_queue_entry(qp->kern_qp.sq_buf, idx, qp->attrs.sq_size,
+				SQEBB_SHIFT);
+
+	/* Clear the SQE header section. */
+	*entry = 0;
+
+	qp->kern_qp.swr_tbl[idx] = send_wr->wr_id;
+	flags = send_wr->send_flags;
+	wqe_hdr = FIELD_PREP(
+		ERDMA_SQE_HDR_CE_MASK,
+		((flags & IB_SEND_SIGNALED) || qp->kern_qp.sig_all) ? 1 : 0);
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_SE_MASK,
+			      flags & IB_SEND_SOLICITED ? 1 : 0);
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_FENCE_MASK,
+			      flags & IB_SEND_FENCE ? 1 : 0);
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_INLINE_MASK,
+			      flags & IB_SEND_INLINE ? 1 : 0);
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_QPN_MASK, QP_ID(qp));
+
+	switch (op) {
+	case IB_WR_RDMA_WRITE:
+	case IB_WR_RDMA_WRITE_WITH_IMM:
+		hw_op = ERDMA_OP_WRITE;
+		if (op == IB_WR_RDMA_WRITE_WITH_IMM)
+			hw_op = ERDMA_OP_WRITE_WITH_IMM;
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_OPCODE_MASK, hw_op);
+		rdma_wr = container_of(send_wr, struct ib_rdma_wr, wr);
+		write_sqe = (struct erdma_write_sqe *)entry;
+
+		write_sqe->imm_data = send_wr->ex.imm_data;
+		write_sqe->sink_stag = cpu_to_le32(rdma_wr->rkey);
+		write_sqe->sink_to_h =
+			cpu_to_le32(upper_32_bits(rdma_wr->remote_addr));
+		write_sqe->sink_to_l =
+			cpu_to_le32(lower_32_bits(rdma_wr->remote_addr));
+
+		length_field = &write_sqe->length;
+		wqe_size = sizeof(struct erdma_write_sqe);
+		sgl_offset = wqe_size;
+		break;
+	case IB_WR_RDMA_READ:
+	case IB_WR_RDMA_READ_WITH_INV:
+		read_sqe = (struct erdma_readreq_sqe *)entry;
+		if (unlikely(send_wr->num_sge != 1))
+			return -EINVAL;
+		hw_op = ERDMA_OP_READ;
+		if (op == IB_WR_RDMA_READ_WITH_INV) {
+			hw_op = ERDMA_OP_READ_WITH_INV;
+			read_sqe->invalid_stag =
+				cpu_to_le32(send_wr->ex.invalidate_rkey);
+		}
+
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_OPCODE_MASK, hw_op);
+		rdma_wr = container_of(send_wr, struct ib_rdma_wr, wr);
+		read_sqe->length = cpu_to_le32(send_wr->sg_list[0].length);
+		read_sqe->sink_stag = cpu_to_le32(send_wr->sg_list[0].lkey);
+		read_sqe->sink_to_l =
+			cpu_to_le32(lower_32_bits(send_wr->sg_list[0].addr));
+		read_sqe->sink_to_h =
+			cpu_to_le32(upper_32_bits(send_wr->sg_list[0].addr));
+
+		sge = get_queue_entry(qp->kern_qp.sq_buf, idx + 1,
+				      qp->attrs.sq_size, SQEBB_SHIFT);
+		sge->addr = rdma_wr->remote_addr;
+		sge->lkey = rdma_wr->rkey;
+		sge->length = send_wr->sg_list[0].length;
+		wqe_size = sizeof(struct erdma_readreq_sqe) +
+			   send_wr->num_sge * sizeof(struct ib_sge);
+
+		goto out;
+	case IB_WR_SEND:
+	case IB_WR_SEND_WITH_IMM:
+	case IB_WR_SEND_WITH_INV:
+		send_sqe = (struct erdma_send_sqe *)entry;
+		hw_op = ERDMA_OP_SEND;
+		if (op == IB_WR_SEND_WITH_IMM) {
+			hw_op = ERDMA_OP_SEND_WITH_IMM;
+			send_sqe->imm_data = send_wr->ex.imm_data;
+		} else if (op == IB_WR_SEND_WITH_INV) {
+			hw_op = ERDMA_OP_SEND_WITH_INV;
+			send_sqe->invalid_stag =
+				cpu_to_le32(send_wr->ex.invalidate_rkey);
+		}
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_OPCODE_MASK, hw_op);
+		length_field = &send_sqe->length;
+		wqe_size = sizeof(struct erdma_send_sqe);
+		sgl_offset = wqe_size;
+
+		break;
+	case IB_WR_REG_MR:
+		wqe_hdr |=
+			FIELD_PREP(ERDMA_SQE_HDR_OPCODE_MASK, ERDMA_OP_REG_MR);
+		regmr_sge = (struct erdma_reg_mr_sqe *)entry;
+		mr = to_emr(reg_wr(send_wr)->mr);
+
+		mr->access = ERDMA_MR_ACC_LR |
+			     to_erdma_access_flags(reg_wr(send_wr)->access);
+		regmr_sge->addr = cpu_to_le64(mr->ibmr.iova);
+		regmr_sge->length = cpu_to_le32(mr->ibmr.length);
+		regmr_sge->stag = cpu_to_le32(mr->ibmr.lkey);
+		attrs = FIELD_PREP(ERDMA_SQE_MR_MODE_MASK, 0) |
+			FIELD_PREP(ERDMA_SQE_MR_ACCESS_MASK, mr->access) |
+			FIELD_PREP(ERDMA_SQE_MR_MTT_CNT_MASK,
+				   mr->mem.mtt_nents);
+
+		if (mr->mem.mtt_nents < ERDMA_MAX_INLINE_MTT_ENTRIES) {
+			attrs |= FIELD_PREP(ERDMA_SQE_MR_MTT_TYPE_MASK, 0);
+			/* Copy SGLs to SQE content to accelerate */
+			memcpy(get_queue_entry(qp->kern_qp.sq_buf, idx + 1,
+					       qp->attrs.sq_size, SQEBB_SHIFT),
+			       mr->mem.mtt_buf, MTT_SIZE(mr->mem.mtt_nents));
+			wqe_size = sizeof(struct erdma_reg_mr_sqe) +
+				   MTT_SIZE(mr->mem.mtt_nents);
+		} else {
+			attrs |= FIELD_PREP(ERDMA_SQE_MR_MTT_TYPE_MASK, 1);
+			wqe_size = sizeof(struct erdma_reg_mr_sqe);
+		}
+
+		regmr_sge->attrs = cpu_to_le32(attrs);
+		goto out;
+	case IB_WR_LOCAL_INV:
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_OPCODE_MASK,
+				      ERDMA_OP_LOCAL_INV);
+		regmr_sge = (struct erdma_reg_mr_sqe *)entry;
+		regmr_sge->stag = cpu_to_le32(send_wr->ex.invalidate_rkey);
+		wqe_size = sizeof(struct erdma_reg_mr_sqe);
+		goto out;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	if (flags & IB_SEND_INLINE) {
+		ret = fill_inline_data(qp, send_wr, idx, sgl_offset,
+				       length_field);
+		if (ret < 0)
+			return -EINVAL;
+		wqe_size += ret;
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_SGL_LEN_MASK, ret);
+	} else {
+		ret = fill_sgl(qp, send_wr, idx, sgl_offset, length_field);
+		if (ret)
+			return -EINVAL;
+		wqe_size += send_wr->num_sge * sizeof(struct ib_sge);
+		wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_SGL_LEN_MASK,
+				      send_wr->num_sge);
+	}
+
+out:
+	wqebb_cnt = SQEBB_COUNT(wqe_size);
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_WQEBB_CNT_MASK, wqebb_cnt - 1);
+	*pi += wqebb_cnt;
+	wqe_hdr |= FIELD_PREP(ERDMA_SQE_HDR_WQEBB_INDEX_MASK, *pi);
+
+	*entry = wqe_hdr;
+
+	return 0;
+}
+
+static void kick_sq_db(struct erdma_qp *qp, u16 pi)
+{
+	u64 db_data = FIELD_PREP(ERDMA_SQE_HDR_QPN_MASK, QP_ID(qp)) |
+		      FIELD_PREP(ERDMA_SQE_HDR_WQEBB_INDEX_MASK, pi);
+
+	*(u64 *)qp->kern_qp.sq_db_info = db_data;
+	writeq(db_data, qp->kern_qp.hw_sq_db);
+}
+
+int erdma_post_send(struct ib_qp *ibqp, const struct ib_send_wr *send_wr,
+		    const struct ib_send_wr **bad_send_wr)
+{
+	struct erdma_qp *qp = to_eqp(ibqp);
+	int ret = 0;
+	const struct ib_send_wr *wr = send_wr;
+	unsigned long flags;
+	u16 sq_pi;
+
+	if (!send_wr)
+		return -EINVAL;
+
+	spin_lock_irqsave(&qp->lock, flags);
+	sq_pi = qp->kern_qp.sq_pi;
+
+	while (wr) {
+		if ((u16)(sq_pi - qp->kern_qp.sq_ci) >= qp->attrs.sq_size) {
+			ret = -ENOMEM;
+			*bad_send_wr = send_wr;
+			break;
+		}
+
+		ret = erdma_push_one_sqe(qp, &sq_pi, wr);
+		if (ret) {
+			*bad_send_wr = wr;
+			break;
+		}
+		qp->kern_qp.sq_pi = sq_pi;
+		kick_sq_db(qp, sq_pi);
+
+		wr = wr->next;
+	}
+	spin_unlock_irqrestore(&qp->lock, flags);
+
+	return ret;
+}
+
+static int erdma_post_recv_one(struct erdma_qp *qp,
+			       const struct ib_recv_wr *recv_wr)
+{
+	struct erdma_rqe *rqe =
+		get_queue_entry(qp->kern_qp.rq_buf, qp->kern_qp.rq_pi,
+				qp->attrs.rq_size, RQE_SHIFT);
+
+	rqe->qe_idx = cpu_to_le16(qp->kern_qp.rq_pi + 1);
+	rqe->qpn = cpu_to_le32(QP_ID(qp));
+
+	if (recv_wr->num_sge == 0) {
+		rqe->length = 0;
+	} else if (recv_wr->num_sge == 1) {
+		rqe->stag = cpu_to_le32(recv_wr->sg_list[0].lkey);
+		rqe->to = cpu_to_le64(recv_wr->sg_list[0].addr);
+		rqe->length = cpu_to_le32(recv_wr->sg_list[0].length);
+	} else {
+		return -EINVAL;
+	}
+
+	*(u64 *)qp->kern_qp.rq_db_info = *(u64 *)rqe;
+	writeq(*(u64 *)rqe, qp->kern_qp.hw_rq_db);
+
+	qp->kern_qp.rwr_tbl[qp->kern_qp.rq_pi & (qp->attrs.rq_size - 1)] =
+		recv_wr->wr_id;
+	qp->kern_qp.rq_pi++;
+
+	return 0;
+}
+
+int erdma_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *recv_wr,
+		    const struct ib_recv_wr **bad_recv_wr)
+{
+	const struct ib_recv_wr *wr = recv_wr;
+	struct erdma_qp *qp = to_eqp(ibqp);
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&qp->lock, flags);
+
+	while (wr) {
+		ret = erdma_post_recv_one(qp, wr);
+		if (ret) {
+			*bad_recv_wr = wr;
+			break;
+		}
+		wr = wr->next;
+	}
+
+	spin_unlock_irqrestore(&qp->lock, flags);
+	return ret;
+}

drivers/infiniband/hw/erdma/erdma_verbs.c

···
+// SPDX-License-Identifier: GPL-2.0
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+/* Authors: Bernard Metzler <[e-mail]> */
+/* Copyright (c) 2008-2019, IBM Corporation */
+
+/* Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved. */
+
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+#include <linux/vmalloc.h>
+#include <net/addrconf.h>
+#include <rdma/erdma-abi.h>
+#include <rdma/ib_umem.h>
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
+#include <rdma/uverbs_ioctl.h>
+
+#include "erdma.h"
+#include "erdma_cm.h"
+#include "erdma_hw.h"
+#include "erdma_verbs.h"
+
+static int create_qp_cmd(struct erdma_dev *dev, struct erdma_qp *qp)
+{
+	struct erdma_cmdq_create_qp_req req;
+	struct erdma_pd *pd = to_epd(qp->ibqp.pd);
+	struct erdma_uqp *user_qp;
+	u64 resp0, resp1;
+	int err;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_CREATE_QP);
+
+	req.cfg0 = FIELD_PREP(ERDMA_CMD_CREATE_QP_SQ_DEPTH_MASK,
+			      ilog2(qp->attrs.sq_size)) |
+		   FIELD_PREP(ERDMA_CMD_CREATE_QP_QPN_MASK, QP_ID(qp));
+	req.cfg1 = FIELD_PREP(ERDMA_CMD_CREATE_QP_RQ_DEPTH_MASK,
+			      ilog2(qp->attrs.rq_size)) |
+		   FIELD_PREP(ERDMA_CMD_CREATE_QP_PD_MASK, pd->pdn);
+
+	if (rdma_is_kernel_res(&qp->ibqp.res)) {
+		u32 pgsz_range = ilog2(SZ_1M) - PAGE_SHIFT;
+
+		req.sq_cqn_mtt_cfg =
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_PAGE_SIZE_MASK,
+				   pgsz_range) |
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_CQN_MASK, qp->scq->cqn);
+		req.rq_cqn_mtt_cfg =
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_PAGE_SIZE_MASK,
+				   pgsz_range) |
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_CQN_MASK, qp->rcq->cqn);
+
+		req.sq_mtt_cfg =
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_PAGE_OFFSET_MASK, 0) |
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_CNT_MASK, 1) |
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_TYPE_MASK,
+				   ERDMA_MR_INLINE_MTT);
+		req.rq_mtt_cfg = req.sq_mtt_cfg;
+
+		req.rq_buf_addr = qp->kern_qp.rq_buf_dma_addr;
+		req.sq_buf_addr = qp->kern_qp.sq_buf_dma_addr;
+		req.sq_db_info_dma_addr = qp->kern_qp.sq_buf_dma_addr +
+					  (qp->attrs.sq_size << SQEBB_SHIFT);
+		req.rq_db_info_dma_addr = qp->kern_qp.rq_buf_dma_addr +
+					  (qp->attrs.rq_size << RQE_SHIFT);
+	} else {
+		user_qp = &qp->user_qp;
+		req.sq_cqn_mtt_cfg = FIELD_PREP(
+			ERDMA_CMD_CREATE_QP_PAGE_SIZE_MASK,
+			ilog2(user_qp->sq_mtt.page_size) - PAGE_SHIFT);
+		req.sq_cqn_mtt_cfg |=
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_CQN_MASK, qp->scq->cqn);
+
+		req.rq_cqn_mtt_cfg = FIELD_PREP(
+			ERDMA_CMD_CREATE_QP_PAGE_SIZE_MASK,
+			ilog2(user_qp->rq_mtt.page_size) - PAGE_SHIFT);
+		req.rq_cqn_mtt_cfg |=
+			FIELD_PREP(ERDMA_CMD_CREATE_QP_CQN_MASK, qp->rcq->cqn);
+
+		req.sq_mtt_cfg = user_qp->sq_mtt.page_offset;
+		req.sq_mtt_cfg |= FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_CNT_MASK,
+					     user_qp->sq_mtt.mtt_nents) |
+				  FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_TYPE_MASK,
+					     user_qp->sq_mtt.mtt_type);
+
+		req.rq_mtt_cfg = user_qp->rq_mtt.page_offset;
+		req.rq_mtt_cfg |= FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_CNT_MASK,
+					     user_qp->rq_mtt.mtt_nents) |
+				  FIELD_PREP(ERDMA_CMD_CREATE_QP_MTT_TYPE_MASK,
+					     user_qp->rq_mtt.mtt_type);
+
+		req.sq_buf_addr = user_qp->sq_mtt.mtt_entry[0];
+		req.rq_buf_addr = user_qp->rq_mtt.mtt_entry[0];
+
+		req.sq_db_info_dma_addr = user_qp->sq_db_info_dma_addr;
+		req.rq_db_info_dma_addr = user_qp->rq_db_info_dma_addr;
+	}
+
+	err = erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), &resp0,
+				  &resp1);
+	if (!err)
+		qp->attrs.cookie =
+			FIELD_GET(ERDMA_CMDQ_CREATE_QP_RESP_COOKIE_MASK, resp0);
+
+	return err;
+}
+
+static int regmr_cmd(struct erdma_dev *dev, struct erdma_mr *mr)
+{
+	struct erdma_cmdq_reg_mr_req req;
+	struct erdma_pd *pd = to_epd(mr->ibmr.pd);
+	u64 *phy_addr;
+	int i;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA, CMDQ_OPCODE_REG_MR);
+
+	req.cfg0 = FIELD_PREP(ERDMA_CMD_MR_VALID_MASK, mr->valid) |
+		   FIELD_PREP(ERDMA_CMD_MR_KEY_MASK, mr->ibmr.lkey & 0xFF) |
+		   FIELD_PREP(ERDMA_CMD_MR_MPT_IDX_MASK, mr->ibmr.lkey >> 8);
+	req.cfg1 = FIELD_PREP(ERDMA_CMD_REGMR_PD_MASK, pd->pdn) |
+		   FIELD_PREP(ERDMA_CMD_REGMR_TYPE_MASK, mr->type) |
+		   FIELD_PREP(ERDMA_CMD_REGMR_RIGHT_MASK, mr->access) |
+		   FIELD_PREP(ERDMA_CMD_REGMR_ACC_MODE_MASK, 0);
+	req.cfg2 = FIELD_PREP(ERDMA_CMD_REGMR_PAGESIZE_MASK,
+			      ilog2(mr->mem.page_size)) |
+		   FIELD_PREP(ERDMA_CMD_REGMR_MTT_TYPE_MASK, mr->mem.mtt_type) |
+		   FIELD_PREP(ERDMA_CMD_REGMR_MTT_CNT_MASK, mr->mem.page_cnt);
+
+	if (mr->type == ERDMA_MR_TYPE_DMA)
+		goto post_cmd;
+
+	if (mr->type == ERDMA_MR_TYPE_NORMAL) {
+		req.start_va = mr->mem.va;
+		req.size = mr->mem.len;
+	}
+
+	if (mr->type == ERDMA_MR_TYPE_FRMR ||
+	    mr->mem.mtt_type == ERDMA_MR_INDIRECT_MTT) {
+		phy_addr = req.phy_addr;
+		*phy_addr = mr->mem.mtt_entry[0];
+	} else {
+		phy_addr = req.phy_addr;
+		for (i = 0; i < mr->mem.mtt_nents; i++)
+			*phy_addr++ = mr->mem.mtt_entry[i];
+	}
+
+post_cmd:
+	return erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				   NULL);
+}
+
+static int create_cq_cmd(struct erdma_dev *dev, struct erdma_cq *cq)
+{
+	struct erdma_cmdq_create_cq_req req;
+	u32 page_size;
+	struct erdma_mem *mtt;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_CREATE_CQ);
+
+	req.cfg0 = FIELD_PREP(ERDMA_CMD_CREATE_CQ_CQN_MASK, cq->cqn) |
+		   FIELD_PREP(ERDMA_CMD_CREATE_CQ_DEPTH_MASK, ilog2(cq->depth));
+	req.cfg1 = FIELD_PREP(ERDMA_CMD_CREATE_CQ_EQN_MASK, cq->assoc_eqn);
+
+	if (rdma_is_kernel_res(&cq->ibcq.res)) {
+		page_size = SZ_32M;
+		req.cfg0 |= FIELD_PREP(ERDMA_CMD_CREATE_CQ_PAGESIZE_MASK,
+				       ilog2(page_size) - PAGE_SHIFT);
+		req.qbuf_addr_l = lower_32_bits(cq->kern_cq.qbuf_dma_addr);
+		req.qbuf_addr_h = upper_32_bits(cq->kern_cq.qbuf_dma_addr);
+
+		req.cfg1 |= FIELD_PREP(ERDMA_CMD_CREATE_CQ_MTT_CNT_MASK, 1) |
+			    FIELD_PREP(ERDMA_CMD_CREATE_CQ_MTT_TYPE_MASK,
+				       ERDMA_MR_INLINE_MTT);
+
+		req.first_page_offset = 0;
+		req.cq_db_info_addr =
+			cq->kern_cq.qbuf_dma_addr + (cq->depth << CQE_SHIFT);
+	} else {
+		mtt = &cq->user_cq.qbuf_mtt;
+		req.cfg0 |= FIELD_PREP(ERDMA_CMD_CREATE_CQ_PAGESIZE_MASK,
+				       ilog2(mtt->page_size) - PAGE_SHIFT);
+		if (mtt->mtt_nents == 1) {
+			req.qbuf_addr_l = lower_32_bits(*(u64 *)mtt->mtt_buf);
+			req.qbuf_addr_h = upper_32_bits(*(u64 *)mtt->mtt_buf);
+		} else {
+			req.qbuf_addr_l = lower_32_bits(mtt->mtt_entry[0]);
+			req.qbuf_addr_h = upper_32_bits(mtt->mtt_entry[0]);
+		}
+		req.cfg1 |= FIELD_PREP(ERDMA_CMD_CREATE_CQ_MTT_CNT_MASK,
+				       mtt->mtt_nents);
+		req.cfg1 |= FIELD_PREP(ERDMA_CMD_CREATE_CQ_MTT_TYPE_MASK,
+				       mtt->mtt_type);
+
+		req.first_page_offset = mtt->page_offset;
+		req.cq_db_info_addr = cq->user_cq.db_info_dma_addr;
+	}
+
+	return erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				   NULL);
+}
+
+static int erdma_alloc_idx(struct erdma_resource_cb *res_cb)
+{
+	int idx;
+	unsigned long flags;
+
+	spin_lock_irqsave(&res_cb->lock, flags);
+	idx = find_next_zero_bit(res_cb->bitmap, res_cb->max_cap,
+				 res_cb->next_alloc_idx);
+	if (idx == res_cb->max_cap) {
+		idx = find_first_zero_bit(res_cb->bitmap, res_cb->max_cap);
+		if (idx == res_cb->max_cap) {
+			res_cb->next_alloc_idx = 1;
+			spin_unlock_irqrestore(&res_cb->lock, flags);
+			return -ENOSPC;
+		}
+	}
+
+	set_bit(idx, res_cb->bitmap);
+	res_cb->next_alloc_idx = idx + 1;
+	spin_unlock_irqrestore(&res_cb->lock, flags);
+
+	return idx;
+}
+
+static inline void erdma_free_idx(struct erdma_resource_cb *res_cb, u32 idx)
+{
+	unsigned long flags;
+	u32 used;
+
+	spin_lock_irqsave(&res_cb->lock, flags);
+	used = __test_and_clear_bit(idx, res_cb->bitmap);
+	spin_unlock_irqrestore(&res_cb->lock, flags);
+	WARN_ON(!used);
+}
+
+static struct rdma_user_mmap_entry *
+erdma_user_mmap_entry_insert(struct erdma_ucontext *uctx, void *address,
+			     u32 size, u8 mmap_flag, u64 *mmap_offset)
+{
+	struct erdma_user_mmap_entry *entry =
+		kzalloc(sizeof(*entry), GFP_KERNEL);
+	int ret;
+
+	if (!entry)
+		return NULL;
+
+	entry->address = (u64)address;
+	entry->mmap_flag = mmap_flag;
+
+	size = PAGE_ALIGN(size);
+
+	ret = rdma_user_mmap_entry_insert(&uctx->ibucontext, &entry->rdma_entry,
+					  size);
+	if (ret) {
+		kfree(entry);
+		return NULL;
+	}
+
+	*mmap_offset = rdma_user_mmap_get_offset(&entry->rdma_entry);
+
+	return &entry->rdma_entry;
+}
+
+int erdma_query_device(struct ib_device *ibdev, struct ib_device_attr *attr,
+		       struct ib_udata *unused)
+{
+	struct erdma_dev *dev = to_edev(ibdev);
+
+	memset(attr, 0, sizeof(*attr));
+
+	attr->max_mr_size = dev->attrs.max_mr_size;
+	attr->vendor_id = PCI_VENDOR_ID_ALIBABA;
+	attr->vendor_part_id = dev->pdev->device;
+	attr->hw_ver = dev->pdev->revision;
+	attr->max_qp = dev->attrs.max_qp;
+	attr->max_qp_wr = min(dev->attrs.max_send_wr, dev->attrs.max_recv_wr);
+	attr->max_qp_rd_atom = dev->attrs.max_ord;
+	attr->max_qp_init_rd_atom = dev->attrs.max_ird;
+	attr->max_res_rd_atom = dev->attrs.max_qp * dev->attrs.max_ird;
+	attr->device_cap_flags = IB_DEVICE_MEM_MGT_EXTENSIONS;
+	attr->kernel_cap_flags = IBK_LOCAL_DMA_LKEY;
+	ibdev->local_dma_lkey = dev->attrs.local_dma_key;
+	attr->max_send_sge = dev->attrs.max_send_sge;
+	attr->max_recv_sge = dev->attrs.max_recv_sge;
+	attr->max_sge_rd = dev->attrs.max_sge_rd;
+	attr->max_cq = dev->attrs.max_cq;
+	attr->max_cqe = dev->attrs.max_cqe;
+	attr->max_mr = dev->attrs.max_mr;
+	attr->max_pd = dev->attrs.max_pd;
+	attr->max_mw = dev->attrs.max_mw;
+	attr->max_fast_reg_page_list_len = ERDMA_MAX_FRMR_PA;
+	attr->page_size_cap = ERDMA_PAGE_SIZE_SUPPORT;
+	attr->fw_ver = dev->attrs.fw_version;
+
+	if (dev->netdev)
+		addrconf_addr_eui48((u8 *)&attr->sys_image_guid,
+				    dev->netdev->dev_addr);
+
+	return 0;
+}
+
+int erdma_query_gid(struct ib_device *ibdev, u32 port, int idx,
+		    union ib_gid *gid)
+{
+	struct erdma_dev *dev = to_edev(ibdev);
+
+	memset(gid, 0, sizeof(*gid));
+	ether_addr_copy(gid->raw, dev->attrs.peer_addr);
+
+	return 0;
+}
+
+int erdma_query_port(struct ib_device *ibdev, u32 port,
+		     struct ib_port_attr *attr)
+{
+	struct erdma_dev *dev = to_edev(ibdev);
+	struct net_device *ndev = dev->netdev;
+
+	memset(attr, 0, sizeof(*attr));
+
+	attr->gid_tbl_len = 1;
+	attr->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_DEVICE_MGMT_SUP;
+	attr->max_msg_sz = -1;
+
+	if (!ndev)
+		goto out;
+
+	ib_get_eth_speed(ibdev, port, &attr->active_speed, &attr->active_width);
+	attr->max_mtu = ib_mtu_int_to_enum(ndev->mtu);
+	attr->active_mtu = ib_mtu_int_to_enum(ndev->mtu);
+	if (netif_running(ndev) && netif_carrier_ok(ndev))
+		dev->state = IB_PORT_ACTIVE;
+	else
+		dev->state = IB_PORT_DOWN;
+	attr->state = dev->state;
+
+out:
+	if (dev->state == IB_PORT_ACTIVE)
+		attr->phys_state = IB_PORT_PHYS_STATE_LINK_UP;
+	else
+		attr->phys_state = IB_PORT_PHYS_STATE_DISABLED;
+
+	return 0;
+}
+
+int erdma_get_port_immutable(struct ib_device *ibdev, u32 port,
+			     struct ib_port_immutable *port_immutable)
+{
+	port_immutable->gid_tbl_len = 1;
+	port_immutable->core_cap_flags = RDMA_CORE_PORT_IWARP;
+
+	return 0;
+}
+
+int erdma_alloc_pd(struct ib_pd *ibpd, struct ib_udata *udata)
+{
+	struct erdma_pd *pd = to_epd(ibpd);
+	struct erdma_dev *dev = to_edev(ibpd->device);
+	int pdn;
+
+	pdn = erdma_alloc_idx(&dev->res_cb[ERDMA_RES_TYPE_PD]);
+	if (pdn < 0)
+		return pdn;
+
+	pd->pdn = pdn;
+
+	return 0;
+}
+
+int erdma_dealloc_pd(struct ib_pd *ibpd, struct ib_udata *udata)
+{
+	struct erdma_pd *pd = to_epd(ibpd);
+	struct erdma_dev *dev = to_edev(ibpd->device);
+
+	erdma_free_idx(&dev->res_cb[ERDMA_RES_TYPE_PD], pd->pdn);
+
+	return 0;
+}
+
+static int erdma_qp_validate_cap(struct erdma_dev *dev,
+				 struct ib_qp_init_attr *attrs)
+{
+	if ((attrs->cap.max_send_wr > dev->attrs.max_send_wr) ||
+	    (attrs->cap.max_recv_wr > dev->attrs.max_recv_wr) ||
+	    (attrs->cap.max_send_sge > dev->attrs.max_send_sge) ||
+	    (attrs->cap.max_recv_sge > dev->attrs.max_recv_sge) ||
+	    (attrs->cap.max_inline_data > ERDMA_MAX_INLINE) ||
+	    !attrs->cap.max_send_wr || !attrs->cap.max_recv_wr) {
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int erdma_qp_validate_attr(struct erdma_dev *dev,
+				  struct ib_qp_init_attr *attrs)
+{
+	if (attrs->qp_type != IB_QPT_RC)
+		return -EOPNOTSUPP;
+
+	if (attrs->srq)
+		return -EOPNOTSUPP;
+
+	if (!attrs->send_cq || !attrs->recv_cq)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static void free_kernel_qp(struct erdma_qp *qp)
+{
+	struct erdma_dev *dev = qp->dev;
+
+	vfree(qp->kern_qp.swr_tbl);
+	vfree(qp->kern_qp.rwr_tbl);
+
+	if (qp->kern_qp.sq_buf)
+		dma_free_coherent(
+			&dev->pdev->dev,
+			WARPPED_BUFSIZE(qp->attrs.sq_size << SQEBB_SHIFT),
+			qp->kern_qp.sq_buf, qp->kern_qp.sq_buf_dma_addr);
+
+	if (qp->kern_qp.rq_buf)
+		dma_free_coherent(
+			&dev->pdev->dev,
+			WARPPED_BUFSIZE(qp->attrs.rq_size << RQE_SHIFT),
+			qp->kern_qp.rq_buf, qp->kern_qp.rq_buf_dma_addr);
+}
+
+static int init_kernel_qp(struct erdma_dev *dev, struct erdma_qp *qp,
+			  struct ib_qp_init_attr *attrs)
+{
+	struct erdma_kqp *kqp = &qp->kern_qp;
+	int size;
+
+	if (attrs->sq_sig_type == IB_SIGNAL_ALL_WR)
+		kqp->sig_all = 1;
+
+	kqp->sq_pi = 0;
+	kqp->sq_ci = 0;
+	kqp->rq_pi = 0;
+	kqp->rq_ci = 0;
+	kqp->hw_sq_db =
+		dev->func_bar + (ERDMA_SDB_SHARED_PAGE_INDEX << PAGE_SHIFT);
+	kqp->hw_rq_db = dev->func_bar + ERDMA_BAR_RQDB_SPACE_OFFSET;
+
+	kqp->swr_tbl = vmalloc(qp->attrs.sq_size * sizeof(u64));
+	kqp->rwr_tbl = vmalloc(qp->attrs.rq_size * sizeof(u64));
+	if (!kqp->swr_tbl || !kqp->rwr_tbl)
+		goto err_out;
+
+	size = (qp->attrs.sq_size << SQEBB_SHIFT) + ERDMA_EXTRA_BUFFER_SIZE;
+	kqp->sq_buf = dma_alloc_coherent(&dev->pdev->dev, size,
+					 &kqp->sq_buf_dma_addr, GFP_KERNEL);
+	if (!kqp->sq_buf)
+		goto err_out;
+
+	size = (qp->attrs.rq_size << RQE_SHIFT) + ERDMA_EXTRA_BUFFER_SIZE;
+	kqp->rq_buf = dma_alloc_coherent(&dev->pdev->dev, size,
+					 &kqp->rq_buf_dma_addr, GFP_KERNEL);
+	if (!kqp->rq_buf)
+		goto err_out;
+
+	kqp->sq_db_info = kqp->sq_buf + (qp->attrs.sq_size << SQEBB_SHIFT);
+	kqp->rq_db_info = kqp->rq_buf + (qp->attrs.rq_size << RQE_SHIFT);
+
+	return 0;
+
+err_out:
+	free_kernel_qp(qp);
+	return -ENOMEM;
+}
+
+static int get_mtt_entries(struct erdma_dev *dev, struct erdma_mem *mem,
+			   u64 start, u64 len, int access, u64 virt,
+			   unsigned long req_page_size, u8 force_indirect_mtt)
+{
+	struct ib_block_iter biter;
+	uint64_t *phy_addr = NULL;
+	int ret = 0;
+
+	mem->umem = ib_umem_get(&dev->ibdev, start, len, access);
+	if (IS_ERR(mem->umem)) {
+		ret = PTR_ERR(mem->umem);
+		mem->umem = NULL;
+		return ret;
+	}
+
+	mem->va = virt;
+	mem->len = len;
+	mem->page_size = ib_umem_find_best_pgsz(mem->umem, req_page_size, virt);
+	mem->page_offset = start & (mem->page_size - 1);
+	mem->mtt_nents = ib_umem_num_dma_blocks(mem->umem, mem->page_size);
+	mem->page_cnt = mem->mtt_nents;
+
+	if (mem->page_cnt > ERDMA_MAX_INLINE_MTT_ENTRIES ||
+	    force_indirect_mtt) {
+		mem->mtt_type = ERDMA_MR_INDIRECT_MTT;
+		mem->mtt_buf =
+			alloc_pages_exact(MTT_SIZE(mem->page_cnt), GFP_KERNEL);
+		if (!mem->mtt_buf) {
+			ret = -ENOMEM;
+			goto error_ret;
+		}
+		phy_addr = mem->mtt_buf;
+	} else {
+		mem->mtt_type = ERDMA_MR_INLINE_MTT;
+		phy_addr = mem->mtt_entry;
+	}
+
+	rdma_umem_for_each_dma_block(mem->umem, &biter, mem->page_size) {
+		*phy_addr = rdma_block_iter_dma_address(&biter);
+		phy_addr++;
+	}
+
+	if (mem->mtt_type == ERDMA_MR_INDIRECT_MTT) {
+		mem->mtt_entry[0] =
+			dma_map_single(&dev->pdev->dev, mem->mtt_buf,
+				       MTT_SIZE(mem->page_cnt), DMA_TO_DEVICE);
+		if (dma_mapping_error(&dev->pdev->dev, mem->mtt_entry[0])) {
+			free_pages_exact(mem->mtt_buf, MTT_SIZE(mem->page_cnt));
+			mem->mtt_buf = NULL;
+			ret = -ENOMEM;
+			goto error_ret;
+		}
+	}
+
+	return 0;
+
+error_ret:
+	if (mem->umem) {
+		ib_umem_release(mem->umem);
+		mem->umem = NULL;
+	}
+
+	return ret;
+}
+
+static void put_mtt_entries(struct erdma_dev *dev, struct erdma_mem *mem)
+{
+	if (mem->mtt_buf) {
+		dma_unmap_single(&dev->pdev->dev, mem->mtt_entry[0],
+				 MTT_SIZE(mem->page_cnt), DMA_TO_DEVICE);
+		free_pages_exact(mem->mtt_buf, MTT_SIZE(mem->page_cnt));
+	}
+
+	if (mem->umem) {
+		ib_umem_release(mem->umem);
+		mem->umem = NULL;
+	}
+}
+
+static int erdma_map_user_dbrecords(struct erdma_ucontext *ctx,
+				    u64 dbrecords_va,
+				    struct erdma_user_dbrecords_page **dbr_page,
+				    dma_addr_t *dma_addr)
+{
+	struct erdma_user_dbrecords_page *page = NULL;
+	int rv = 0;
+
+	mutex_lock(&ctx->dbrecords_page_mutex);
+
+	list_for_each_entry(page, &ctx->dbrecords_page_list, list)
+		if (page->va == (dbrecords_va & PAGE_MASK))
+			goto found;
+
+	page = kmalloc(sizeof(*page), GFP_KERNEL);
+	if (!page) {
+		rv = -ENOMEM;
+		goto out;
+	}
+
+	page->va = (dbrecords_va & PAGE_MASK);
+	page->refcnt = 0;
+
+	page->umem = ib_umem_get(ctx->ibucontext.device,
+				 dbrecords_va & PAGE_MASK, PAGE_SIZE, 0);
+	if (IS_ERR(page->umem)) {
+		rv = PTR_ERR(page->umem);
+		kfree(page);
+		goto out;
+	}
+
+	list_add(&page->list, &ctx->dbrecords_page_list);
+
+found:
+	*dma_addr = sg_dma_address(page->umem->sgt_append.sgt.sgl) +
+		    (dbrecords_va & ~PAGE_MASK);
+	*dbr_page = page;
+	page->refcnt++;
+
+out:
+	mutex_unlock(&ctx->dbrecords_page_mutex);
+	return rv;
+}
+
+static void
+erdma_unmap_user_dbrecords(struct erdma_ucontext *ctx,
+			   struct erdma_user_dbrecords_page **dbr_page)
+{
+	if (!ctx || !(*dbr_page))
+		return;
+
+	mutex_lock(&ctx->dbrecords_page_mutex);
+	if (--(*dbr_page)->refcnt == 0) {
+		list_del(&(*dbr_page)->list);
+		ib_umem_release((*dbr_page)->umem);
+		kfree(*dbr_page);
+	}
+
+	*dbr_page = NULL;
+	mutex_unlock(&ctx->dbrecords_page_mutex);
+}
+
+static int init_user_qp(struct erdma_qp *qp, struct erdma_ucontext *uctx,
+			u64 va, u32 len, u64 db_info_va)
+{
+	dma_addr_t db_info_dma_addr;
+	u32 rq_offset;
+	int ret;
+
+	if (len < (PAGE_ALIGN(qp->attrs.sq_size * SQEBB_SIZE) +
+		   qp->attrs.rq_size * RQE_SIZE))
+		return -EINVAL;
+
+	ret = get_mtt_entries(qp->dev, &qp->user_qp.sq_mtt, va,
+			      qp->attrs.sq_size << SQEBB_SHIFT, 0, va,
+			      (SZ_1M - SZ_4K), 1);
+	if (ret)
+		return ret;
+
+	rq_offset = PAGE_ALIGN(qp->attrs.sq_size << SQEBB_SHIFT);
+	qp->user_qp.rq_offset = rq_offset;
+
+	ret = get_mtt_entries(qp->dev, &qp->user_qp.rq_mtt, va + rq_offset,
+			      qp->attrs.rq_size << RQE_SHIFT, 0, va + rq_offset,
+			      (SZ_1M - SZ_4K), 1);
+	if (ret)
+		goto put_sq_mtt;
+
+	ret = erdma_map_user_dbrecords(uctx, db_info_va,
+				       &qp->user_qp.user_dbr_page,
+				       &db_info_dma_addr);
+	if (ret)
+		goto put_rq_mtt;
+
+	qp->user_qp.sq_db_info_dma_addr = db_info_dma_addr;
+	qp->user_qp.rq_db_info_dma_addr = db_info_dma_addr + ERDMA_DB_SIZE;
+
+	return 0;
+
+put_rq_mtt:
+	put_mtt_entries(qp->dev, &qp->user_qp.rq_mtt);
+
+put_sq_mtt:
+	put_mtt_entries(qp->dev, &qp->user_qp.sq_mtt);
+
+	return ret;
+}
+
+static void free_user_qp(struct erdma_qp *qp, struct erdma_ucontext *uctx)
+{
+	put_mtt_entries(qp->dev, &qp->user_qp.sq_mtt);
+	put_mtt_entries(qp->dev, &qp->user_qp.rq_mtt);
+	erdma_unmap_user_dbrecords(uctx, &qp->user_qp.user_dbr_page);
+}
+
+int erdma_create_qp(struct ib_qp *ibqp, struct ib_qp_init_attr *attrs,
+		    struct ib_udata *udata)
+{
+	struct erdma_qp *qp = to_eqp(ibqp);
+	struct erdma_dev *dev = to_edev(ibqp->device);
+	struct erdma_ucontext *uctx = rdma_udata_to_drv_context(
+		udata, struct erdma_ucontext, ibucontext);
+	struct erdma_ureq_create_qp ureq;
+	struct erdma_uresp_create_qp uresp;
+	int ret;
+
+	ret = erdma_qp_validate_cap(dev, attrs);
+	if (ret)
+		goto err_out;
+
+	ret = erdma_qp_validate_attr(dev, attrs);
+	if (ret)
+		goto err_out;
+
+	qp->scq = to_ecq(attrs->send_cq);
+	qp->rcq = to_ecq(attrs->recv_cq);
+	qp->dev = dev;
+	qp->attrs.cc = dev->attrs.cc;
+
+	init_rwsem(&qp->state_lock);
+	kref_init(&qp->ref);
+	init_completion(&qp->safe_free);
+
+	ret = xa_alloc_cyclic(&dev->qp_xa, &qp->ibqp.qp_num, qp,
+			      XA_LIMIT(1, dev->attrs.max_qp - 1),
+			      &dev->next_alloc_qpn, GFP_KERNEL);
+	if (ret < 0) {
+		ret = -ENOMEM;
+		goto err_out;
+	}
+
+	qp->attrs.sq_size = roundup_pow_of_two(attrs->cap.max_send_wr *
+					       ERDMA_MAX_WQEBB_PER_SQE);
+	qp->attrs.rq_size = roundup_pow_of_two(attrs->cap.max_recv_wr);
+
+	if (uctx) {
+		ret = ib_copy_from_udata(&ureq, udata,
+					 min(sizeof(ureq), udata->inlen));
+		if (ret)
+			goto err_out_xa;
+
+		ret = init_user_qp(qp, uctx, ureq.qbuf_va, ureq.qbuf_len,
+				   ureq.db_record_va);
+		if (ret)
+			goto err_out_xa;
+
+		memset(&uresp, 0, sizeof(uresp));
+
+		uresp.num_sqe = qp->attrs.sq_size;
+		uresp.num_rqe = qp->attrs.rq_size;
+		uresp.qp_id = QP_ID(qp);
+		uresp.rq_offset = qp->user_qp.rq_offset;
+
+		ret = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+		if (ret)
+			goto err_out_cmd;
+	} else {
+		init_kernel_qp(dev, qp, attrs);
+	}
+
+	qp->attrs.max_send_sge = attrs->cap.max_send_sge;
+	qp->attrs.max_recv_sge = attrs->cap.max_recv_sge;
+	qp->attrs.state = ERDMA_QP_STATE_IDLE;
+
+	ret = create_qp_cmd(dev, qp);
+	if (ret)
+		goto err_out_cmd;
+
+	spin_lock_init(&qp->lock);
+
+	return 0;
+
+err_out_cmd:
+	if (uctx)
+		free_user_qp(qp, uctx);
+	else
+		free_kernel_qp(qp);
+err_out_xa:
+	xa_erase(&dev->qp_xa, QP_ID(qp));
+err_out:
+	return ret;
+}
+
+static int erdma_create_stag(struct erdma_dev *dev, u32 *stag)
+{
+	int stag_idx;
+
+	stag_idx = erdma_alloc_idx(&dev->res_cb[ERDMA_RES_TYPE_STAG_IDX]);
+	if (stag_idx < 0)
+		return stag_idx;
+
+	/* For now, we always let key field be zero. */
+	*stag = (stag_idx << 8);
+
+	return 0;
+}
+
+struct ib_mr *erdma_get_dma_mr(struct ib_pd *ibpd, int acc)
+{
+	struct erdma_dev *dev = to_edev(ibpd->device);
+	struct erdma_mr *mr;
+	u32 stag;
+	int ret;
+
+	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
+	if (!mr)
+		return ERR_PTR(-ENOMEM);
+
+	ret = erdma_create_stag(dev, &stag);
+	if (ret)
+		goto out_free;
+
+	mr->type = ERDMA_MR_TYPE_DMA;
+
+	mr->ibmr.lkey = stag;
+	mr->ibmr.rkey = stag;
+	mr->ibmr.pd = ibpd;
+	mr->access = ERDMA_MR_ACC_LR | to_erdma_access_flags(acc);
+	ret = regmr_cmd(dev, mr);
+	if (ret)
+		goto out_remove_stag;
+
+	return &mr->ibmr;
+
+out_remove_stag:
+	erdma_free_idx(&dev->res_cb[ERDMA_RES_TYPE_STAG_IDX],
+		       mr->ibmr.lkey >> 8);
+
+out_free:
+	kfree(mr);
+
+	return ERR_PTR(ret);
+}
+
+struct ib_mr *erdma_ib_alloc_mr(struct ib_pd *ibpd, enum ib_mr_type mr_type,
+				u32 max_num_sg)
+{
+	struct erdma_mr *mr;
+	struct erdma_dev *dev = to_edev(ibpd->device);
+	int ret;
+	u32 stag;
+
+	if (mr_type != IB_MR_TYPE_MEM_REG)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	if (max_num_sg > ERDMA_MR_MAX_MTT_CNT)
+		return ERR_PTR(-EINVAL);
+
+	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
+	if (!mr)
+		return ERR_PTR(-ENOMEM);
+
+	ret = erdma_create_stag(dev, &stag);
+	if (ret)
+		goto out_free;
+
+	mr->type = ERDMA_MR_TYPE_FRMR;
+
+	mr->ibmr.lkey = stag;
+	mr->ibmr.rkey = stag;
+	mr->ibmr.pd = ibpd;
+	/* update it in FRMR. */
+	mr->access = ERDMA_MR_ACC_LR | ERDMA_MR_ACC_LW | ERDMA_MR_ACC_RR |
+		     ERDMA_MR_ACC_RW;
+
+	mr->mem.page_size = PAGE_SIZE; /* update it later. */
+	mr->mem.page_cnt = max_num_sg;
+	mr->mem.mtt_type = ERDMA_MR_INDIRECT_MTT;
+	mr->mem.mtt_buf =
+		alloc_pages_exact(MTT_SIZE(mr->mem.page_cnt), GFP_KERNEL);
+	if (!mr->mem.mtt_buf) {
+		ret = -ENOMEM;
+		goto out_remove_stag;
+	}
+
+	mr->mem.mtt_entry[0] =
+		dma_map_single(&dev->pdev->dev, mr->mem.mtt_buf,
+			       MTT_SIZE(mr->mem.page_cnt), DMA_TO_DEVICE);
+	if (dma_mapping_error(&dev->pdev->dev, mr->mem.mtt_entry[0])) {
+		ret = -ENOMEM;
+		goto out_free_mtt;
+	}
+
+	ret = regmr_cmd(dev, mr);
+	if (ret)
+		goto out_dma_unmap;
+
+	return &mr->ibmr;
+
+out_dma_unmap:
+	dma_unmap_single(&dev->pdev->dev, mr->mem.mtt_entry[0],
+			 MTT_SIZE(mr->mem.page_cnt), DMA_TO_DEVICE);
+out_free_mtt:
+	free_pages_exact(mr->mem.mtt_buf, MTT_SIZE(mr->mem.page_cnt));
+
+out_remove_stag:
+	erdma_free_idx(&dev->res_cb[ERDMA_RES_TYPE_STAG_IDX],
+		       mr->ibmr.lkey >> 8);
+
+out_free:
+	kfree(mr);
+
+	return ERR_PTR(ret);
+}
+
+static int erdma_set_page(struct ib_mr *ibmr, u64 addr)
+{
+	struct erdma_mr *mr = to_emr(ibmr);
+
+	if (mr->mem.mtt_nents >= mr->mem.page_cnt)
+		return -1;
+
+	*((u64 *)mr->mem.mtt_buf + mr->mem.mtt_nents) = addr;
+	mr->mem.mtt_nents++;
+
+	return 0;
+}
+
+int erdma_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents,
+		    unsigned int *sg_offset)
+{
+	struct erdma_mr *mr = to_emr(ibmr);
+	int num;
+
+	mr->mem.mtt_nents = 0;
+
+	num = ib_sg_to_pages(&mr->ibmr, sg, sg_nents, sg_offset,
+			     erdma_set_page);
+
+	return num;
+}
+
+struct ib_mr *erdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
+				u64 virt, int access, struct ib_udata *udata)
+{
+	struct erdma_mr *mr = NULL;
+	struct erdma_dev *dev = to_edev(ibpd->device);
+	u32 stag;
+	int ret;
+
+	if (!len || len > dev->attrs.max_mr_size)
+		return ERR_PTR(-EINVAL);
+
+	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
+	if (!mr)
+		return ERR_PTR(-ENOMEM);
+
+	ret = get_mtt_entries(dev, &mr->mem, start, len, access, virt,
+			      SZ_2G - SZ_4K, 0);
+	if (ret)
+		goto err_out_free;
+
+	ret = erdma_create_stag(dev, &stag);
+	if (ret)
+		goto err_out_put_mtt;
+
+	mr->ibmr.lkey = mr->ibmr.rkey = stag;
+	mr->ibmr.pd = ibpd;
+	mr->mem.va = virt;
+	mr->mem.len = len;
+	mr->access = ERDMA_MR_ACC_LR | to_erdma_access_flags(access);
+	mr->valid = 1;
+	mr->type = ERDMA_MR_TYPE_NORMAL;
+
+	ret = regmr_cmd(dev, mr);
+	if (ret)
+		goto err_out_mr;
+
+	return &mr->ibmr;
+
+err_out_mr:
+	erdma_free_idx(&dev->res_cb[ERDMA_RES_TYPE_STAG_IDX],
+		       mr->ibmr.lkey >> 8);
+
+err_out_put_mtt:
+	put_mtt_entries(dev, &mr->mem);
+
+err_out_free:
+	kfree(mr);
+
+	return ERR_PTR(ret);
+}
+
+int erdma_dereg_mr(struct ib_mr *ibmr, struct ib_udata *udata)
+{
+	struct erdma_mr *mr;
+	struct erdma_dev *dev = to_edev(ibmr->device);
+	struct erdma_cmdq_dereg_mr_req req;
+	int ret;
+
+	mr = to_emr(ibmr);
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_DEREG_MR);
+
+	req.cfg = FIELD_PREP(ERDMA_CMD_MR_MPT_IDX_MASK, ibmr->lkey >> 8) |
+		  FIELD_PREP(ERDMA_CMD_MR_KEY_MASK, ibmr->lkey & 0xFF);
+
+	ret = erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				  NULL);
+	if (ret)
+		return ret;
+
+	erdma_free_idx(&dev->res_cb[ERDMA_RES_TYPE_STAG_IDX], ibmr->lkey >> 8);
+
+	put_mtt_entries(dev, &mr->mem);
+
+	kfree(mr);
+	return 0;
+}
+
+int erdma_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata)
+{
+	struct erdma_cq *cq = to_ecq(ibcq);
+	struct erdma_dev *dev = to_edev(ibcq->device);
+	struct erdma_ucontext *ctx = rdma_udata_to_drv_context(
+		udata, struct erdma_ucontext, ibucontext);
+	int err;
+	struct erdma_cmdq_destroy_cq_req req;
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_DESTROY_CQ);
+	req.cqn = cq->cqn;
+
+	err = erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				  NULL);
+	if (err)
+		return err;
+
+	if (rdma_is_kernel_res(&cq->ibcq.res)) {
+		dma_free_coherent(&dev->pdev->dev,
+				  WARPPED_BUFSIZE(cq->depth << CQE_SHIFT),
+				  cq->kern_cq.qbuf, cq->kern_cq.qbuf_dma_addr);
+	} else {
+		erdma_unmap_user_dbrecords(ctx, &cq->user_cq.user_dbr_page);
+		put_mtt_entries(dev, &cq->user_cq.qbuf_mtt);
+	}
+
+	xa_erase(&dev->cq_xa, cq->cqn);
+
+	return 0;
+}
+
+int erdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata)
+{
+	struct erdma_qp *qp = to_eqp(ibqp);
+	struct erdma_dev *dev = to_edev(ibqp->device);
+	struct erdma_ucontext *ctx = rdma_udata_to_drv_context(
+		udata, struct erdma_ucontext, ibucontext);
+	struct erdma_qp_attrs qp_attrs;
+	int err;
+	struct erdma_cmdq_destroy_qp_req req;
+
+	down_write(&qp->state_lock);
+	qp_attrs.state = ERDMA_QP_STATE_ERROR;
+	erdma_modify_qp_internal(qp, &qp_attrs, ERDMA_QP_ATTR_STATE);
+	up_write(&qp->state_lock);
+
+	erdma_cmdq_build_reqhdr(&req.hdr, CMDQ_SUBMOD_RDMA,
+				CMDQ_OPCODE_DESTROY_QP);
+	req.qpn = QP_ID(qp);
+
+	err = erdma_post_cmd_wait(&dev->cmdq, (u64 *)&req, sizeof(req), NULL,
+				  NULL);
+	if (err)
+		return err;
+
+	erdma_qp_put(qp);
+	wait_for_completion(&qp->safe_free);
+
+	if (rdma_is_kernel_res(&qp->ibqp.res)) {
+		vfree(qp->kern_qp.swr_tbl);
+		vfree(qp->kern_qp.rwr_tbl);
+		dma_free_coherent(
+			&dev->pdev->dev,
+			WARPPED_BUFSIZE(qp->attrs.rq_size << RQE_SHIFT),
+			qp->kern_qp.rq_buf, qp->kern_qp.rq_buf_dma_addr);
+		dma_free_coherent(
+			&dev->pdev->dev,
+			WARPPED_BUFSIZE(qp->attrs.sq_size << SQEBB_SHIFT),
+			qp->kern_qp.sq_buf, qp->kern_qp.sq_buf_dma_addr);
+	} else {
+		put_mtt_entries(dev, &qp->user_qp.sq_mtt);
+		put_mtt_entries(dev, &qp->user_qp.rq_mtt);
+		erdma_unmap_user_dbrecords(ctx, &qp->user_qp.user_dbr_page);
+	}
+
+	if (qp->cep)
+		erdma_cep_put(qp->cep);
+	xa_erase(&dev->qp_xa, QP_ID(qp));
+
+	return 0;
+}
+
+void erdma_qp_get_ref(struct ib_qp *ibqp)
+{
+	erdma_qp_get(to_eqp(ibqp));
+}
+
+void erdma_qp_put_ref(struct ib_qp *ibqp)
+{
+	erdma_qp_put(to_eqp(ibqp));
+}
+
+int erdma_mmap(struct ib_ucontext *ctx, struct vm_area_struct *vma)
+{
+	struct rdma_user_mmap_entry *rdma_entry;
+	struct erdma_user_mmap_entry *entry;
+	pgprot_t prot;
+	int err;
+
+	rdma_entry = rdma_user_mmap_entry_get(ctx, vma);
+	if (!rdma_entry)
+		return -EINVAL;
+
+	entry = to_emmap(rdma_entry);
+
+	switch (entry->mmap_flag) {
+	case ERDMA_MMAP_IO_NC:
+		/* map doorbell. */
+		prot = pgprot_device(vma->vm_page_prot);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	err = rdma_user_mmap_io(ctx, vma, PFN_DOWN(entry->address), PAGE_SIZE,
+				prot, rdma_entry);
+
+	rdma_user_mmap_entry_put(rdma_entry);
+	return err;
+}
+
+void erdma_mmap_free(struct rdma_user_mmap_entry *rdma_entry)
+{
+	struct erdma_user_mmap_entry *entry = to_emmap(rdma_entry);
+
+	kfree(entry);
+}
+
+#define ERDMA_SDB_PAGE 0
+#define ERDMA_SDB_ENTRY 1
+#define ERDMA_SDB_SHARED 2
+
+static void alloc_db_resources(struct erdma_dev *dev,
+			       struct erdma_ucontext *ctx)
+{
+	u32 bitmap_idx;
+	struct erdma_devattr *attrs = &dev->attrs;
+
+	if (attrs->disable_dwqe)
+		goto alloc_normal_db;
+
+	/* Try to alloc independent SDB page. */
+	spin_lock(&dev->db_bitmap_lock);
+	bitmap_idx = find_first_zero_bit(dev->sdb_page, attrs->dwqe_pages);
+	if (bitmap_idx != attrs->dwqe_pages) {
+		set_bit(bitmap_idx, dev->sdb_page);
+		spin_unlock(&dev->db_bitmap_lock);
+
+		ctx->sdb_type = ERDMA_SDB_PAGE;
+		ctx->sdb_idx = bitmap_idx;
+		ctx->sdb_page_idx = bitmap_idx;
+		ctx->sdb = dev->func_bar_addr + ERDMA_BAR_SQDB_SPACE_OFFSET +
+			   (bitmap_idx << PAGE_SHIFT);
+		ctx->sdb_page_off = 0;
+
+		return;
+	}
+
+	bitmap_idx = find_first_zero_bit(dev->sdb_entry, attrs->dwqe_entries);
+	if (bitmap_idx != attrs->dwqe_entries) {
+		set_bit(bitmap_idx, dev->sdb_entry);
+		spin_unlock(&dev->db_bitmap_lock);
+
+		ctx->sdb_type = ERDMA_SDB_ENTRY;
+		ctx->sdb_idx = bitmap_idx;
+		ctx->sdb_page_idx = attrs->dwqe_pages +
+				    bitmap_idx / ERDMA_DWQE_TYPE1_CNT_PER_PAGE;
+		ctx->sdb_page_off = bitmap_idx % ERDMA_DWQE_TYPE1_CNT_PER_PAGE;
+
+		ctx->sdb = dev->func_bar_addr + ERDMA_BAR_SQDB_SPACE_OFFSET +
+			   (ctx->sdb_page_idx << PAGE_SHIFT);
+
+		return;
+	}
+
+	spin_unlock(&dev->db_bitmap_lock);
+
+alloc_normal_db:
+	ctx->sdb_type = ERDMA_SDB_SHARED;
+	ctx->sdb_idx = 0;
+	ctx->sdb_page_idx = ERDMA_SDB_SHARED_PAGE_INDEX;
+	ctx->sdb_page_off = 0;
+
+	ctx->sdb = dev->func_bar_addr + (ctx->sdb_page_idx << PAGE_SHIFT);
+}
+
+static void erdma_uctx_user_mmap_entries_remove(struct erdma_ucontext *uctx)
+{
+	rdma_user_mmap_entry_remove(uctx->sq_db_mmap_entry);
+	rdma_user_mmap_entry_remove(uctx->rq_db_mmap_entry);
+	rdma_user_mmap_entry_remove(uctx->cq_db_mmap_entry);
+}
+
+int erdma_alloc_ucontext(struct ib_ucontext *ibctx, struct ib_udata *udata)
+{
+	struct erdma_ucontext *ctx = to_ectx(ibctx);
+	struct erdma_dev *dev = to_edev(ibctx->device);
+	int ret;
+	struct erdma_uresp_alloc_ctx uresp = {};
+
+	if (atomic_inc_return(&dev->num_ctx) > ERDMA_MAX_CONTEXT) {
+		ret = -ENOMEM;
+		goto err_out;
+	}
+
+	INIT_LIST_HEAD(&ctx->dbrecords_page_list);
+	mutex_init(&ctx->dbrecords_page_mutex);
+
+	alloc_db_resources(dev, ctx);
+
+	ctx->rdb = dev->func_bar_addr + ERDMA_BAR_RQDB_SPACE_OFFSET;
+	ctx->cdb = dev->func_bar_addr + ERDMA_BAR_CQDB_SPACE_OFFSET;
+
+	if (udata->outlen < sizeof(uresp)) {
+		ret = -EINVAL;
+		goto err_out;
+	}
+
+	ctx->sq_db_mmap_entry = erdma_user_mmap_entry_insert(
+		ctx, (void *)ctx->sdb, PAGE_SIZE, ERDMA_MMAP_IO_NC, &uresp.sdb);
+	if (!ctx->sq_db_mmap_entry) {
+		ret = -ENOMEM;
+		goto err_out;
+	}
+
+	ctx->rq_db_mmap_entry = erdma_user_mmap_entry_insert(
+		ctx, (void *)ctx->rdb, PAGE_SIZE, ERDMA_MMAP_IO_NC, &uresp.rdb);
+	if (!ctx->rq_db_mmap_entry) {
+		ret = -EINVAL;
+		goto err_out;
+	}
+
+	ctx->cq_db_mmap_entry = erdma_user_mmap_entry_insert(
+		ctx, (void *)ctx->cdb, PAGE_SIZE, ERDMA_MMAP_IO_NC, &uresp.cdb);
+	if (!ctx->cq_db_mmap_entry) {
+		ret = -EINVAL;
+		goto err_out;
+	}
+
+	uresp.dev_id = dev->pdev->device;
+	uresp.sdb_type = ctx->sdb_type;
+	uresp.sdb_offset = ctx->sdb_page_off;
+
+	ret = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+	if (ret)
+		goto err_out;
+
+	return 0;
+
+err_out:
+	erdma_uctx_user_mmap_entries_remove(ctx);
+	atomic_dec(&dev->num_ctx);
+	return ret;
+}
+
+void erdma_dealloc_ucontext(struct ib_ucontext *ibctx)
+{
+	struct erdma_ucontext *ctx = to_ectx(ibctx);
+	struct erdma_dev *dev = to_edev(ibctx->device);
+
+	spin_lock(&dev->db_bitmap_lock);
+	if (ctx->sdb_type == ERDMA_SDB_PAGE)
+		clear_bit(ctx->sdb_idx, dev->sdb_page);
+	else if (ctx->sdb_type == ERDMA_SDB_ENTRY)
+		clear_bit(ctx->sdb_idx, dev->sdb_entry);
+
+	erdma_uctx_user_mmap_entries_remove(ctx);
+
+	spin_unlock(&dev->db_bitmap_lock);
+
+	atomic_dec(&dev->num_ctx);
+}
+
+static int ib_qp_state_to_erdma_qp_state[IB_QPS_ERR + 1] = {
+	[IB_QPS_RESET] = ERDMA_QP_STATE_IDLE,
+	[IB_QPS_INIT] = ERDMA_QP_STATE_IDLE,
+	[IB_QPS_RTR] = ERDMA_QP_STATE_RTR,
+	[IB_QPS_RTS] = ERDMA_QP_STATE_RTS,
+	[IB_QPS_SQD] = ERDMA_QP_STATE_CLOSING,
+	[IB_QPS_SQE] = ERDMA_QP_STATE_TERMINATE,
+	[IB_QPS_ERR] = ERDMA_QP_STATE_ERROR
+};
+
+int erdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
+		    struct ib_udata *udata)
+{
+	struct erdma_qp_attrs new_attrs;
+	enum erdma_qp_attr_mask erdma_attr_mask = 0;
+	struct erdma_qp *qp = to_eqp(ibqp);
+	int ret = 0;
+
+	if (attr_mask & ~IB_QP_ATTR_STANDARD_BITS)
+		return -EOPNOTSUPP;
+
+	memset(&new_attrs, 0, sizeof(new_attrs));
+
+	if (attr_mask & IB_QP_STATE) {
+		new_attrs.state = ib_qp_state_to_erdma_qp_state[attr->qp_state];
+
+		erdma_attr_mask |= ERDMA_QP_ATTR_STATE;
+	}
+
+	down_write(&qp->state_lock);
+
+	ret = erdma_modify_qp_internal(qp, &new_attrs, erdma_attr_mask);
+
+	up_write(&qp->state_lock);
+
+	return ret;
+}
+
+int erdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
+		   int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr)
+{
+	struct erdma_qp *qp;
+	struct erdma_dev *dev;
+
+	if (ibqp && qp_attr && qp_init_attr) {
+		qp = to_eqp(ibqp);
+		dev = to_edev(ibqp->device);
+	} else {
+		return -EINVAL;
+	}
+
+	qp_attr->cap.max_inline_data = ERDMA_MAX_INLINE;
+	qp_init_attr->cap.max_inline_data = ERDMA_MAX_INLINE;
+
+	qp_attr->cap.max_send_wr = qp->attrs.sq_size;
+	qp_attr->cap.max_recv_wr = qp->attrs.rq_size;
+	qp_attr->cap.max_send_sge = qp->attrs.max_send_sge;
+	qp_attr->cap.max_recv_sge = qp->attrs.max_recv_sge;
+
+	qp_attr->path_mtu = ib_mtu_int_to_enum(dev->netdev->mtu);
+	qp_attr->max_rd_atomic = qp->attrs.irq_size;
+	qp_attr->max_dest_rd_atomic = qp->attrs.orq_size;
+
+	qp_attr->qp_access_flags = IB_ACCESS_LOCAL_WRITE |
+				   IB_ACCESS_REMOTE_WRITE |
+				   IB_ACCESS_REMOTE_READ;
+
+	qp_init_attr->cap = qp_attr->cap;
+
+	return 0;
+}
+
+static int erdma_init_user_cq(struct erdma_ucontext *ctx, struct erdma_cq *cq,
+			      struct erdma_ureq_create_cq *ureq)
+{
+	int ret;
+	struct erdma_dev *dev = to_edev(cq->ibcq.device);
+
+	ret = get_mtt_entries(dev, &cq->user_cq.qbuf_mtt, ureq->qbuf_va,
+			      ureq->qbuf_len, 0, ureq->qbuf_va, SZ_64M - SZ_4K,
+			      1);
+	if (ret)
+		return ret;
+
+	ret = erdma_map_user_dbrecords(ctx, ureq->db_record_va,
+				       &cq->user_cq.user_dbr_page,
+				       &cq->user_cq.db_info_dma_addr);
+	if (ret)
+		put_mtt_entries(dev, &cq->user_cq.qbuf_mtt);
+
+	return ret;
+}
+
+static int erdma_init_kernel_cq(struct erdma_cq *cq)
+{
+	struct erdma_dev *dev = to_edev(cq->ibcq.device);
+
+	cq->kern_cq.qbuf =
+		dma_alloc_coherent(&dev->pdev->dev,
+				   WARPPED_BUFSIZE(cq->depth << CQE_SHIFT),
+				   &cq->kern_cq.qbuf_dma_addr, GFP_KERNEL);
+	if (!cq->kern_cq.qbuf)
+		return -ENOMEM;
+
+	cq->kern_cq.db_record =
+		(u64 *)(cq->kern_cq.qbuf + (cq->depth << CQE_SHIFT));
+	spin_lock_init(&cq->kern_cq.lock);
+	/* use default cqdb addr */
+	cq->kern_cq.db = dev->func_bar + ERDMA_BAR_CQDB_SPACE_OFFSET;
+
+	return 0;
+}
+
+int erdma_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
+		    struct ib_udata *udata)
+{
+	struct erdma_cq *cq = to_ecq(ibcq);
+	struct erdma_dev *dev = to_edev(ibcq->device);
+	unsigned int depth = attr->cqe;
+	int ret;
+	struct erdma_ucontext *ctx = rdma_udata_to_drv_context(
+		udata, struct erdma_ucontext, ibucontext);
+
+	if (depth > dev->attrs.max_cqe)
+		return -EINVAL;
+
+	depth = roundup_pow_of_two(depth);
+	cq->ibcq.cqe = depth;
+	cq->depth = depth;
+	cq->assoc_eqn = attr->comp_vector + 1;
+
+	ret = xa_alloc_cyclic(&dev->cq_xa, &cq->cqn, cq,
+			      XA_LIMIT(1, dev->attrs.max_cq - 1),
+			      &dev->next_alloc_cqn, GFP_KERNEL);
+	if (ret < 0)
+		return ret;
+
+	if (!rdma_is_kernel_res(&ibcq->res)) {
+		struct erdma_ureq_create_cq ureq;
+		struct erdma_uresp_create_cq uresp;
+
+		ret = ib_copy_from_udata(&ureq, udata,
+					 min(udata->inlen, sizeof(ureq)));
+		if (ret)
+			goto err_out_xa;
+
+		ret = erdma_init_user_cq(ctx, cq, &ureq);
+		if (ret)
+			goto err_out_xa;
+
+		uresp.cq_id = cq->cqn;
+		uresp.num_cqe = depth;
+
+		ret = ib_copy_to_udata(udata, &uresp,
+				       min(sizeof(uresp), udata->outlen));
+		if (ret)
+			goto err_free_res;
+	} else {
+		ret = erdma_init_kernel_cq(cq);
+		if (ret)
+			goto err_out_xa;
+	}
+
+	ret = create_cq_cmd(dev, cq);
+	if (ret)
+		goto err_free_res;
+
+	return 0;
+
+err_free_res:
+	if (!rdma_is_kernel_res(&ibcq->res)) {
+		erdma_unmap_user_dbrecords(ctx, &cq->user_cq.user_dbr_page);
+		put_mtt_entries(dev, &cq->user_cq.qbuf_mtt);
+	} else {
+		dma_free_coherent(&dev->pdev->dev,
+				  WARPPED_BUFSIZE(depth << CQE_SHIFT),
+				  cq->kern_cq.qbuf, cq->kern_cq.qbuf_dma_addr);
+	}
+
+err_out_xa:
+	xa_erase(&dev->cq_xa, cq->cqn);
+
+	return ret;
+}
+
+void erdma_port_event(struct erdma_dev *dev, enum ib_event_type reason)
+{
+	struct ib_event event;
+
+	event.device = &dev->ibdev;
+	event.element.port_num = 1;
+	event.event = reason;
+
+	ib_dispatch_event(&event);
+}

drivers/infiniband/hw/erdma/erdma_verbs.h

···
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+
+/* Authors: Cheng Xu <[e-mail]> */
+/*          Kai Shen <[e-mail]> */
+/* Copyright (c) 2020-2022, Alibaba Group. */
+
+#ifndef __ERDMA_VERBS_H__
+#define __ERDMA_VERBS_H__
+
+#include <linux/errno.h>
+
+#include <rdma/ib_verbs.h>
+#include <rdma/ib_user_verbs.h>
+#include <rdma/iw_cm.h>
+
+#include "erdma.h"
+#include "erdma_cm.h"
+#include "erdma_hw.h"
+
+/* RDMA Capability. */
+#define ERDMA_MAX_PD (128 * 1024)
+#define ERDMA_MAX_SEND_WR 4096
+#define ERDMA_MAX_ORD 128
+#define ERDMA_MAX_IRD 128
+#define ERDMA_MAX_SGE_RD 1
+#define ERDMA_MAX_CONTEXT (128 * 1024)
+#define ERDMA_MAX_SEND_SGE 6
+#define ERDMA_MAX_RECV_SGE 1
+#define ERDMA_MAX_INLINE (sizeof(struct erdma_sge) * (ERDMA_MAX_SEND_SGE))
+#define ERDMA_MAX_FRMR_PA 512
+
+enum {
+	ERDMA_MMAP_IO_NC = 0, /* no cache */
+};
+
+struct erdma_user_mmap_entry {
+	struct rdma_user_mmap_entry rdma_entry;
+	u64 address;
+	u8 mmap_flag;
+};
+
+struct erdma_ucontext {
+	struct ib_ucontext ibucontext;
+
+	u32 sdb_type;
+	u32 sdb_idx;
+	u32 sdb_page_idx;
+	u32 sdb_page_off;
+	u64 sdb;
+	u64 rdb;
+	u64 cdb;
+
+	struct rdma_user_mmap_entry *sq_db_mmap_entry;
+	struct rdma_user_mmap_entry *rq_db_mmap_entry;
+	struct rdma_user_mmap_entry *cq_db_mmap_entry;
+
+	/* doorbell records */
+	struct list_head dbrecords_page_list;
+	struct mutex dbrecords_page_mutex;
+};
+
+struct erdma_pd {
+	struct ib_pd ibpd;
+	u32 pdn;
+};
+
+/*
+ * MemoryRegion definition.
+ */
+#define ERDMA_MAX_INLINE_MTT_ENTRIES 4
+#define MTT_SIZE(mtt_cnt) (mtt_cnt << 3) /* per mtt takes 8 Bytes. */
+#define ERDMA_MR_MAX_MTT_CNT 524288
+#define ERDMA_MTT_ENTRY_SIZE 8
+
+#define ERDMA_MR_TYPE_NORMAL 0
+#define ERDMA_MR_TYPE_FRMR 1
+#define ERDMA_MR_TYPE_DMA 2
+
+#define ERDMA_MR_INLINE_MTT 0
+#define ERDMA_MR_INDIRECT_MTT 1
+
+#define ERDMA_MR_ACC_LR BIT(0)
+#define ERDMA_MR_ACC_LW BIT(1)
+#define ERDMA_MR_ACC_RR BIT(2)
+#define ERDMA_MR_ACC_RW BIT(3)
+
+static inline u8 to_erdma_access_flags(int access)
+{
+	return (access & IB_ACCESS_REMOTE_READ ? ERDMA_MR_ACC_RR : 0) |
+	       (access & IB_ACCESS_LOCAL_WRITE ? ERDMA_MR_ACC_LW : 0) |
+	       (access & IB_ACCESS_REMOTE_WRITE ? ERDMA_MR_ACC_RW : 0);
+}
+
+struct erdma_mem {
+	struct ib_umem *umem;
+	void *mtt_buf;
+	u32 mtt_type;
+	u32 page_size;
+	u32 page_offset;
+	u32 page_cnt;
+	u32 mtt_nents;
+
+	u64 va;
+	u64 len;
+
+	u64 mtt_entry[ERDMA_MAX_INLINE_MTT_ENTRIES];
+};
+
+struct erdma_mr {
+	struct ib_mr ibmr;
+	struct erdma_mem mem;
+	u8 type;
+	u8 access;
+	u8 valid;
+};
+
+struct erdma_user_dbrecords_page {
+	struct list_head list;
+	struct ib_umem *umem;
+	u64 va;
+	int refcnt;
+};
+
+struct erdma_uqp {
+	struct erdma_mem sq_mtt;
+	struct erdma_mem rq_mtt;
+
+	dma_addr_t sq_db_info_dma_addr;
+	dma_addr_t rq_db_info_dma_addr;
+
+	struct erdma_user_dbrecords_page *user_dbr_page;
+
+	u32 rq_offset;
+};
+
+struct erdma_kqp {
+	u16 sq_pi;
+	u16 sq_ci;
+
+	u16 rq_pi;
+	u16 rq_ci;
+
+	u64 *swr_tbl;
+	u64 *rwr_tbl;
+
+	void __iomem *hw_sq_db;
+	void __iomem *hw_rq_db;
+
+	void *sq_buf;
+	dma_addr_t sq_buf_dma_addr;
+
+	void *rq_buf;
+	dma_addr_t rq_buf_dma_addr;
+
+	void *sq_db_info;
+	void *rq_db_info;
+
+	u8 sig_all;
+};
+
+enum erdma_qp_state {
+	ERDMA_QP_STATE_IDLE = 0,
+	ERDMA_QP_STATE_RTR = 1,
+	ERDMA_QP_STATE_RTS = 2,
+	ERDMA_QP_STATE_CLOSING = 3,
+	ERDMA_QP_STATE_TERMINATE = 4,
+	ERDMA_QP_STATE_ERROR = 5,
+	ERDMA_QP_STATE_UNDEF = 7,
+	ERDMA_QP_STATE_COUNT = 8
+};
+
+enum erdma_qp_attr_mask {
+	ERDMA_QP_ATTR_STATE = (1 << 0),
+	ERDMA_QP_ATTR_LLP_HANDLE = (1 << 2),
+	ERDMA_QP_ATTR_ORD = (1 << 3),
+	ERDMA_QP_ATTR_IRD = (1 << 4),
+	ERDMA_QP_ATTR_SQ_SIZE = (1 << 5),
+	ERDMA_QP_ATTR_RQ_SIZE = (1 << 6),
+	ERDMA_QP_ATTR_MPA = (1 << 7)
+};
+
+struct erdma_qp_attrs {
+	enum erdma_qp_state state;
+	enum erdma_cc_alg cc; /* Congestion control algorithm */
+	u32 sq_size;
+	u32 rq_size;
+	u32 orq_size;
+	u32 irq_size;
+	u32 max_send_sge;
+	u32 max_recv_sge;
+	u32 cookie;
+#define ERDMA_QP_ACTIVE 0
+#define ERDMA_QP_PASSIVE 1
+	u8 qp_type;
+	u8 pd_len;
+};
+
+struct erdma_qp {
+	struct ib_qp ibqp;
+	struct kref ref;
+	struct completion safe_free;
+	struct erdma_dev *dev;
+	struct erdma_cep *cep;
+	struct rw_semaphore state_lock;
+
+	union {
+		struct erdma_kqp kern_qp;
+		struct erdma_uqp user_qp;
+	};
+
+	struct erdma_cq *scq;
+	struct erdma_cq *rcq;
+
+	struct erdma_qp_attrs attrs;
+	spinlock_t lock;
+};
+
+struct erdma_kcq_info {
+	void *qbuf;
+	dma_addr_t qbuf_dma_addr;
+	u32 ci;
+	u32 cmdsn;
+	u32 notify_cnt;
+
+	spinlock_t lock;
+	u8 __iomem *db;
+	u64 *db_record;
+};
+
+struct erdma_ucq_info {
+	struct erdma_mem qbuf_mtt;
+	struct erdma_user_dbrecords_page *user_dbr_page;
+	dma_addr_t db_info_dma_addr;
+};
+
+struct erdma_cq {
+	struct ib_cq ibcq;
+	u32 cqn;
+
+	u32 depth;
+	u32 assoc_eqn;
+
+	union {
+		struct erdma_kcq_info kern_cq;
+		struct erdma_ucq_info user_cq;
+	};
+};
+
+#define QP_ID(qp) ((qp)->ibqp.qp_num)
+
+static inline struct erdma_qp *find_qp_by_qpn(struct erdma_dev *dev, int id)
+{
+	return (struct erdma_qp *)xa_load(&dev->qp_xa, id);
+}
+
+static inline struct erdma_cq *find_cq_by_cqn(struct erdma_dev *dev, int id)
+{
+	return (struct erdma_cq *)xa_load(&dev->cq_xa, id);
+}
+
+void erdma_qp_get(struct erdma_qp *qp);
+void erdma_qp_put(struct erdma_qp *qp);
+int erdma_modify_qp_internal(struct erdma_qp *qp, struct erdma_qp_attrs *attrs,
+			     enum erdma_qp_attr_mask mask);
+void erdma_qp_llp_close(struct erdma_qp *qp);
+void erdma_qp_cm_drop(struct erdma_qp *qp);
+
+static inline struct erdma_ucontext *to_ectx(struct ib_ucontext *ibctx)
+{
+	return container_of(ibctx, struct erdma_ucontext, ibucontext);
+}
+
+static inline struct erdma_pd *to_epd(struct ib_pd *pd)
+{
+	return container_of(pd, struct erdma_pd, ibpd);
+}
+
+static inline struct erdma_mr *to_emr(struct ib_mr *ibmr)
+{
+	return container_of(ibmr, struct erdma_mr, ibmr);
+}
+
+static inline struct erdma_qp *to_eqp(struct ib_qp *qp)
+{
+	return container_of(qp, struct erdma_qp, ibqp);
+}
+
+static inline struct erdma_cq *to_ecq(struct ib_cq *ibcq)
+{
+	return container_of(ibcq, struct erdma_cq, ibcq);
+}
+
+static inline struct erdma_user_mmap_entry *
+to_emmap(struct rdma_user_mmap_entry *ibmmap)
+{
+	return container_of(ibmmap, struct erdma_user_mmap_entry, rdma_entry);
+}
+
+int erdma_alloc_ucontext(struct ib_ucontext *ibctx, struct ib_udata *data);
+void erdma_dealloc_ucontext(struct ib_ucontext *ibctx);
+int erdma_query_device(struct ib_device *dev, struct ib_device_attr *attr,
+		       struct ib_udata *data);
+int erdma_get_port_immutable(struct ib_device *dev, u32 port,
+			     struct ib_port_immutable *ib_port_immutable);
+int erdma_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
+		    struct ib_udata *data);
+int erdma_query_port(struct ib_device *dev, u32 port,
+		     struct ib_port_attr *attr);
+int erdma_query_gid(struct ib_device *dev, u32 port, int idx,
+		    union ib_gid *gid);
+int erdma_alloc_pd(struct ib_pd *ibpd, struct ib_udata *data);
+int erdma_dealloc_pd(struct ib_pd *ibpd, struct ib_udata *udata);
+int erdma_create_qp(struct ib_qp *ibqp, struct ib_qp_init_attr *attr,
+		    struct ib_udata *data);
+int erdma_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int mask,
+		   struct ib_qp_init_attr *init_attr);
+int erdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int mask,
+		    struct ib_udata *data);
+int erdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata);
+int erdma_destroy_cq(struct ib_cq *ibcq, struct ib_udata *udata);
+int erdma_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags);
+struct ib_mr *erdma_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
+				u64 virt, int access, struct ib_udata *udata);
+struct ib_mr *erdma_get_dma_mr(struct ib_pd *ibpd, int rights);
+int erdma_dereg_mr(struct ib_mr *ibmr, struct ib_udata *data);
+int erdma_mmap(struct ib_ucontext *ctx, struct vm_area_struct *vma);
+void erdma_mmap_free(struct rdma_user_mmap_entry *rdma_entry);
+void erdma_qp_get_ref(struct ib_qp *ibqp);
+void erdma_qp_put_ref(struct ib_qp *ibqp);
+struct ib_qp *erdma_get_ibqp(struct ib_device *dev, int id);
+int erdma_post_send(struct ib_qp *ibqp, const struct ib_send_wr *send_wr,
+		    const struct ib_send_wr **bad_send_wr);
+int erdma_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *recv_wr,
+		    const struct ib_recv_wr **bad_recv_wr);
+int erdma_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc);
+struct ib_mr *erdma_ib_alloc_mr(struct ib_pd *ibpd, enum ib_mr_type mr_type,
+				u32 max_num_sg);
+int erdma_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents,
+		    unsigned int *sg_offset);
+void erdma_port_event(struct erdma_dev *dev, enum ib_event_type reason);
+
+#endif

drivers/infiniband/hw/hfi1/Kconfig

···
 # SPDX-License-Identifier: GPL-2.0-only
 config INFINIBAND_HFI1
 	tristate "Cornelis OPX Gen1 support"
-	depends on X86_64 && INFINIBAND_RDMAVT && I2C
+	depends on X86_64 && INFINIBAND_RDMAVT && I2C && !UML
 	select MMU_NOTIFIER
 	select CRC32
 	select I2C_ALGOBIT

drivers/infiniband/hw/hfi1/file_ops.c

···
 		goto done;
 
 	ret = init_user_ctxt(fd, uctxt);
-	if (ret)
+	if (ret) {
+		hfi1_free_ctxt_rcv_groups(uctxt);
 		goto done;
+	}
 
 	user_init(uctxt);
 

drivers/infiniband/hw/hfi1/ipoib_tx.c

···
 				kzalloc_node(sizeof(*tx->sdma_hdr),
 					     GFP_KERNEL, priv->dd->node);
 
-		netif_tx_napi_add(dev, &txq->napi,
-				  hfi1_ipoib_poll_tx_ring,
-				  NAPI_POLL_WEIGHT);
+		netif_napi_add_tx(dev, &txq->napi, hfi1_ipoib_poll_tx_ring);
 	}
 
 	return 0;

drivers/infiniband/hw/hfi1/netdev_rx.c

···
 		 * right now.
 		 */
 		set_bit(NAPI_STATE_NO_BUSY_POLL, &rxq->napi.state);
-		netif_napi_add(dev, &rxq->napi, hfi1_netdev_rx_napi, 64);
+		netif_napi_add_weight(dev, &rxq->napi, hfi1_netdev_rx_napi, 64);
 		rc = msix_netdev_request_rcd_irq(rxq->rcd);
 		if (rc)
 			goto bail_context_irq_failure;

drivers/infiniband/hw/hfi1/pio_copy.c

···
 }
 
 /*
- * Read nbytes from "from" and and place them in the low bytes
+ * Read nbytes from "from" and place them in the low bytes
  * of pbuf->carry.  Other bytes are left as-is.  Any previous
  * value in pbuf->carry is lost.
  *

drivers/infiniband/hw/hns/hns_roce_device.h

···
 	const struct hns_roce_hw *hw;
 	void			*priv;
 	struct workqueue_struct *irq_workq;
+	struct work_struct ecc_work;
 	const struct hns_roce_dfx_hw *dfx;
 	u32 func_num;
 	u32 is_vf;

drivers/infiniband/hw/hns/hns_roce_hw_v2.c

···
 	CMD_RST_PRC_EBUSY,
 };
 
+enum ecc_resource_type {
+	ECC_RESOURCE_QPC,
+	ECC_RESOURCE_CQC,
+	ECC_RESOURCE_MPT,
+	ECC_RESOURCE_SRQC,
+	ECC_RESOURCE_GMV,
+	ECC_RESOURCE_QPC_TIMER,
+	ECC_RESOURCE_CQC_TIMER,
+	ECC_RESOURCE_SCCC,
+	ECC_RESOURCE_COUNT,
+};
+
+static const struct {
+	const char *name;
+	u8 read_bt0_op;
+	u8 write_bt0_op;
+} fmea_ram_res[] = {
+	{ "ECC_RESOURCE_QPC",
+	  HNS_ROCE_CMD_READ_QPC_BT0, HNS_ROCE_CMD_WRITE_QPC_BT0 },
+	{ "ECC_RESOURCE_CQC",
+	  HNS_ROCE_CMD_READ_CQC_BT0, HNS_ROCE_CMD_WRITE_CQC_BT0 },
+	{ "ECC_RESOURCE_MPT",
+	  HNS_ROCE_CMD_READ_MPT_BT0, HNS_ROCE_CMD_WRITE_MPT_BT0 },
+	{ "ECC_RESOURCE_SRQC",
+	  HNS_ROCE_CMD_READ_SRQC_BT0, HNS_ROCE_CMD_WRITE_SRQC_BT0 },
+	/* ECC_RESOURCE_GMV is handled by cmdq, not mailbox */
+	{ "ECC_RESOURCE_GMV",
+	  0, 0 },
+	{ "ECC_RESOURCE_QPC_TIMER",
+	  HNS_ROCE_CMD_READ_QPC_TIMER_BT0, HNS_ROCE_CMD_WRITE_QPC_TIMER_BT0 },
+	{ "ECC_RESOURCE_CQC_TIMER",
+	  HNS_ROCE_CMD_READ_CQC_TIMER_BT0, HNS_ROCE_CMD_WRITE_CQC_TIMER_BT0 },
+	{ "ECC_RESOURCE_SCCC",
+	  HNS_ROCE_CMD_READ_SCCC_BT0, HNS_ROCE_CMD_WRITE_SCCC_BT0 },
+};
+
 static inline void set_data_seg_v2(struct hns_roce_v2_wqe_data_seg *dseg,
 				   struct ib_sge *sg)
 {
···
 		!!(eq->cons_index & eq->entries)) ? aeqe : NULL;
 }
 
-static int hns_roce_v2_aeq_int(struct hns_roce_dev *hr_dev,
-			       struct hns_roce_eq *eq)
+static irqreturn_t hns_roce_v2_aeq_int(struct hns_roce_dev *hr_dev,
+				       struct hns_roce_eq *eq)
 {
 	struct device *dev = hr_dev->dev;
 	struct hns_roce_aeqe *aeqe = next_aeqe_sw_v2(eq);
-	int aeqe_found = 0;
+	irqreturn_t aeqe_found = IRQ_NONE;
 	int event_type;
 	u32 queue_num;
 	int sub_type;
···
 		eq->event_type = event_type;
 		eq->sub_type = sub_type;
 		++eq->cons_index;
-		aeqe_found = 1;
+		aeqe_found = IRQ_HANDLED;
 
 		hns_roce_v2_init_irq_work(hr_dev, eq, queue_num);
 
···
 	}
 
 	update_eq_db(eq);
-	return aeqe_found;
+
+	return IRQ_RETVAL(aeqe_found);
 }
 
 static struct hns_roce_ceqe *next_ceqe_sw_v2(struct hns_roce_eq *eq)
···
 		!!(eq->cons_index & eq->entries)) ? ceqe : NULL;
 }
 
-static int hns_roce_v2_ceq_int(struct hns_roce_dev *hr_dev,
-			       struct hns_roce_eq *eq)
+static irqreturn_t hns_roce_v2_ceq_int(struct hns_roce_dev *hr_dev,
+				       struct hns_roce_eq *eq)
 {
 	struct hns_roce_ceqe *ceqe = next_ceqe_sw_v2(eq);
-	int ceqe_found = 0;
+	irqreturn_t ceqe_found = IRQ_NONE;
 	u32 cqn;
 
 	while (ceqe) {
···
 		hns_roce_cq_completion(hr_dev, cqn);
 
 		++eq->cons_index;
-		ceqe_found = 1;
+		ceqe_found = IRQ_HANDLED;
 
 		ceqe = next_ceqe_sw_v2(eq);
 	}
 
 	update_eq_db(eq);
 
-	return ceqe_found;
+	return IRQ_RETVAL(ceqe_found);
 }
 
 static irqreturn_t hns_roce_v2_msix_interrupt_eq(int irq, void *eq_ptr)
 {
 	struct hns_roce_eq *eq = eq_ptr;
 	struct hns_roce_dev *hr_dev = eq->hr_dev;
-	int int_work;
+	irqreturn_t int_work;
 
 	if (eq->type_flag == HNS_ROCE_CEQ)
 		/* Completion event interrupt */
···
 	return IRQ_RETVAL(int_work);
 }
 
-static irqreturn_t hns_roce_v2_msix_interrupt_abn(int irq, void *dev_id)
+static irqreturn_t abnormal_interrupt_basic(struct hns_roce_dev *hr_dev,
+					    u32 int_st)
 {
-	struct hns_roce_dev *hr_dev = dev_id;
-	struct device *dev = hr_dev->dev;
-	int int_work = 0;
-	u32 int_st;
+	struct pci_dev *pdev = hr_dev->pci_dev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	const struct hnae3_ae_ops *ops = ae_dev->ops;
+	irqreturn_t int_work = IRQ_NONE;
 	u32 int_en;
 
-	/* Abnormal interrupt */
-	int_st = roce_read(hr_dev, ROCEE_VF_ABN_INT_ST_REG);
 	int_en = roce_read(hr_dev, ROCEE_VF_ABN_INT_EN_REG);
 
 	if (int_st & BIT(HNS_ROCE_V2_VF_INT_ST_AEQ_OVERFLOW_S)) {
-		struct pci_dev *pdev = hr_dev->pci_dev;
-		struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
-		const struct hnae3_ae_ops *ops = ae_dev->ops;
+		dev_err(hr_dev->dev, "AEQ overflow!\n");
 
-		dev_err(dev, "AEQ overflow!\n");
-
-		int_st |= 1 << HNS_ROCE_V2_VF_INT_ST_AEQ_OVERFLOW_S;
-		roce_write(hr_dev, ROCEE_VF_ABN_INT_ST_REG, int_st);
+		roce_write(hr_dev, ROCEE_VF_ABN_INT_ST_REG,
+			   1 << HNS_ROCE_V2_VF_INT_ST_AEQ_OVERFLOW_S);
 
 		/* Set reset level for reset_event() */
 		if (ops->set_default_reset_request)
···
 		int_en |= 1 << HNS_ROCE_V2_VF_ABN_INT_EN_S;
 		roce_write(hr_dev, ROCEE_VF_ABN_INT_EN_REG, int_en);
 
-		int_work = 1;
-	} else if (int_st & BIT(HNS_ROCE_V2_VF_INT_ST_RAS_INT_S)) {
-		dev_err(dev, "RAS interrupt!\n");
-
-		int_st |= 1 << HNS_ROCE_V2_VF_INT_ST_RAS_INT_S;
-		roce_write(hr_dev, ROCEE_VF_ABN_INT_ST_REG, int_st);
-
-		int_en |= 1 << HNS_ROCE_V2_VF_ABN_INT_EN_S;
-		roce_write(hr_dev, ROCEE_VF_ABN_INT_EN_REG, int_en);
-
-		int_work = 1;
+		int_work = IRQ_HANDLED;
 	} else {
-		dev_err(dev, "There is no abnormal irq found!\n");
+		dev_err(hr_dev->dev, "there is no basic abn irq found.\n");
+	}
+
+	return IRQ_RETVAL(int_work);
+}
+
+static int fmea_ram_ecc_query(struct hns_roce_dev *hr_dev,
+			       struct fmea_ram_ecc *ecc_info)
+{
+	struct hns_roce_cmq_desc desc;
+	struct hns_roce_cmq_req *req = (struct hns_roce_cmq_req *)desc.data;
+	int ret;
+
+	hns_roce_cmq_setup_basic_desc(&desc, HNS_ROCE_QUERY_RAM_ECC, true);
+	ret = hns_roce_cmq_send(hr_dev, &desc, 1);
+	if (ret)
+		return ret;
+
+	ecc_info->is_ecc_err = hr_reg_read(req, QUERY_RAM_ECC_1BIT_ERR);
+	ecc_info->res_type = hr_reg_read(req, QUERY_RAM_ECC_RES_TYPE);
+	ecc_info->index = hr_reg_read(req, QUERY_RAM_ECC_TAG);
+
+	return 0;
+}
+
+static int fmea_recover_gmv(struct hns_roce_dev *hr_dev, u32 idx)
+{
+	struct hns_roce_cmq_desc desc;
+	struct hns_roce_cmq_req *req = (struct hns_roce_cmq_req *)desc.data;
+	u32 addr_upper;
+	u32 addr_low;
+	int ret;
+
+	hns_roce_cmq_setup_basic_desc(&desc, HNS_ROCE_OPC_CFG_GMV_BT, true);
+	hr_reg_write(req, CFG_GMV_BT_IDX, idx);
+
+	ret = hns_roce_cmq_send(hr_dev, &desc, 1);
+	if (ret) {
+		dev_err(hr_dev->dev,
+			"failed to execute cmd to read gmv, ret = %d.\n", ret);
+		return ret;
+	}
+
+	addr_low =  hr_reg_read(req, CFG_GMV_BT_BA_L);
+	addr_upper = hr_reg_read(req, CFG_GMV_BT_BA_H);
+
+	hns_roce_cmq_setup_basic_desc(&desc, HNS_ROCE_OPC_CFG_GMV_BT, false);
+	hr_reg_write(req, CFG_GMV_BT_BA_L, addr_low);
+	hr_reg_write(req, CFG_GMV_BT_BA_H, addr_upper);
+	hr_reg_write(req, CFG_GMV_BT_IDX, idx);
+
+	return hns_roce_cmq_send(hr_dev, &desc, 1);
+}
+
+static u64 fmea_get_ram_res_addr(u32 res_type, __le64 *data)
+{
+	if (res_type == ECC_RESOURCE_QPC_TIMER ||
+	    res_type == ECC_RESOURCE_CQC_TIMER ||
+	    res_type == ECC_RESOURCE_SCCC)
+		return le64_to_cpu(*data);
+
+	return le64_to_cpu(*data) << PAGE_SHIFT;
+}
+
+static int fmea_recover_others(struct hns_roce_dev *hr_dev, u32 res_type,
+			       u32 index)
+{
+	u8 write_bt0_op = fmea_ram_res[res_type].write_bt0_op;
+	u8 read_bt0_op = fmea_ram_res[res_type].read_bt0_op;
+	struct hns_roce_cmd_mailbox *mailbox;
+	u64 addr;
+	int ret;
+
+	mailbox = hns_roce_alloc_cmd_mailbox(hr_dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	ret = hns_roce_cmd_mbox(hr_dev, 0, mailbox->dma, read_bt0_op, index);
+	if (ret) {
+		dev_err(hr_dev->dev,
+			"failed to execute cmd to read fmea ram, ret = %d.\n",
+			ret);
+		goto out;
+	}
+
+	addr = fmea_get_ram_res_addr(res_type, mailbox->buf);
+
+	ret = hns_roce_cmd_mbox(hr_dev, addr, 0, write_bt0_op, index);
+	if (ret)
+		dev_err(hr_dev->dev,
+			"failed to execute cmd to write fmea ram, ret = %d.\n",
+			ret);
+
+out:
+	hns_roce_free_cmd_mailbox(hr_dev, mailbox);
+	return ret;
+}
+
+static void fmea_ram_ecc_recover(struct hns_roce_dev *hr_dev,
+				 struct fmea_ram_ecc *ecc_info)
+{
+	u32 res_type = ecc_info->res_type;
+	u32 index = ecc_info->index;
+	int ret;
+
+	BUILD_BUG_ON(ARRAY_SIZE(fmea_ram_res) != ECC_RESOURCE_COUNT);
+
+	if (res_type >= ECC_RESOURCE_COUNT) {
+		dev_err(hr_dev->dev, "unsupported fmea ram ecc type %u.\n",
+			res_type);
+		return;
+	}
+
+	if (res_type == ECC_RESOURCE_GMV)
+		ret = fmea_recover_gmv(hr_dev, index);
+	else
+		ret = fmea_recover_others(hr_dev, res_type, index);
+	if (ret)
+		dev_err(hr_dev->dev,
+			"failed to recover %s, index = %u, ret = %d.\n",
+			fmea_ram_res[res_type].name, index, ret);
+}
+
+static void fmea_ram_ecc_work(struct work_struct *ecc_work)
+{
+	struct hns_roce_dev *hr_dev =
+		container_of(ecc_work, struct hns_roce_dev, ecc_work);
+	struct fmea_ram_ecc ecc_info = {};
+
+	if (fmea_ram_ecc_query(hr_dev, &ecc_info)) {
+		dev_err(hr_dev->dev, "failed to query fmea ram ecc.\n");
+		return;
+	}
+
+	if (!ecc_info.is_ecc_err) {
+		dev_err(hr_dev->dev, "there is no fmea ram ecc err found.\n");
+		return;
+	}
+
+	fmea_ram_ecc_recover(hr_dev, &ecc_info);
+}
+
+static irqreturn_t hns_roce_v2_msix_interrupt_abn(int irq, void *dev_id)
+{
+	struct hns_roce_dev *hr_dev = dev_id;
+	irqreturn_t int_work = IRQ_NONE;
+	u32 int_st;
+
+	int_st = roce_read(hr_dev, ROCEE_VF_ABN_INT_ST_REG);
+
+	if (int_st) {
+		int_work = abnormal_interrupt_basic(hr_dev, int_st);
+	} else if (hr_dev->pci_dev->revision >= PCI_REVISION_ID_HIP09) {
+		queue_work(hr_dev->irq_workq, &hr_dev->ecc_work);
+		int_work = IRQ_HANDLED;
+	} else {
+		dev_err(hr_dev->dev, "there is no abnormal irq found.\n");
 	}
 
 	return IRQ_RETVAL(int_work);
···
 			goto err_create_eq_fail;
 		}
 	}
+
+	INIT_WORK(&hr_dev->ecc_work, fmea_ram_ecc_work);
 
 	hr_dev->irq_workq = alloc_ordered_workqueue("hns_roce_irq_workq", 0);
 	if (!hr_dev->irq_workq) {

drivers/infiniband/hw/hns/hns_roce_hw_v2.h

···
 	HNS_ROCE_OPC_CFG_GMV_TBL			= 0x850f,
 	HNS_ROCE_OPC_CFG_GMV_BT				= 0x8510,
 	HNS_ROCE_OPC_EXT_CFG				= 0x8512,
+	HNS_ROCE_QUERY_RAM_ECC				= 0x8513,
 	HNS_SWITCH_PARAMETER_CFG			= 0x1033,
 };
 
···
 #define CFG_GMV_BT_BA_H CMQ_REQ_FIELD_LOC(51, 32)
 #define CFG_GMV_BT_IDX CMQ_REQ_FIELD_LOC(95, 64)
 
+/* Fields of HNS_ROCE_QUERY_RAM_ECC */
+#define QUERY_RAM_ECC_1BIT_ERR CMQ_REQ_FIELD_LOC(31, 0)
+#define QUERY_RAM_ECC_RES_TYPE CMQ_REQ_FIELD_LOC(63, 32)
+#define QUERY_RAM_ECC_TAG CMQ_REQ_FIELD_LOC(95, 64)
+
 struct hns_roce_cfg_sgid_tb {
 	__le32	table_idx_rsv;
 	__le32	vf_sgid_l;
···
 	struct list_head node; /* all dips are on a list */
 };
 
+struct fmea_ram_ecc {
+	u32	is_ecc_err;
+	u32	res_type;
+	u32	index;
+};
+
 /* only for RNR timeout issue of HIP08 */
 #define HNS_ROCE_CLOCK_ADJUST 1000
 #define HNS_ROCE_MAX_CQ_PERIOD 65
···
 #define HNS_ROCE_V2_ASYNC_EQE_NUM		0x1000
 
 #define HNS_ROCE_V2_VF_INT_ST_AEQ_OVERFLOW_S	0
-#define HNS_ROCE_V2_VF_INT_ST_RAS_INT_S		1
 
 #define HNS_ROCE_EQ_DB_CMD_AEQ			0x0
 #define HNS_ROCE_EQ_DB_CMD_AEQ_ARMED		0x1

drivers/infiniband/hw/irdma/cm.c

···
 	list_for_each_entry (listen_node, &cm_core->listen_list, list) {
 		memcpy(listen_addr, listen_node->loc_addr, sizeof(listen_addr));
 		listen_port = listen_node->loc_port;
+		if (listen_port != dst_port ||
+		    !(listener_state & listen_node->listener_state))
+			continue;
 		/* compare node pair, return node handle if a match */
-		if ((!memcmp(listen_addr, dst_addr, sizeof(listen_addr)) ||
-		     !memcmp(listen_addr, ip_zero, sizeof(listen_addr))) &&
-		    listen_port == dst_port &&
-		    vlan_id == listen_node->vlan_id &&
-		    (listener_state & listen_node->listener_state)) {
+		if (!memcmp(listen_addr, ip_zero, sizeof(listen_addr)) ||
+		    (!memcmp(listen_addr, dst_addr, sizeof(listen_addr)) &&
+		     vlan_id == listen_node->vlan_id)) {
 			refcount_inc(&listen_node->refcnt);
 			spin_unlock_irqrestore(&cm_core->listen_list_lock,
 					       flags);

drivers/infiniband/hw/irdma/ctrl.c

···
 
 		sd_diff = sd_needed - hmc_fpm_misc->max_sds;
 		if (sd_diff > 128) {
-			if (qpwanted > 128 && sd_diff > 144)
+			if (!(loop_count % 2) && qpwanted > 128) {
 				qpwanted /= 2;
-			mrwanted /= 2;
-			pblewanted /= 2;
+			} else {
+				mrwanted /= 2;
+				pblewanted /= 2;
+			}
 			continue;
 		}
 		if (dev->cqp->hmc_profile != IRDMA_HMC_PROFILE_FAVOR_VF &&

drivers/infiniband/hw/irdma/hw.c

···
 				iwqp->last_aeq = info->ae_id;
 			spin_unlock_irqrestore(&iwqp->lock, flags);
 			ctx_info = &iwqp->ctx_info;
-			if (rdma_protocol_roce(&iwqp->iwdev->ibdev, 1))
-				ctx_info->roce_info->err_rq_idx_valid = true;
-			else
-				ctx_info->iwarp_info->err_rq_idx_valid = true;
 		} else {
 			if (info->ae_id != IRDMA_AE_CQ_OPERATION_ERROR)
 				continue;
···
 		case IRDMA_AE_LCE_FUNCTION_CATASTROPHIC:
 		case IRDMA_AE_LCE_CQ_CATASTROPHIC:
 		case IRDMA_AE_UDA_XMIT_DGRAM_TOO_LONG:
-			if (rdma_protocol_roce(&iwdev->ibdev, 1))
-				ctx_info->roce_info->err_rq_idx_valid = false;
-			else
-				ctx_info->iwarp_info->err_rq_idx_valid = false;
-			fallthrough;
 		default:
-			ibdev_err(&iwdev->ibdev, "abnormal ae_id = 0x%x bool qp=%d qp_id = %d\n",
-				  info->ae_id, info->qp, info->qp_cq_id);
+			ibdev_err(&iwdev->ibdev, "abnormal ae_id = 0x%x bool qp=%d qp_id = %d, ae_src=%d\n",
+				  info->ae_id, info->qp, info->qp_cq_id, info->ae_src);
 			if (rdma_protocol_roce(&iwdev->ibdev, 1)) {
-				if (!info->sq && ctx_info->roce_info->err_rq_idx_valid) {
+				ctx_info->roce_info->err_rq_idx_valid = info->rq;
+				if (info->rq) {
 					ctx_info->roce_info->err_rq_idx = info->wqe_idx;
 					irdma_sc_qp_setctx_roce(&iwqp->sc_qp, iwqp->host_ctx.va,
 								ctx_info);
···
 				irdma_cm_disconn(iwqp);
 				break;
 			}
-			if (!info->sq && ctx_info->iwarp_info->err_rq_idx_valid) {
+			ctx_info->iwarp_info->err_rq_idx_valid = info->rq;
+			if (info->rq) {
 				ctx_info->iwarp_info->err_rq_idx = info->wqe_idx;
 				ctx_info->tcp_info_valid = false;
 				ctx_info->iwarp_info_valid = true;
···
 	int status;
 	u32 qpcnt;
 
-	if (rf->rdma_ver == IRDMA_GEN_1)
-		qpcnt = rsrc_limits_table[rf->limits_sel].qplimit * 2;
-	else
-		qpcnt = rsrc_limits_table[rf->limits_sel].qplimit;
+	qpcnt = rsrc_limits_table[rf->limits_sel].qplimit;
 
 	rf->sd_type = IRDMA_SD_TYPE_DIRECT;
 	status = irdma_cfg_fpm_val(&rf->sc_dev, qpcnt);
···
 			  rf->obj_mem.pa);
 	rf->obj_mem.va = NULL;
 	if (rf->rdma_ver != IRDMA_GEN_1) {
-		kfree(rf->allocated_ws_nodes);
+		bitmap_free(rf->allocated_ws_nodes);
 		rf->allocated_ws_nodes = NULL;
 	}
 	kfree(rf->ceqlist);
···
 	u32 ret;
 
 	if (rf->rdma_ver != IRDMA_GEN_1) {
-		rf->allocated_ws_nodes =
-			kcalloc(BITS_TO_LONGS(IRDMA_MAX_WS_NODES),
-				sizeof(unsigned long), GFP_KERNEL);
+		rf->allocated_ws_nodes = bitmap_zalloc(IRDMA_MAX_WS_NODES,
+						       GFP_KERNEL);
 		if (!rf->allocated_ws_nodes)
 			return -ENOMEM;
 
···
 	return 0;
 
 mem_rsrc_kzalloc_fail:
-	kfree(rf->allocated_ws_nodes);
+	bitmap_free(rf->allocated_ws_nodes);
 	rf->allocated_ws_nodes = NULL;
 
 	return ret;

drivers/infiniband/hw/irdma/main.h

···
 #define	IRDMA_NO_QSET	0xffff
 
 #define IW_CFG_FPM_QP_COUNT		32768
-#define IRDMA_MAX_PAGES_PER_FMR		512
+#define IRDMA_MAX_PAGES_PER_FMR		262144
 #define IRDMA_MIN_PAGES_PER_FMR		1
 #define IRDMA_CQP_COMPL_RQ_WQE_FLUSHED	2
 #define IRDMA_CQP_COMPL_SQ_WQE_FLUSHED	3

drivers/infiniband/hw/irdma/utils.c

···
 };
 
 static const struct irdma_cqp_err_info irdma_noncrit_err_list[] = {
+	{0xffff, 0x8002, "Invalid State"},
 	{0xffff, 0x8006, "Flush No Wqe Pending"},
 	{0xffff, 0x8007, "Modify QP Bad Close"},
 	{0xffff, 0x8009, "LLP Closed"},

drivers/infiniband/hw/irdma/verbs.c

···
 	spin_unlock_irqrestore(&iwcq->lock, flags);
 
 	irdma_cq_wq_destroy(iwdev->rf, cq);
-	irdma_cq_free_rsrc(iwdev->rf, iwcq);
 
 	spin_lock_irqsave(&iwceq->ce_lock, flags);
 	irdma_sc_cleanup_ceqes(cq, ceq);
 	spin_unlock_irqrestore(&iwceq->ce_lock, flags);
+	irdma_cq_free_rsrc(iwdev->rf, iwcq);
 
 	return 0;
 }
···
 	palloc = &iwpbl->pble_alloc;
 	iwmr->page_cnt = max_num_sg;
 	err_code = irdma_get_pble(iwdev->rf->pble_rsrc, palloc, iwmr->page_cnt,
-				  true);
+				  false);
 	if (err_code)
 		goto err_get_pble;
 
···
 	if (unlikely(iwmr->npages == iwmr->page_cnt))
 		return -ENOMEM;
 
-	pbl = palloc->level1.addr;
-	pbl[iwmr->npages++] = addr;
+	if (palloc->level == PBLE_LEVEL_2) {
+		struct irdma_pble_info *palloc_info =
+			palloc->level2.leaf + (iwmr->npages >> PBLE_512_SHIFT);
+
+		palloc_info->addr[iwmr->npages & (PBLE_PER_PAGE - 1)] = addr;
+	} else {
+		pbl = palloc->level1.addr;
+		pbl[iwmr->npages] = addr;
+	}
+	iwmr->npages++;
 
 	return 0;
 }

drivers/infiniband/hw/mlx5/cq.c

···
 			    "Requestor" : "Responder", cq->mcq.cqn);
 		mlx5_ib_dbg(dev, "syndrome 0x%x, vendor syndrome 0x%x\n",
 			    err_cqe->syndrome, err_cqe->vendor_err_synd);
+		if (wc->status != IB_WC_WR_FLUSH_ERR &&
+		    (*cur_qp)->type == MLX5_IB_QPT_REG_UMR)
+			dev->umrc.state = MLX5_UMR_STATE_RECOVER;
+
 		if (opcode == MLX5_CQE_REQ_ERR) {
 			wq = &(*cur_qp)->sq;
 			wqe_ctr = be16_to_cpu(cqe64->wqe_counter);

drivers/infiniband/hw/mlx5/fs.c

···
 #define MLX5_FS_MAX_TYPES	 6
 #define MLX5_FS_MAX_ENTRIES	 BIT(16)
 
-static struct mlx5_ib_flow_prio *_get_prio(struct mlx5_flow_namespace *ns,
+static bool mlx5_ib_shared_ft_allowed(struct ib_device *device)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+
+	return MLX5_CAP_GEN(dev->mdev, shared_object_to_user_object_allowed);
+}
+
+static struct mlx5_ib_flow_prio *_get_prio(struct mlx5_ib_dev *dev,
+					   struct mlx5_flow_namespace *ns,
 					   struct mlx5_ib_flow_prio *prio,
 					   int priority,
 					   int num_entries, int num_groups,
···
 	struct mlx5_flow_table_attr ft_attr = {};
 	struct mlx5_flow_table *ft;
 
+	if (mlx5_ib_shared_ft_allowed(&dev->ib_dev))
+		ft_attr.uid = MLX5_SHARED_RESOURCE_UID;
 	ft_attr.prio = priority;
 	ft_attr.max_fte = num_entries;
 	ft_attr.flags = flags;
···
 
 	ft = prio->flow_table;
 	if (!ft)
-		return _get_prio(ns, prio, priority, max_table_size, num_groups,
-				 flags);
+		return _get_prio(dev, ns, prio, priority, max_table_size,
+				 num_groups, flags);
 
 	return prio;
 }
···
 
 	prio = &dev->flow_db->opfcs[type];
 	if (!prio->flow_table) {
-		prio = _get_prio(ns, prio, priority,
+		prio = _get_prio(dev, ns, prio, priority,
 				 dev->num_ports * MAX_OPFC_RULES, 1, 0);
 		if (IS_ERR(prio)) {
 			err = PTR_ERR(prio);
···
 }
 
 static struct mlx5_ib_flow_prio *
-_get_flow_table(struct mlx5_ib_dev *dev,
-		struct mlx5_ib_flow_matcher *fs_matcher,
+_get_flow_table(struct mlx5_ib_dev *dev, u16 user_priority,
+		enum mlx5_flow_namespace_type ns_type,
 		bool mcast)
 {
 	struct mlx5_flow_namespace *ns = NULL;
···
 	if (mcast)
 		priority = MLX5_IB_FLOW_MCAST_PRIO;
 	else
-		priority = ib_prio_to_core_prio(fs_matcher->priority, false);
+		priority = ib_prio_to_core_prio(user_priority, false);
 
 	esw_encap = mlx5_eswitch_get_encap_mode(dev->mdev) !=
 		DEVLINK_ESWITCH_ENCAP_MODE_NONE;
-	switch (fs_matcher->ns_type) {
+	switch (ns_type) {
 	case MLX5_FLOW_NAMESPACE_BYPASS:
 		max_table_size = BIT(
 			MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev, log_max_ft_size));
···
 					       reformat_l3_tunnel_to_l2) &&
 		    esw_encap)
 			flags |= MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT;
-		priority = fs_matcher->priority;
+		priority = user_priority;
 		break;
 	case MLX5_FLOW_NAMESPACE_RDMA_RX:
 		max_table_size = BIT(
 			MLX5_CAP_FLOWTABLE_RDMA_RX(dev->mdev, log_max_ft_size));
-		priority = fs_matcher->priority;
+		priority = user_priority;
 		break;
 	case MLX5_FLOW_NAMESPACE_RDMA_TX:
 		max_table_size = BIT(
 			MLX5_CAP_FLOWTABLE_RDMA_TX(dev->mdev, log_max_ft_size));
-		priority = fs_matcher->priority;
+		priority = user_priority;
 		break;
 	default:
 		break;
···
 
 	max_table_size = min_t(int, max_table_size, MLX5_FS_MAX_ENTRIES);
 
-	ns = mlx5_get_flow_namespace(dev->mdev, fs_matcher->ns_type);
+	ns = mlx5_get_flow_namespace(dev->mdev, ns_type);
 	if (!ns)
 		return ERR_PTR(-EOPNOTSUPP);
 
-	switch (fs_matcher->ns_type) {
+	switch (ns_type) {
 	case MLX5_FLOW_NAMESPACE_BYPASS:
 		prio = &dev->flow_db->prios[priority];
 		break;
···
 	if (prio->flow_table)
 		return prio;
 
-	return _get_prio(ns, prio, priority, max_table_size,
+	return _get_prio(dev, ns, prio, priority, max_table_size,
 			 MLX5_FS_MAX_TYPES, flags);
 }
 
···
 	mcast = raw_fs_is_multicast(fs_matcher, cmd_in);
 	mutex_lock(&dev->flow_db->lock);
 
-	ft_prio = _get_flow_table(dev, fs_matcher, mcast);
+	ft_prio = _get_flow_table(dev, fs_matcher->priority,
+				  fs_matcher->ns_type, mcast);
 	if (IS_ERR(ft_prio)) {
 		err = PTR_ERR(ft_prio);
 		goto unlock;
···
 	return 0;
 }
 
+static int steering_anchor_cleanup(struct ib_uobject *uobject,
+				   enum rdma_remove_reason why,
+				   struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_steering_anchor *obj = uobject->object;
+
+	if (atomic_read(&obj->usecnt))
+		return -EBUSY;
+
+	mutex_lock(&obj->dev->flow_db->lock);
+	put_flow_table(obj->dev, obj->ft_prio, true);
+	mutex_unlock(&obj->dev->flow_db->lock);
+
+	kfree(obj);
+	return 0;
+}
+
 static int mlx5_ib_matcher_ns(struct uverbs_attr_bundle *attrs,
 			      struct mlx5_ib_flow_matcher *obj)
 {
···
 		if (err)
 			return err;
 
-		if (flags) {
-			mlx5_ib_ft_type_to_namespace(
+		if (flags)
+			return mlx5_ib_ft_type_to_namespace(
 				MLX5_IB_UAPI_FLOW_TABLE_TYPE_NIC_TX,
 				&obj->ns_type);
-			return 0;
-		}
 	}
 
 	obj->ns_type = MLX5_FLOW_NAMESPACE_BYPASS;
···
 
 end:
 	kfree(obj);
+	return err;
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_STEERING_ANCHOR_CREATE)(
+	struct uverbs_attr_bundle *attrs)
+{
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(
+		attrs, MLX5_IB_ATTR_STEERING_ANCHOR_CREATE_HANDLE);
+	struct mlx5_ib_dev *dev = mlx5_udata_to_mdev(&attrs->driver_udata);
+	enum mlx5_ib_uapi_flow_table_type ib_uapi_ft_type;
+	enum mlx5_flow_namespace_type ns_type;
+	struct mlx5_ib_steering_anchor *obj;
+	struct mlx5_ib_flow_prio *ft_prio;
+	u16 priority;
+	u32 ft_id;
+	int err;
+
+	if (!capable(CAP_NET_RAW))
+		return -EPERM;
+
+	err = uverbs_get_const(&ib_uapi_ft_type, attrs,
+			       MLX5_IB_ATTR_STEERING_ANCHOR_FT_TYPE);
+	if (err)
+		return err;
+
+	err = mlx5_ib_ft_type_to_namespace(ib_uapi_ft_type, &ns_type);
+	if (err)
+		return err;
+
+	err = uverbs_copy_from(&priority, attrs,
+			       MLX5_IB_ATTR_STEERING_ANCHOR_PRIORITY);
+	if (err)
+		return err;
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return -ENOMEM;
+
+	mutex_lock(&dev->flow_db->lock);
+	ft_prio = _get_flow_table(dev, priority, ns_type, 0);
+	if (IS_ERR(ft_prio)) {
+		mutex_unlock(&dev->flow_db->lock);
+		err = PTR_ERR(ft_prio);
+		goto free_obj;
+	}
+
+	ft_prio->refcount++;
+	ft_id = mlx5_flow_table_id(ft_prio->flow_table);
+	mutex_unlock(&dev->flow_db->lock);
+
+	err = uverbs_copy_to(attrs, MLX5_IB_ATTR_STEERING_ANCHOR_FT_ID,
+			     &ft_id, sizeof(ft_id));
+	if (err)
+		goto put_flow_table;
+
+	uobj->object = obj;
+	obj->dev = dev;
+	obj->ft_prio = ft_prio;
+	atomic_set(&obj->usecnt, 0);
+
+	return 0;
+
+put_flow_table:
+	mutex_lock(&dev->flow_db->lock);
+	put_flow_table(dev, ft_prio, true);
+	mutex_unlock(&dev->flow_db->lock);
+free_obj:
+	kfree(obj);
+
 	return err;
 }
 
···
 			    &UVERBS_METHOD(MLX5_IB_METHOD_FLOW_MATCHER_CREATE),
 			    &UVERBS_METHOD(MLX5_IB_METHOD_FLOW_MATCHER_DESTROY));
 
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_STEERING_ANCHOR_CREATE,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_STEERING_ANCHOR_CREATE_HANDLE,
+			MLX5_IB_OBJECT_STEERING_ANCHOR,
+			UVERBS_ACCESS_NEW,
+			UA_MANDATORY),
+	UVERBS_ATTR_CONST_IN(MLX5_IB_ATTR_STEERING_ANCHOR_FT_TYPE,
+			     enum mlx5_ib_uapi_flow_table_type,
+			     UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_STEERING_ANCHOR_PRIORITY,
+			   UVERBS_ATTR_TYPE(u16),
+			   UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_STEERING_ANCHOR_FT_ID,
+			   UVERBS_ATTR_TYPE(u32),
+			   UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD_DESTROY(
+	MLX5_IB_METHOD_STEERING_ANCHOR_DESTROY,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_STEERING_ANCHOR_DESTROY_HANDLE,
+			MLX5_IB_OBJECT_STEERING_ANCHOR,
+			UVERBS_ACCESS_DESTROY,
+			UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_OBJECT(
+	MLX5_IB_OBJECT_STEERING_ANCHOR,
+	UVERBS_TYPE_ALLOC_IDR(steering_anchor_cleanup),
+	&UVERBS_METHOD(MLX5_IB_METHOD_STEERING_ANCHOR_CREATE),
+	&UVERBS_METHOD(MLX5_IB_METHOD_STEERING_ANCHOR_DESTROY));
+
 const struct uapi_definition mlx5_ib_flow_defs[] = {
 	UAPI_DEF_CHAIN_OBJ_TREE_NAMED(
 		MLX5_IB_OBJECT_FLOW_MATCHER),
···
 		&mlx5_ib_fs),
 	UAPI_DEF_CHAIN_OBJ_TREE(UVERBS_OBJECT_FLOW_ACTION,
 				&mlx5_ib_flow_actions),
+	UAPI_DEF_CHAIN_OBJ_TREE_NAMED(
+		MLX5_IB_OBJECT_STEERING_ANCHOR,
+		UAPI_DEF_IS_OBJ_SUPPORTED(mlx5_ib_shared_ft_allowed)),
 	{},
 };
 

drivers/infiniband/hw/mlx5/main.c

···
 {
 	int err;
 
-	err = mlx5_mr_cache_cleanup(dev);
+	err = mlx5_mkey_cache_cleanup(dev);
 	if (err)
 		mlx5_ib_warn(dev, "mr cache cleanup failed\n");
 
···
 	if (ret)
 		return ret;
 
-	ret = mlx5_mr_cache_init(dev);
+	ret = mlx5_mkey_cache_init(dev);
 	if (ret) {
 		mlx5_ib_warn(dev, "mr cache init failed %d\n", ret);
 		mlx5r_umr_resource_cleanup(dev);

drivers/infiniband/hw/mlx5/mlx5_ib.h

···
 	u8			match_criteria_enable;
 };
 
+struct mlx5_ib_steering_anchor {
+	struct mlx5_ib_flow_prio *ft_prio;
+	struct mlx5_ib_dev *dev;
+	atomic_t usecnt;
+};
+
 struct mlx5_ib_pp {
 	u16 index;
 	struct mlx5_core_dev *mdev;
···
 	unsigned int ndescs;
 	struct wait_queue_head wait;
 	refcount_t usecount;
+	struct mlx5_cache_ent *cache_ent;
 };
 
 #define MLX5_IB_MTT_PRESENT (MLX5_IB_MTT_READ | MLX5_IB_MTT_WRITE)
···
 	struct ib_mr ibmr;
 	struct mlx5_ib_mkey mmkey;
 
-	/* User MR data */
-	struct mlx5_cache_ent *cache_ent;
-	/* Everything after cache_ent is zero'd when MR allocated */
 	struct ib_umem *umem;
 
 	union {
-		/* Used only while the MR is in the cache */
-		struct {
-			u32 out[MLX5_ST_SZ_DW(create_mkey_out)];
-			struct mlx5_async_work cb_work;
-			/* Cache list element */
-			struct list_head list;
-		};
-
 		/* Used only by kernel MRs (umem == NULL) */
 		struct {
 			void *descs;
···
 	};
 };
 
-/* Zero the fields in the mr that are variant depending on usage */
-static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr)
-{
-	memset_after(mr, 0, cache_ent);
-}
-
 static inline bool is_odp_mr(struct mlx5_ib_mr *mr)
 {
 	return IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING) && mr->umem &&
···
 	struct completion	done;
 };
 
+enum {
+	MLX5_UMR_STATE_ACTIVE,
+	MLX5_UMR_STATE_RECOVER,
+	MLX5_UMR_STATE_ERR,
+};
+
 struct umr_common {
 	struct ib_pd	*pd;
 	struct ib_cq	*cq;
 	struct ib_qp	*qp;
-	/* control access to UMR QP
+	/* Protects from UMR QP overflow
 	 */
 	struct semaphore	sem;
+	/* Protects from using UMR while the UMR is not active
+	 */
+	struct mutex lock;
+	unsigned int state;
 };
 
 struct mlx5_cache_ent {
-	struct list_head	head;
-	/* sync access to the cahce entry
-	 */
-	spinlock_t		lock;
-
+	struct xarray		mkeys;
+	unsigned long		stored;
+	unsigned long		reserved;
 
 	char                    name[4];
 	u32                     order;
···
 	u8 fill_to_high_water:1;
 
 	/*
-	 * - available_mrs is the length of list head, ie the number of MRs
-	 *   available for immediate allocation.
-	 * - total_mrs is available_mrs plus all in use MRs that could be
-	 *   returned to the cache.
-	 * - limit is the low water mark for available_mrs, 2* limit is the
+	 * - limit is the low water mark for stored mkeys, 2* limit is the
 	 *   upper water mark.
-	 * - pending is the number of MRs currently being created
 	 */
-	u32 total_mrs;
-	u32 available_mrs;
+	u32 in_use;
 	u32 limit;
-	u32 pending;
 
 	/* Statistics */
 	u32                     miss;
···
 	struct delayed_work	dwork;
 };
 
-struct mlx5_mr_cache {
+struct mlx5r_async_create_mkey {
+	union {
+		u32 in[MLX5_ST_SZ_BYTES(create_mkey_in)];
+		u32 out[MLX5_ST_SZ_DW(create_mkey_out)];
+	};
+	struct mlx5_async_work cb_work;
+	struct mlx5_cache_ent *ent;
+	u32 mkey;
+};
+
+struct mlx5_mkey_cache {
 	struct workqueue_struct *wq;
-	struct mlx5_cache_ent	ent[MAX_MR_CACHE_ENTRIES];
+	struct mlx5_cache_ent	ent[MAX_MKEY_CACHE_ENTRIES];
 	struct dentry		*root;
 	unsigned long		last_add;
 };
···
 	struct mlx5_ib_resources	devr;
 
 	atomic_t			mkey_var;
-	struct mlx5_mr_cache		cache;
+	struct mlx5_mkey_cache		cache;
 	struct timer_list		delay_timer;
 	/* Prevents soft lock on massive reg MRs */
 	struct mutex			slow_path_mutex;
···
 			  u64 access_flags);
 void mlx5_ib_copy_pas(u64 *old, u64 *new, int step, int num);
 int mlx5_ib_get_cqe_size(struct ib_cq *ibcq);
-int mlx5_mr_cache_init(struct mlx5_ib_dev *dev);
-int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev);
+int mlx5_mkey_cache_init(struct mlx5_ib_dev *dev);
+int mlx5_mkey_cache_cleanup(struct mlx5_ib_dev *dev);
 
 struct mlx5_ib_mr *mlx5_mr_cache_alloc(struct mlx5_ib_dev *dev,
 				       struct mlx5_cache_ent *ent,
···
 void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *ibdev);
 int __init mlx5_ib_odp_init(void);
 void mlx5_ib_odp_cleanup(void);
-void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent);
+void mlx5_odp_init_mkey_cache_entry(struct mlx5_cache_ent *ent);
 void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries,
 			   struct mlx5_ib_mr *mr, int flags);
 
···
 static inline void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *ibdev) {}
 static inline int mlx5_ib_odp_init(void) { return 0; }
 static inline void mlx5_ib_odp_cleanup(void)				    {}
-static inline void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent) {}
+static inline void mlx5_odp_init_mkey_cache_entry(struct mlx5_cache_ent *ent) {}
 static inline void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries,
 					 struct mlx5_ib_mr *mr, int flags) {}
 

drivers/infiniband/hw/mlx5/mr.c

···
 	MLX5_SET64(mkc, mkc, start_addr, start_addr);
 }
 
-static void assign_mkey_variant(struct mlx5_ib_dev *dev,
-				struct mlx5_ib_mkey *mkey, u32 *in)
+static void assign_mkey_variant(struct mlx5_ib_dev *dev, u32 *mkey, u32 *in)
 {
 	u8 key = atomic_inc_return(&dev->mkey_var);
 	void *mkc;
 
 	mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry);
 	MLX5_SET(mkc, mkc, mkey_7_0, key);
-	mkey->key = key;
+	*mkey = key;
 }
 
 static int mlx5_ib_create_mkey(struct mlx5_ib_dev *dev,
···
 {
 	int ret;
 
-	assign_mkey_variant(dev, mkey, in);
+	assign_mkey_variant(dev, &mkey->key, in);
 	ret = mlx5_core_create_mkey(dev->mdev, &mkey->key, in, inlen);
 	if (!ret)
 		init_waitqueue_head(&mkey->wait);
···
 	return ret;
 }
 
-static int
-mlx5_ib_create_mkey_cb(struct mlx5_ib_dev *dev,
-		       struct mlx5_ib_mkey *mkey,
-		       struct mlx5_async_ctx *async_ctx,
-		       u32 *in, int inlen, u32 *out, int outlen,
-		       struct mlx5_async_work *context)
+static int mlx5_ib_create_mkey_cb(struct mlx5r_async_create_mkey *async_create)
 {
-	MLX5_SET(create_mkey_in, in, opcode, MLX5_CMD_OP_CREATE_MKEY);
-	assign_mkey_variant(dev, mkey, in);
-	return mlx5_cmd_exec_cb(async_ctx, in, inlen, out, outlen,
-				create_mkey_callback, context);
+	struct mlx5_ib_dev *dev = async_create->ent->dev;
+	size_t inlen = MLX5_ST_SZ_BYTES(create_mkey_in);
+	size_t outlen = MLX5_ST_SZ_BYTES(create_mkey_out);
+
+	MLX5_SET(create_mkey_in, async_create->in, opcode,
+		 MLX5_CMD_OP_CREATE_MKEY);
+	assign_mkey_variant(dev, &async_create->mkey, async_create->in);
+	return mlx5_cmd_exec_cb(&dev->async_ctx, async_create->in, inlen,
+				async_create->out, outlen, create_mkey_callback,
+				&async_create->cb_work);
 }
 
-static int mr_cache_max_order(struct mlx5_ib_dev *dev);
+static int mkey_cache_max_order(struct mlx5_ib_dev *dev);
 static void queue_adjust_cache_locked(struct mlx5_cache_ent *ent);
 
 static int destroy_mkey(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
···
 	mlx5_cmd_out_err(dev->mdev, MLX5_CMD_OP_CREATE_MKEY, 0, out);
 }
 
+
+static int push_mkey(struct mlx5_cache_ent *ent, bool limit_pendings,
+		     void *to_store)
+{
+	XA_STATE(xas, &ent->mkeys, 0);
+	void *curr;
+
+	xa_lock_irq(&ent->mkeys);
+	if (limit_pendings &&
+	    (ent->reserved - ent->stored) > MAX_PENDING_REG_MR) {
+		xa_unlock_irq(&ent->mkeys);
+		return -EAGAIN;
+	}
+	while (1) {
+		/*
+		 * This is cmpxchg (NULL, XA_ZERO_ENTRY) however this version
+		 * doesn't transparently unlock. Instead we set the xas index to
+		 * the current value of reserved every iteration.
+		 */
+		xas_set(&xas, ent->reserved);
+		curr = xas_load(&xas);
+		if (!curr) {
+			if (to_store && ent->stored == ent->reserved)
+				xas_store(&xas, to_store);
+			else
+				xas_store(&xas, XA_ZERO_ENTRY);
+			if (xas_valid(&xas)) {
+				ent->reserved++;
+				if (to_store) {
+					if (ent->stored != ent->reserved)
+						__xa_store(&ent->mkeys,
+							   ent->stored,
+							   to_store,
+							   GFP_KERNEL);
+					ent->stored++;
+					queue_adjust_cache_locked(ent);
+					WRITE_ONCE(ent->dev->cache.last_add,
+						   jiffies);
+				}
+			}
+		}
+		xa_unlock_irq(&ent->mkeys);
+
+		/*
+		 * Notice xas_nomem() must always be called as it cleans
+		 * up any cached allocation.
+		 */
+		if (!xas_nomem(&xas, GFP_KERNEL))
+			break;
+		xa_lock_irq(&ent->mkeys);
+	}
+	if (xas_error(&xas))
+		return xas_error(&xas);
+	if (WARN_ON(curr))
+		return -EINVAL;
+	return 0;
+}
+
+static void undo_push_reserve_mkey(struct mlx5_cache_ent *ent)
+{
+	void *old;
+
+	ent->reserved--;
+	old = __xa_erase(&ent->mkeys, ent->reserved);
+	WARN_ON(old);
+}
+
+static void push_to_reserved(struct mlx5_cache_ent *ent, u32 mkey)
+{
+	void *old;
+
+	old = __xa_store(&ent->mkeys, ent->stored, xa_mk_value(mkey), 0);
+	WARN_ON(old);
+	ent->stored++;
+}
+
+static u32 pop_stored_mkey(struct mlx5_cache_ent *ent)
+{
+	void *old, *xa_mkey;
+
+	ent->stored--;
+	ent->reserved--;
+
+	if (ent->stored == ent->reserved) {
+		xa_mkey = __xa_erase(&ent->mkeys, ent->stored);
+		WARN_ON(!xa_mkey);
+		return (u32)xa_to_value(xa_mkey);
+	}
+
+	xa_mkey = __xa_store(&ent->mkeys, ent->stored, XA_ZERO_ENTRY,
+			     GFP_KERNEL);
+	WARN_ON(!xa_mkey || xa_is_err(xa_mkey));
+	old = __xa_erase(&ent->mkeys, ent->reserved);
+	WARN_ON(old);
+	return (u32)xa_to_value(xa_mkey);
+}
+
 static void create_mkey_callback(int status, struct mlx5_async_work *context)
 {
-	struct mlx5_ib_mr *mr =
-		container_of(context, struct mlx5_ib_mr, cb_work);
-	struct mlx5_cache_ent *ent = mr->cache_ent;
+	struct mlx5r_async_create_mkey *mkey_out =
+		container_of(context, struct mlx5r_async_create_mkey, cb_work);
+	struct mlx5_cache_ent *ent = mkey_out->ent;
 	struct mlx5_ib_dev *dev = ent->dev;
 	unsigned long flags;
 
 	if (status) {
-		create_mkey_warn(dev, status, mr->out);
-		kfree(mr);
-		spin_lock_irqsave(&ent->lock, flags);
-		ent->pending--;
+		create_mkey_warn(dev, status, mkey_out->out);
+		kfree(mkey_out);
+		xa_lock_irqsave(&ent->mkeys, flags);
+		undo_push_reserve_mkey(ent);
 		WRITE_ONCE(dev->fill_delay, 1);
-		spin_unlock_irqrestore(&ent->lock, flags);
+		xa_unlock_irqrestore(&ent->mkeys, flags);
 		mod_timer(&dev->delay_timer, jiffies + HZ);
 		return;
 	}
 
-	mr->mmkey.type = MLX5_MKEY_MR;
-	mr->mmkey.key |= mlx5_idx_to_mkey(
-		MLX5_GET(create_mkey_out, mr->out, mkey_index));
-	init_waitqueue_head(&mr->mmkey.wait);
-
+	mkey_out->mkey |= mlx5_idx_to_mkey(
+		MLX5_GET(create_mkey_out, mkey_out->out, mkey_index));
 	WRITE_ONCE(dev->cache.last_add, jiffies);
 
-	spin_lock_irqsave(&ent->lock, flags);
-	list_add_tail(&mr->list, &ent->head);
-	ent->available_mrs++;
-	ent->total_mrs++;
+	xa_lock_irqsave(&ent->mkeys, flags);
+	push_to_reserved(ent, mkey_out->mkey);
 	/* If we are doing fill_to_high_water then keep going. */
 	queue_adjust_cache_locked(ent);
-	ent->pending--;
-	spin_unlock_irqrestore(&ent->lock, flags);
+	xa_unlock_irqrestore(&ent->mkeys, flags);
+	kfree(mkey_out);
 }
 
 static int get_mkc_octo_size(unsigned int access_mode, unsigned int ndescs)
···
 	return ret;
 }
 
-static struct mlx5_ib_mr *alloc_cache_mr(struct mlx5_cache_ent *ent, void *mkc)
+static void set_cache_mkc(struct mlx5_cache_ent *ent, void *mkc)
 {
-	struct mlx5_ib_mr *mr;
-
-	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
-	if (!mr)
-		return NULL;
-	mr->cache_ent = ent;
-
 	set_mkc_access_pd_addr_fields(mkc, 0, 0, ent->dev->umrc.pd);
 	MLX5_SET(mkc, mkc, free, 1);
 	MLX5_SET(mkc, mkc, umr_en, 1);
···
 	MLX5_SET(mkc, mkc, translations_octword_size,
 		 get_mkc_octo_size(ent->access_mode, ent->ndescs));
 	MLX5_SET(mkc, mkc, log_page_size, ent->page);
-	return mr;
 }
 
 /* Asynchronously schedule new MRs to be populated in the cache. */
 static int add_keys(struct mlx5_cache_ent *ent, unsigned int num)
 {
-	size_t inlen = MLX5_ST_SZ_BYTES(create_mkey_in);
-	struct mlx5_ib_mr *mr;
+	struct mlx5r_async_create_mkey *async_create;
 	void *mkc;
-	u32 *in;
 	int err = 0;
 	int i;
 
-	in = kzalloc(inlen, GFP_KERNEL);
-	if (!in)
-		return -ENOMEM;
-
-	mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry);
 	for (i = 0; i < num; i++) {
-		mr = alloc_cache_mr(ent, mkc);
-		if (!mr) {
-			err = -ENOMEM;
-			break;
-		}
-		spin_lock_irq(&ent->lock);
-		if (ent->pending >= MAX_PENDING_REG_MR) {
-			err = -EAGAIN;
-			spin_unlock_irq(&ent->lock);
-			kfree(mr);
-			break;
-		}
-		ent->pending++;
-		spin_unlock_irq(&ent->lock);
-		err = mlx5_ib_create_mkey_cb(ent->dev, &mr->mmkey,
-					     &ent->dev->async_ctx, in, inlen,
-					     mr->out, sizeof(mr->out),
-					     &mr->cb_work);
+		async_create = kzalloc(sizeof(struct mlx5r_async_create_mkey),
+				       GFP_KERNEL);
+		if (!async_create)
+			return -ENOMEM;
+		mkc = MLX5_ADDR_OF(create_mkey_in, async_create->in,
+				   memory_key_mkey_entry);
+		set_cache_mkc(ent, mkc);
+		async_create->ent = ent;
+
+		err = push_mkey(ent, true, NULL);
+		if (err)
+			goto free_async_create;
+
+		err = mlx5_ib_create_mkey_cb(async_create);
 		if (err) {
-			spin_lock_irq(&ent->lock);
-			ent->pending--;
-			spin_unlock_irq(&ent->lock);
 			mlx5_ib_warn(ent->dev, "create mkey failed %d\n", err);
-			kfree(mr);
-			break;
+			goto err_undo_reserve;
 		}
 	}
 
-	kfree(in);
+	return 0;
+
+err_undo_reserve:
+	xa_lock_irq(&ent->mkeys);
+	undo_push_reserve_mkey(ent);
+	xa_unlock_irq(&ent->mkeys);
+free_async_create:
+	kfree(async_create);
 	return err;
 }
 
 /* Synchronously create a MR in the cache */
-static struct mlx5_ib_mr *create_cache_mr(struct mlx5_cache_ent *ent)
+static int create_cache_mkey(struct mlx5_cache_ent *ent, u32 *mkey)
 {
 	size_t inlen = MLX5_ST_SZ_BYTES(create_mkey_in);
-	struct mlx5_ib_mr *mr;
 	void *mkc;
 	u32 *in;
 	int err;
 
 	in = kzalloc(inlen, GFP_KERNEL);
 	if (!in)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 	mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry);
+	set_cache_mkc(ent, mkc);
 
-	mr = alloc_cache_mr(ent, mkc);
-	if (!mr) {
-		err = -ENOMEM;
-		goto free_in;
-	}
-
-	err = mlx5_core_create_mkey(ent->dev->mdev, &mr->mmkey.key, in, inlen);
+	err = mlx5_core_create_mkey(ent->dev->mdev, mkey, in, inlen);
 	if (err)
-		goto free_mr;
+		goto free_in;
 
-	init_waitqueue_head(&mr->mmkey.wait);
-	mr->mmkey.type = MLX5_MKEY_MR;
 	WRITE_ONCE(ent->dev->cache.last_add, jiffies);
-	spin_lock_irq(&ent->lock);
-	ent->total_mrs++;
-	spin_unlock_irq(&ent->lock);
-	kfree(in);
-	return mr;
-free_mr:
-	kfree(mr);
 free_in:
 	kfree(in);
-	return ERR_PTR(err);
+	return err;
 }
 
 static void remove_cache_mr_locked(struct mlx5_cache_ent *ent)
 {
-	struct mlx5_ib_mr *mr;
+	u32 mkey;
 
-	lockdep_assert_held(&ent->lock);
-	if (list_empty(&ent->head))
+	lockdep_assert_held(&ent->mkeys.xa_lock);
+	if (!ent->stored)
 		return;
-	mr = list_first_entry(&ent->head, struct mlx5_ib_mr, list);
-	list_del(&mr->list);
-	ent->available_mrs--;
-	ent->total_mrs--;
-	spin_unlock_irq(&ent->lock);
-	mlx5_core_destroy_mkey(ent->dev->mdev, mr->mmkey.key);
-	kfree(mr);
-	spin_lock_irq(&ent->lock);
+	mkey = pop_stored_mkey(ent);
+	xa_unlock_irq(&ent->mkeys);
+	mlx5_core_destroy_mkey(ent->dev->mdev, mkey);
+	xa_lock_irq(&ent->mkeys);
 }
 
 static int resize_available_mrs(struct mlx5_cache_ent *ent, unsigned int target,
 				bool limit_fill)
+	 __acquires(&ent->mkeys) __releases(&ent->mkeys)
 {
 	int err;
 
-	lockdep_assert_held(&ent->lock);
+	lockdep_assert_held(&ent->mkeys.xa_lock);
 
 	while (true) {
 		if (limit_fill)
 			target = ent->limit * 2;
-		if (target == ent->available_mrs + ent->pending)
+		if (target == ent->reserved)
 			return 0;
-		if (target > ent->available_mrs + ent->pending) {
-			u32 todo = target - (ent->available_mrs + ent->pending);
+		if (target > ent->reserved) {
+			u32 todo = target - ent->reserved;
 
-			spin_unlock_irq(&ent->lock);
+			xa_unlock_irq(&ent->mkeys);
 			err = add_keys(ent, todo);
 			if (err == -EAGAIN)
 				usleep_range(3000, 5000);
-			spin_lock_irq(&ent->lock);
+			xa_lock_irq(&ent->mkeys);
 			if (err) {
 				if (err != -EAGAIN)
 					return err;
···
 
 	/*
 	 * Target is the new value of total_mrs the user requests, however we
-	 * cannot free MRs that are in use. Compute the target value for
-	 * available_mrs.
+	 * cannot free MRs that are in use. Compute the target value for stored
+	 * mkeys.
 	 */
-	spin_lock_irq(&ent->lock);
-	if (target < ent->total_mrs - ent->available_mrs) {
+	xa_lock_irq(&ent->mkeys);
+	if (target < ent->in_use) {
 		err = -EINVAL;
 		goto err_unlock;
 	}
-	target = target - (ent->total_mrs - ent->available_mrs);
+	target = target - ent->in_use;
 	if (target < ent->limit || target > ent->limit*2) {
 		err = -EINVAL;
 		goto err_unlock;
···
 	err = resize_available_mrs(ent, target, false);
 	if (err)
 		goto err_unlock;
-	spin_unlock_irq(&ent->lock);
+	xa_unlock_irq(&ent->mkeys);
 
 	return count;
 
 err_unlock:
-	spin_unlock_irq(&ent->lock);
+	xa_unlock_irq(&ent->mkeys);
 	return err;
 }
 
···
 	char lbuf[20];
 	int err;
 
-	err = snprintf(lbuf, sizeof(lbuf), "%d\n", ent->total_mrs);
+	err = snprintf(lbuf, sizeof(lbuf), "%ld\n", ent->stored + ent->in_use);
 	if (err < 0)
 		return err;
 
···
 	 * Upon set we immediately fill the cache to high water mark implied by
 	 * the limit.
 	 */
-	spin_lock_irq(&ent->lock);
+	xa_lock_irq(&ent->mkeys);
 	ent->limit = var;
 	err = resize_available_mrs(ent, 0, true);
-	spin_unlock_irq(&ent->lock);
+	xa_unlock_irq(&ent->mkeys);
 	if (err)
 		return err;
 	return count;
···
 	.read	= limit_read,
 };
 
-static bool someone_adding(struct mlx5_mr_cache *cache)
+static bool someone_adding(struct mlx5_mkey_cache *cache)
 {
 	unsigned int i;
 
-	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) {
+	for (i = 0; i < MAX_MKEY_CACHE_ENTRIES; i++) {
 		struct mlx5_cache_ent *ent = &cache->ent[i];
 		bool ret;
 
-		spin_lock_irq(&ent->lock);
-		ret = ent->available_mrs < ent->limit;
-		spin_unlock_irq(&ent->lock);
+		xa_lock_irq(&ent->mkeys);
+		ret = ent->stored < ent->limit;
+		xa_unlock_irq(&ent->mkeys);
 		if (ret)
 			return true;
 	}
···
  */
 static void queue_adjust_cache_locked(struct mlx5_cache_ent *ent)
 {
-	lockdep_assert_held(&ent->lock);
+	lockdep_assert_held(&ent->mkeys.xa_lock);
 
 	if (ent->disabled || READ_ONCE(ent->dev->fill_delay))
 		return;
-	if (ent->available_mrs < ent->limit) {
+	if (ent->stored < ent->limit) {
 		ent->fill_to_high_water = true;
 		mod_delayed_work(ent->dev->cache.wq, &ent->dwork, 0);
 	} else if (ent->fill_to_high_water &&
-		   ent->available_mrs + ent->pending < 2 * ent->limit) {
+		   ent->reserved < 2 * ent->limit) {
 		/*
 		 * Once we start populating due to hitting a low water mark
 		 * continue until we pass the high water mark.
 		 */
 		mod_delayed_work(ent->dev->cache.wq, &ent->dwork, 0);
-	} else if (ent->available_mrs == 2 * ent->limit) {
+	} else if (ent->stored == 2 * ent->limit) {
 		ent->fill_to_high_water = false;
-	} else if (ent->available_mrs > 2 * ent->limit) {
+	} else if (ent->stored > 2 * ent->limit) {
 		/* Queue deletion of excess entries */
 		ent->fill_to_high_water = false;
-		if (ent->pending)
+		if (ent->stored != ent->reserved)
 			queue_delayed_work(ent->dev->cache.wq, &ent->dwork,
 					   msecs_to_jiffies(1000));
 		else
···
 static void __cache_work_func(struct mlx5_cache_ent *ent)
 {
 	struct mlx5_ib_dev *dev = ent->dev;
-	struct mlx5_mr_cache *cache = &dev->cache;
+	struct mlx5_mkey_cache *cache = &dev->cache;
 	int err;
 
-	spin_lock_irq(&ent->lock);
+	xa_lock_irq(&ent->mkeys);
 	if (ent->disabled)
 		goto out;
 
-	if (ent->fill_to_high_water &&
-	    ent->available_mrs + ent->pending < 2 * ent->limit &&
+	if (ent->fill_to_high_water && ent->reserved < 2 * ent->limit &&
 	    !READ_ONCE(dev->fill_delay)) {
-		spin_unlock_irq(&ent->lock);
+		xa_unlock_irq(&ent->mkeys);
 		err = add_keys(ent, 1);
-		spin_lock_irq(&ent->lock);
+		xa_lock_irq(&ent->mkeys);
 		if (ent->disabled)
 			goto out;
 		if (err) {
 			/*
-			 * EAGAIN only happens if pending is positive, so we
-			 * will be rescheduled from reg_mr_callback(). The only
+			 * EAGAIN only happens if there are pending MRs, so we
+			 * will be rescheduled when storing them. The only
 			 * failure path here is ENOMEM.
 			 */
 			if (err != -EAGAIN) {
···
 						   msecs_to_jiffies(1000));
 			}
 		}
-	} else if (ent->available_mrs > 2 * ent->limit) {
+	} else if (ent->stored > 2 * ent->limit) {
 		bool need_delay;
 
 		/*
···
 		 * the garbage collection work to try to run in next cycle, in
 		 * order to free CPU resources to other tasks.
 		 */
-		spin_unlock_irq(&ent->lock);
+		xa_unlock_irq(&ent->mkeys);
 		need_delay = need_resched() || someone_adding(cache) ||
 			     !time_after(jiffies,
 					 READ_ONCE(cache->last_add) + 300 * HZ);
-		spin_lock_irq(&ent->lock);
+		xa_lock_irq(&ent->mkeys);
 		if (ent->disabled)
 			goto out;
 		if (need_delay) {
···
 		queue_adjust_cache_locked(ent);
 	}
 out:
-	spin_unlock_irq(&ent->lock);
+	xa_unlock_irq(&ent->mkeys);
 }
 
 static void delayed_cache_work_func(struct work_struct *work)
···
 				       int access_flags)
 {
 	struct mlx5_ib_mr *mr;
+	int err;
 
-	/* Matches access in alloc_cache_mr() */
 	if (!mlx5r_umr_can_reconfig(dev, 0, access_flags))
 		return ERR_PTR(-EOPNOTSUPP);
 
-	spin_lock_irq(&ent->lock);
-	if (list_empty(&ent->head)) {
+	mr = kzalloc(sizeof(*mr), GFP_KERNEL);
+	if (!mr)
+		return ERR_PTR(-ENOMEM);
+
+	xa_lock_irq(&ent->mkeys);
+	ent->in_use++;
+
+	if (!ent->stored) {
 		queue_adjust_cache_locked(ent);
 		ent->miss++;
-		spin_unlock_irq(&ent->lock);
-		mr = create_cache_mr(ent);
-		if (IS_ERR(mr))
-			return mr;
+		xa_unlock_irq(&ent->mkeys);
+		err = create_cache_mkey(ent, &mr->mmkey.key);
+		if (err) {
+			xa_lock_irq(&ent->mkeys);
+			ent->in_use--;
+			xa_unlock_irq(&ent->mkeys);
+			kfree(mr);
+			return ERR_PTR(err);
+		}
 	} else {
-		mr = list_first_entry(&ent->head, struct mlx5_ib_mr, list);
-		list_del(&mr->list);
-		ent->available_mrs--;
+		mr->mmkey.key = pop_stored_mkey(ent);
 		queue_adjust_cache_locked(ent);
-		spin_unlock_irq(&ent->lock);
-
-		mlx5_clear_mr(mr);
+		xa_unlock_irq(&ent->mkeys);
 	}
+	mr->mmkey.cache_ent = ent;
+	mr->mmkey.type = MLX5_MKEY_MR;
+	init_waitqueue_head(&mr->mmkey.wait);
 	return mr;
-}
-
-static void mlx5_mr_cache_free(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
-{
-	struct mlx5_cache_ent *ent = mr->cache_ent;
-
-	WRITE_ONCE(dev->cache.last_add, jiffies);
-	spin_lock_irq(&ent->lock);
-	list_add_tail(&mr->list, &ent->head);
-	ent->available_mrs++;
-	queue_adjust_cache_locked(ent);
-	spin_unlock_irq(&ent->lock);
 }
 
 static void clean_keys(struct mlx5_ib_dev *dev, int c)
 {
-	struct mlx5_mr_cache *cache = &dev->cache;
+	struct mlx5_mkey_cache *cache = &dev->cache;
 	struct mlx5_cache_ent *ent = &cache->ent[c];
-	struct mlx5_ib_mr *tmp_mr;
-	struct mlx5_ib_mr *mr;
-	LIST_HEAD(del_list);
+	u32 mkey;
 
 	cancel_delayed_work(&ent->dwork);
-	while (1) {
-		spin_lock_irq(&ent->lock);
-		if (list_empty(&ent->head)) {
-			spin_unlock_irq(&ent->lock);
-			break;
-		}
-		mr = list_first_entry(&ent->head, struct mlx5_ib_mr, list);
-		list_move(&mr->list, &del_list);
-		ent->available_mrs--;
-		ent->total_mrs--;
-		spin_unlock_irq(&ent->lock);
-		mlx5_core_destroy_mkey(dev->mdev, mr->mmkey.key);
+	xa_lock_irq(&ent->mkeys);
+	while (ent->stored) {
+		mkey = pop_stored_mkey(ent);
+		xa_unlock_irq(&ent->mkeys);
+		mlx5_core_destroy_mkey(dev->mdev, mkey);
+		xa_lock_irq(&ent->mkeys);
 	}
-
-	list_for_each_entry_safe(mr, tmp_mr, &del_list, list) {
-		list_del(&mr->list);
-		kfree(mr);
-	}
+	xa_unlock_irq(&ent->mkeys);
 }
 
-static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev)
+static void mlx5_mkey_cache_debugfs_cleanup(struct mlx5_ib_dev *dev)
 {
 	if (!mlx5_debugfs_root || dev->is_rep)
 		return;
···
 	dev->cache.root = NULL;
 }
 
-static void mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev)
+static void mlx5_mkey_cache_debugfs_init(struct mlx5_ib_dev *dev)
 {
-	struct mlx5_mr_cache *cache = &dev->cache;
+	struct mlx5_mkey_cache *cache = &dev->cache;
 	struct mlx5_cache_ent *ent;
 	struct dentry *dir;
 	int i;
···
 
 	cache->root = debugfs_create_dir("mr_cache", mlx5_debugfs_get_dev_root(dev->mdev));
 
-	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) {
+	for (i = 0; i < MAX_MKEY_CACHE_ENTRIES; i++) {
 		ent = &cache->ent[i];
 		sprintf(ent->name, "%d", ent->order);
 		dir = debugfs_create_dir(ent->name, cache->root);
 		debugfs_create_file("size", 0600, dir, ent, &size_fops);
 		debugfs_create_file("limit", 0600, dir, ent, &limit_fops);
-		debugfs_create_u32("cur", 0400, dir, &ent->available_mrs);
+		debugfs_create_ulong("cur", 0400, dir, &ent->stored);
 		debugfs_create_u32("miss", 0600, dir, &ent->miss);
 	}
 }
···
 	WRITE_ONCE(dev->fill_delay, 0);
 }
 
-int mlx5_mr_cache_init(struct mlx5_ib_dev *dev)
+int mlx5_mkey_cache_init(struct mlx5_ib_dev *dev)
 {
-	struct mlx5_mr_cache *cache = &dev->cache;
+	struct mlx5_mkey_cache *cache = &dev->cache;
 	struct mlx5_cache_ent *ent;
 	int i;
 
···
 
 	mlx5_cmd_init_async_ctx(dev->mdev, &dev->async_ctx);
 	timer_setup(&dev->delay_timer, delay_time_func, 0);
-	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) {
+	for (i = 0; i < MAX_MKEY_CACHE_ENTRIES; i++) {
 		ent = &cache->ent[i];
-		INIT_LIST_HEAD(&ent->head);
-		spin_lock_init(&ent->lock);
+		xa_init_flags(&ent->mkeys, XA_FLAGS_LOCK_IRQ);
 		ent->order = i + 2;
 		ent->dev = dev;
 		ent->limit = 0;
 
 		INIT_DELAYED_WORK(&ent->dwork, delayed_cache_work_func);
 
-		if (i > MR_CACHE_LAST_STD_ENTRY) {
-			mlx5_odp_init_mr_cache_entry(ent);
+		if (i > MKEY_CACHE_LAST_STD_ENTRY) {
+			mlx5_odp_init_mkey_cache_entry(ent);
 			continue;
 		}
 
-		if (ent->order > mr_cache_max_order(dev))
+		if (ent->order > mkey_cache_max_order(dev))
 			continue;
 
 		ent->page = PAGE_SHIFT;
···
 			ent->limit = dev->mdev->profile.mr_cache[i].limit;
 		else
 			ent->limit = 0;
-		spin_lock_irq(&ent->lock);
+		xa_lock_irq(&ent->mkeys);
 		queue_adjust_cache_locked(ent);
-		spin_unlock_irq(&ent->lock);
+		xa_unlock_irq(&ent->mkeys);
 	}
 
-	mlx5_mr_cache_debugfs_init(dev);
+	mlx5_mkey_cache_debugfs_init(dev);
 
 	return 0;
 }
 
-int mlx5_mr_cache_cleanup(struct mlx5_ib_dev *dev)
+int mlx5_mkey_cache_cleanup(struct mlx5_ib_dev *dev)
 {
 	unsigned int i;
 
 	if (!dev->cache.wq)
 		return 0;
 
-	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++) {
+	for (i = 0; i < MAX_MKEY_CACHE_ENTRIES; i++) {
 		struct mlx5_cache_ent *ent = &dev->cache.ent[i];
 
-		spin_lock_irq(&ent->lock);
+		xa_lock_irq(&ent->mkeys);
 		ent->disabled = true;
-		spin_unlock_irq(&ent->lock);
+		xa_unlock_irq(&ent->mkeys);
 		cancel_delayed_work_sync(&ent->dwork);
 	}
 
-	mlx5_mr_cache_debugfs_cleanup(dev);
+	mlx5_mkey_cache_debugfs_cleanup(dev);
 	mlx5_cmd_cleanup_async_ctx(&dev->async_ctx);
 
-	for (i = 0; i < MAX_MR_CACHE_ENTRIES; i++)
+	for (i = 0; i < MAX_MKEY_CACHE_ENTRIES; i++)
 		clean_keys(dev, i);
 
 	destroy_workqueue(dev->cache.wq);
···
 	return (npages + 1) / 2;
 }
 
-static int mr_cache_max_order(struct mlx5_ib_dev *dev)
+static int mkey_cache_max_order(struct mlx5_ib_dev *dev)
 {
 	if (MLX5_CAP_GEN(dev->mdev, umr_extended_translation_offset))
-		return MR_CACHE_LAST_STD_ENTRY + 2;
+		return MKEY_CACHE_LAST_STD_ENTRY + 2;
 	return MLX5_MAX_UMR_SHIFT;
 }
 
-static struct mlx5_cache_ent *mr_cache_ent_from_order(struct mlx5_ib_dev *dev,
-						      unsigned int order)
+static struct mlx5_cache_ent *mkey_cache_ent_from_order(struct mlx5_ib_dev *dev,
+							unsigned int order)
 {
-	struct mlx5_mr_cache *cache = &dev->cache;
+	struct mlx5_mkey_cache *cache = &dev->cache;
 
 	if (order < cache->ent[0].order)
 		return &cache->ent[0];
 	order = order - cache->ent[0].order;
-	if (order > MR_CACHE_LAST_STD_ENTRY)
+	if (order > MKEY_CACHE_LAST_STD_ENTRY)
 		return NULL;
 	return &cache->ent[order];
 }
···
 						     0, iova);
 	if (WARN_ON(!page_size))
 		return ERR_PTR(-EINVAL);
-	ent = mr_cache_ent_from_order(
+	ent = mkey_cache_ent_from_order(
 		dev, order_base_2(ib_umem_num_dma_blocks(umem, page_size)));
 	/*
 	 * Matches access in alloc_cache_mr(). If the MR can't come from the
···
 	struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device);
 
 	/* We only track the allocated sizes of MRs from the cache */
-	if (!mr->cache_ent)
+	if (!mr->mmkey.cache_ent)
 		return false;
 	if (!mlx5r_umr_can_load_pas(dev, new_umem->length))
 		return false;
···
 		mlx5_umem_find_best_pgsz(new_umem, mkc, log_page_size, 0, iova);
 	if (WARN_ON(!*page_size))
 		return false;
-	return (1ULL << mr->cache_ent->order) >=
+	return (1ULL << mr->mmkey.cache_ent->order) >=
 	       ib_umem_num_dma_blocks(new_umem, *page_size);
 }
 
···
 	}
 
 	/* Stop DMA */
-	if (mr->cache_ent) {
-		if (mlx5r_umr_revoke_mr(mr)) {
-			spin_lock_irq(&mr->cache_ent->lock);
-			mr->cache_ent->total_mrs--;
-			spin_unlock_irq(&mr->cache_ent->lock);
-			mr->cache_ent = NULL;
-		}
+	if (mr->mmkey.cache_ent) {
+		xa_lock_irq(&mr->mmkey.cache_ent->mkeys);
+		mr->mmkey.cache_ent->in_use--;
+		xa_unlock_irq(&mr->mmkey.cache_ent->mkeys);
+
+		if (mlx5r_umr_revoke_mr(mr) ||
+		    push_mkey(mr->mmkey.cache_ent, false,
+			      xa_mk_value(mr->mmkey.key)))
+			mr->mmkey.cache_ent = NULL;
 	}
-	if (!mr->cache_ent) {
+	if (!mr->mmkey.cache_ent) {
 		rc = destroy_mkey(to_mdev(mr->ibmr.device), mr);
 		if (rc)
 			return rc;
···
 			mlx5_ib_free_odp_mr(mr);
 	}
 
-	if (mr->cache_ent) {
-		mlx5_mr_cache_free(dev, mr);
-	} else {
+	if (!mr->mmkey.cache_ent)
 		mlx5_free_priv_descs(mr);
-		kfree(mr);
-	}
+
+	kfree(mr);
 	return 0;
 }
 

drivers/infiniband/hw/mlx5/odp.c

···
 	return err;
 }
 
-void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent)
+void mlx5_odp_init_mkey_cache_entry(struct mlx5_cache_ent *ent)
 {
 	if (!(ent->dev->odp_caps.general_caps & IB_ODP_SUPPORT_IMPLICIT))
 		return;

drivers/infiniband/hw/mlx5/umr.c

···
 	dev->umrc.pd = pd;
 
 	sema_init(&dev->umrc.sem, MAX_UMR_WR);
+	mutex_init(&dev->umrc.lock);
 
 	return 0;
 
···
 	ib_destroy_qp(dev->umrc.qp);
 	ib_free_cq(dev->umrc.cq);
 	ib_dealloc_pd(dev->umrc.pd);
+}
+
+static int mlx5r_umr_recover(struct mlx5_ib_dev *dev)
+{
+	struct umr_common *umrc = &dev->umrc;
+	struct ib_qp_attr attr;
+	int err;
+
+	attr.qp_state = IB_QPS_RESET;
+	err = ib_modify_qp(umrc->qp, &attr, IB_QP_STATE);
+	if (err) {
+		mlx5_ib_dbg(dev, "Couldn't modify UMR QP\n");
+		goto err;
+	}
+
+	err = mlx5r_umr_qp_rst2rts(dev, umrc->qp);
+	if (err)
+		goto err;
+
+	umrc->state = MLX5_UMR_STATE_ACTIVE;
+	return 0;
+
+err:
+	umrc->state = MLX5_UMR_STATE_ERR;
+	return err;
 }
 
 static int mlx5r_umr_post_send(struct ib_qp *ibqp, u32 mkey, struct ib_cqe *cqe,
···
 
 	id.ib_cqe = cqe;
 	mlx5r_finish_wqe(qp, ctrl, seg, size, cur_edge, idx, id.wr_id, 0,
-			 MLX5_FENCE_MODE_NONE, MLX5_OPCODE_UMR);
+			 MLX5_FENCE_MODE_INITIATOR_SMALL, MLX5_OPCODE_UMR);
 
 	mlx5r_ring_db(qp, 1, ctrl);
 
···
 	mlx5r_umr_init_context(&umr_context);
 
 	down(&umrc->sem);
-	err = mlx5r_umr_post_send(umrc->qp, mkey, &umr_context.cqe, wqe,
-				  with_data);
-	if (err)
-		mlx5_ib_warn(dev, "UMR post send failed, err %d\n", err);
-	else {
-		wait_for_completion(&umr_context.done);
-		if (umr_context.status != IB_WC_SUCCESS) {
-			mlx5_ib_warn(dev, "reg umr failed (%u)\n",
-				     umr_context.status);
+	while (true) {
+		mutex_lock(&umrc->lock);
+		if (umrc->state == MLX5_UMR_STATE_ERR) {
+			mutex_unlock(&umrc->lock);
 			err = -EFAULT;
+			break;
 		}
+
+		if (umrc->state == MLX5_UMR_STATE_RECOVER) {
+			mutex_unlock(&umrc->lock);
+			usleep_range(3000, 5000);
+			continue;
+		}
+
+		err = mlx5r_umr_post_send(umrc->qp, mkey, &umr_context.cqe, wqe,
+					  with_data);
+		mutex_unlock(&umrc->lock);
+		if (err) {
+			mlx5_ib_warn(dev, "UMR post send failed, err %d\n",
+				     err);
+			break;
+		}
+
+		wait_for_completion(&umr_context.done);
+
+		if (umr_context.status == IB_WC_SUCCESS)
+			break;
+
+		if (umr_context.status == IB_WC_WR_FLUSH_ERR)
+			continue;
+
+		WARN_ON_ONCE(1);
+		mlx5_ib_warn(dev,
+			"reg umr failed (%u). Trying to recover and resubmit the flushed WQEs\n",
+			umr_context.status);
+		mutex_lock(&umrc->lock);
+		err = mlx5r_umr_recover(dev);
+		mutex_unlock(&umrc->lock);
+		if (err)
+			mlx5_ib_warn(dev, "couldn't recover UMR, err %d\n",
+				     err);
+		err = -EFAULT;
+		break;
 	}
 	up(&umrc->sem);
 	return err;

drivers/infiniband/hw/qedr/verbs.c

···
 		else
 			DP_ERR(dev, "roce alloc tid returned error %d\n", rc);
 
-		goto err0;
+		goto err1;
 	}
 
 	/* Index only, 18 bit long, lkey = itid << 8 | key */
···
 	rc = dev->ops->rdma_register_tid(dev->rdma_ctx, &mr->hw_mr);
 	if (rc) {
 		DP_ERR(dev, "roce register tid returned an error %d\n", rc);
-		goto err1;
+		goto err2;
 	}
 
 	mr->ibmr.lkey = mr->hw_mr.itid << 8 | mr->hw_mr.key;
···
 	DP_DEBUG(dev, QEDR_MSG_MR, "alloc frmr: %x\n", mr->ibmr.lkey);
 	return mr;
 
-err1:
+err2:
 	dev->ops->rdma_free_tid(dev->rdma_ctx, mr->hw_mr.itid);
+err1:
+	qedr_free_pbl(dev, &mr->info.pbl_info, mr->info.pbl_table);
 err0:
 	kfree(mr);
 	return ERR_PTR(rc);

drivers/infiniband/hw/qib/qib.h

···
  * These 7 values (SDR, DDR, and QDR may be ORed for auto-speed
  * negotiation) are used for the 3rd argument to path_f_set_ib_cfg
  * with cmd QIB_IB_CFG_SPD_ENB, by direct calls or via sysfs.  They
- * are also the the possible values for qib_link_speed_enabled and active
+ * are also the possible values for qib_link_speed_enabled and active
  * The values were chosen to match values used within the IB spec.
  */
 #define QIB_IB_SDR 1

drivers/infiniband/hw/qib/qib_file_ops.c

···
 		kinfo->spi_tidcnt += dd->rcvtidcnt % subctxt_cnt;
 	/*
 	 * for this use, may be cfgctxts summed over all chips that
-	 * are are configured and present
+	 * are configured and present
 	 */
 	kinfo->spi_nctxts = dd->cfgctxts;
 	/* unit (chip/board) our context is on */
···
 		ret = -EPERM;
 		goto bail;
 	}
-	/* don't allow them to later change to writeable with mprotect */
+	/* don't allow them to later change to writable with mprotect */
 	vma->vm_flags &= ~VM_MAYWRITE;
 
 	start = vma->vm_start;
···
 			goto bail;
 		}
 		/*
-		 * Don't allow permission to later change to writeable
+		 * Don't allow permission to later change to writable
 		 * with mprotect.
 		 */
 		vma->vm_flags &= ~VM_MAYWRITE;

drivers/infiniband/hw/qib/qib_iba7220.c

···
 /*
  * This file contains almost all the chip-specific register information and
  * access functions for the QLogic QLogic_IB 7220 PCI-Express chip, with the
- * exception of SerDes support, which in in qib_sd7220.c.
+ * exception of SerDes support, which in qib_sd7220.c.
  */
 
 /* Below uses machine-generated qib_chipnum_regs.h file */

drivers/infiniband/hw/qib/qib_iba7322.c

···
 
 	qib_7322_free_irq(dd);
 	kfree(dd->cspec->cntrs);
-	kfree(dd->cspec->sendchkenable);
-	kfree(dd->cspec->sendgrhchk);
-	kfree(dd->cspec->sendibchk);
+	bitmap_free(dd->cspec->sendchkenable);
+	bitmap_free(dd->cspec->sendgrhchk);
+	bitmap_free(dd->cspec->sendibchk);
 	kfree(dd->cspec->msix_entries);
 	for (i = 0; i < dd->num_pports; i++) {
 		unsigned long flags;
···
 	features = qib_7322_boardname(dd);
 
 	/* now that piobcnt2k and 4k set, we can allocate these */
-	sbufcnt = dd->piobcnt2k + dd->piobcnt4k +
-		NUM_VL15_BUFS + BITS_PER_LONG - 1;
-	sbufcnt /= BITS_PER_LONG;
-	dd->cspec->sendchkenable =
-		kmalloc_array(sbufcnt, sizeof(*dd->cspec->sendchkenable),
-			      GFP_KERNEL);
-	dd->cspec->sendgrhchk =
-		kmalloc_array(sbufcnt, sizeof(*dd->cspec->sendgrhchk),
-			      GFP_KERNEL);
-	dd->cspec->sendibchk =
-		kmalloc_array(sbufcnt, sizeof(*dd->cspec->sendibchk),
-			      GFP_KERNEL);
+	sbufcnt = dd->piobcnt2k + dd->piobcnt4k + NUM_VL15_BUFS;
+
+	dd->cspec->sendchkenable = bitmap_zalloc(sbufcnt, GFP_KERNEL);
+	dd->cspec->sendgrhchk = bitmap_zalloc(sbufcnt, GFP_KERNEL);
+	dd->cspec->sendibchk = bitmap_zalloc(sbufcnt, GFP_KERNEL);
 	if (!dd->cspec->sendchkenable || !dd->cspec->sendgrhchk ||
 		!dd->cspec->sendibchk) {
 		ret = -ENOMEM;

drivers/infiniband/hw/qib/qib_init.c

···
 	if (!qib_cpulist_count) {
 		u32 count = num_online_cpus();
 
-		qib_cpulist = kcalloc(BITS_TO_LONGS(count), sizeof(long),
-				      GFP_KERNEL);
+		qib_cpulist = bitmap_zalloc(count, GFP_KERNEL);
 		if (qib_cpulist)
 			qib_cpulist_count = count;
 	}
···
 #endif
 
 	qib_cpulist_count = 0;
-	kfree(qib_cpulist);
+	bitmap_free(qib_cpulist);
 
 	WARN_ON(!xa_empty(&qib_dev_table));
 	qib_dev_cleanup();

drivers/infiniband/hw/qib/qib_sd7220.c

···
 		/* Need to release */
 		u64 pollval;
 		/*
-		 * The only writeable bits are the request and CS.
+		 * The only writable bits are the request and CS.
 		 * Both should be clear
 		 */
 		u64 newval = 0;

drivers/infiniband/hw/usnic/usnic_uiom.c

···
 	if (err)
 		goto out_free_dev;
 
-	if (!iommu_capable(dev->bus, IOMMU_CAP_CACHE_COHERENCY)) {
+	if (!device_iommu_capable(dev, IOMMU_CAP_CACHE_COHERENCY)) {
 		usnic_err("IOMMU of %s does not support cache coherency\n",
 				dev_name(dev));
 		err = -EINVAL;

drivers/infiniband/sw/rxe/rxe_comp.c

···
 {
 	struct rxe_qp *qp = from_timer(qp, t, retrans_timer);
 
+	pr_debug("%s: fired for qp#%d\n", __func__, qp->elem.index);
+
 	if (qp->valid) {
 		qp->comp.timeout = 1;
 		rxe_run_task(&qp->comp.task, 1);
···
 	struct sk_buff *skb = NULL;
 	struct rxe_pkt_info *pkt = NULL;
 	enum comp_state state;
-	int ret = 0;
+	int ret;
 
 	if (!rxe_get(qp))
 		return -EAGAIN;
 
-	if (!qp->valid || qp->req.state == QP_STATE_ERROR ||
-	    qp->req.state == QP_STATE_RESET) {
+	if (!qp->valid || qp->comp.state == QP_STATE_ERROR ||
+	    qp->comp.state == QP_STATE_RESET) {
 		rxe_drain_resp_pkts(qp, qp->valid &&
-				    qp->req.state == QP_STATE_ERROR);
-		ret = -EAGAIN;
-		goto done;
+				    qp->comp.state == QP_STATE_ERROR);
+		goto exit;
 	}
 
 	if (qp->comp.timeout) {
···
 		qp->comp.timeout_retry = 0;
 	}
 
-	if (qp->req.need_retry) {
-		ret = -EAGAIN;
-		goto done;
-	}
+	if (qp->req.need_retry)
+		goto exit;
 
 	state = COMPST_GET_ACK;
 
···
 			    qp->qp_timeout_jiffies)
 				mod_timer(&qp->retrans_timer,
 					  jiffies + qp->qp_timeout_jiffies);
-			ret = -EAGAIN;
-			goto done;
+			goto exit;
 
 		case COMPST_ERROR_RETRY:
 			/* we come here if the retry timer fired and we did
···
 			 */
 
 			/* there is nothing to retry in this case */
-			if (!wqe || (wqe->state == wqe_state_posted)) {
-				ret = -EAGAIN;
-				goto done;
-			}
+			if (!wqe || (wqe->state == wqe_state_posted))
+				goto exit;
 
 			/* if we've started a retry, don't start another
 			 * retry sequence, unless this is a timeout.
···
 			break;
 
 		case COMPST_RNR_RETRY:
+			/* we come here if we received an RNR NAK */
 			if (qp->comp.rnr_retry > 0) {
 				if (qp->comp.rnr_retry != 7)
 					qp->comp.rnr_retry--;
 
-				qp->req.need_retry = 1;
+				/* don't start a retry flow until the
+				 * rnr timer has fired
+				 */
+				qp->req.wait_for_rnr_timer = 1;
 				pr_debug("qp#%d set rnr nak timer\n",
 					 qp_num(qp));
 				mod_timer(&qp->rnr_nak_timer,
 					  jiffies + rnrnak_jiffies(aeth_syn(pkt)
 						& ~AETH_TYPE_MASK));
-				ret = -EAGAIN;
-				goto done;
+				goto exit;
 			} else {
 				rxe_counter_inc(rxe,
 						RXE_CNT_RNR_RETRY_EXCEEDED);
···
 			WARN_ON_ONCE(wqe->status == IB_WC_SUCCESS);
 			do_complete(qp, wqe);
 			rxe_qp_error(qp);
-			ret = -EAGAIN;
-			goto done;
+			goto exit;
 		}
 	}
 
+	/* A non-zero return value will cause rxe_do_task to
+	 * exit its loop and end the tasklet. A zero return
+	 * will continue looping and return to rxe_completer
+	 */
 done:
+	ret = 0;
+	goto out;
+exit:
+	ret = -EAGAIN;
+out:
 	if (pkt)
 		free_pkt(pkt);
 	rxe_put(qp);

drivers/infiniband/sw/rxe/rxe_cq.c

···
 	}
 
 	if (cqe > rxe->attr.max_cqe) {
-		pr_warn("cqe(%d) > max_cqe(%d)\n",
-			cqe, rxe->attr.max_cqe);
+		pr_debug("cqe(%d) > max_cqe(%d)\n",
+				cqe, rxe->attr.max_cqe);
 		goto err1;
 	}
 
 	if (cq) {
 		count = queue_count(cq->queue, QUEUE_TYPE_TO_CLIENT);
 		if (cqe < count) {
-			pr_warn("cqe(%d) < current # elements in queue (%d)",
-				cqe, count);
+			pr_debug("cqe(%d) < current # elements in queue (%d)",
+					cqe, count);
 			goto err1;
 		}
 	}

drivers/infiniband/sw/rxe/rxe_loc.h

···
 			 enum rxe_mr_lookup_type type);
 int mr_check_range(struct rxe_mr *mr, u64 iova, size_t length);
 int advance_dma_data(struct rxe_dma_info *dma, unsigned int length);
-int rxe_invalidate_mr(struct rxe_qp *qp, u32 rkey);
+int rxe_invalidate_mr(struct rxe_qp *qp, u32 key);
 int rxe_reg_fast_mr(struct rxe_qp *qp, struct rxe_send_wqe *wqe);
-int rxe_mr_set_page(struct ib_mr *ibmr, u64 addr);
 int rxe_dereg_mr(struct ib_mr *ibmr, struct ib_udata *udata);
 void rxe_mr_cleanup(struct rxe_pool_elem *elem);
 
···
 		max_sge * sizeof(struct ib_sge);
 }
 
-void free_rd_atomic_resource(struct rxe_qp *qp, struct resp_res *res);
+void free_rd_atomic_resource(struct resp_res *res);
 
 static inline void rxe_advance_resp_resource(struct rxe_qp *qp)
 {

drivers/infiniband/sw/rxe/rxe_mr.c

···
 
 int mr_check_range(struct rxe_mr *mr, u64 iova, size_t length)
 {
-	struct rxe_map_set *set = mr->cur_map_set;
+
 
 	switch (mr->type) {
 	case IB_MR_TYPE_DMA:
···
 
 	case IB_MR_TYPE_USER:
 	case IB_MR_TYPE_MEM_REG:
-		if (iova < set->iova || length > set->length ||
-		    iova > set->iova + set->length - length)
+		if (iova < mr->iova || length > mr->length ||
+		    iova > mr->iova + mr->length - length)
 			return -EFAULT;
 		return 0;
 
···
 	mr->map_shift = ilog2(RXE_BUF_PER_MAP);
 }
 
-static void rxe_mr_free_map_set(int num_map, struct rxe_map_set *set)
+static int rxe_mr_alloc(struct rxe_mr *mr, int num_buf)
 {
 	int i;
+	int num_map;
+	struct rxe_map **map = mr->map;
 
-	for (i = 0; i < num_map; i++)
-		kfree(set->map[i]);
+	num_map = (num_buf + RXE_BUF_PER_MAP - 1) / RXE_BUF_PER_MAP;
 
-	kfree(set->map);
-	kfree(set);
-}
-
-static int rxe_mr_alloc_map_set(int num_map, struct rxe_map_set **setp)
-{
-	int i;
-	struct rxe_map_set *set;
-
-	set = kmalloc(sizeof(*set), GFP_KERNEL);
-	if (!set)
-		goto err_out;
-
-	set->map = kmalloc_array(num_map, sizeof(struct rxe_map *), GFP_KERNEL);
-	if (!set->map)
-		goto err_free_set;
+	mr->map = kmalloc_array(num_map, sizeof(*map), GFP_KERNEL);
+	if (!mr->map)
+		goto err1;
 
 	for (i = 0; i < num_map; i++) {
-		set->map[i] = kmalloc(sizeof(struct rxe_map), GFP_KERNEL);
-		if (!set->map[i])
-			goto err_free_map;
+		mr->map[i] = kmalloc(sizeof(**map), GFP_KERNEL);
+		if (!mr->map[i])
+			goto err2;
 	}
 
-	*setp = set;
-
-	return 0;
-
-err_free_map:
-	for (i--; i >= 0; i--)
-		kfree(set->map[i]);
-
-	kfree(set->map);
-err_free_set:
-	kfree(set);
-err_out:
-	return -ENOMEM;
-}
-
-/**
- * rxe_mr_alloc() - Allocate memory map array(s) for MR
- * @mr: Memory region
- * @num_buf: Number of buffer descriptors to support
- * @both: If non zero allocate both mr->map and mr->next_map
- *	  else just allocate mr->map. Used for fast MRs
- *
- * Return: 0 on success else an error
- */
-static int rxe_mr_alloc(struct rxe_mr *mr, int num_buf, int both)
-{
-	int ret;
-	int num_map;
-
 	BUILD_BUG_ON(!is_power_of_2(RXE_BUF_PER_MAP));
-	num_map = (num_buf + RXE_BUF_PER_MAP - 1) / RXE_BUF_PER_MAP;
 
 	mr->map_shift = ilog2(RXE_BUF_PER_MAP);
 	mr->map_mask = RXE_BUF_PER_MAP - 1;
+
 	mr->num_buf = num_buf;
-	mr->max_buf = num_map * RXE_BUF_PER_MAP;
 	mr->num_map = num_map;
-
-	ret = rxe_mr_alloc_map_set(num_map, &mr->cur_map_set);
-	if (ret)
-		return -ENOMEM;
-
-	if (both) {
-		ret = rxe_mr_alloc_map_set(num_map, &mr->next_map_set);
-		if (ret)
-			goto err_free;
-	}
+	mr->max_buf = num_map * RXE_BUF_PER_MAP;
 
 	return 0;
 
-err_free:
-	rxe_mr_free_map_set(mr->num_map, mr->cur_map_set);
-	mr->cur_map_set = NULL;
+err2:
+	for (i--; i >= 0; i--)
+		kfree(mr->map[i]);
+
+	kfree(mr->map);
+err1:
 	return -ENOMEM;
 }
 
···
 int rxe_mr_init_user(struct rxe_pd *pd, u64 start, u64 length, u64 iova,
 		     int access, struct rxe_mr *mr)
 {
-	struct rxe_map_set	*set;
 	struct rxe_map		**map;
 	struct rxe_phys_buf	*buf = NULL;
 	struct ib_umem		*umem;
···
 	int			num_buf;
 	void			*vaddr;
 	int err;
+	int i;
 
 	umem = ib_umem_get(pd->ibpd.device, start, length, access);
 	if (IS_ERR(umem)) {
···
 
 	rxe_mr_init(access, mr);
 
-	err = rxe_mr_alloc(mr, num_buf, 0);
+	err = rxe_mr_alloc(mr, num_buf);
 	if (err) {
 		pr_warn("%s: Unable to allocate memory for map\n",
 				__func__);
 		goto err_release_umem;
 	}
 
-	set = mr->cur_map_set;
-	set->page_shift = PAGE_SHIFT;
-	set->page_mask = PAGE_SIZE - 1;
+	mr->page_shift = PAGE_SHIFT;
+	mr->page_mask = PAGE_SIZE - 1;
 
-	num_buf = 0;
-	map = set->map;
-
+	num_buf			= 0;
+	map = mr->map;
 	if (length > 0) {
 		buf = map[0]->buf;
 
···
 				pr_warn("%s: Unable to get virtual address\n",
 						__func__);
 				err = -ENOMEM;
-				goto err_release_umem;
+				goto err_cleanup_map;
 			}
 
 			buf->addr = (uintptr_t)vaddr;
 			buf->size = PAGE_SIZE;
 			num_buf++;
 			buf++;
+
 		}
 	}
 
 	mr->ibmr.pd = &pd->ibpd;
 	mr->umem = umem;
 	mr->access = access;
+	mr->length = length;
+	mr->iova = iova;
+	mr->va = start;
+	mr->offset = ib_umem_offset(umem);
 	mr->state = RXE_MR_STATE_VALID;
 	mr->type = IB_MR_TYPE_USER;
 
-	set->length = length;
-	set->iova = iova;
-	set->va = start;
-	set->offset = ib_umem_offset(umem);
-
 	return 0;
 
+err_cleanup_map:
+	for (i = 0; i < mr->num_map; i++)
+		kfree(mr->map[i]);
+	kfree(mr->map);
 err_release_umem:
 	ib_umem_release(umem);
 err_out:
···
 	/* always allow remote access for FMRs */
 	rxe_mr_init(IB_ACCESS_REMOTE, mr);
 
-	err = rxe_mr_alloc(mr, max_pages, 1);
+	err = rxe_mr_alloc(mr, max_pages);
 	if (err)
 		goto err1;
 
···
 static void lookup_iova(struct rxe_mr *mr, u64 iova, int *m_out, int *n_out,
 			size_t *offset_out)
 {
-	struct rxe_map_set *set = mr->cur_map_set;
-	size_t offset = iova - set->iova + set->offset;
+	size_t offset = iova - mr->iova + mr->offset;
 	int			map_index;
 	int			buf_index;
 	u64			length;
-	struct rxe_map *map;
 
-	if (likely(set->page_shift)) {
-		*offset_out = offset & set->page_mask;
-		offset >>= set->page_shift;
+	if (likely(mr->page_shift)) {
+		*offset_out = offset & mr->page_mask;
+		offset >>= mr->page_shift;
 		*n_out = offset & mr->map_mask;
 		*m_out = offset >> mr->map_shift;
 	} else {
 		map_index = 0;
 		buf_index = 0;
 
-		map = set->map[map_index];
-		length = map->buf[buf_index].size;
+		length = mr->map[map_index]->buf[buf_index].size;
 
 		while (offset >= length) {
 			offset -= length;
···
 				map_index++;
 				buf_index = 0;
 			}
-			map = set->map[map_index];
-			length = map->buf[buf_index].size;
+			length = mr->map[map_index]->buf[buf_index].size;
 		}
 
 		*m_out = map_index;
···
 		goto out;
 	}
 
-	if (!mr->cur_map_set) {
+	if (!mr->map) {
 		addr = (void *)(uintptr_t)iova;
 		goto out;
 	}
···
 
 	lookup_iova(mr, iova, &m, &n, &offset);
 
-	if (offset + length > mr->cur_map_set->map[m]->buf[n].size) {
+	if (offset + length > mr->map[m]->buf[n].size) {
 		pr_warn("crosses page boundary\n");
 		addr = NULL;
 		goto out;
 	}
 
-	addr = (void *)(uintptr_t)mr->cur_map_set->map[m]->buf[n].addr + offset;
+	addr = (void *)(uintptr_t)mr->map[m]->buf[n].addr + offset;
 
 out:
 	return addr;
···
 		return 0;
 	}
 
-	WARN_ON_ONCE(!mr->cur_map_set);
+	WARN_ON_ONCE(!mr->map);
 
 	err = mr_check_range(mr, iova, length);
 	if (err) {
···
 
 	lookup_iova(mr, iova, &m, &i, &offset);
 
-	map = mr->cur_map_set->map + m;
+	map = mr->map + m;
 	buf	= map[0]->buf + i;
 
 	while (length > 0) {
···
 	return mr;
 }
 
-int rxe_invalidate_mr(struct rxe_qp *qp, u32 rkey)
+int rxe_invalidate_mr(struct rxe_qp *qp, u32 key)
 {
 	struct rxe_dev *rxe = to_rdev(qp->ibqp.device);
 	struct rxe_mr *mr;
 	int ret;
 
-	mr = rxe_pool_get_index(&rxe->mr_pool, rkey >> 8);
+	mr = rxe_pool_get_index(&rxe->mr_pool, key >> 8);
 	if (!mr) {
-		pr_err("%s: No MR for rkey %#x\n", __func__, rkey);
+		pr_err("%s: No MR for key %#x\n", __func__, key);
 		ret = -EINVAL;
 		goto err;
 	}
 
-	if (rkey != mr->rkey) {
-		pr_err("%s: rkey (%#x) doesn't match mr->rkey (%#x)\n",
-			__func__, rkey, mr->rkey);
+	if (mr->rkey ? (key != mr->rkey) : (key != mr->lkey)) {
+		pr_err("%s: wr key (%#x) doesn't match mr key (%#x)\n",
+			__func__, key, (mr->rkey ? mr->rkey : mr->lkey));
 		ret = -EINVAL;
 		goto err_drop_ref;
 	}
···
 int rxe_reg_fast_mr(struct rxe_qp *qp, struct rxe_send_wqe *wqe)
 {
 	struct rxe_mr *mr = to_rmr(wqe->wr.wr.reg.mr);
-	u32 key = wqe->wr.wr.reg.key & 0xff;
+	u32 key = wqe->wr.wr.reg.key;
 	u32 access = wqe->wr.wr.reg.access;
-	struct rxe_map_set *set;
 
 	/* user can only register MR in free state */
 	if (unlikely(mr->state != RXE_MR_STATE_FREE)) {
···
 		return -EINVAL;
 	}
 
+	/* user is only allowed to change key portion of l/rkey */
+	if (unlikely((mr->lkey & ~0xff) != (key & ~0xff))) {
+		pr_warn("%s: key = 0x%x has wrong index mr->lkey = 0x%x\n",
+			__func__, key, mr->lkey);
+		return -EINVAL;
+	}
+
 	mr->access = access;
-	mr->lkey = (mr->lkey & ~0xff) | key;
-	mr->rkey = (access & IB_ACCESS_REMOTE) ? mr->lkey : 0;
+	mr->lkey = key;
+	mr->rkey = (access & IB_ACCESS_REMOTE) ? key : 0;
+	mr->iova = wqe->wr.wr.reg.mr->iova;
 	mr->state = RXE_MR_STATE_VALID;
-
-	set = mr->cur_map_set;
-	mr->cur_map_set = mr->next_map_set;
-	mr->cur_map_set->iova = wqe->wr.wr.reg.mr->iova;
-	mr->next_map_set = set;
-
-	return 0;
-}
-
-int rxe_mr_set_page(struct ib_mr *ibmr, u64 addr)
-{
-	struct rxe_mr *mr = to_rmr(ibmr);
-	struct rxe_map_set *set = mr->next_map_set;
-	struct rxe_map *map;
-	struct rxe_phys_buf *buf;
-
-	if (unlikely(set->nbuf == mr->num_buf))
-		return -ENOMEM;
-
-	map = set->map[set->nbuf / RXE_BUF_PER_MAP];
-	buf = &map->buf[set->nbuf % RXE_BUF_PER_MAP];
-
-	buf->addr = addr;
-	buf->size = ibmr->page_size;
-	set->nbuf++;
 
 	return 0;
 }
···
 	if (atomic_read(&mr->num_mw) > 0)
 		return -EINVAL;
 
-	rxe_put(mr);
+	rxe_cleanup(mr);
 
 	return 0;
 }
···
 void rxe_mr_cleanup(struct rxe_pool_elem *elem)
 {
 	struct rxe_mr *mr = container_of(elem, typeof(*mr), elem);
+	int i;
 
 	rxe_put(mr_pd(mr));
-
 	ib_umem_release(mr->umem);
 
-	if (mr->cur_map_set)
-		rxe_mr_free_map_set(mr->num_map, mr->cur_map_set);
+	if (mr->map) {
+		for (i = 0; i < mr->num_map; i++)
+			kfree(mr->map[i]);
 
-	if (mr->next_map_set)
-		rxe_mr_free_map_set(mr->num_map, mr->next_map_set);
+		kfree(mr->map);
+	}
 }

drivers/infiniband/sw/rxe/rxe_mw.c

···
 			RXE_MW_STATE_FREE : RXE_MW_STATE_VALID;
 	spin_lock_init(&mw->lock);
 
+	rxe_finalize(mw);
+
 	return 0;
 }
 
···
 {
 	struct rxe_mw *mw = to_rmw(ibmw);
 
-	rxe_put(mw);
+	rxe_cleanup(mw);
 
 	return 0;
 }
···
 static int rxe_check_bind_mw(struct rxe_qp *qp, struct rxe_send_wqe *wqe,
 			 struct rxe_mw *mw, struct rxe_mr *mr)
 {
-	u32 key = wqe->wr.wr.mw.rkey & 0xff;
-
 	if (mw->ibmw.type == IB_MW_TYPE_1) {
 		if (unlikely(mw->state != RXE_MW_STATE_VALID)) {
 			pr_err_once(
···
 		}
 	}
 
-	if (unlikely(key == (mw->rkey & 0xff))) {
-		pr_err_once("attempt to bind MW with same key\n");
-		return -EINVAL;
-	}
-
 	/* remaining checks only apply to a nonzero MR */
 	if (!mr)
 		return 0;
···
 		      (IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_ATOMIC)) &&
 		     !(mr->access & IB_ACCESS_LOCAL_WRITE))) {
 		pr_err_once(
-			"attempt to bind an writeable MW to an MR without local write access\n");
+			"attempt to bind an Writable MW to an MR without local write access\n");
 		return -EINVAL;
 	}
 
 	/* C10-75 */
 	if (mw->access & IB_ZERO_BASED) {
-		if (unlikely(wqe->wr.wr.mw.length > mr->cur_map_set->length)) {
+		if (unlikely(wqe->wr.wr.mw.length > mr->length)) {
 			pr_err_once(
 				"attempt to bind a ZB MW outside of the MR\n");
 			return -EINVAL;
 		}
 	} else {
-		if (unlikely((wqe->wr.wr.mw.addr < mr->cur_map_set->iova) ||
+		if (unlikely((wqe->wr.wr.mw.addr < mr->iova) ||
 			     ((wqe->wr.wr.mw.addr + wqe->wr.wr.mw.length) >
-			      (mr->cur_map_set->iova + mr->cur_map_set->length)))) {
+			      (mr->iova + mr->length)))) {
 			pr_err_once(
 				"attempt to bind a VA MW outside of the MR\n");
 			return -EINVAL;

drivers/infiniband/sw/rxe/rxe_param.h

···
 	RXE_INFLIGHT_SKBS_PER_QP_HIGH	= 64,
 	RXE_INFLIGHT_SKBS_PER_QP_LOW	= 16,
 
+	/* Max number of interations of each tasklet
+	 * before yielding the cpu to let other
+	 * work make progress
+	 */
+	RXE_MAX_ITERATIONS		= 1024,
+
 	/* Delay before calling arbiter timer */
 	RXE_NSEC_ARB_TIMER_DELAY	= 200,
 

drivers/infiniband/sw/rxe/rxe_pool.c

···
 
 #include "rxe.h"
 
+#define RXE_POOL_TIMEOUT	(200)
 #define RXE_POOL_ALIGN		(16)
 
 static const struct rxe_type_info {
···
 	elem->pool = pool;
 	elem->obj = obj;
 	kref_init(&elem->ref_cnt);
+	init_completion(&elem->complete);
 
-	err = xa_alloc_cyclic(&pool->xa, &elem->index, elem, pool->limit,
+	/* allocate index in array but leave pointer as NULL so it
+	 * can't be looked up until rxe_finalize() is called
+	 */
+	err = xa_alloc_cyclic(&pool->xa, &elem->index, NULL, pool->limit,
 			      &pool->next, GFP_KERNEL);
-	if (err)
+	if (err < 0)
 		goto err_free;
 
 	return obj;
···
 	return NULL;
 }
 
-int __rxe_add_to_pool(struct rxe_pool *pool, struct rxe_pool_elem *elem)
+int __rxe_add_to_pool(struct rxe_pool *pool, struct rxe_pool_elem *elem,
+				bool sleepable)
 {
 	int err;
+	gfp_t gfp_flags;
 
 	if (WARN_ON(pool->type == RXE_TYPE_MR))
 		return -EINVAL;
···
 	elem->pool = pool;
 	elem->obj = (u8 *)elem - pool->elem_offset;
 	kref_init(&elem->ref_cnt);
+	init_completion(&elem->complete);
 
-	err = xa_alloc_cyclic(&pool->xa, &elem->index, elem, pool->limit,
-			      &pool->next, GFP_KERNEL);
-	if (err)
+	/* AH objects are unique in that the create_ah verb
+	 * can be called in atomic context. If the create_ah
+	 * call is not sleepable use GFP_ATOMIC.
+	 */
+	gfp_flags = sleepable ? GFP_KERNEL : GFP_ATOMIC;
+
+	if (sleepable)
+		might_sleep();
+	err = xa_alloc_cyclic(&pool->xa, &elem->index, NULL, pool->limit,
+			      &pool->next, gfp_flags);
+	if (err < 0)
 		goto err_cnt;
 
 	return 0;
···
 {
 	struct rxe_pool_elem *elem;
 	struct xarray *xa = &pool->xa;
-	unsigned long flags;
 	void *obj;
 
-	xa_lock_irqsave(xa, flags);
+	rcu_read_lock();
 	elem = xa_load(xa, index);
 	if (elem && kref_get_unless_zero(&elem->ref_cnt))
 		obj = elem->obj;
 	else
 		obj = NULL;
-	xa_unlock_irqrestore(xa, flags);
+	rcu_read_unlock();
 
 	return obj;
 }
···
 static void rxe_elem_release(struct kref *kref)
 {
 	struct rxe_pool_elem *elem = container_of(kref, typeof(*elem), ref_cnt);
-	struct rxe_pool *pool = elem->pool;
 
-	xa_erase(&pool->xa, elem->index);
+	complete(&elem->complete);
+}
+
+int __rxe_cleanup(struct rxe_pool_elem *elem, bool sleepable)
+{
+	struct rxe_pool *pool = elem->pool;
+	struct xarray *xa = &pool->xa;
+	static int timeout = RXE_POOL_TIMEOUT;
+	int ret, err = 0;
+	void *xa_ret;
+
+	if (sleepable)
+		might_sleep();
+
+	/* erase xarray entry to prevent looking up
+	 * the pool elem from its index
+	 */
+	xa_ret = xa_erase(xa, elem->index);
+	WARN_ON(xa_err(xa_ret));
+
+	/* if this is the last call to rxe_put complete the
+	 * object. It is safe to touch obj->elem after this since
+	 * it is freed below
+	 */
+	__rxe_put(elem);
+
+	/* wait until all references to the object have been
+	 * dropped before final object specific cleanup and
+	 * return to rdma-core
+	 */
+	if (sleepable) {
+		if (!completion_done(&elem->complete) && timeout) {
+			ret = wait_for_completion_timeout(&elem->complete,
+					timeout);
+
+			/* Shouldn't happen. There are still references to
+			 * the object but, rather than deadlock, free the
+			 * object or pass back to rdma-core.
+			 */
+			if (WARN_ON(!ret))
+				err = -EINVAL;
+		}
+	} else {
+		unsigned long until = jiffies + timeout;
+
+		/* AH objects are unique in that the destroy_ah verb
+		 * can be called in atomic context. This delay
+		 * replaces the wait_for_completion call above
+		 * when the destroy_ah call is not sleepable
+		 */
+		while (!completion_done(&elem->complete) &&
+				time_before(jiffies, until))
+			mdelay(1);
+
+		if (WARN_ON(!completion_done(&elem->complete)))
+			err = -EINVAL;
+	}
 
 	if (pool->cleanup)
 		pool->cleanup(elem);
 
 	if (pool->type == RXE_TYPE_MR)
-		kfree(elem->obj);
+		kfree_rcu(elem->obj);
 
 	atomic_dec(&pool->num_elem);
+
+	return err;
 }
 
 int __rxe_get(struct rxe_pool_elem *elem)
···
 int __rxe_put(struct rxe_pool_elem *elem)
 {
 	return kref_put(&elem->ref_cnt, rxe_elem_release);
+}
+
+void __rxe_finalize(struct rxe_pool_elem *elem)
+{
+	void *xa_ret;
+
+	xa_ret = xa_store(&elem->pool->xa, elem->index, elem, GFP_KERNEL);
+	WARN_ON(xa_err(xa_ret));
 }

drivers/infiniband/sw/rxe/rxe_pool.h

···
 	void			*obj;
 	struct kref		ref_cnt;
 	struct list_head	list;
+	struct completion	complete;
 	u32			index;
 };
 
···
 void *rxe_alloc(struct rxe_pool *pool);
 
 /* connect already allocated object to pool */
-int __rxe_add_to_pool(struct rxe_pool *pool, struct rxe_pool_elem *elem);
-
-#define rxe_add_to_pool(pool, obj) __rxe_add_to_pool(pool, &(obj)->elem)
+int __rxe_add_to_pool(struct rxe_pool *pool, struct rxe_pool_elem *elem,
+				bool sleepable);
+#define rxe_add_to_pool(pool, obj) __rxe_add_to_pool(pool, &(obj)->elem, true)
+#define rxe_add_to_pool_ah(pool, obj, sleepable) __rxe_add_to_pool(pool, \
+				&(obj)->elem, sleepable)
 
 /* lookup an indexed object from index. takes a reference on object */
 void *rxe_pool_get_index(struct rxe_pool *pool, u32 index);
 
 int __rxe_get(struct rxe_pool_elem *elem);
-
 #define rxe_get(obj) __rxe_get(&(obj)->elem)
 
 int __rxe_put(struct rxe_pool_elem *elem);
-
 #define rxe_put(obj) __rxe_put(&(obj)->elem)
 
+int __rxe_cleanup(struct rxe_pool_elem *elem, bool sleepable);
+#define rxe_cleanup(obj) __rxe_cleanup(&(obj)->elem, true)
+#define rxe_cleanup_ah(obj, sleepable) __rxe_cleanup(&(obj)->elem, sleepable)
+
 #define rxe_read(obj) kref_read(&(obj)->elem.ref_cnt)
+
+void __rxe_finalize(struct rxe_pool_elem *elem);
+#define rxe_finalize(obj) __rxe_finalize(&(obj)->elem)
 
 #endif /* RXE_POOL_H */

drivers/infiniband/sw/rxe/rxe_qp.c

···
 		for (i = 0; i < qp->attr.max_dest_rd_atomic; i++) {
 			struct resp_res *res = &qp->resp.resources[i];
 
-			free_rd_atomic_resource(qp, res);
+			free_rd_atomic_resource(res);
 		}
 		kfree(qp->resp.resources);
 		qp->resp.resources = NULL;
 	}
 }
 
-void free_rd_atomic_resource(struct rxe_qp *qp, struct resp_res *res)
+void free_rd_atomic_resource(struct resp_res *res)
 {
-	if (res->type == RXE_ATOMIC_MASK)
-		kfree_skb(res->atomic.skb);
 	res->type = 0;
 }
 
···
 	if (qp->resp.resources) {
 		for (i = 0; i < qp->attr.max_dest_rd_atomic; i++) {
 			res = &qp->resp.resources[i];
-			free_rd_atomic_resource(qp, res);
+			free_rd_atomic_resource(res);
 		}
 	}
 }
···
 	}
 
 	spin_lock_init(&qp->state_lock);
+
+	spin_lock_init(&qp->req.task.state_lock);
+	spin_lock_init(&qp->resp.task.state_lock);
+	spin_lock_init(&qp->comp.task.state_lock);
+
+	spin_lock_init(&qp->sq.sq_lock);
+	spin_lock_init(&qp->rq.producer_lock);
+	spin_lock_init(&qp->rq.consumer_lock);
 
 	atomic_set(&qp->ssn, 0);
 	atomic_set(&qp->skb_out, 0);
···
 					       QUEUE_TYPE_FROM_CLIENT);
 
 	qp->req.state		= QP_STATE_RESET;
+	qp->comp.state		= QP_STATE_RESET;
 	qp->req.opcode		= -1;
 	qp->comp.opcode		= -1;
 
-	spin_lock_init(&qp->sq.sq_lock);
 	skb_queue_head_init(&qp->req_pkts);
 
 	rxe_init_task(rxe, &qp->req.task, qp,
···
 			return err;
 		}
 	}
-
-	spin_lock_init(&qp->rq.producer_lock);
-	spin_lock_init(&qp->rq.consumer_lock);
 
 	skb_queue_head_init(&qp->resp_pkts);
 
···
 
 	/* move qp to the reset state */
 	qp->req.state = QP_STATE_RESET;
+	qp->comp.state = QP_STATE_RESET;
 	qp->resp.state = QP_STATE_RESET;
 
 	/* let state machines reset themselves drain work and packet queues
···
 	atomic_set(&qp->ssn, 0);
 	qp->req.opcode = -1;
 	qp->req.need_retry = 0;
+	qp->req.wait_for_rnr_timer = 0;
 	qp->req.noack_pkts = 0;
 	qp->resp.msn = 0;
 	qp->resp.opcode = -1;
···
 {
 	qp->req.state = QP_STATE_ERROR;
 	qp->resp.state = QP_STATE_ERROR;
+	qp->comp.state = QP_STATE_ERROR;
 	qp->attr.qp_state = IB_QPS_ERR;
 
 	/* drain work and packet queues */
···
 			pr_debug("qp#%d state -> INIT\n", qp_num(qp));
 			qp->req.state = QP_STATE_INIT;
 			qp->resp.state = QP_STATE_INIT;
+			qp->comp.state = QP_STATE_INIT;
 			break;
 
 		case IB_QPS_RTR:
···
 		case IB_QPS_RTS:
 			pr_debug("qp#%d state -> RTS\n", qp_num(qp));
 			qp->req.state = QP_STATE_READY;
+			qp->comp.state = QP_STATE_READY;
 			break;
 
 		case IB_QPS_SQD:
···
 	if (qp->rq.queue)
 		rxe_queue_cleanup(qp->rq.queue);
 
-	atomic_dec(&qp->scq->num_wq);
-	if (qp->scq)
+	if (qp->scq) {
+		atomic_dec(&qp->scq->num_wq);
 		rxe_put(qp->scq);
+	}
 
-	atomic_dec(&qp->rcq->num_wq);
-	if (qp->rcq)
+	if (qp->rcq) {
+		atomic_dec(&qp->rcq->num_wq);
 		rxe_put(qp->rcq);
+	}
 
 	if (qp->pd)
 		rxe_put(qp->pd);

drivers/infiniband/sw/rxe/rxe_queue.h

···
 #ifndef RXE_QUEUE_H
 #define RXE_QUEUE_H
 
-/* for definition of shared struct rxe_queue_buf */
-#include <uapi/rdma/rdma_user_rxe.h>
-
 /* Implements a simple circular buffer that is shared between user
  * and the driver and can be resized. The requested element size is
  * rounded up to a power of 2 and the number of elements in the buffer
···
 	QUEUE_TYPE_TO_DRIVER,
 	QUEUE_TYPE_FROM_DRIVER,
 };
+
+struct rxe_queue_buf;
 
 struct rxe_queue {
 	struct rxe_dev		*rxe;

drivers/infiniband/sw/rxe/rxe_req.c

···
 		       u32 opcode);
 
 static inline void retry_first_write_send(struct rxe_qp *qp,
-					  struct rxe_send_wqe *wqe,
-					  unsigned int mask, int npsn)
+					  struct rxe_send_wqe *wqe, int npsn)
 {
 	int i;
 
···
 			if (mask & WR_WRITE_OR_SEND_MASK) {
 				npsn = (qp->comp.psn - wqe->first_psn) &
 					BTH_PSN_MASK;
-				retry_first_write_send(qp, wqe, mask, npsn);
+				retry_first_write_send(qp, wqe, npsn);
 			}
 
 			if (mask & WR_READ_MASK) {
···
 {
 	struct rxe_qp *qp = from_timer(qp, t, rnr_nak_timer);
 
-	pr_debug("qp#%d rnr nak timer fired\n", qp_num(qp));
+	pr_debug("%s: fired for qp#%d\n", __func__, qp_num(qp));
+
+	/* request a send queue retry */
+	qp->req.need_retry = 1;
+	qp->req.wait_for_rnr_timer = 0;
 	rxe_run_task(&qp->req.task, 1);
 }
 
···
 		     (wqe->state != wqe_state_processing)))
 		return NULL;
 
-	if (unlikely((wqe->wr.send_flags & IB_SEND_FENCE) &&
-						     (index != cons))) {
-		qp->req.wait_fence = 1;
-		return NULL;
-	}
-
 	wqe->mask = wr_opcode_mask(wqe->wr.opcode, qp);
 	return wqe;
+}
+
+/**
+ * rxe_wqe_is_fenced - check if next wqe is fenced
+ * @qp: the queue pair
+ * @wqe: the next wqe
+ *
+ * Returns: 1 if wqe needs to wait
+ *	    0 if wqe is ready to go
+ */
+static int rxe_wqe_is_fenced(struct rxe_qp *qp, struct rxe_send_wqe *wqe)
+{
+	/* Local invalidate fence (LIF) see IBA 10.6.5.1
+	 * Requires ALL previous operations on the send queue
+	 * are complete. Make mandatory for the rxe driver.
+	 */
+	if (wqe->wr.opcode == IB_WR_LOCAL_INV)
+		return qp->req.wqe_index != queue_get_consumer(qp->sq.queue,
+						QUEUE_TYPE_FROM_CLIENT);
+
+	/* Fence see IBA 10.8.3.3
+	 * Requires that all previous read and atomic operations
+	 * are complete.
+	 */
+	return (wqe->wr.send_flags & IB_SEND_FENCE) &&
+		atomic_read(&qp->req.rd_atomic) != qp->attr.max_rd_atomic;
 }
 
 static int next_opcode_rc(struct rxe_qp *qp, u32 opcode, int fits)
···
 	wqe->status = IB_WC_SUCCESS;
 	qp->req.wqe_index = queue_next_index(qp->sq.queue, qp->req.wqe_index);
 
-	if ((wqe->wr.send_flags & IB_SEND_SIGNALED) ||
-	    qp->sq_sig_type == IB_SIGNAL_ALL_WR)
-		rxe_run_task(&qp->comp.task, 1);
+	/* There is no ack coming for local work requests
+	 * which can lead to a deadlock. So go ahead and complete
+	 * it now.
+	 */
+	rxe_run_task(&qp->comp.task, 1);
 
 	return 0;
 }
···
 	u32 payload;
 	int mtu;
 	int opcode;
+	int err;
 	int ret;
 	struct rxe_send_wqe rollback_wqe;
 	u32 rollback_psn;
···
 	if (!rxe_get(qp))
 		return -EAGAIN;
 
-next_wqe:
-	if (unlikely(!qp->valid || qp->req.state == QP_STATE_ERROR))
+	if (unlikely(!qp->valid))
 		goto exit;
+
+	if (unlikely(qp->req.state == QP_STATE_ERROR)) {
+		wqe = req_next_wqe(qp);
+		if (wqe)
+			/*
+			 * Generate an error completion for error qp state
+			 */
+			goto err;
+		else
+			goto exit;
+	}
 
 	if (unlikely(qp->req.state == QP_STATE_RESET)) {
 		qp->req.wqe_index = queue_get_consumer(q,
···
 		qp->req.need_rd_atomic = 0;
 		qp->req.wait_psn = 0;
 		qp->req.need_retry = 0;
+		qp->req.wait_for_rnr_timer = 0;
 		goto exit;
 	}
 
-	if (unlikely(qp->req.need_retry)) {
+	/* we come here if the retransmit timer has fired
+	 * or if the rnr timer has fired. If the retransmit
+	 * timer fires while we are processing an RNR NAK wait
+	 * until the rnr timer has fired before starting the
+	 * retry flow
+	 */
+	if (unlikely(qp->req.need_retry && !qp->req.wait_for_rnr_timer)) {
 		req_retry(qp);
 		qp->req.need_retry = 0;
 	}
···
 	if (unlikely(!wqe))
 		goto exit;
 
+	if (rxe_wqe_is_fenced(qp, wqe)) {
+		qp->req.wait_fence = 1;
+		goto exit;
+	}
+
 	if (wqe->mask & WR_LOCAL_OP_MASK) {
-		ret = rxe_do_local_ops(qp, wqe);
-		if (unlikely(ret))
+		err = rxe_do_local_ops(qp, wqe);
+		if (unlikely(err))
 			goto err;
 		else
-			goto next_wqe;
+			goto done;
 	}
 
 	if (unlikely(qp_type(qp) == IB_QPT_RC &&
···
 						       qp->req.wqe_index);
 			wqe->state = wqe_state_done;
 			wqe->status = IB_WC_SUCCESS;
-			__rxe_do_task(&qp->comp.task);
-			rxe_put(qp);
-			return 0;
+			rxe_run_task(&qp->comp.task, 0);
+			goto done;
 		}
 		payload = mtu;
 	}
···
 	if (unlikely(!av)) {
 		pr_err("qp#%d Failed no address vector\n", qp_num(qp));
 		wqe->status = IB_WC_LOC_QP_OP_ERR;
-		goto err_drop_ah;
+		goto err;
 	}
 
 	skb = init_req_packet(qp, av, wqe, opcode, payload, &pkt);
 	if (unlikely(!skb)) {
 		pr_err("qp#%d Failed allocating skb\n", qp_num(qp));
 		wqe->status = IB_WC_LOC_QP_OP_ERR;
-		goto err_drop_ah;
+		if (ah)
+			rxe_put(ah);
+		goto err;
 	}
 
-	ret = finish_packet(qp, av, wqe, &pkt, skb, payload);
-	if (unlikely(ret)) {
+	err = finish_packet(qp, av, wqe, &pkt, skb, payload);
+	if (unlikely(err)) {
 		pr_debug("qp#%d Error during finish packet\n", qp_num(qp));
-		if (ret == -EFAULT)
+		if (err == -EFAULT)
 			wqe->status = IB_WC_LOC_PROT_ERR;
 		else
 			wqe->status = IB_WC_LOC_QP_OP_ERR;
 		kfree_skb(skb);
-		goto err_drop_ah;
+		if (ah)
+			rxe_put(ah);
+		goto err;
 	}
 
 	if (ah)
···
 	save_state(wqe, qp, &rollback_wqe, &rollback_psn);
 	update_wqe_state(qp, wqe, &pkt);
 	update_wqe_psn(qp, wqe, &pkt, payload);
-	ret = rxe_xmit_packet(qp, &pkt, skb);
-	if (ret) {
+
+	err = rxe_xmit_packet(qp, &pkt, skb);
+	if (err) {
 		qp->need_req_skb = 1;
 
 		rollback_state(wqe, qp, &rollback_wqe, rollback_psn);
 
-		if (ret == -EAGAIN) {
+		if (err == -EAGAIN) {
 			rxe_run_task(&qp->req.task, 1);
 			goto exit;
 		}
···
 
 	update_state(qp, &pkt);
 
-	goto next_wqe;
-
-err_drop_ah:
-	if (ah)
-		rxe_put(ah);
+	/* A non-zero return value will cause rxe_do_task to
+	 * exit its loop and end the tasklet. A zero return
+	 * will continue looping and return to rxe_requester
+	 */
+done:
+	ret = 0;
+	goto out;
 err:
+	/* update wqe_index for each wqe completion */
+	qp->req.wqe_index = queue_next_index(qp->sq.queue, qp->req.wqe_index);
 	wqe->state = wqe_state_error;
-	__rxe_do_task(&qp->comp.task);
-
+	qp->req.state = QP_STATE_ERROR;
+	rxe_run_task(&qp->comp.task, 0);
 exit:
+	ret = -EAGAIN;
+out:
 	rxe_put(qp);
-	return -EAGAIN;
+
+	return ret;
 }

drivers/infiniband/sw/rxe/rxe_resp.c

···
 	RESPST_CHK_RKEY,
 	RESPST_EXECUTE,
 	RESPST_READ_REPLY,
+	RESPST_ATOMIC_REPLY,
 	RESPST_COMPLETE,
 	RESPST_ACKNOWLEDGE,
 	RESPST_CLEANUP,
···
 	[RESPST_CHK_RKEY]			= "CHK_RKEY",
 	[RESPST_EXECUTE]			= "EXECUTE",
 	[RESPST_READ_REPLY]			= "READ_REPLY",
+	[RESPST_ATOMIC_REPLY]			= "ATOMIC_REPLY",
 	[RESPST_COMPLETE]			= "COMPLETE",
 	[RESPST_ACKNOWLEDGE]			= "ACKNOWLEDGE",
 	[RESPST_CLEANUP]			= "CLEANUP",
···
 	if (rkey_is_mw(rkey)) {
 		mw = rxe_lookup_mw(qp, access, rkey);
 		if (!mw) {
-			pr_err("%s: no MW matches rkey %#x\n", __func__, rkey);
+			pr_debug("%s: no MW matches rkey %#x\n",
+					__func__, rkey);
 			state = RESPST_ERR_RKEY_VIOLATION;
 			goto err;
 		}
···
 	} else {
 		mr = lookup_mr(qp->pd, access, rkey, RXE_LOOKUP_REMOTE);
 		if (!mr) {
-			pr_err("%s: no MR matches rkey %#x\n", __func__, rkey);
+			pr_debug("%s: no MR matches rkey %#x\n",
+					__func__, rkey);
 			state = RESPST_ERR_RKEY_VIOLATION;
 			goto err;
 		}
···
 	return rc;
 }
 
+static struct resp_res *rxe_prepare_res(struct rxe_qp *qp,
+					struct rxe_pkt_info *pkt,
+					int type)
+{
+	struct resp_res *res;
+	u32 pkts;
+
+	res = &qp->resp.resources[qp->resp.res_head];
+	rxe_advance_resp_resource(qp);
+	free_rd_atomic_resource(res);
+
+	res->type = type;
+	res->replay = 0;
+
+	switch (type) {
+	case RXE_READ_MASK:
+		res->read.va = qp->resp.va + qp->resp.offset;
+		res->read.va_org = qp->resp.va + qp->resp.offset;
+		res->read.resid = qp->resp.resid;
+		res->read.length = qp->resp.resid;
+		res->read.rkey = qp->resp.rkey;
+
+		pkts = max_t(u32, (reth_len(pkt) + qp->mtu - 1)/qp->mtu, 1);
+		res->first_psn = pkt->psn;
+		res->cur_psn = pkt->psn;
+		res->last_psn = (pkt->psn + pkts - 1) & BTH_PSN_MASK;
+
+		res->state = rdatm_res_state_new;
+		break;
+	case RXE_ATOMIC_MASK:
+		res->first_psn = pkt->psn;
+		res->last_psn = pkt->psn;
+		res->cur_psn = pkt->psn;
+		break;
+	}
+
+	return res;
+}
+
 /* Guarantee atomicity of atomic operations at the machine level. */
 static DEFINE_SPINLOCK(atomic_ops_lock);
 
-static enum resp_states process_atomic(struct rxe_qp *qp,
-				       struct rxe_pkt_info *pkt)
+static enum resp_states atomic_reply(struct rxe_qp *qp,
+					 struct rxe_pkt_info *pkt)
 {
 	u64 *vaddr;
 	enum resp_states ret;
 	struct rxe_mr *mr = qp->resp.mr;
+	struct resp_res *res = qp->resp.res;
+	u64 value;
 
-	if (mr->state != RXE_MR_STATE_VALID) {
-		ret = RESPST_ERR_RKEY_VIOLATION;
-		goto out;
+	if (!res) {
+		res = rxe_prepare_res(qp, pkt, RXE_ATOMIC_MASK);
+		qp->resp.res = res;
 	}
 
-	vaddr = iova_to_vaddr(mr, qp->resp.va + qp->resp.offset, sizeof(u64));
+	if (!res->replay) {
+		if (mr->state != RXE_MR_STATE_VALID) {
+			ret = RESPST_ERR_RKEY_VIOLATION;
+			goto out;
+		}
 
-	/* check vaddr is 8 bytes aligned. */
-	if (!vaddr || (uintptr_t)vaddr & 7) {
-		ret = RESPST_ERR_MISALIGNED_ATOMIC;
-		goto out;
+		vaddr = iova_to_vaddr(mr, qp->resp.va + qp->resp.offset,
+					sizeof(u64));
+
+		/* check vaddr is 8 bytes aligned. */
+		if (!vaddr || (uintptr_t)vaddr & 7) {
+			ret = RESPST_ERR_MISALIGNED_ATOMIC;
+			goto out;
+		}
+
+		spin_lock_bh(&atomic_ops_lock);
+		res->atomic.orig_val = value = *vaddr;
+
+		if (pkt->opcode == IB_OPCODE_RC_COMPARE_SWAP) {
+			if (value == atmeth_comp(pkt))
+				value = atmeth_swap_add(pkt);
+		} else {
+			value += atmeth_swap_add(pkt);
+		}
+
+		*vaddr = value;
+		spin_unlock_bh(&atomic_ops_lock);
+
+		qp->resp.msn++;
+
+		/* next expected psn, read handles this separately */
+		qp->resp.psn = (pkt->psn + 1) & BTH_PSN_MASK;
+		qp->resp.ack_psn = qp->resp.psn;
+
+		qp->resp.opcode = pkt->opcode;
+		qp->resp.status = IB_WC_SUCCESS;
 	}
 
-	spin_lock_bh(&atomic_ops_lock);
-
-	qp->resp.atomic_orig = *vaddr;
-
-	if (pkt->opcode == IB_OPCODE_RC_COMPARE_SWAP) {
-		if (*vaddr == atmeth_comp(pkt))
-			*vaddr = atmeth_swap_add(pkt);
-	} else {
-		*vaddr += atmeth_swap_add(pkt);
-	}
-
-	spin_unlock_bh(&atomic_ops_lock);
-
-	ret = RESPST_NONE;
+	ret = RESPST_ACKNOWLEDGE;
 out:
 	return ret;
 }
 
 static struct sk_buff *prepare_ack_packet(struct rxe_qp *qp,
-					  struct rxe_pkt_info *pkt,
 					  struct rxe_pkt_info *ack,
 					  int opcode,
 					  int payload,
···
 	}
 
 	if (ack->mask & RXE_ATMACK_MASK)
-		atmack_set_orig(ack, qp->resp.atomic_orig);
+		atmack_set_orig(ack, qp->resp.res->atomic.orig_val);
 
 	err = rxe_prepare(&qp->pri_av, ack, skb);
 	if (err) {
···
 	}
 
 	return skb;
-}
-
-static struct resp_res *rxe_prepare_read_res(struct rxe_qp *qp,
-					struct rxe_pkt_info *pkt)
-{
-	struct resp_res *res;
-	u32 pkts;
-
-	res = &qp->resp.resources[qp->resp.res_head];
-	rxe_advance_resp_resource(qp);
-	free_rd_atomic_resource(qp, res);
-
-	res->type = RXE_READ_MASK;
-	res->replay = 0;
-	res->read.va = qp->resp.va + qp->resp.offset;
-	res->read.va_org = qp->resp.va + qp->resp.offset;
-	res->read.resid = qp->resp.resid;
-	res->read.length = qp->resp.resid;
-	res->read.rkey = qp->resp.rkey;
-
-	pkts = max_t(u32, (reth_len(pkt) + qp->mtu - 1)/qp->mtu, 1);
-	res->first_psn = pkt->psn;
-	res->cur_psn = pkt->psn;
-	res->last_psn = (pkt->psn + pkts - 1) & BTH_PSN_MASK;
-
-	res->state = rdatm_res_state_new;
-
-	return res;
 }
 
 /**
···
 	struct rxe_mr *mr;
 
 	if (!res) {
-		res = rxe_prepare_read_res(qp, req_pkt);
+		res = rxe_prepare_res(qp, req_pkt, RXE_READ_MASK);
 		qp->resp.res = res;
 	}
 
···
 
 	payload = min_t(int, res->read.resid, mtu);
 
-	skb = prepare_ack_packet(qp, req_pkt, &ack_pkt, opcode, payload,
+	skb = prepare_ack_packet(qp, &ack_pkt, opcode, payload,
 				 res->cur_psn, AETH_ACK_UNLIMITED);
 	if (!skb)
 		return RESPST_ERR_RNR;
···
 		qp->resp.msn++;
 		return RESPST_READ_REPLY;
 	} else if (pkt->mask & RXE_ATOMIC_MASK) {
-		err = process_atomic(qp, pkt);
-		if (err)
-			return err;
+		return RESPST_ATOMIC_REPLY;
 	} else {
 		/* Unreachable */
 		WARN_ON_ONCE(1);
···
 		return RESPST_CLEANUP;
 }
 
-static int send_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt,
-		    u8 syndrome, u32 psn)
+static int send_ack(struct rxe_qp *qp, u8 syndrome, u32 psn)
 {
 	int err = 0;
 	struct rxe_pkt_info ack_pkt;
 	struct sk_buff *skb;
 
-	skb = prepare_ack_packet(qp, pkt, &ack_pkt, IB_OPCODE_RC_ACKNOWLEDGE,
+	skb = prepare_ack_packet(qp, &ack_pkt, IB_OPCODE_RC_ACKNOWLEDGE,
 				 0, psn, syndrome);
 	if (!skb) {
 		err = -ENOMEM;
···
 	return err;
 }
 
-static int send_atomic_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt,
-			   u8 syndrome)
+static int send_atomic_ack(struct rxe_qp *qp, u8 syndrome, u32 psn)
 {
-	int rc = 0;
+	int err = 0;
 	struct rxe_pkt_info ack_pkt;
 	struct sk_buff *skb;
-	struct resp_res *res;
 
-	skb = prepare_ack_packet(qp, pkt, &ack_pkt,
-				 IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE, 0, pkt->psn,
-				 syndrome);
+	skb = prepare_ack_packet(qp, &ack_pkt, IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE,
+				 0, psn, syndrome);
 	if (!skb) {
-		rc = -ENOMEM;
+		err = -ENOMEM;
 		goto out;
 	}
 
-	res = &qp->resp.resources[qp->resp.res_head];
-	free_rd_atomic_resource(qp, res);
-	rxe_advance_resp_resource(qp);
+	err = rxe_xmit_packet(qp, &ack_pkt, skb);
+	if (err)
+		pr_err_ratelimited("Failed sending atomic ack\n");
 
-	skb_get(skb);
-	res->type = RXE_ATOMIC_MASK;
-	res->atomic.skb = skb;
-	res->first_psn = ack_pkt.psn;
-	res->last_psn  = ack_pkt.psn;
-	res->cur_psn   = ack_pkt.psn;
-
-	rc = rxe_xmit_packet(qp, &ack_pkt, skb);
-	if (rc) {
-		pr_err_ratelimited("Failed sending ack\n");
-		rxe_put(qp);
-	}
+	/* have to clear this since it is used to trigger
+	 * long read replies
+	 */
+	qp->resp.res = NULL;
 out:
-	return rc;
+	return err;
 }
 
 static enum resp_states acknowledge(struct rxe_qp *qp,
···
 		return RESPST_CLEANUP;
 
 	if (qp->resp.aeth_syndrome != AETH_ACK_UNLIMITED)
-		send_ack(qp, pkt, qp->resp.aeth_syndrome, pkt->psn);
+		send_ack(qp, qp->resp.aeth_syndrome, pkt->psn);
 	else if (pkt->mask & RXE_ATOMIC_MASK)
-		send_atomic_ack(qp, pkt, AETH_ACK_UNLIMITED);
+		send_atomic_ack(qp, AETH_ACK_UNLIMITED, pkt->psn);
 	else if (bth_ack(pkt))
-		send_ack(qp, pkt, AETH_ACK_UNLIMITED, pkt->psn);
+		send_ack(qp, AETH_ACK_UNLIMITED, pkt->psn);
 
 	return RESPST_CLEANUP;
 }
···
 	if (pkt->mask & RXE_SEND_MASK ||
 	    pkt->mask & RXE_WRITE_MASK) {
 		/* SEND. Ack again and cleanup. C9-105. */
-		send_ack(qp, pkt, AETH_ACK_UNLIMITED, prev_psn);
+		send_ack(qp, AETH_ACK_UNLIMITED, prev_psn);
 		return RESPST_CLEANUP;
 	} else if (pkt->mask & RXE_READ_MASK) {
 		struct resp_res *res;
···
 		/* Find the operation in our list of responder resources. */
 		res = find_resource(qp, pkt->psn);
 		if (res) {
-			skb_get(res->atomic.skb);
-			/* Resend the result. */
-			rc = rxe_xmit_packet(qp, pkt, res->atomic.skb);
-			if (rc) {
-				pr_err("Failed resending result. This flow is not handled - skb ignored\n");
-				rc = RESPST_CLEANUP;
-				goto out;
-			}
+			res->replay = 1;
+			res->cur_psn = pkt->psn;
+			qp->resp.res = res;
+			rc = RESPST_ATOMIC_REPLY;
+			goto out;
 		}
 
 		/* Resource not found. Class D error. Drop the request. */
···
 	struct rxe_dev *rxe = to_rdev(qp->ibqp.device);
 	enum resp_states state;
 	struct rxe_pkt_info *pkt = NULL;
-	int ret = 0;
+	int ret;
 
 	if (!rxe_get(qp))
 		return -EAGAIN;
 
 	qp->resp.aeth_syndrome = AETH_ACK_UNLIMITED;
 
-	if (!qp->valid) {
-		ret = -EINVAL;
-		goto done;
-	}
+	if (!qp->valid)
+		goto exit;
 
 	switch (qp->resp.state) {
 	case QP_STATE_RESET:
···
 		case RESPST_READ_REPLY:
 			state = read_reply(qp, pkt);
 			break;
+		case RESPST_ATOMIC_REPLY:
+			state = atomic_reply(qp, pkt);
+			break;
 		case RESPST_ACKNOWLEDGE:
 			state = acknowledge(qp, pkt);
 			break;
···
 			break;
 		case RESPST_ERR_PSN_OUT_OF_SEQ:
 			/* RC only - Class B. Drop packet. */
-			send_ack(qp, pkt, AETH_NAK_PSN_SEQ_ERROR, qp->resp.psn);
+			send_ack(qp, AETH_NAK_PSN_SEQ_ERROR, qp->resp.psn);
 			state = RESPST_CLEANUP;
 			break;
 
···
 			if (qp_type(qp) == IB_QPT_RC) {
 				rxe_counter_inc(rxe, RXE_CNT_SND_RNR);
 				/* RC - class B */
-				send_ack(qp, pkt, AETH_RNR_NAK |
+				send_ack(qp, AETH_RNR_NAK |
 					 (~AETH_TYPE_MASK &
 					 qp->attr.min_rnr_timer),
 					 pkt->psn);
···
 
 		case RESPST_ERROR:
 			qp->resp.goto_error = 0;
-			pr_warn("qp#%d moved to error state\n", qp_num(qp));
+			pr_debug("qp#%d moved to error state\n", qp_num(qp));
 			rxe_qp_error(qp);
 			goto exit;
 
···
 		}
 	}
 
+	/* A non-zero return value will cause rxe_do_task to
+	 * exit its loop and end the tasklet. A zero return
+	 * will continue looping and return to rxe_responder
+	 */
+done:
+	ret = 0;
+	goto out;
 exit:
 	ret = -EAGAIN;
-done:
+out:
 	rxe_put(qp);
 	return ret;
 }

drivers/infiniband/sw/rxe/rxe_task.c

···
 #include <linux/interrupt.h>
 #include <linux/hardirq.h>
 
-#include "rxe_task.h"
+#include "rxe.h"
 
 int __rxe_do_task(struct rxe_task *task)
 
···
 	int cont;
 	int ret;
 	struct rxe_task *task = from_tasklet(task, t, tasklet);
+	unsigned int iterations = RXE_MAX_ITERATIONS;
 
 	spin_lock_bh(&task->state_lock);
 	switch (task->state) {
···
 		spin_lock_bh(&task->state_lock);
 		switch (task->state) {
 		case TASK_STATE_BUSY:
-			if (ret)
+			if (ret) {
 				task->state = TASK_STATE_START;
-			else
+			} else if (iterations--) {
 				cont = 1;
+			} else {
+				/* reschedule the tasklet and exit
+				 * the loop to give up the cpu
+				 */
+				tasklet_schedule(&task->tasklet);
+				task->state = TASK_STATE_START;
+			}
 			break;
 
-		/* soneone tried to run the task since the last time we called
+		/* someone tried to run the task since the last time we called
 		 * func, so we will call one more time regardless of the
 		 * return value
 		 */

drivers/infiniband/sw/rxe/rxe_verbs.c

···
 {
 	struct rxe_ucontext *uc = to_ruc(ibuc);
 
-	rxe_put(uc);
+	rxe_cleanup(uc);
 }
 
 static int rxe_port_immutable(struct ib_device *dev, u32 port_num,
···
 {
 	struct rxe_pd *pd = to_rpd(ibpd);
 
-	rxe_put(pd);
+	rxe_cleanup(pd);
 	return 0;
 }
 
···
 	if (err)
 		return err;
 
-	err = rxe_add_to_pool(&rxe->ah_pool, ah);
+	err = rxe_add_to_pool_ah(&rxe->ah_pool, ah,
+			init_attr->flags & RDMA_CREATE_AH_SLEEPABLE);
 	if (err)
 		return err;
 
···
 		err = copy_to_user(&uresp->ah_num, &ah->ah_num,
 					 sizeof(uresp->ah_num));
 		if (err) {
-			rxe_put(ah);
+			rxe_cleanup(ah);
 			return -EFAULT;
 		}
 	} else if (ah->is_user) {
···
 	}
 
 	rxe_init_av(init_attr->ah_attr, &ah->av);
+	rxe_finalize(ah);
+
 	return 0;
 }
 
···
 {
 	struct rxe_ah *ah = to_rah(ibah);
 
-	rxe_put(ah);
+	rxe_cleanup_ah(ah, flags & RDMA_DESTROY_AH_SLEEPABLE);
+
 	return 0;
 }
 
···
 
 	err = rxe_srq_from_init(rxe, srq, init, udata, uresp);
 	if (err)
-		goto err_put;
+		goto err_cleanup;
 
 	return 0;
 
-err_put:
-	rxe_put(srq);
+err_cleanup:
+	rxe_cleanup(srq);
+
 	return err;
 }
 
···
 {
 	struct rxe_srq *srq = to_rsrq(ibsrq);
 
-	rxe_put(srq);
+	rxe_cleanup(srq);
 	return 0;
 }
 
···
 	if (err)
 		goto qp_init;
 
+	rxe_finalize(qp);
 	return 0;
 
 qp_init:
-	rxe_put(qp);
+	rxe_cleanup(qp);
 	return err;
 }
 
···
 	if (ret)
 		return ret;
 
-	rxe_put(qp);
+	rxe_cleanup(qp);
 	return 0;
 }
 
···
 
 	rxe_cq_disable(cq);
 
-	rxe_put(cq);
+	rxe_cleanup(cq);
 	return 0;
 }
 
···
 
 	rxe_get(pd);
 	rxe_mr_init_dma(pd, access, mr);
+	rxe_finalize(mr);
 
 	return &mr->ibmr;
 }
···
 	if (err)
 		goto err3;
 
+	rxe_finalize(mr);
+
 	return &mr->ibmr;
 
 err3:
 	rxe_put(pd);
-	rxe_put(mr);
+	rxe_cleanup(mr);
 err2:
 	return ERR_PTR(err);
 }
···
 	if (err)
 		goto err2;
 
+	rxe_finalize(mr);
+
 	return &mr->ibmr;
 
 err2:
 	rxe_put(pd);
-	rxe_put(mr);
+	rxe_cleanup(mr);
 err1:
 	return ERR_PTR(err);
 }
 
-/* build next_map_set from scatterlist
- * The IB_WR_REG_MR WR will swap map_sets
- */
+static int rxe_set_page(struct ib_mr *ibmr, u64 addr)
+{
+	struct rxe_mr *mr = to_rmr(ibmr);
+	struct rxe_map *map;
+	struct rxe_phys_buf *buf;
+
+	if (unlikely(mr->nbuf == mr->num_buf))
+		return -ENOMEM;
+
+	map = mr->map[mr->nbuf / RXE_BUF_PER_MAP];
+	buf = &map->buf[mr->nbuf % RXE_BUF_PER_MAP];
+
+	buf->addr = addr;
+	buf->size = ibmr->page_size;
+	mr->nbuf++;
+
+	return 0;
+}
+
 static int rxe_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg,
 			 int sg_nents, unsigned int *sg_offset)
 {
 	struct rxe_mr *mr = to_rmr(ibmr);
-	struct rxe_map_set *set = mr->next_map_set;
 	int n;
 
-	set->nbuf = 0;
+	mr->nbuf = 0;
 
-	n = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, rxe_mr_set_page);
+	n = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, rxe_set_page);
 
-	set->va = ibmr->iova;
-	set->iova = ibmr->iova;
-	set->length = ibmr->length;
-	set->page_shift = ilog2(ibmr->page_size);
-	set->page_mask = ibmr->page_size - 1;
-	set->offset = set->iova & set->page_mask;
+	mr->va = ibmr->iova;
+	mr->iova = ibmr->iova;
+	mr->length = ibmr->length;
+	mr->page_shift = ilog2(ibmr->page_size);
+	mr->page_mask = ibmr->page_size - 1;
+	mr->offset = mr->iova & mr->page_mask;
 
 	return n;
 }

drivers/infiniband/sw/rxe/rxe_verbs.h

···
 
 #include <linux/interrupt.h>
 #include <linux/workqueue.h>
-#include <rdma/rdma_user_rxe.h>
 #include "rxe_pool.h"
 #include "rxe_task.h"
 #include "rxe_hw_counters.h"
···
 	int			need_rd_atomic;
 	int			wait_psn;
 	int			need_retry;
+	int			wait_for_rnr_timer;
 	int			noack_pkts;
 	struct rxe_task		task;
 };
 
 struct rxe_comp_info {
+	enum rxe_qp_state	state;
 	u32			psn;
 	int			opcode;
 	int			timeout;
···
 
 	union {
 		struct {
-			struct sk_buff	*skb;
+			u64		orig_val;
 		} atomic;
 		struct {
 			u64		va_org;
···
 	u32			resid;
 	u32			rkey;
 	u32			length;
-	u64			atomic_orig;
 
 	/* SRQ only */
 	struct {
···
 	struct rxe_phys_buf	buf[RXE_BUF_PER_MAP];
 };
 
-struct rxe_map_set {
-	struct rxe_map		**map;
-	u64			va;
-	u64			iova;
-	size_t			length;
-	u32			offset;
-	u32			nbuf;
-	int			page_shift;
-	int			page_mask;
-};
-
 static inline int rkey_is_mw(u32 rkey)
 {
 	u32 index = rkey >> 8;
···
 	u32			rkey;
 	enum rxe_mr_state	state;
 	enum ib_mr_type		type;
+	u64			va;
+	u64			iova;
+	size_t			length;
+	u32			offset;
 	int			access;
 
+	int			page_shift;
+	int			page_mask;
 	int			map_shift;
 	int			map_mask;
 
 	u32			num_buf;
+	u32			nbuf;
 
 	u32			max_buf;
 	u32			num_map;
 
 	atomic_t		num_mw;
 
-	struct rxe_map_set	*cur_map_set;
-	struct rxe_map_set	*next_map_set;
+	struct rxe_map		**map;
 };
 
 enum rxe_mw_state {

drivers/infiniband/sw/siw/siw_cm.c

···
 	enum mpa_v2_ctrl mpa_p2p_mode = MPA_V2_RDMA_NO_RTR;
 
 	rv = siw_recv_mpa_rr(cep);
-	if (rv != -EAGAIN)
-		siw_cancel_mpatimer(cep);
 	if (rv)
 		goto out_err;
+
+	siw_cancel_mpatimer(cep);
 
 	rep = &cep->mpa.hdr;
 
···
 	}
 
 out_err:
-	siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, -EINVAL);
+	if (rv != -EAGAIN)
+		siw_cm_upcall(cep, IW_CM_EVENT_CONNECT_REPLY, -EINVAL);
 
 	return rv;
 }

drivers/infiniband/sw/siw/siw_verbs.c

···
 err_out:
 	siw_dbg(base_cq->device, "CQ creation failed: %d", rv);
 
-	if (cq && cq->queue) {
+	if (cq->queue) {
 		struct siw_ucontext *ctx =
 			rdma_udata_to_drv_context(udata, struct siw_ucontext,
 						  base_ucontext);

drivers/infiniband/ulp/ipoib/ipoib_ib.c

···
 	 * if he sets the device address back to be based on GID index 0,
 	 * he no longer wishs to control it.
 	 *
-	 * If the user doesn't control the the device address,
+	 * If the user doesn't control the device address,
 	 * IPOIB_FLAG_DEV_ADDR_SET is set and ib_find_gid failed it means
 	 * the port GUID has changed and GID at index 0 has changed
 	 * so we need to change priv->local_gid and priv->dev->dev_addr

drivers/infiniband/ulp/ipoib/ipoib_main.c

···
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 
-	netif_napi_add(dev, &priv->recv_napi, ipoib_rx_poll, IPOIB_NUM_WC);
-	netif_napi_add(dev, &priv->send_napi, ipoib_tx_poll, MAX_SEND_CQE);
+	netif_napi_add_weight(dev, &priv->recv_napi, ipoib_rx_poll,
+			      IPOIB_NUM_WC);
+	netif_napi_add_weight(dev, &priv->send_napi, ipoib_tx_poll,
+			      MAX_SEND_CQE);
 }
 
 static void ipoib_napi_del(struct net_device *dev)

drivers/infiniband/ulp/iser/iser_verbs.c

···
 	device = ib_conn->device;
 	ib_dev = device->ib_device;
 
+	/* +1 for drain */
 	if (ib_conn->pi_support)
 		max_send_wr = ISER_QP_SIG_MAX_REQ_DTOS + 1;
 	else
···
 	init_attr.qp_context = (void *)ib_conn;
 	init_attr.send_cq = ib_conn->cq;
 	init_attr.recv_cq = ib_conn->cq;
-	init_attr.cap.max_recv_wr = ISER_QP_MAX_RECV_DTOS;
+	/* +1 for drain */
+	init_attr.cap.max_recv_wr = ISER_QP_MAX_RECV_DTOS + 1;
 	init_attr.cap.max_send_sge = 2;
 	init_attr.cap.max_recv_sge = 1;
 	init_attr.sq_sig_type = IB_SIGNAL_REQ_WR;
···
 				 iser_conn, err);
 
 		/* block until all flush errors are consumed */
-		ib_drain_sq(ib_conn->qp);
+		ib_drain_qp(ib_conn->qp);
 	}
 
 	return 1;

drivers/infiniband/ulp/rtrs/rtrs-clt-stats.c

···
 
 void rtrs_clt_inc_failover_cnt(struct rtrs_clt_stats *stats)
 {
-	struct rtrs_clt_stats_pcpu *s;
-
-	s = get_cpu_ptr(stats->pcpu_stats);
-	s->rdma.failover_cnt++;
-	put_cpu_ptr(stats->pcpu_stats);
+	this_cpu_inc(stats->pcpu_stats->rdma.failover_cnt);
 }
 
 int rtrs_clt_stats_migration_from_cnt_to_str(struct rtrs_clt_stats *stats, char *buf)
···
 static inline void rtrs_clt_update_rdma_stats(struct rtrs_clt_stats *stats,
 					       size_t size, int d)
 {
-	struct rtrs_clt_stats_pcpu *s;
-
-	s = get_cpu_ptr(stats->pcpu_stats);
-	s->rdma.dir[d].cnt++;
-	s->rdma.dir[d].size_total += size;
-	put_cpu_ptr(stats->pcpu_stats);
+	this_cpu_inc(stats->pcpu_stats->rdma.dir[d].cnt);
+	this_cpu_add(stats->pcpu_stats->rdma.dir[d].size_total, size);
 }
 
 void rtrs_clt_update_all_stats(struct rtrs_clt_io_req *req, int dir)

drivers/infiniband/ulp/rtrs/rtrs-clt.c

···
 	struct rtrs_clt_path *(*next_path)(struct path_it *it);
 };
 
-/**
- * list_next_or_null_rr_rcu - get next list element in round-robin fashion.
+/*
+ * rtrs_clt_get_next_path_or_null - get clt path from the list or return NULL
  * @head:	the head for the list.
- * @ptr:        the list head to take the next element from.
- * @type:       the type of the struct this is embedded in.
- * @memb:       the name of the list_head within the struct.
+ * @clt_path:	The element to take the next clt_path from.
  *
- * Next element returned in round-robin fashion, i.e. head will be skipped,
+ * Next clt path returned in round-robin fashion, i.e. head will be skipped,
  * but if list is observed as empty, NULL will be returned.
  *
- * This primitive may safely run concurrently with the _rcu list-mutation
+ * This function may safely run concurrently with the _rcu list-mutation
  * primitives such as list_add_rcu() as long as it's guarded by rcu_read_lock().
  */
-#define list_next_or_null_rr_rcu(head, ptr, type, memb) \
-({ \
-	list_next_or_null_rcu(head, ptr, type, memb) ?: \
-		list_next_or_null_rcu(head, READ_ONCE((ptr)->next), \
-				      type, memb); \
-})
+static inline struct rtrs_clt_path *
+rtrs_clt_get_next_path_or_null(struct list_head *head, struct rtrs_clt_path *clt_path)
+{
+	return list_next_or_null_rcu(head, &clt_path->s.entry, typeof(*clt_path), s.entry) ?:
+				     list_next_or_null_rcu(head,
+							   READ_ONCE((&clt_path->s.entry)->next),
+							   typeof(*clt_path), s.entry);
+}
 
 /**
  * get_next_path_rr() - Returns path in round-robin fashion.
···
 		path = list_first_or_null_rcu(&clt->paths_list,
 					      typeof(*path), s.entry);
 	else
-		path = list_next_or_null_rr_rcu(&clt->paths_list,
-						&path->s.entry,
-						typeof(*path),
-						s.entry);
+		path = rtrs_clt_get_next_path_or_null(&clt->paths_list, path);
+
 	rcu_assign_pointer(*ppcpu_path, path);
 
 	return path;
···
 	unsigned int chunk_bits;
 	int err, i;
 
-	clt->permits_map = kcalloc(BITS_TO_LONGS(clt->queue_depth),
-				   sizeof(long), GFP_KERNEL);
+	clt->permits_map = bitmap_zalloc(clt->queue_depth, GFP_KERNEL);
 	if (!clt->permits_map) {
 		err = -ENOMEM;
 		goto out_err;
···
 	return 0;
 
 err_map:
-	kfree(clt->permits_map);
+	bitmap_free(clt->permits_map);
 	clt->permits_map = NULL;
 out_err:
 	return err;
···
 
 static void free_permits(struct rtrs_clt_sess *clt)
 {
-	if (clt->permits_map) {
-		size_t sz = clt->queue_depth;
-
+	if (clt->permits_map)
 		wait_event(clt->permits_wait,
-			   find_first_bit(clt->permits_map, sz) >= sz);
-	}
-	kfree(clt->permits_map);
+			   bitmap_empty(clt->permits_map, clt->queue_depth));
+
+	bitmap_free(clt->permits_map);
 	clt->permits_map = NULL;
 	kfree(clt->permits);
 	clt->permits = NULL;
···
 	 * removed.  If @sess is the last element, then @next is NULL.
 	 */
 	rcu_read_lock();
-	next = list_next_or_null_rr_rcu(&clt->paths_list, &clt_path->s.entry,
-					typeof(*next), s.entry);
+	next = rtrs_clt_get_next_path_or_null(&clt->paths_list, clt_path);
 	rcu_read_unlock();
 
 	/*

drivers/infiniband/ulp/rtrs/rtrs-pri.h

···
 #define RTRS_PROTO_VER_STRING __stringify(RTRS_PROTO_VER_MAJOR) "." \
 			       __stringify(RTRS_PROTO_VER_MINOR)
 
+/*
+ * Max IB immediate data size is 2^28 (MAX_IMM_PAYL_BITS)
+ * and the minimum chunk size is 4096 (2^12).
+ * So the maximum sess_queue_depth is 65536 (2^16) in theory.
+ * But mempool_create, create_qp and ib_post_send fail with
+ * "cannot allocate memory" error if sess_queue_depth is too big.
+ * Therefore the pratical max value of sess_queue_depth is
+ * somewhere between 1 and 65534 and it depends on the system.
+ */
+#define MAX_SESS_QUEUE_DEPTH 65535
+
 enum rtrs_imm_const {
 	MAX_IMM_TYPE_BITS = 4,
 	MAX_IMM_TYPE_MASK = ((1 << MAX_IMM_TYPE_BITS) - 1),
···
 
 	MAX_PATHS_NUM = 128,
 
-	/*
-	 * Max IB immediate data size is 2^28 (MAX_IMM_PAYL_BITS)
-	 * and the minimum chunk size is 4096 (2^12).
-	 * So the maximum sess_queue_depth is 65536 (2^16) in theory.
-	 * But mempool_create, create_qp and ib_post_send fail with
-	 * "cannot allocate memory" error if sess_queue_depth is too big.
-	 * Therefore the pratical max value of sess_queue_depth is
-	 * somewhere between 1 and 65534 and it depends on the system.
-	 */
-	MAX_SESS_QUEUE_DEPTH = 65535,
 	MIN_CHUNK_SIZE = 8192,
 
 	RTRS_HB_INTERVAL_MS = 5000,

drivers/infiniband/ulp/rtrs/rtrs-srv-stats.c

···
 int rtrs_srv_reset_rdma_stats(struct rtrs_srv_stats *stats, bool enable)
 {
 	if (enable) {
-		struct rtrs_srv_stats_rdma_stats *r = &stats->rdma_stats;
+		int cpu;
+		struct rtrs_srv_stats_rdma_stats *r;
 
-		memset(r, 0, sizeof(*r));
+		for_each_possible_cpu(cpu) {
+			r = per_cpu_ptr(stats->rdma_stats, cpu);
+			memset(r, 0, sizeof(*r));
+		}
+
 		return 0;
 	}
 
···
 
 ssize_t rtrs_srv_stats_rdma_to_str(struct rtrs_srv_stats *stats, char *page)
 {
-	struct rtrs_srv_stats_rdma_stats *r = &stats->rdma_stats;
+	int cpu;
+	struct rtrs_srv_stats_rdma_stats sum;
+	struct rtrs_srv_stats_rdma_stats *r;
 
-	return sysfs_emit(page, "%lld %lld %lld %lldn %u\n",
-			  (s64)atomic64_read(&r->dir[READ].cnt),
-			  (s64)atomic64_read(&r->dir[READ].size_total),
-			  (s64)atomic64_read(&r->dir[WRITE].cnt),
-			  (s64)atomic64_read(&r->dir[WRITE].size_total), 0);
+	memset(&sum, 0, sizeof(sum));
+
+	for_each_possible_cpu(cpu) {
+		r = per_cpu_ptr(stats->rdma_stats, cpu);
+
+		sum.dir[READ].cnt	  += r->dir[READ].cnt;
+		sum.dir[READ].size_total  += r->dir[READ].size_total;
+		sum.dir[WRITE].cnt	  += r->dir[WRITE].cnt;
+		sum.dir[WRITE].size_total += r->dir[WRITE].size_total;
+	}
+
+	return sysfs_emit(page, "%llu %llu %llu %llu\n",
+			  sum.dir[READ].cnt, sum.dir[READ].size_total,
+			  sum.dir[WRITE].cnt, sum.dir[WRITE].size_total);
 }

drivers/infiniband/ulp/rtrs/rtrs-srv-sysfs.c

···
 
 	stats = container_of(kobj, struct rtrs_srv_stats, kobj_stats);
 
+	free_percpu(stats->rdma_stats);
+
 	kfree(stats);
 }
 

drivers/infiniband/ulp/rtrs/rtrs-srv.c

···
 #define pr_fmt(fmt) KBUILD_MODNAME " L" __stringify(__LINE__) ": " fmt
 
 #include <linux/module.h>
-#include <linux/mempool.h>
 
 #include "rtrs-srv.h"
 #include "rtrs-log.h"
···
 #define DEFAULT_SESS_QUEUE_DEPTH 512
 #define MAX_HDR_SIZE PAGE_SIZE
 
-/* We guarantee to serve 10 paths at least */
-#define CHUNK_POOL_SZ 10
-
 static struct rtrs_rdma_dev_pd dev_pd;
-static mempool_t *chunk_pool;
 struct class *rtrs_dev_class;
 static struct rtrs_srv_ib_ctx ib_ctx;
 
···
 
 	WARN_ON(refcount_read(&srv->refcount));
 	for (i = 0; i < srv->queue_depth; i++)
-		mempool_free(srv->chunks[i], chunk_pool);
+		__free_pages(srv->chunks[i], get_order(max_chunk_size));
 	kfree(srv->chunks);
 	mutex_destroy(&srv->paths_mutex);
 	mutex_destroy(&srv->paths_ev_mutex);
···
 		goto err_free_srv;
 
 	for (i = 0; i < srv->queue_depth; i++) {
-		srv->chunks[i] = mempool_alloc(chunk_pool, GFP_KERNEL);
+		srv->chunks[i] = alloc_pages(GFP_KERNEL,
+					     get_order(max_chunk_size));
 		if (!srv->chunks[i])
 			goto err_free_chunks;
 	}
···
 
 err_free_chunks:
 	while (i--)
-		mempool_free(srv->chunks[i], chunk_pool);
+		__free_pages(srv->chunks[i], get_order(max_chunk_size));
 	kfree(srv->chunks);
 
 err_free_srv:
···
 		kobject_del(&srv_path->kobj);
 		kobject_put(&srv_path->kobj);
 	} else {
+		free_percpu(srv_path->stats->rdma_stats);
 		kfree(srv_path->stats);
 		kfree(srv_path);
 	}
···
 	if (!srv_path->stats)
 		goto err_free_sess;
 
+	srv_path->stats->rdma_stats = alloc_percpu(struct rtrs_srv_stats_rdma_stats);
+	if (!srv_path->stats->rdma_stats)
+		goto err_free_stats;
+
 	srv_path->stats->srv_path = srv_path;
 
 	srv_path->dma_addr = kcalloc(srv->queue_depth,
 				     sizeof(*srv_path->dma_addr),
 				     GFP_KERNEL);
 	if (!srv_path->dma_addr)
-		goto err_free_stats;
+		goto err_free_percpu;
 
 	srv_path->s.con = kcalloc(con_num, sizeof(*srv_path->s.con),
 				  GFP_KERNEL);
···
 	kfree(srv_path->s.con);
 err_free_dma_addr:
 	kfree(srv_path->dma_addr);
+err_free_percpu:
+	free_percpu(srv_path->stats->rdma_stats);
 err_free_stats:
 	kfree(srv_path->stats);
 err_free_sess:
···
 		       err);
 		return err;
 	}
-	chunk_pool = mempool_create_page_pool(sess_queue_depth * CHUNK_POOL_SZ,
-					      get_order(max_chunk_size));
-	if (!chunk_pool)
-		return -ENOMEM;
 	rtrs_dev_class = class_create(THIS_MODULE, "rtrs-server");
 	if (IS_ERR(rtrs_dev_class)) {
 		err = PTR_ERR(rtrs_dev_class);
-		goto out_chunk_pool;
+		goto out_err;
 	}
 	rtrs_wq = alloc_workqueue("rtrs_server_wq", 0, 0);
 	if (!rtrs_wq) {
···
 
 out_dev_class:
 	class_destroy(rtrs_dev_class);
-out_chunk_pool:
-	mempool_destroy(chunk_pool);
-
+out_err:
 	return err;
 }
 
···
 {
 	destroy_workqueue(rtrs_wq);
 	class_destroy(rtrs_dev_class);
-	mempool_destroy(chunk_pool);
 	rtrs_rdma_dev_pd_deinit(&dev_pd);
 }
 

drivers/infiniband/ulp/rtrs/rtrs-srv.h

···
 
 #include <linux/device.h>
 #include <linux/refcount.h>
+#include <linux/percpu.h>
 #include "rtrs-pri.h"
 
 /*
···
  */
 struct rtrs_srv_stats_rdma_stats {
 	struct {
-		atomic64_t	cnt;
-		atomic64_t	size_total;
+		u64 cnt;
+		u64 size_total;
 	} dir[2];
 };
 
 struct rtrs_srv_stats {
-	struct kobject				kobj_stats;
-	struct rtrs_srv_stats_rdma_stats	rdma_stats;
-	struct rtrs_srv_path			*srv_path;
+	struct kobject					kobj_stats;
+	struct rtrs_srv_stats_rdma_stats __percpu	*rdma_stats;
+	struct rtrs_srv_path				*srv_path;
 };
 
 struct rtrs_srv_con {
···
 static inline void rtrs_srv_update_rdma_stats(struct rtrs_srv_stats *s,
 					      size_t size, int d)
 {
-	atomic64_inc(&s->rdma_stats.dir[d].cnt);
-	atomic64_add(size, &s->rdma_stats.dir[d].size_total);
+	this_cpu_inc(s->rdma_stats->dir[d].cnt);
+	this_cpu_add(s->rdma_stats->dir[d].size_total, size);
 }
 
 /* functions which are implemented in rtrs-srv-stats.c */

drivers/infiniband/ulp/srpt/ib_srpt.c

···
 	if (ret)
 		return ret;
 
-	sport->port_guid_id.wwn.priv = sport;
-	srpt_format_guid(sport->port_guid_id.name,
-			 sizeof(sport->port_guid_id.name),
+	srpt_format_guid(sport->guid_name, ARRAY_SIZE(sport->guid_name),
 			 &sport->gid.global.interface_id);
-	sport->port_gid_id.wwn.priv = sport;
-	snprintf(sport->port_gid_id.name, sizeof(sport->port_gid_id.name),
+	snprintf(sport->gid_name, ARRAY_SIZE(sport->gid_name),
 		 "0x%016llx%016llx",
 		 be64_to_cpu(sport->gid.global.subnet_prefix),
 		 be64_to_cpu(sport->gid.global.interface_id));
···
 	ch->zw_cqe.done = srpt_zerolength_write_done;
 	INIT_WORK(&ch->release_work, srpt_release_channel_work);
 	ch->sport = sport;
-	if (ib_cm_id) {
-		ch->ib_cm.cm_id = ib_cm_id;
-		ib_cm_id->context = ch;
-	} else {
+	if (rdma_cm_id) {
 		ch->using_rdma_cm = true;
 		ch->rdma_cm.cm_id = rdma_cm_id;
 		rdma_cm_id->context = ch;
+	} else {
+		ch->ib_cm.cm_id = ib_cm_id;
+		ib_cm_id->context = ch;
 	}
 	/*
 	 * ch->rq_size should be at least as large as the initiator queue
···
 	tag_num = ch->rq_size;
 	tag_size = 1; /* ib_srpt does not use se_sess->sess_cmd_map */
 
-	mutex_lock(&sport->port_guid_id.mutex);
-	list_for_each_entry(stpg, &sport->port_guid_id.tpg_list, entry) {
-		if (!IS_ERR_OR_NULL(ch->sess))
-			break;
-		ch->sess = target_setup_session(&stpg->tpg, tag_num,
+	if (sport->guid_id) {
+		mutex_lock(&sport->guid_id->mutex);
+		list_for_each_entry(stpg, &sport->guid_id->tpg_list, entry) {
+			if (!IS_ERR_OR_NULL(ch->sess))
+				break;
+			ch->sess = target_setup_session(&stpg->tpg, tag_num,
 						tag_size, TARGET_PROT_NORMAL,
 						ch->sess_name, ch, NULL);
+		}
+		mutex_unlock(&sport->guid_id->mutex);
 	}
-	mutex_unlock(&sport->port_guid_id.mutex);
 
-	mutex_lock(&sport->port_gid_id.mutex);
-	list_for_each_entry(stpg, &sport->port_gid_id.tpg_list, entry) {
-		if (!IS_ERR_OR_NULL(ch->sess))
-			break;
-		ch->sess = target_setup_session(&stpg->tpg, tag_num,
+	if (sport->gid_id) {
+		mutex_lock(&sport->gid_id->mutex);
+		list_for_each_entry(stpg, &sport->gid_id->tpg_list, entry) {
+			if (!IS_ERR_OR_NULL(ch->sess))
+				break;
+			ch->sess = target_setup_session(&stpg->tpg, tag_num,
 					tag_size, TARGET_PROT_NORMAL, i_port_id,
 					ch, NULL);
-		if (!IS_ERR_OR_NULL(ch->sess))
-			break;
-		/* Retry without leading "0x" */
-		ch->sess = target_setup_session(&stpg->tpg, tag_num,
+			if (!IS_ERR_OR_NULL(ch->sess))
+				break;
+			/* Retry without leading "0x" */
+			ch->sess = target_setup_session(&stpg->tpg, tag_num,
 						tag_size, TARGET_PROT_NORMAL,
 						i_port_id + 2, ch, NULL);
+		}
+		mutex_unlock(&sport->gid_id->mutex);
 	}
-	mutex_unlock(&sport->port_gid_id.mutex);
 
 	if (IS_ERR_OR_NULL(ch->sess)) {
 		WARN_ON_ONCE(ch->sess == NULL);
···
 	return 0;
 }
 
-static struct se_wwn *__srpt_lookup_wwn(const char *name)
+struct port_and_port_id {
+	struct srpt_port *sport;
+	struct srpt_port_id **port_id;
+};
+
+static struct port_and_port_id __srpt_lookup_port(const char *name)
 {
 	struct ib_device *dev;
 	struct srpt_device *sdev;
···
 		for (i = 0; i < dev->phys_port_cnt; i++) {
 			sport = &sdev->port[i];
 
-			if (strcmp(sport->port_guid_id.name, name) == 0)
-				return &sport->port_guid_id.wwn;
-			if (strcmp(sport->port_gid_id.name, name) == 0)
-				return &sport->port_gid_id.wwn;
+			if (strcmp(sport->guid_name, name) == 0) {
+				kref_get(&sdev->refcnt);
+				return (struct port_and_port_id){
+					sport, &sport->guid_id};
+			}
+			if (strcmp(sport->gid_name, name) == 0) {
+				kref_get(&sdev->refcnt);
+				return (struct port_and_port_id){
+					sport, &sport->gid_id};
+			}
 		}
 	}
 
-	return NULL;
+	return (struct port_and_port_id){};
 }
 
-static struct se_wwn *srpt_lookup_wwn(const char *name)
+/**
+ * srpt_lookup_port() - Look up an RDMA port by name
+ * @name: ASCII port name
+ *
+ * Increments the RDMA port reference count if an RDMA port pointer is returned.
+ * The caller must drop that reference count by calling srpt_port_put_ref().
+ */
+static struct port_and_port_id srpt_lookup_port(const char *name)
 {
-	struct se_wwn *wwn;
+	struct port_and_port_id papi;
 
 	spin_lock(&srpt_dev_lock);
-	wwn = __srpt_lookup_wwn(name);
+	papi = __srpt_lookup_port(name);
 	spin_unlock(&srpt_dev_lock);
 
-	return wwn;
+	return papi;
 }
 
 static void srpt_free_srq(struct srpt_device *sdev)
···
 	return ret;
 }
 
+static void srpt_free_sdev(struct kref *refcnt)
+{
+	struct srpt_device *sdev = container_of(refcnt, typeof(*sdev), refcnt);
+
+	kfree(sdev);
+}
+
+static void srpt_sdev_put(struct srpt_device *sdev)
+{
+	kref_put(&sdev->refcnt, srpt_free_sdev);
+}
+
 /**
  * srpt_add_one - InfiniBand device addition callback function
  * @device: Describes a HCA.
···
 	if (!sdev)
 		return -ENOMEM;
 
+	kref_init(&sdev->refcnt);
 	sdev->device = device;
 	mutex_init(&sdev->sdev_mutex);
 
···
 		sport->port_attrib.srp_sq_size = DEF_SRPT_SQ_SIZE;
 		sport->port_attrib.use_srq = false;
 		INIT_WORK(&sport->work, srpt_refresh_port_work);
-		mutex_init(&sport->port_guid_id.mutex);
-		INIT_LIST_HEAD(&sport->port_guid_id.tpg_list);
-		mutex_init(&sport->port_gid_id.mutex);
-		INIT_LIST_HEAD(&sport->port_gid_id.tpg_list);
 
 		ret = srpt_refresh_port(sport);
 		if (ret) {
···
 	srpt_free_srq(sdev);
 	ib_dealloc_pd(sdev->pd);
 free_dev:
-	kfree(sdev);
+	srpt_sdev_put(sdev);
 	pr_info("%s(%s) failed.\n", __func__, dev_name(&device->dev));
 	return ret;
 }
···
 
 	ib_dealloc_pd(sdev->pd);
 
-	kfree(sdev);
+	srpt_sdev_put(sdev);
 }
 
 static struct ib_client srpt_client = {
···
 {
 	struct srpt_port *sport = wwn->priv;
 
-	if (wwn == &sport->port_guid_id.wwn)
-		return &sport->port_guid_id;
-	if (wwn == &sport->port_gid_id.wwn)
-		return &sport->port_gid_id;
+	if (sport->guid_id && &sport->guid_id->wwn == wwn)
+		return sport->guid_id;
+	if (sport->gid_id && &sport->gid_id->wwn == wwn)
+		return sport->gid_id;
 	WARN_ON_ONCE(true);
 	return NULL;
 }
···
 				      struct config_group *group,
 				      const char *name)
 {
-	return srpt_lookup_wwn(name) ? : ERR_PTR(-EINVAL);
+	struct port_and_port_id papi = srpt_lookup_port(name);
+	struct srpt_port *sport = papi.sport;
+	struct srpt_port_id *port_id;
+
+	if (!papi.port_id)
+		return ERR_PTR(-EINVAL);
+	if (*papi.port_id) {
+		/* Attempt to create a directory that already exists. */
+		WARN_ON_ONCE(true);
+		return &(*papi.port_id)->wwn;
+	}
+	port_id = kzalloc(sizeof(*port_id), GFP_KERNEL);
+	if (!port_id) {
+		srpt_sdev_put(sport->sdev);
+		return ERR_PTR(-ENOMEM);
+	}
+	mutex_init(&port_id->mutex);
+	INIT_LIST_HEAD(&port_id->tpg_list);
+	port_id->wwn.priv = sport;
+	memcpy(port_id->name, port_id == sport->guid_id ? sport->guid_name :
+	       sport->gid_name, ARRAY_SIZE(port_id->name));
+
+	*papi.port_id = port_id;
+
+	return &port_id->wwn;
 }
 
 /**
···
  */
 static void srpt_drop_tport(struct se_wwn *wwn)
 {
+	struct srpt_port_id *port_id = container_of(wwn, typeof(*port_id), wwn);
+	struct srpt_port *sport = wwn->priv;
+
+	if (sport->guid_id == port_id)
+		sport->guid_id = NULL;
+	else if (sport->gid_id == port_id)
+		sport->gid_id = NULL;
+	else
+		WARN_ON_ONCE(true);
+
+	srpt_sdev_put(sport->sdev);
+	kfree(port_id);
 }
 
 static ssize_t srpt_wwn_version_show(struct config_item *item, char *buf)

drivers/infiniband/ulp/srpt/ib_srpt.h

···
 };
 
 /**
- * struct srpt_port_id - information about an RDMA port name
+ * struct srpt_port_id - LIO RDMA port information
  * @mutex:	Protects @tpg_list changes.
  * @tpg_list:	TPGs associated with the RDMA port name.
  * @wwn:	WWN associated with the RDMA port name.
···
 };
 
 /**
- * struct srpt_port - information associated by SRPT with a single IB port
+ * struct srpt_port - SRPT RDMA port information
  * @sdev:      backpointer to the HCA information.
  * @mad_agent: per-port management datagram processing information.
  * @enabled:   Whether or not this target port is enabled.
···
  * @lid:       cached value of the port's lid.
  * @gid:       cached value of the port's gid.
  * @work:      work structure for refreshing the aforementioned cached values.
- * @port_guid_id: target port GUID
- * @port_gid_id: target port GID
+ * @guid_name: port name in GUID format.
+ * @guid_id:   LIO target port information for the port name in GUID format.
+ * @gid_name:  port name in GID format.
+ * @gid_id:    LIO target port information for the port name in GID format.
  * @port_attrib:   Port attributes that can be accessed through configfs.
  * @refcount:	   Number of objects associated with this port.
  * @freed_channels: Completion that will be signaled once @refcount becomes 0.
···
 	u32			lid;
 	union ib_gid		gid;
 	struct work_struct	work;
-	struct srpt_port_id	port_guid_id;
-	struct srpt_port_id	port_gid_id;
+	char			guid_name[64];
+	struct srpt_port_id	*guid_id;
+	char			gid_name[64];
+	struct srpt_port_id	*gid_id;
 	struct srpt_port_attrib port_attrib;
 	atomic_t		refcount;
 	struct completion	*freed_channels;
···
 
 /**
  * struct srpt_device - information associated by SRPT with a single HCA
+ * @refcnt:	   Reference count for this device.
  * @device:        Backpointer to the struct ib_device managed by the IB core.
  * @pd:            IB protection domain.
  * @lkey:          L_Key (local key) with write access to all local memory.
···
  * @port:          Information about the ports owned by this HCA.
  */
 struct srpt_device {
+	struct kref		refcnt;
 	struct ib_device	*device;
 	struct ib_pd		*pd;
 	u32			lkey;

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

···
 
 static int mlx5_cmd_stub_create_flow_table(struct mlx5_flow_root_namespace *ns,
 					   struct mlx5_flow_table *ft,
-					   unsigned int size,
+					   struct mlx5_flow_table_attr *ft_attr,
 					   struct mlx5_flow_table *next_ft)
 {
-	ft->max_fte = size ? roundup_pow_of_two(size) : 1;
+	int max_fte = ft_attr->max_fte;
+
+	ft->max_fte = max_fte ? roundup_pow_of_two(max_fte) : 1;
 
 	return 0;
 }
···
 
 static int mlx5_cmd_create_flow_table(struct mlx5_flow_root_namespace *ns,
 				      struct mlx5_flow_table *ft,
-				      unsigned int size,
+				      struct mlx5_flow_table_attr *ft_attr,
 				      struct mlx5_flow_table *next_ft)
 {
 	int en_encap = !!(ft->flags & MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT);
···
 	u32 out[MLX5_ST_SZ_DW(create_flow_table_out)] = {};
 	u32 in[MLX5_ST_SZ_DW(create_flow_table_in)] = {};
 	struct mlx5_core_dev *dev = ns->dev;
+	unsigned int size;
 	int err;
 
-	if (size != POOL_NEXT_SIZE)
-		size = roundup_pow_of_two(size);
-	size = mlx5_ft_pool_get_avail_sz(dev, ft->type, size);
+	if (ft_attr->max_fte != POOL_NEXT_SIZE)
+		size = roundup_pow_of_two(ft_attr->max_fte);
+	size = mlx5_ft_pool_get_avail_sz(dev, ft->type, ft_attr->max_fte);
 	if (!size)
 		return -ENOSPC;
 
 	MLX5_SET(create_flow_table_in, in, opcode,
 		 MLX5_CMD_OP_CREATE_FLOW_TABLE);
 
+	MLX5_SET(create_flow_table_in, in, uid, ft_attr->uid);
 	MLX5_SET(create_flow_table_in, in, table_type, ft->type);
 	MLX5_SET(create_flow_table_in, in, flow_table_context.level, ft->level);
 	MLX5_SET(create_flow_table_in, in, flow_table_context.log_size, size ? ilog2(size) : 0);

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

···
 struct mlx5_flow_cmds {
 	int (*create_flow_table)(struct mlx5_flow_root_namespace *ns,
 				 struct mlx5_flow_table *ft,
-				 unsigned int size,
+				 struct mlx5_flow_table_attr *ft_attr,
 				 struct mlx5_flow_table *next_ft);
 	int (*destroy_flow_table)(struct mlx5_flow_root_namespace *ns,
 				  struct mlx5_flow_table *ft);

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

···
 			      find_next_chained_ft(fs_prio);
 	ft->def_miss_action = ns->def_miss_action;
 	ft->ns = ns;
-	err = root->cmds->create_flow_table(root, ft, ft_attr->max_fte, next_ft);
+	err = root->cmds->create_flow_table(root, ft, ft_attr, next_ft);
 	if (err)
 		goto free_ft;
 
···
 	return __mlx5_create_flow_table(ns, ft_attr, FS_FT_OP_MOD_NORMAL, 0);
 }
 EXPORT_SYMBOL(mlx5_create_flow_table);
+
+u32 mlx5_flow_table_id(struct mlx5_flow_table *ft)
+{
+	return ft->id;
+}
+EXPORT_SYMBOL(mlx5_flow_table_id);
 
 struct mlx5_flow_table *
 mlx5_create_vport_flow_table(struct mlx5_flow_namespace *ns,

drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c

···
 
 	MLX5_SET(create_flow_table_in, in, opcode, MLX5_CMD_OP_CREATE_FLOW_TABLE);
 	MLX5_SET(create_flow_table_in, in, table_type, attr->table_type);
+	MLX5_SET(create_flow_table_in, in, uid, attr->uid);
 
 	ft_mdev = MLX5_ADDR_OF(create_flow_table_in, in, flow_table_context);
 	MLX5_SET(flow_table_context, ft_mdev, termination_table, attr->term_tbl);

drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c

···
 					     tbl->table_type);
 }
 
-static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl)
+static int dr_table_create_sw_owned_tbl(struct mlx5dr_table *tbl, u16 uid)
 {
 	bool en_encap = !!(tbl->flags & MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT);
 	bool en_decap = !!(tbl->flags & MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
···
 	ft_attr.sw_owner = true;
 	ft_attr.decap_en = en_decap;
 	ft_attr.reformat_en = en_encap;
+	ft_attr.uid = uid;
 
 	ret = mlx5dr_cmd_create_flow_table(tbl->dmn->mdev, &ft_attr,
 					   NULL, &tbl->table_id);
···
 	return ret;
 }
 
-struct mlx5dr_table *mlx5dr_table_create(struct mlx5dr_domain *dmn, u32 level, u32 flags)
+struct mlx5dr_table *mlx5dr_table_create(struct mlx5dr_domain *dmn, u32 level,
+					 u32 flags, u16 uid)
 {
 	struct mlx5dr_table *tbl;
 	int ret;
···
 	if (ret)
 		goto free_tbl;
 
-	ret = dr_table_create_sw_owned_tbl(tbl);
+	ret = dr_table_create_sw_owned_tbl(tbl, uid);
 	if (ret)
 		goto uninit_tbl;
 

drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h

···
 
 struct mlx5dr_cmd_create_flow_table_attr {
 	u32 table_type;
+	u16 uid;
 	u64 icm_addr_rx;
 	u64 icm_addr_tx;
 	u8 level;

drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c

···
 
 static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns,
 					 struct mlx5_flow_table *ft,
-					 unsigned int size,
+					 struct mlx5_flow_table_attr *ft_attr,
 					 struct mlx5_flow_table *next_ft)
 {
 	struct mlx5dr_table *tbl;
···
 
 	if (mlx5_dr_is_fw_table(ft->flags))
 		return mlx5_fs_cmd_get_fw_cmds()->create_flow_table(ns, ft,
-								    size,
+								    ft_attr,
 								    next_ft);
 	flags = ft->flags;
 	/* turn off encap/decap if not supported for sw-str by fw */
···
 		flags = ft->flags & ~(MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
 				      MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);
 
-	tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain, ft->level, flags);
+	tbl = mlx5dr_table_create(ns->fs_dr_domain.dr_domain, ft->level, flags,
+				  ft_attr->uid);
 	if (!tbl) {
 		mlx5_core_err(ns->dev, "Failed creating dr flow_table\n");
 		return -EINVAL;

drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h

···
 			    struct mlx5dr_domain *peer_dmn);
 
 struct mlx5dr_table *
-mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level, u32 flags);
+mlx5dr_table_create(struct mlx5dr_domain *domain, u32 level, u32 flags,
+		    u16 uid);
 
 struct mlx5dr_table *
 mlx5dr_table_get_from_fs_ft(struct mlx5_flow_table *ft);

include/linux/mlx5/driver.h

···
 };
 
 enum {
-	MR_CACHE_LAST_STD_ENTRY = 20,
+	MKEY_CACHE_LAST_STD_ENTRY = 20,
 	MLX5_IMR_MTT_CACHE_ENTRY,
 	MLX5_IMR_KSM_CACHE_ENTRY,
-	MAX_MR_CACHE_ENTRIES
+	MAX_MKEY_CACHE_ENTRIES
 };
 
 struct mlx5_profile {
···
 	struct {
 		int	size;
 		int	limit;
-	} mr_cache[MAX_MR_CACHE_ENTRIES];
+	} mr_cache[MAX_MKEY_CACHE_ENTRIES];
 };
 
 struct mlx5_hca_cap {

include/linux/mlx5/fs.h

···
 	int max_fte;
 	u32 level;
 	u32 flags;
+	u16 uid;
 	struct mlx5_flow_table *next_ft;
 
 	struct {
···
 void mlx5_packet_reformat_dealloc(struct mlx5_core_dev *dev,
 				  struct mlx5_pkt_reformat *reformat);
 
+u32 mlx5_flow_table_id(struct mlx5_flow_table *ft);
 #endif

include/linux/mlx5/mlx5_ifc.h

···
 };
 
 struct mlx5_ifc_cmd_hca_cap_bits {
-	u8         reserved_at_0[0x1f];
+	u8         reserved_at_0[0x10];
+	u8         shared_object_to_user_object_allowed[0x1];
+	u8         reserved_at_13[0xe];
 	u8         vhca_resource_manager[0x1];
 
 	u8         hca_cap_2[0x1];
···
 
 struct mlx5_ifc_create_flow_table_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_at_10[0x10];
+	u8         uid[0x10];
 
 	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];

include/rdma/ib_verbs.h

···
 
 /**
  * ib_lid_cpu16 - Return lid in 16bit CPU encoding.
- *     In the current implementation the only way to get
+ *     In the current implementation the only way to
  *     get the 32bit lid is from other sources for OPA.
  *     For IB, lids will always be 16bits so cast the
  *     value accordingly.

include/rdma/rdma_cm.h

···
 	enum rdma_ucm_port_space ps;
 	enum ib_qp_type		 qp_type;
 	u32			 port_num;
+	struct work_struct net_work;
 };
 
 struct rdma_cm_id *

include/uapi/rdma/erdma-abi.h

···
+/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */
+/*
+ * Copyright (c) 2020-2022, Alibaba Group.
+ */
+
+#ifndef __ERDMA_USER_H__
+#define __ERDMA_USER_H__
+
+#include <linux/types.h>
+
+#define ERDMA_ABI_VERSION       1
+
+struct erdma_ureq_create_cq {
+	__aligned_u64 db_record_va;
+	__aligned_u64 qbuf_va;
+	__u32 qbuf_len;
+	__u32 rsvd0;
+};
+
+struct erdma_uresp_create_cq {
+	__u32 cq_id;
+	__u32 num_cqe;
+};
+
+struct erdma_ureq_create_qp {
+	__aligned_u64 db_record_va;
+	__aligned_u64 qbuf_va;
+	__u32 qbuf_len;
+	__u32 rsvd0;
+};
+
+struct erdma_uresp_create_qp {
+	__u32 qp_id;
+	__u32 num_sqe;
+	__u32 num_rqe;
+	__u32 rq_offset;
+};
+
+struct erdma_uresp_alloc_ctx {
+	__u32 dev_id;
+	__u32 pad;
+	__u32 sdb_type;
+	__u32 sdb_offset;
+	__aligned_u64 sdb;
+	__aligned_u64 rdb;
+	__aligned_u64 cdb;
+};
+
+#endif

include/uapi/rdma/ib_user_ioctl_verbs.h

···
 	RDMA_DRIVER_QIB,
 	RDMA_DRIVER_EFA,
 	RDMA_DRIVER_SIW,
+	RDMA_DRIVER_ERDMA,
 };
 
 enum ib_uverbs_gid_type {

include/uapi/rdma/mlx5_user_ioctl_cmds.h

···
 	MLX5_IB_OBJECT_VAR,
 	MLX5_IB_OBJECT_PP,
 	MLX5_IB_OBJECT_UAR,
+	MLX5_IB_OBJECT_STEERING_ANCHOR,
 };
 
 enum mlx5_ib_flow_matcher_create_attrs {
···
 enum mlx5_ib_flow_matcher_methods {
 	MLX5_IB_METHOD_FLOW_MATCHER_CREATE = (1U << UVERBS_ID_NS_SHIFT),
 	MLX5_IB_METHOD_FLOW_MATCHER_DESTROY,
+};
+
+enum mlx5_ib_flow_steering_anchor_create_attrs {
+	MLX5_IB_ATTR_STEERING_ANCHOR_CREATE_HANDLE = (1U << UVERBS_ID_NS_SHIFT),
+	MLX5_IB_ATTR_STEERING_ANCHOR_FT_TYPE,
+	MLX5_IB_ATTR_STEERING_ANCHOR_PRIORITY,
+	MLX5_IB_ATTR_STEERING_ANCHOR_FT_ID,
+};
+
+enum mlx5_ib_flow_steering_anchor_destroy_attrs {
+	MLX5_IB_ATTR_STEERING_ANCHOR_DESTROY_HANDLE = (1U << UVERBS_ID_NS_SHIFT),
+};
+
+enum mlx5_ib_steering_anchor_methods {
+	MLX5_IB_METHOD_STEERING_ANCHOR_CREATE = (1U << UVERBS_ID_NS_SHIFT),
+	MLX5_IB_METHOD_STEERING_ANCHOR_DESTROY,
 };
 
 enum mlx5_ib_device_query_context_attrs {