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

Pull rdma updates from Doug Ledford:
 "Initial roundup of 4.5 merge window patches

   - Remove usage of ib_query_device and instead store attributes in
     ib_device struct

   - Move iopoll out of block and into lib, rename to irqpoll, and use
     in several places in the rdma stack as our new completion queue
     polling library mechanism.  Update the other block drivers that
     already used iopoll to use the new mechanism too.

   - Replace the per-entry GID table locks with a single GID table lock

   - IPoIB multicast cleanup

   - Cleanups to the IB MR facility

   - Add support for 64bit extended IB counters

   - Fix for netlink oops while parsing RDMA nl messages

   - RoCEv2 support for the core IB code

   - mlx4 RoCEv2 support

   - mlx5 RoCEv2 support

   - Cross Channel support for mlx5

   - Timestamp support for mlx5

   - Atomic support for mlx5

   - Raw QP support for mlx5

   - MAINTAINERS update for mlx4/mlx5

   - Misc ocrdma, qib, nes, usNIC, cxgb3, cxgb4, mlx4, mlx5 updates

   - Add support for remote invalidate to the iSER driver (pushed
     through the RDMA tree due to dependencies, acknowledged by nab)

   - Update to NFSoRDMA (pushed through the RDMA tree due to
     dependencies, acknowledged by Bruce)"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (169 commits)
  IB/mlx5: Unify CQ create flags check
  IB/mlx5: Expose Raw Packet QP to user space consumers
  {IB, net}/mlx5: Move the modify QP operation table to mlx5_ib
  IB/mlx5: Support setting Ethernet priority for Raw Packet QPs
  IB/mlx5: Add Raw Packet QP query functionality
  IB/mlx5: Add create and destroy functionality for Raw Packet QP
  IB/mlx5: Refactor mlx5_ib_qp to accommodate other QP types
  IB/mlx5: Allocate a Transport Domain for each ucontext
  net/mlx5_core: Warn on unsupported events of QP/RQ/SQ
  net/mlx5_core: Add RQ and SQ event handling
  net/mlx5_core: Export transport objects
  IB/mlx5: Expose CQE version to user-space
  IB/mlx5: Add CQE version 1 support to user QPs and SRQs
  IB/mlx5: Fix data validation in mlx5_ib_alloc_ucontext
  IB/sa: Fix netlink local service GFP crash
  IB/srpt: Remove redundant wc array
  IB/qib: Improve ipoib UD performance
  IB/mlx4: Advertise RoCE v2 support
  IB/mlx4: Create and use another QP1 for RoCEv2
  IB/mlx4: Enable send of RoCE QP1 packets with IP/UDP headers
  ...

24 files changed, 2150 insertions, 533 deletions

diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile
index d43a899..f818538 100644
--- a/drivers/infiniband/core/Makefile
+++ b/drivers/infiniband/core/Makefile
@@ -8,7 +8,7 @@ obj-$(CONFIG_INFINIBAND_USER_MAD) +=	ib_umad.o
 obj-$(CONFIG_INFINIBAND_USER_ACCESS) +=	ib_uverbs.o ib_ucm.o \
 					$(user_access-y)
 
-ib_core-y :=			packer.o ud_header.o verbs.o sysfs.o \
+ib_core-y :=			packer.o ud_header.o verbs.o cq.o sysfs.o \
 				device.o fmr_pool.o cache.o netlink.o \
 				roce_gid_mgmt.o
 ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
@@ -24,6 +24,8 @@ iw_cm-y :=			iwcm.o iwpm_util.o iwpm_msg.o
 
 rdma_cm-y :=			cma.o
 
+rdma_cm-$(CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS) += cma_configfs.o
+
 rdma_ucm-y :=			ucma.o
 
 ib_addr-y :=			addr.o
diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 34b1ada..337353d 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -121,7 +121,8 @@ int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
 }
 EXPORT_SYMBOL(rdma_copy_addr);
 
-int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr,
+int rdma_translate_ip(const struct sockaddr *addr,
+		      struct rdma_dev_addr *dev_addr,
 		      u16 *vlan_id)
 {
 	struct net_device *dev;
@@ -139,7 +140,7 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr,
 	switch (addr->sa_family) {
 	case AF_INET:
 		dev = ip_dev_find(dev_addr->net,
-			((struct sockaddr_in *) addr)->sin_addr.s_addr);
+			((const struct sockaddr_in *)addr)->sin_addr.s_addr);
 
 		if (!dev)
 			return ret;
@@ -154,7 +155,7 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr,
 		rcu_read_lock();
 		for_each_netdev_rcu(dev_addr->net, dev) {
 			if (ipv6_chk_addr(dev_addr->net,
-					  &((struct sockaddr_in6 *) addr)->sin6_addr,
+					  &((const struct sockaddr_in6 *)addr)->sin6_addr,
 					  dev, 1)) {
 				ret = rdma_copy_addr(dev_addr, dev, NULL);
 				if (vlan_id)
@@ -198,7 +199,8 @@ static void queue_req(struct addr_req *req)
 	mutex_unlock(&lock);
 }
 
-static int dst_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr, void *daddr)
+static int dst_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr,
+			const void *daddr)
 {
 	struct neighbour *n;
 	int ret;
@@ -222,8 +224,9 @@ static int dst_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr, v
 }
 
 static int addr4_resolve(struct sockaddr_in *src_in,
-			 struct sockaddr_in *dst_in,
-			 struct rdma_dev_addr *addr)
+			 const struct sockaddr_in *dst_in,
+			 struct rdma_dev_addr *addr,
+			 struct rtable **prt)
 {
 	__be32 src_ip = src_in->sin_addr.s_addr;
 	__be32 dst_ip = dst_in->sin_addr.s_addr;
@@ -243,33 +246,29 @@ static int addr4_resolve(struct sockaddr_in *src_in,
 	src_in->sin_family = AF_INET;
 	src_in->sin_addr.s_addr = fl4.saddr;
 
-	if (rt->dst.dev->flags & IFF_LOOPBACK) {
-		ret = rdma_translate_ip((struct sockaddr *)dst_in, addr, NULL);
-		if (!ret)
-			memcpy(addr->dst_dev_addr, addr->src_dev_addr, MAX_ADDR_LEN);
-		goto put;
-	}
+	/* If there's a gateway, we're definitely in RoCE v2 (as RoCE v1 isn't
+	 * routable) and we could set the network type accordingly.
+	 */
+	if (rt->rt_uses_gateway)
+		addr->network = RDMA_NETWORK_IPV4;
 
-	/* If the device does ARP internally, return 'done' */
-	if (rt->dst.dev->flags & IFF_NOARP) {
-		ret = rdma_copy_addr(addr, rt->dst.dev, NULL);
-		goto put;
-	}
+	addr->hoplimit = ip4_dst_hoplimit(&rt->dst);
 
-	ret = dst_fetch_ha(&rt->dst, addr, &fl4.daddr);
-put:
-	ip_rt_put(rt);
+	*prt = rt;
+	return 0;
 out:
 	return ret;
 }
 
 #if IS_ENABLED(CONFIG_IPV6)
 static int addr6_resolve(struct sockaddr_in6 *src_in,
-			 struct sockaddr_in6 *dst_in,
-			 struct rdma_dev_addr *addr)
+			 const struct sockaddr_in6 *dst_in,
+			 struct rdma_dev_addr *addr,
+			 struct dst_entry **pdst)
 {
 	struct flowi6 fl6;
 	struct dst_entry *dst;
+	struct rt6_info *rt;
 	int ret;
 
 	memset(&fl6, 0, sizeof fl6);
@@ -281,6 +280,7 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,
 	if ((ret = dst->error))
 		goto put;
 
+	rt = (struct rt6_info *)dst;
 	if (ipv6_addr_any(&fl6.saddr)) {
 		ret = ipv6_dev_get_saddr(addr->net, ip6_dst_idev(dst)->dev,
 					 &fl6.daddr, 0, &fl6.saddr);
@@ -291,43 +291,111 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,
 		src_in->sin6_addr = fl6.saddr;
 	}
 
-	if (dst->dev->flags & IFF_LOOPBACK) {
-		ret = rdma_translate_ip((struct sockaddr *)dst_in, addr, NULL);
-		if (!ret)
-			memcpy(addr->dst_dev_addr, addr->src_dev_addr, MAX_ADDR_LEN);
-		goto put;
-	}
+	/* If there's a gateway, we're definitely in RoCE v2 (as RoCE v1 isn't
+	 * routable) and we could set the network type accordingly.
+	 */
+	if (rt->rt6i_flags & RTF_GATEWAY)
+		addr->network = RDMA_NETWORK_IPV6;
 
-	/* If the device does ARP internally, return 'done' */
-	if (dst->dev->flags & IFF_NOARP) {
-		ret = rdma_copy_addr(addr, dst->dev, NULL);
-		goto put;
-	}
+	addr->hoplimit = ip6_dst_hoplimit(dst);
 
-	ret = dst_fetch_ha(dst, addr, &fl6.daddr);
+	*pdst = dst;
+	return 0;
 put:
 	dst_release(dst);
 	return ret;
 }
 #else
 static int addr6_resolve(struct sockaddr_in6 *src_in,
-			 struct sockaddr_in6 *dst_in,
-			 struct rdma_dev_addr *addr)
+			 const struct sockaddr_in6 *dst_in,
+			 struct rdma_dev_addr *addr,
+			 struct dst_entry **pdst)
 {
 	return -EADDRNOTAVAIL;
 }
 #endif
 
+static int addr_resolve_neigh(struct dst_entry *dst,
+			      const struct sockaddr *dst_in,
+			      struct rdma_dev_addr *addr)
+{
+	if (dst->dev->flags & IFF_LOOPBACK) {
+		int ret;
+
+		ret = rdma_translate_ip(dst_in, addr, NULL);
+		if (!ret)
+			memcpy(addr->dst_dev_addr, addr->src_dev_addr,
+			       MAX_ADDR_LEN);
+
+		return ret;
+	}
+
+	/* If the device doesn't do ARP internally */
+	if (!(dst->dev->flags & IFF_NOARP)) {
+		const struct sockaddr_in *dst_in4 =
+			(const struct sockaddr_in *)dst_in;
+		const struct sockaddr_in6 *dst_in6 =
+			(const struct sockaddr_in6 *)dst_in;
+
+		return dst_fetch_ha(dst, addr,
+				    dst_in->sa_family == AF_INET ?
+				    (const void *)&dst_in4->sin_addr.s_addr :
+				    (const void *)&dst_in6->sin6_addr);
+	}
+
+	return rdma_copy_addr(addr, dst->dev, NULL);
+}
+
 static int addr_resolve(struct sockaddr *src_in,
-			struct sockaddr *dst_in,
-			struct rdma_dev_addr *addr)
+			const struct sockaddr *dst_in,
+			struct rdma_dev_addr *addr,
+			bool resolve_neigh)
 {
+	struct net_device *ndev;
+	struct dst_entry *dst;
+	int ret;
+
 	if (src_in->sa_family == AF_INET) {
-		return addr4_resolve((struct sockaddr_in *) src_in,
-			(struct sockaddr_in *) dst_in, addr);
-	} else
-		return addr6_resolve((struct sockaddr_in6 *) src_in,
-			(struct sockaddr_in6 *) dst_in, addr);
+		struct rtable *rt = NULL;
+		const struct sockaddr_in *dst_in4 =
+			(const struct sockaddr_in *)dst_in;
+
+		ret = addr4_resolve((struct sockaddr_in *)src_in,
+				    dst_in4, addr, &rt);
+		if (ret)
+			return ret;
+
+		if (resolve_neigh)
+			ret = addr_resolve_neigh(&rt->dst, dst_in, addr);
+
+		ndev = rt->dst.dev;
+		dev_hold(ndev);
+
+		ip_rt_put(rt);
+	} else {
+		const struct sockaddr_in6 *dst_in6 =
+			(const struct sockaddr_in6 *)dst_in;
+
+		ret = addr6_resolve((struct sockaddr_in6 *)src_in,
+				    dst_in6, addr,
+				    &dst);
+		if (ret)
+			return ret;
+
+		if (resolve_neigh)
+			ret = addr_resolve_neigh(dst, dst_in, addr);
+
+		ndev = dst->dev;
+		dev_hold(ndev);
+
+		dst_release(dst);
+	}
+
+	addr->bound_dev_if = ndev->ifindex;
+	addr->net = dev_net(ndev);
+	dev_put(ndev);
+
+	return ret;
 }
 
 static void process_req(struct work_struct *work)
@@ -343,7 +411,8 @@ static void process_req(struct work_struct *work)
 		if (req->status == -ENODATA) {
 			src_in = (struct sockaddr *) &req->src_addr;
 			dst_in = (struct sockaddr *) &req->dst_addr;
-			req->status = addr_resolve(src_in, dst_in, req->addr);
+			req->status = addr_resolve(src_in, dst_in, req->addr,
+						   true);
 			if (req->status && time_after_eq(jiffies, req->timeout))
 				req->status = -ETIMEDOUT;
 			else if (req->status == -ENODATA)
@@ -403,7 +472,7 @@ int rdma_resolve_ip(struct rdma_addr_client *client,
 	req->client = client;
 	atomic_inc(&client->refcount);
 
-	req->status = addr_resolve(src_in, dst_in, addr);
+	req->status = addr_resolve(src_in, dst_in, addr, true);
 	switch (req->status) {
 	case 0:
 		req->timeout = jiffies;
@@ -425,6 +494,26 @@ err:
 }
 EXPORT_SYMBOL(rdma_resolve_ip);
 
+int rdma_resolve_ip_route(struct sockaddr *src_addr,
+			  const struct sockaddr *dst_addr,
+			  struct rdma_dev_addr *addr)
+{
+	struct sockaddr_storage ssrc_addr = {};
+	struct sockaddr *src_in = (struct sockaddr *)&ssrc_addr;
+
+	if (src_addr) {
+		if (src_addr->sa_family != dst_addr->sa_family)
+			return -EINVAL;
+
+		memcpy(src_in, src_addr, rdma_addr_size(src_addr));
+	} else {
+		src_in->sa_family = dst_addr->sa_family;
+	}
+
+	return addr_resolve(src_in, dst_addr, addr, false);
+}
+EXPORT_SYMBOL(rdma_resolve_ip_route);
+
 void rdma_addr_cancel(struct rdma_dev_addr *addr)
 {
 	struct addr_req *req, *temp_req;
@@ -456,8 +545,10 @@ static void resolve_cb(int status, struct sockaddr *src_addr,
 	complete(&((struct resolve_cb_context *)context)->comp);
 }
 
-int rdma_addr_find_dmac_by_grh(const union ib_gid *sgid, const union ib_gid *dgid,
-			       u8 *dmac, u16 *vlan_id, int if_index)
+int rdma_addr_find_l2_eth_by_grh(const union ib_gid *sgid,
+				 const union ib_gid *dgid,
+				 u8 *dmac, u16 *vlan_id, int *if_index,
+				 int *hoplimit)
 {
 	int ret = 0;
 	struct rdma_dev_addr dev_addr;
@@ -475,7 +566,8 @@ int rdma_addr_find_dmac_by_grh(const union ib_gid *sgid, const union ib_gid *dgi
 	rdma_gid2ip(&dgid_addr._sockaddr, dgid);
 
 	memset(&dev_addr, 0, sizeof(dev_addr));
-	dev_addr.bound_dev_if = if_index;
+	if (if_index)
+		dev_addr.bound_dev_if = *if_index;
 	dev_addr.net = &init_net;
 
 	ctx.addr = &dev_addr;
@@ -491,12 +583,16 @@ int rdma_addr_find_dmac_by_grh(const union ib_gid *sgid, const union ib_gid *dgi
 	dev = dev_get_by_index(&init_net, dev_addr.bound_dev_if);
 	if (!dev)
 		return -ENODEV;
+	if (if_index)
+		*if_index = dev_addr.bound_dev_if;
 	if (vlan_id)
 		*vlan_id = rdma_vlan_dev_vlan_id(dev);
+	if (hoplimit)
+		*hoplimit = dev_addr.hoplimit;
 	dev_put(dev);
 	return ret;
 }
-EXPORT_SYMBOL(rdma_addr_find_dmac_by_grh);
+EXPORT_SYMBOL(rdma_addr_find_l2_eth_by_grh);
 
 int rdma_addr_find_smac_by_sgid(union ib_gid *sgid, u8 *smac, u16 *vlan_id)
 {
diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index 89bebea..53343ff 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -64,6 +64,7 @@ enum gid_attr_find_mask {
 	GID_ATTR_FIND_MASK_GID          = 1UL << 0,
 	GID_ATTR_FIND_MASK_NETDEV	= 1UL << 1,
 	GID_ATTR_FIND_MASK_DEFAULT	= 1UL << 2,
+	GID_ATTR_FIND_MASK_GID_TYPE	= 1UL << 3,
 };
 
 enum gid_table_entry_props {
@@ -81,10 +82,6 @@ enum gid_table_write_action {
 };
 
 struct ib_gid_table_entry {
-	/* This lock protects an entry from being
-	 * read and written simultaneously.
-	 */
-	rwlock_t	    lock;
 	unsigned long	    props;
 	union ib_gid        gid;
 	struct ib_gid_attr  attr;
@@ -109,28 +106,86 @@ struct ib_gid_table {
 	 * are locked by this lock.
 	 **/
 	struct mutex         lock;
+	/* This lock protects the table entries from being
+	 * read and written simultaneously.
+	 */
+	rwlock_t	     rwlock;
 	struct ib_gid_table_entry *data_vec;
 };
 
+static void dispatch_gid_change_event(struct ib_device *ib_dev, u8 port)
+{
+	if (rdma_cap_roce_gid_table(ib_dev, port)) {
+		struct ib_event event;
+
+		event.device		= ib_dev;
+		event.element.port_num	= port;
+		event.event		= IB_EVENT_GID_CHANGE;
+
+		ib_dispatch_event(&event);
+	}
+}
+
+static const char * const gid_type_str[] = {
+	[IB_GID_TYPE_IB]	= "IB/RoCE v1",
+	[IB_GID_TYPE_ROCE_UDP_ENCAP]	= "RoCE v2",
+};
+
+const char *ib_cache_gid_type_str(enum ib_gid_type gid_type)
+{
+	if (gid_type < ARRAY_SIZE(gid_type_str) && gid_type_str[gid_type])
+		return gid_type_str[gid_type];
+
+	return "Invalid GID type";
+}
+EXPORT_SYMBOL(ib_cache_gid_type_str);
+
+int ib_cache_gid_parse_type_str(const char *buf)
+{
+	unsigned int i;
+	size_t len;
+	int err = -EINVAL;
+
+	len = strlen(buf);
+	if (len == 0)
+		return -EINVAL;
+
+	if (buf[len - 1] == '\n')
+		len--;
+
+	for (i = 0; i < ARRAY_SIZE(gid_type_str); ++i)
+		if (gid_type_str[i] && !strncmp(buf, gid_type_str[i], len) &&
+		    len == strlen(gid_type_str[i])) {
+			err = i;
+			break;
+		}
+
+	return err;
+}
+EXPORT_SYMBOL(ib_cache_gid_parse_type_str);
+
+/* This function expects that rwlock will be write locked in all
+ * scenarios and that lock will be locked in sleep-able (RoCE)
+ * scenarios.
+ */
 static int write_gid(struct ib_device *ib_dev, u8 port,
 		     struct ib_gid_table *table, int ix,
 		     const union ib_gid *gid,
 		     const struct ib_gid_attr *attr,
 		     enum gid_table_write_action action,
 		     bool  default_gid)
+	__releases(&table->rwlock) __acquires(&table->rwlock)
 {
 	int ret = 0;
 	struct net_device *old_net_dev;
-	unsigned long flags;
 
 	/* in rdma_cap_roce_gid_table, this funciton should be protected by a
 	 * sleep-able lock.
 	 */
-	write_lock_irqsave(&table->data_vec[ix].lock, flags);
 
 	if (rdma_cap_roce_gid_table(ib_dev, port)) {
 		table->data_vec[ix].props |= GID_TABLE_ENTRY_INVALID;
-		write_unlock_irqrestore(&table->data_vec[ix].lock, flags);
+		write_unlock_irq(&table->rwlock);
 		/* GID_TABLE_WRITE_ACTION_MODIFY currently isn't supported by
 		 * RoCE providers and thus only updates the cache.
 		 */
@@ -140,7 +195,7 @@ static int write_gid(struct ib_device *ib_dev, u8 port,
 		else if (action == GID_TABLE_WRITE_ACTION_DEL)
 			ret = ib_dev->del_gid(ib_dev, port, ix,
 					      &table->data_vec[ix].context);
-		write_lock_irqsave(&table->data_vec[ix].lock, flags);
+		write_lock_irq(&table->rwlock);
 	}
 
 	old_net_dev = table->data_vec[ix].attr.ndev;
@@ -162,17 +217,6 @@ static int write_gid(struct ib_device *ib_dev, u8 port,
 
 	table->data_vec[ix].props &= ~GID_TABLE_ENTRY_INVALID;
 
-	write_unlock_irqrestore(&table->data_vec[ix].lock, flags);
-
-	if (!ret && rdma_cap_roce_gid_table(ib_dev, port)) {
-		struct ib_event event;
-
-		event.device		= ib_dev;
-		event.element.port_num	= port;
-		event.event		= IB_EVENT_GID_CHANGE;
-
-		ib_dispatch_event(&event);
-	}
 	return ret;
 }
 
@@ -201,41 +245,58 @@ static int del_gid(struct ib_device *ib_dev, u8 port,
 			 GID_TABLE_WRITE_ACTION_DEL, default_gid);
 }
 
+/* rwlock should be read locked */
 static int find_gid(struct ib_gid_table *table, const union ib_gid *gid,
 		    const struct ib_gid_attr *val, bool default_gid,
-		    unsigned long mask)
+		    unsigned long mask, int *pempty)
 {
-	int i;
+	int i = 0;
+	int found = -1;
+	int empty = pempty ? -1 : 0;
 
-	for (i = 0; i < table->sz; i++) {
-		unsigned long flags;
-		struct ib_gid_attr *attr = &table->data_vec[i].attr;
+	while (i < table->sz && (found < 0 || empty < 0)) {
+		struct ib_gid_table_entry *data = &table->data_vec[i];
+		struct ib_gid_attr *attr = &data->attr;
+		int curr_index = i;
 
-		read_lock_irqsave(&table->data_vec[i].lock, flags);
+		i++;
 
-		if (table->data_vec[i].props & GID_TABLE_ENTRY_INVALID)
-			goto next;
+		if (data->props & GID_TABLE_ENTRY_INVALID)
+			continue;
+
+		if (empty < 0)
+			if (!memcmp(&data->gid, &zgid, sizeof(*gid)) &&
+			    !memcmp(attr, &zattr, sizeof(*attr)) &&
+			    !data->props)
+				empty = curr_index;
+
+		if (found >= 0)
+			continue;
+
+		if (mask & GID_ATTR_FIND_MASK_GID_TYPE &&
+		    attr->gid_type != val->gid_type)
+			continue;
 
 		if (mask & GID_ATTR_FIND_MASK_GID &&
-		    memcmp(gid, &table->data_vec[i].gid, sizeof(*gid)))
-			goto next;
+		    memcmp(gid, &data->gid, sizeof(*gid)))
+			continue;
 
 		if (mask & GID_ATTR_FIND_MASK_NETDEV &&
 		    attr->ndev != val->ndev)
-			goto next;
+			continue;
 
 		if (mask & GID_ATTR_FIND_MASK_DEFAULT &&
-		    !!(table->data_vec[i].props & GID_TABLE_ENTRY_DEFAULT) !=
+		    !!(data->props & GID_TABLE_ENTRY_DEFAULT) !=
 		    default_gid)
-			goto next;
+			continue;
 
-		read_unlock_irqrestore(&table->data_vec[i].lock, flags);
-		return i;
-next:
-		read_unlock_irqrestore(&table->data_vec[i].lock, flags);
+		found = curr_index;
 	}
 
-	return -1;
+	if (pempty)
+		*pempty = empty;
+
+	return found;
 }
 
 static void make_default_gid(struct  net_device *dev, union ib_gid *gid)
@@ -252,6 +313,7 @@ int ib_cache_gid_add(struct ib_device *ib_dev, u8 port,
 	int ix;
 	int ret = 0;
 	struct net_device *idev;
+	int empty;
 
 	table = ports_table[port - rdma_start_port(ib_dev)];
 
@@ -275,22 +337,25 @@ int ib_cache_gid_add(struct ib_device *ib_dev, u8 port,
 	}
 
 	mutex_lock(&table->lock);
+	write_lock_irq(&table->rwlock);
 
 	ix = find_gid(table, gid, attr, false, GID_ATTR_FIND_MASK_GID |
-		      GID_ATTR_FIND_MASK_NETDEV);
+		      GID_ATTR_FIND_MASK_GID_TYPE |
+		      GID_ATTR_FIND_MASK_NETDEV, &empty);
 	if (ix >= 0)
 		goto out_unlock;
 
-	ix = find_gid(table, &zgid, NULL, false, GID_ATTR_FIND_MASK_GID |
-		      GID_ATTR_FIND_MASK_DEFAULT);
-	if (ix < 0) {
+	if (empty < 0) {
 		ret = -ENOSPC;
 		goto out_unlock;
 	}
 
-	add_gid(ib_dev, port, table, ix, gid, attr, false);
+	ret = add_gid(ib_dev, port, table, empty, gid, attr, false);
+	if (!ret)
+		dispatch_gid_change_event(ib_dev, port);
 
 out_unlock:
+	write_unlock_irq(&table->rwlock);
 	mutex_unlock(&table->lock);
 	return ret;
 }
@@ -305,17 +370,22 @@ int ib_cache_gid_del(struct ib_device *ib_dev, u8 port,
 	table = ports_table[port - rdma_start_port(ib_dev)];
 
 	mutex_lock(&table->lock);
+	write_lock_irq(&table->rwlock);
 
 	ix = find_gid(table, gid, attr, false,
 		      GID_ATTR_FIND_MASK_GID	  |
+		      GID_ATTR_FIND_MASK_GID_TYPE |
 		      GID_ATTR_FIND_MASK_NETDEV	  |
-		      GID_ATTR_FIND_MASK_DEFAULT);
+		      GID_ATTR_FIND_MASK_DEFAULT,
+		      NULL);
 	if (ix < 0)
 		goto out_unlock;
 
-	del_gid(ib_dev, port, table, ix, false);
+	if (!del_gid(ib_dev, port, table, ix, false))
+		dispatch_gid_change_event(ib_dev, port);
 
 out_unlock:
+	write_unlock_irq(&table->rwlock);
 	mutex_unlock(&table->lock);
 	return 0;
 }
@@ -326,16 +396,24 @@ int ib_cache_gid_del_all_netdev_gids(struct ib_device *ib_dev, u8 port,
 	struct ib_gid_table **ports_table = ib_dev->cache.gid_cache;
 	struct ib_gid_table *table;
 	int ix;
+	bool deleted = false;
 
 	table  = ports_table[port - rdma_start_port(ib_dev)];
 
 	mutex_lock(&table->lock);
+	write_lock_irq(&table->rwlock);
 
 	for (ix = 0; ix < table->sz; ix++)
 		if (table->data_vec[ix].attr.ndev == ndev)
-			del_gid(ib_dev, port, table, ix, false);
+			if (!del_gid(ib_dev, port, table, ix, false))
+				deleted = true;
 
+	write_unlock_irq(&table->rwlock);
 	mutex_unlock(&table->lock);
+
+	if (deleted)
+		dispatch_gid_change_event(ib_dev, port);
+
 	return 0;
 }
 
@@ -344,18 +422,14 @@ static int __ib_cache_gid_get(struct ib_device *ib_dev, u8 port, int index,
 {
 	struct ib_gid_table **ports_table = ib_dev->cache.gid_cache;
 	struct ib_gid_table *table;
-	unsigned long flags;
 
 	table = ports_table[port - rdma_start_port(ib_dev)];
 
 	if (index < 0 || index >= table->sz)
 		return -EINVAL;
 
-	read_lock_irqsave(&table->data_vec[index].lock, flags);
-	if (table->data_vec[index].props & GID_TABLE_ENTRY_INVALID) {
-		read_unlock_irqrestore(&table->data_vec[index].lock, flags);
+	if (table->data_vec[index].props & GID_TABLE_ENTRY_INVALID)
 		return -EAGAIN;
-	}
 
 	memcpy(gid, &table->data_vec[index].gid, sizeof(*gid));
 	if (attr) {
@@ -364,7 +438,6 @@ static int __ib_cache_gid_get(struct ib_device *ib_dev, u8 port, int index,
 			dev_hold(attr->ndev);
 	}
 
-	read_unlock_irqrestore(&table->data_vec[index].lock, flags);
 	return 0;
 }
 
@@ -378,17 +451,21 @@ static int _ib_cache_gid_table_find(struct ib_device *ib_dev,
 	struct ib_gid_table *table;
 	u8 p;
 	int local_index;
+	unsigned long flags;
 
 	for (p = 0; p < ib_dev->phys_port_cnt; p++) {
 		table = ports_table[p];
-		local_index = find_gid(table, gid, val, false, mask);
+		read_lock_irqsave(&table->rwlock, flags);
+		local_index = find_gid(table, gid, val, false, mask, NULL);
 		if (local_index >= 0) {
 			if (index)
 				*index = local_index;
 			if (port)
 				*port = p + rdma_start_port(ib_dev);
+			read_unlock_irqrestore(&table->rwlock, flags);
 			return 0;
 		}
+		read_unlock_irqrestore(&table->rwlock, flags);
 	}
 
 	return -ENOENT;
@@ -396,11 +473,13 @@ static int _ib_cache_gid_table_find(struct ib_device *ib_dev,
 
 static int ib_cache_gid_find(struct ib_device *ib_dev,
 			     const union ib_gid *gid,
+			     enum ib_gid_type gid_type,
 			     struct net_device *ndev, u8 *port,
 			     u16 *index)
 {
-	unsigned long mask = GID_ATTR_FIND_MASK_GID;
-	struct ib_gid_attr gid_attr_val = {.ndev = ndev};
+	unsigned long mask = GID_ATTR_FIND_MASK_GID |
+			     GID_ATTR_FIND_MASK_GID_TYPE;
+	struct ib_gid_attr gid_attr_val = {.ndev = ndev, .gid_type = gid_type};
 
 	if (ndev)
 		mask |= GID_ATTR_FIND_MASK_NETDEV;
@@ -411,14 +490,17 @@ static int ib_cache_gid_find(struct ib_device *ib_dev,
 
 int ib_find_cached_gid_by_port(struct ib_device *ib_dev,
 			       const union ib_gid *gid,
+			       enum ib_gid_type gid_type,
 			       u8 port, struct net_device *ndev,
 			       u16 *index)
 {
 	int local_index;
 	struct ib_gid_table **ports_table = ib_dev->cache.gid_cache;
 	struct ib_gid_table *table;
-	unsigned long mask = GID_ATTR_FIND_MASK_GID;
-	struct ib_gid_attr val = {.ndev = ndev};
+	unsigned long mask = GID_ATTR_FIND_MASK_GID |
+			     GID_ATTR_FIND_MASK_GID_TYPE;
+	struct ib_gid_attr val = {.ndev = ndev, .gid_type = gid_type};
+	unsigned long flags;
 
 	if (port < rdma_start_port(ib_dev) ||
 	    port > rdma_end_port(ib_dev))
@@ -429,13 +511,16 @@ int ib_find_cached_gid_by_port(struct ib_device *ib_dev,
 	if (ndev)
 		mask |= GID_ATTR_FIND_MASK_NETDEV;
 
-	local_index = find_gid(table, gid, &val, false, mask);
+	read_lock_irqsave(&table->rwlock, flags);
+	local_index = find_gid(table, gid, &val, false, mask, NULL);
 	if (local_index >= 0) {
 		if (index)
 			*index = local_index;
+		read_unlock_irqrestore(&table->rwlock, flags);
 		return 0;
 	}
 
+	read_unlock_irqrestore(&table->rwlock, flags);
 	return -ENOENT;
 }
 EXPORT_SYMBOL(ib_find_cached_gid_by_port);
@@ -472,6 +557,7 @@ static int ib_cache_gid_find_by_filter(struct ib_device *ib_dev,
 	struct ib_gid_table **ports_table = ib_dev->cache.gid_cache;
 	struct ib_gid_table *table;
 	unsigned int i;
+	unsigned long flags;
 	bool found = false;
 
 	if (!ports_table)
@@ -484,11 +570,10 @@ static int ib_cache_gid_find_by_filter(struct ib_device *ib_dev,
 
 	table = ports_table[port - rdma_start_port(ib_dev)];
 
+	read_lock_irqsave(&table->rwlock, flags);
 	for (i = 0; i < table->sz; i++) {
 		struct ib_gid_attr attr;
-		unsigned long flags;
 
-		read_lock_irqsave(&table->data_vec[i].lock, flags);
 		if (table->data_vec[i].props & GID_TABLE_ENTRY_INVALID)
 			goto next;
 
@@ -501,11 +586,10 @@ static int ib_cache_gid_find_by_filter(struct ib_device *ib_dev,
 			found = true;
 
 next:
-		read_unlock_irqrestore(&table->data_vec[i].lock, flags);
-
 		if (found)
 			break;
 	}
+	read_unlock_irqrestore(&table->rwlock, flags);
 
 	if (!found)
 		return -ENOENT;
@@ -517,9 +601,9 @@ next:
 
 static struct ib_gid_table *alloc_gid_table(int sz)
 {
-	unsigned int i;
 	struct ib_gid_table *table =
 		kzalloc(sizeof(struct ib_gid_table), GFP_KERNEL);
+
 	if (!table)
 		return NULL;
 
@@ -530,9 +614,7 @@ static struct ib_gid_table *alloc_gid_table(int sz)
 	mutex_init(&table->lock);
 
 	table->sz = sz;
-
-	for (i = 0; i < sz; i++)
-		rwlock_init(&table->data_vec[i].lock);
+	rwlock_init(&table->rwlock);
 
 	return table;
 
@@ -553,30 +635,37 @@ static void cleanup_gid_table_port(struct ib_device *ib_dev, u8 port,
 				   struct ib_gid_table *table)
 {
 	int i;
+	bool deleted = false;
 
 	if (!table)
 		return;
 
+	write_lock_irq(&table->rwlock);
 	for (i = 0; i < table->sz; ++i) {
 		if (memcmp(&table->data_vec[i].gid, &zgid,
 			   sizeof(table->data_vec[i].gid)))
-			del_gid(ib_dev, port, table, i,
-				table->data_vec[i].props &
-				GID_ATTR_FIND_MASK_DEFAULT);
+			if (!del_gid(ib_dev, port, table, i,
+				     table->data_vec[i].props &
+				     GID_ATTR_FIND_MASK_DEFAULT))
+				deleted = true;
 	}
+	write_unlock_irq(&table->rwlock);
+
+	if (deleted)
+		dispatch_gid_change_event(ib_dev, port);
 }
 
 void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port,
 				  struct net_device *ndev,
+				  unsigned long gid_type_mask,
 				  enum ib_cache_gid_default_mode mode)
 {
 	struct ib_gid_table **ports_table = ib_dev->cache.gid_cache;
 	union ib_gid gid;
 	struct ib_gid_attr gid_attr;
+	struct ib_gid_attr zattr_type = zattr;
 	struct ib_gid_table *table;
-	int ix;
-	union ib_gid current_gid;
-	struct ib_gid_attr current_gid_attr = {};
+	unsigned int gid_type;
 
 	table  = ports_table[port - rdma_start_port(ib_dev)];
 
@@ -584,46 +673,82 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port,
 	memset(&gid_attr, 0, sizeof(gid_attr));
 	gid_attr.ndev = ndev;
 
-	mutex_lock(&table->lock);
-	ix = find_gid(table, NULL, NULL, true, GID_ATTR_FIND_MASK_DEFAULT);
-
-	/* Coudn't find default GID location */
-	WARN_ON(ix < 0);
-
-	if (!__ib_cache_gid_get(ib_dev, port, ix,
-				&current_gid, &current_gid_attr) &&
-	    mode == IB_CACHE_GID_DEFAULT_MODE_SET &&
-	    !memcmp(&gid, &current_gid, sizeof(gid)) &&
-	    !memcmp(&gid_attr, &current_gid_attr, sizeof(gid_attr)))
-		goto unlock;
-
-	if ((memcmp(&current_gid, &zgid, sizeof(current_gid)) ||
-	     memcmp(&current_gid_attr, &zattr,
-		    sizeof(current_gid_attr))) &&
-	    del_gid(ib_dev, port, table, ix, true)) {
-		pr_warn("ib_cache_gid: can't delete index %d for default gid %pI6\n",
-			ix, gid.raw);
-		goto unlock;
-	}
+	for (gid_type = 0; gid_type < IB_GID_TYPE_SIZE; ++gid_type) {
+		int ix;
+		union ib_gid current_gid;
+		struct ib_gid_attr current_gid_attr = {};
+
+		if (1UL << gid_type & ~gid_type_mask)
+			continue;
+
+		gid_attr.gid_type = gid_type;
+
+		mutex_lock(&table->lock);
+		write_lock_irq(&table->rwlock);
+		ix = find_gid(table, NULL, &gid_attr, true,
+			      GID_ATTR_FIND_MASK_GID_TYPE |
+			      GID_ATTR_FIND_MASK_DEFAULT,
+			      NULL);
+
+		/* Coudn't find default GID location */
+		WARN_ON(ix < 0);
+
+		zattr_type.gid_type = gid_type;
+
+		if (!__ib_cache_gid_get(ib_dev, port, ix,
+					&current_gid, &current_gid_attr) &&
+		    mode == IB_CACHE_GID_DEFAULT_MODE_SET &&
+		    !memcmp(&gid, &current_gid, sizeof(gid)) &&
+		    !memcmp(&gid_attr, &current_gid_attr, sizeof(gid_attr)))
+			goto release;
+
+		if (memcmp(&current_gid, &zgid, sizeof(current_gid)) ||
+		    memcmp(&current_gid_attr, &zattr_type,
+			   sizeof(current_gid_attr))) {
+			if (del_gid(ib_dev, port, table, ix, true)) {
+				pr_warn("ib_cache_gid: can't delete index %d for default gid %pI6\n",
+					ix, gid.raw);
+				goto release;
+			} else {
+				dispatch_gid_change_event(ib_dev, port);
+			}
+		}
 
-	if (mode == IB_CACHE_GID_DEFAULT_MODE_SET)
-		if (add_gid(ib_dev, port, table, ix, &gid, &gid_attr, true))
-			pr_warn("ib_cache_gid: unable to add default gid %pI6\n",
-				gid.raw);
+		if (mode == IB_CACHE_GID_DEFAULT_MODE_SET) {
+			if (add_gid(ib_dev, port, table, ix, &gid, &gid_attr, true))
+				pr_warn("ib_cache_gid: unable to add default gid %pI6\n",
+					gid.raw);
+			else
+				dispatch_gid_change_event(ib_dev, port);
+		}
 
-unlock:
-	if (current_gid_attr.ndev)
-		dev_put(current_gid_attr.ndev);
-	mutex_unlock(&table->lock);
+release:
+		if (current_gid_attr.ndev)
+			dev_put(current_gid_attr.ndev);
+		write_unlock_irq(&table->rwlock);
+		mutex_unlock(&table->lock);
+	}
 }
 
 static int gid_table_reserve_default(struct ib_device *ib_dev, u8 port,
 				     struct ib_gid_table *table)
 {
-	if (rdma_protocol_roce(ib_dev, port)) {
-		struct ib_gid_table_entry *entry = &table->data_vec[0];
+	unsigned int i;
+	unsigned long roce_gid_type_mask;
+	unsigned int num_default_gids;
+	unsigned int current_gid = 0;
+
+	roce_gid_type_mask = roce_gid_type_mask_support(ib_dev, port);
+	num_default_gids = hweight_long(roce_gid_type_mask);
+	for (i = 0; i < num_default_gids && i < table->sz; i++) {
+		struct ib_gid_table_entry *entry =
+			&table->data_vec[i];
 
 		entry->props |= GID_TABLE_ENTRY_DEFAULT;
+		current_gid = find_next_bit(&roce_gid_type_mask,
+					    BITS_PER_LONG,
+					    current_gid);
+		entry->attr.gid_type = current_gid++;
 	}
 
 	return 0;
@@ -728,20 +853,30 @@ int ib_get_cached_gid(struct ib_device *device,
 		      union ib_gid     *gid,
 		      struct ib_gid_attr *gid_attr)
 {
+	int res;
+	unsigned long flags;
+	struct ib_gid_table **ports_table = device->cache.gid_cache;
+	struct ib_gid_table *table = ports_table[port_num - rdma_start_port(device)];
+
 	if (port_num < rdma_start_port(device) || port_num > rdma_end_port(device))
 		return -EINVAL;
 
-	return __ib_cache_gid_get(device, port_num, index, gid, gid_attr);
+	read_lock_irqsave(&table->rwlock, flags);
+	res = __ib_cache_gid_get(device, port_num, index, gid, gid_attr);
+	read_unlock_irqrestore(&table->rwlock, flags);
+
+	return res;
 }
 EXPORT_SYMBOL(ib_get_cached_gid);
 
 int ib_find_cached_gid(struct ib_device *device,
 		       const union ib_gid *gid,
+		       enum ib_gid_type gid_type,
 		       struct net_device *ndev,
 		       u8               *port_num,
 		       u16              *index)
 {
-	return ib_cache_gid_find(device, gid, ndev, port_num, index);
+	return ib_cache_gid_find(device, gid, gid_type, ndev, port_num, index);
 }
 EXPORT_SYMBOL(ib_find_cached_gid);
 
@@ -956,10 +1091,12 @@ static void ib_cache_update(struct ib_device *device,
 
 	device->cache.pkey_cache[port - rdma_start_port(device)] = pkey_cache;
 	if (!use_roce_gid_table) {
+		write_lock(&table->rwlock);
 		for (i = 0; i < gid_cache->table_len; i++) {
 			modify_gid(device, port, table, i, gid_cache->table + i,
 				   &zattr, false);
 		}
+		write_unlock(&table->rwlock);
 	}
 
 	device->cache.lmc_cache[port - rdma_start_port(device)] = tprops->lmc;
diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c
index 0a26dd6..1d92e09 100644
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -364,7 +364,7 @@ static int cm_init_av_by_path(struct ib_sa_path_rec *path, struct cm_av *av)
 	read_lock_irqsave(&cm.device_lock, flags);
 	list_for_each_entry(cm_dev, &cm.device_list, list) {
 		if (!ib_find_cached_gid(cm_dev->ib_device, &path->sgid,
-					ndev, &p, NULL)) {
+					path->gid_type, ndev, &p, NULL)) {
 			port = cm_dev->port[p-1];
 			break;
 		}
@@ -782,11 +782,11 @@ static void cm_enter_timewait(struct cm_id_private *cm_id_priv)
 	wait_time = cm_convert_to_ms(cm_id_priv->av.timeout);
 
 	/* Check if the device started its remove_one */
-	spin_lock_irq(&cm.lock);
+	spin_lock_irqsave(&cm.lock, flags);
 	if (!cm_dev->going_down)
 		queue_delayed_work(cm.wq, &cm_id_priv->timewait_info->work.work,
 				   msecs_to_jiffies(wait_time));
-	spin_unlock_irq(&cm.lock);
+	spin_unlock_irqrestore(&cm.lock, flags);
 
 	cm_id_priv->timewait_info = NULL;
 }
@@ -1600,6 +1600,8 @@ static int cm_req_handler(struct cm_work *work)
 	struct ib_cm_id *cm_id;
 	struct cm_id_private *cm_id_priv, *listen_cm_id_priv;
 	struct cm_req_msg *req_msg;
+	union ib_gid gid;
+	struct ib_gid_attr gid_attr;
 	int ret;
 
 	req_msg = (struct cm_req_msg *)work->mad_recv_wc->recv_buf.mad;
@@ -1639,11 +1641,31 @@ static int cm_req_handler(struct cm_work *work)
 	cm_format_paths_from_req(req_msg, &work->path[0], &work->path[1]);
 
 	memcpy(work->path[0].dmac, cm_id_priv->av.ah_attr.dmac, ETH_ALEN);
-	ret = cm_init_av_by_path(&work->path[0], &cm_id_priv->av);
+	work->path[0].hop_limit = cm_id_priv->av.ah_attr.grh.hop_limit;
+	ret = ib_get_cached_gid(work->port->cm_dev->ib_device,
+				work->port->port_num,
+				cm_id_priv->av.ah_attr.grh.sgid_index,
+				&gid, &gid_attr);
+	if (!ret) {
+		if (gid_attr.ndev) {
+			work->path[0].ifindex = gid_attr.ndev->ifindex;
+			work->path[0].net = dev_net(gid_attr.ndev);
+			dev_put(gid_attr.ndev);
+		}
+		work->path[0].gid_type = gid_attr.gid_type;
+		ret = cm_init_av_by_path(&work->path[0], &cm_id_priv->av);
+	}
 	if (ret) {
-		ib_get_cached_gid(work->port->cm_dev->ib_device,
-				  work->port->port_num, 0, &work->path[0].sgid,
-				  NULL);
+		int err = ib_get_cached_gid(work->port->cm_dev->ib_device,
+					    work->port->port_num, 0,
+					    &work->path[0].sgid,
+					    &gid_attr);
+		if (!err && gid_attr.ndev) {
+			work->path[0].ifindex = gid_attr.ndev->ifindex;
+			work->path[0].net = dev_net(gid_attr.ndev);
+			dev_put(gid_attr.ndev);
+		}
+		work->path[0].gid_type = gid_attr.gid_type;
 		ib_send_cm_rej(cm_id, IB_CM_REJ_INVALID_GID,
 			       &work->path[0].sgid, sizeof work->path[0].sgid,
 			       NULL, 0);
@@ -3482,6 +3504,7 @@ int ib_cm_notify(struct ib_cm_id *cm_id, enum ib_event_type event)
 EXPORT_SYMBOL(ib_cm_notify);
 
 static void cm_recv_handler(struct ib_mad_agent *mad_agent,
+			    struct ib_mad_send_buf *send_buf,
 			    struct ib_mad_recv_wc *mad_recv_wc)
 {
 	struct cm_port *port = mad_agent->context;
@@ -3731,16 +3754,6 @@ int ib_cm_init_qp_attr(struct ib_cm_id *cm_id,
 }
 EXPORT_SYMBOL(ib_cm_init_qp_attr);
 
-static void cm_get_ack_delay(struct cm_device *cm_dev)
-{
-	struct ib_device_attr attr;
-
-	if (ib_query_device(cm_dev->ib_device, &attr))
-		cm_dev->ack_delay = 0; /* acks will rely on packet life time */
-	else
-		cm_dev->ack_delay = attr.local_ca_ack_delay;
-}
-
 static ssize_t cm_show_counter(struct kobject *obj, struct attribute *attr,
 			       char *buf)
 {
@@ -3852,7 +3865,7 @@ static void cm_add_one(struct ib_device *ib_device)
 		return;
 
 	cm_dev->ib_device = ib_device;
-	cm_get_ack_delay(cm_dev);
+	cm_dev->ack_delay = ib_device->attrs.local_ca_ack_delay;
 	cm_dev->going_down = 0;
 	cm_dev->device = device_create(&cm_class, &ib_device->dev,
 				       MKDEV(0, 0), NULL,
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 2d762a2..9729639 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -38,6 +38,7 @@
 #include <linux/in6.h>
 #include <linux/mutex.h>
 #include <linux/random.h>
+#include <linux/igmp.h>
 #include <linux/idr.h>
 #include <linux/inetdevice.h>
 #include <linux/slab.h>
@@ -60,6 +61,8 @@
 #include <rdma/ib_sa.h>
 #include <rdma/iw_cm.h>
 
+#include "core_priv.h"
+
 MODULE_AUTHOR("Sean Hefty");
 MODULE_DESCRIPTION("Generic RDMA CM Agent");
 MODULE_LICENSE("Dual BSD/GPL");
@@ -150,6 +153,7 @@ struct cma_device {
 	struct completion	comp;
 	atomic_t		refcount;
 	struct list_head	id_list;
+	enum ib_gid_type	*default_gid_type;
 };
 
 struct rdma_bind_list {
@@ -185,6 +189,67 @@ enum {
 	CMA_OPTION_AFONLY,
 };
 
+void cma_ref_dev(struct cma_device *cma_dev)
+{
+	atomic_inc(&cma_dev->refcount);
+}
+
+struct cma_device *cma_enum_devices_by_ibdev(cma_device_filter	filter,
+					     void		*cookie)
+{
+	struct cma_device *cma_dev;
+	struct cma_device *found_cma_dev = NULL;
+
+	mutex_lock(&lock);
+
+	list_for_each_entry(cma_dev, &dev_list, list)
+		if (filter(cma_dev->device, cookie)) {
+			found_cma_dev = cma_dev;
+			break;
+		}
+
+	if (found_cma_dev)
+		cma_ref_dev(found_cma_dev);
+	mutex_unlock(&lock);
+	return found_cma_dev;
+}
+
+int cma_get_default_gid_type(struct cma_device *cma_dev,
+			     unsigned int port)
+{
+	if (port < rdma_start_port(cma_dev->device) ||
+	    port > rdma_end_port(cma_dev->device))
+		return -EINVAL;
+
+	return cma_dev->default_gid_type[port - rdma_start_port(cma_dev->device)];
+}
+
+int cma_set_default_gid_type(struct cma_device *cma_dev,
+			     unsigned int port,
+			     enum ib_gid_type default_gid_type)
+{
+	unsigned long supported_gids;
+
+	if (port < rdma_start_port(cma_dev->device) ||
+	    port > rdma_end_port(cma_dev->device))
+		return -EINVAL;
+
+	supported_gids = roce_gid_type_mask_support(cma_dev->device, port);
+
+	if (!(supported_gids & 1 << default_gid_type))
+		return -EINVAL;
+
+	cma_dev->default_gid_type[port - rdma_start_port(cma_dev->device)] =
+		default_gid_type;
+
+	return 0;
+}
+
+struct ib_device *cma_get_ib_dev(struct cma_device *cma_dev)
+{
+	return cma_dev->device;
+}
+
 /*
  * Device removal can occur at anytime, so we need extra handling to
  * serialize notifying the user of device removal with other callbacks.
@@ -228,6 +293,7 @@ struct rdma_id_private {
 	u8			tos;
 	u8			reuseaddr;
 	u8			afonly;
+	enum ib_gid_type	gid_type;
 };
 
 struct cma_multicast {
@@ -239,6 +305,7 @@ struct cma_multicast {
 	void			*context;
 	struct sockaddr_storage	addr;
 	struct kref		mcref;
+	bool			igmp_joined;
 };
 
 struct cma_work {
@@ -335,18 +402,48 @@ static inline void cma_set_ip_ver(struct cma_hdr *hdr, u8 ip_ver)
 	hdr->ip_version = (ip_ver << 4) | (hdr->ip_version & 0xF);
 }
 
-static void cma_attach_to_dev(struct rdma_id_private *id_priv,
-			      struct cma_device *cma_dev)
+static int cma_igmp_send(struct net_device *ndev, union ib_gid *mgid, bool join)
 {
-	atomic_inc(&cma_dev->refcount);
+	struct in_device *in_dev = NULL;
+
+	if (ndev) {
+		rtnl_lock();
+		in_dev = __in_dev_get_rtnl(ndev);
+		if (in_dev) {
+			if (join)
+				ip_mc_inc_group(in_dev,
+						*(__be32 *)(mgid->raw + 12));
+			else
+				ip_mc_dec_group(in_dev,
+						*(__be32 *)(mgid->raw + 12));
+		}
+		rtnl_unlock();
+	}
+	return (in_dev) ? 0 : -ENODEV;
+}
+
+static void _cma_attach_to_dev(struct rdma_id_private *id_priv,
+			       struct cma_device *cma_dev)
+{
+	cma_ref_dev(cma_dev);
 	id_priv->cma_dev = cma_dev;
+	id_priv->gid_type = 0;
 	id_priv->id.device = cma_dev->device;
 	id_priv->id.route.addr.dev_addr.transport =
 		rdma_node_get_transport(cma_dev->device->node_type);
 	list_add_tail(&id_priv->list, &cma_dev->id_list);
 }
 
-static inline void cma_deref_dev(struct cma_device *cma_dev)
+static void cma_attach_to_dev(struct rdma_id_private *id_priv,
+			      struct cma_device *cma_dev)
+{
+	_cma_attach_to_dev(id_priv, cma_dev);
+	id_priv->gid_type =
+		cma_dev->default_gid_type[id_priv->id.port_num -
+					  rdma_start_port(cma_dev->device)];
+}
+
+void cma_deref_dev(struct cma_device *cma_dev)
 {
 	if (atomic_dec_and_test(&cma_dev->refcount))
 		complete(&cma_dev->comp);
@@ -441,6 +538,7 @@ static int cma_translate_addr(struct sockaddr *addr, struct rdma_dev_addr *dev_a
 }
 
 static inline int cma_validate_port(struct ib_device *device, u8 port,
+				    enum ib_gid_type gid_type,
 				      union ib_gid *gid, int dev_type,
 				      int bound_if_index)
 {
@@ -453,10 +551,25 @@ static inline int cma_validate_port(struct ib_device *device, u8 port,
 	if ((dev_type != ARPHRD_INFINIBAND) && rdma_protocol_ib(device, port))
 		return ret;
 
-	if (dev_type == ARPHRD_ETHER)
+	if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) {
 		ndev = dev_get_by_index(&init_net, bound_if_index);
+		if (ndev && ndev->flags & IFF_LOOPBACK) {
+			pr_info("detected loopback device\n");
+			dev_put(ndev);
 
-	ret = ib_find_cached_gid_by_port(device, gid, port, ndev, NULL);
+			if (!device->get_netdev)
+				return -EOPNOTSUPP;
+
+			ndev = device->get_netdev(device, port);
+			if (!ndev)
+				return -ENODEV;
+		}
+	} else {
+		gid_type = IB_GID_TYPE_IB;
+	}
+
+	ret = ib_find_cached_gid_by_port(device, gid, gid_type, port,
+					 ndev, NULL);
 
 	if (ndev)
 		dev_put(ndev);
@@ -490,7 +603,10 @@ static int cma_acquire_dev(struct rdma_id_private *id_priv,
 		gidp = rdma_protocol_roce(cma_dev->device, port) ?
 		       &iboe_gid : &gid;
 
-		ret = cma_validate_port(cma_dev->device, port, gidp,
+		ret = cma_validate_port(cma_dev->device, port,
+					rdma_protocol_ib(cma_dev->device, port) ?
+					IB_GID_TYPE_IB :
+					listen_id_priv->gid_type, gidp,
 					dev_addr->dev_type,
 					dev_addr->bound_dev_if);
 		if (!ret) {
@@ -509,8 +625,11 @@ static int cma_acquire_dev(struct rdma_id_private *id_priv,
 			gidp = rdma_protocol_roce(cma_dev->device, port) ?
 			       &iboe_gid : &gid;
 
-			ret = cma_validate_port(cma_dev->device, port, gidp,
-						dev_addr->dev_type,
+			ret = cma_validate_port(cma_dev->device, port,
+						rdma_protocol_ib(cma_dev->device, port) ?
+						IB_GID_TYPE_IB :
+						cma_dev->default_gid_type[port - 1],
+						gidp, dev_addr->dev_type,
 						dev_addr->bound_dev_if);
 			if (!ret) {
 				id_priv->id.port_num = port;
@@ -1437,8 +1556,24 @@ static void cma_leave_mc_groups(struct rdma_id_private *id_priv)
 				      id_priv->id.port_num)) {
 			ib_sa_free_multicast(mc->multicast.ib);
 			kfree(mc);
-		} else
+		} else {
+			if (mc->igmp_joined) {
+				struct rdma_dev_addr *dev_addr =
+					&id_priv->id.route.addr.dev_addr;
+				struct net_device *ndev = NULL;
+
+				if (dev_addr->bound_dev_if)
+					ndev = dev_get_by_index(&init_net,
+								dev_addr->bound_dev_if);
+				if (ndev) {
+					cma_igmp_send(ndev,
+						      &mc->multicast.ib->rec.mgid,
+						      false);
+					dev_put(ndev);
+				}
+			}
 			kref_put(&mc->mcref, release_mc);
+		}
 	}
 }
 
@@ -1896,7 +2031,6 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 	struct rdma_id_private *listen_id, *conn_id;
 	struct rdma_cm_event event;
 	int ret;
-	struct ib_device_attr attr;
 	struct sockaddr *laddr = (struct sockaddr *)&iw_event->local_addr;
 	struct sockaddr *raddr = (struct sockaddr *)&iw_event->remote_addr;
 
@@ -1938,13 +2072,6 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id,
 	memcpy(cma_src_addr(conn_id), laddr, rdma_addr_size(laddr));
 	memcpy(cma_dst_addr(conn_id), raddr, rdma_addr_size(raddr));
 
-	ret = ib_query_device(conn_id->id.device, &attr);
-	if (ret) {
-		mutex_unlock(&conn_id->handler_mutex);
-		rdma_destroy_id(new_cm_id);
-		goto out;
-	}
-
 	memset(&event, 0, sizeof event);
 	event.event = RDMA_CM_EVENT_CONNECT_REQUEST;
 	event.param.conn.private_data = iw_event->private_data;
@@ -2051,7 +2178,7 @@ static void cma_listen_on_dev(struct rdma_id_private *id_priv,
 	memcpy(cma_src_addr(dev_id_priv), cma_src_addr(id_priv),
 	       rdma_addr_size(cma_src_addr(id_priv)));
 
-	cma_attach_to_dev(dev_id_priv, cma_dev);
+	_cma_attach_to_dev(dev_id_priv, cma_dev);
 	list_add_tail(&dev_id_priv->listen_list, &id_priv->listen_list);
 	atomic_inc(&id_priv->refcount);
 	dev_id_priv->internal_id = 1;
@@ -2321,8 +2448,23 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv)
 
 	if (addr->dev_addr.bound_dev_if) {
 		ndev = dev_get_by_index(&init_net, addr->dev_addr.bound_dev_if);
+		if (!ndev)
+			return -ENODEV;
+
+		if (ndev->flags & IFF_LOOPBACK) {
+			dev_put(ndev);
+			if (!id_priv->id.device->get_netdev)
+				return -EOPNOTSUPP;
+
+			ndev = id_priv->id.device->get_netdev(id_priv->id.device,
+							      id_priv->id.port_num);
+			if (!ndev)
+				return -ENODEV;
+		}
+
 		route->path_rec->net = &init_net;
-		route->path_rec->ifindex = addr->dev_addr.bound_dev_if;
+		route->path_rec->ifindex = ndev->ifindex;
+		route->path_rec->gid_type = id_priv->gid_type;
 	}
 	if (!ndev) {
 		ret = -ENODEV;
@@ -2336,7 +2478,14 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv)
 	rdma_ip2gid((struct sockaddr *)&id_priv->id.route.addr.dst_addr,
 		    &route->path_rec->dgid);
 
-	route->path_rec->hop_limit = 1;
+	/* Use the hint from IP Stack to select GID Type */
+	if (route->path_rec->gid_type < ib_network_to_gid_type(addr->dev_addr.network))
+		route->path_rec->gid_type = ib_network_to_gid_type(addr->dev_addr.network);
+	if (((struct sockaddr *)&id_priv->id.route.addr.dst_addr)->sa_family != AF_IB)
+		/* TODO: get the hoplimit from the inet/inet6 device */
+		route->path_rec->hop_limit = addr->dev_addr.hoplimit;
+	else
+		route->path_rec->hop_limit = 1;
 	route->path_rec->reversible = 1;
 	route->path_rec->pkey = cpu_to_be16(0xffff);
 	route->path_rec->mtu_selector = IB_SA_EQ;
@@ -3534,12 +3683,23 @@ static int cma_ib_mc_handler(int status, struct ib_sa_multicast *multicast)
 	event.status = status;
 	event.param.ud.private_data = mc->context;
 	if (!status) {
+		struct rdma_dev_addr *dev_addr =
+			&id_priv->id.route.addr.dev_addr;
+		struct net_device *ndev =
+			dev_get_by_index(&init_net, dev_addr->bound_dev_if);
+		enum ib_gid_type gid_type =
+			id_priv->cma_dev->default_gid_type[id_priv->id.port_num -
+			rdma_start_port(id_priv->cma_dev->device)];
+
 		event.event = RDMA_CM_EVENT_MULTICAST_JOIN;
 		ib_init_ah_from_mcmember(id_priv->id.device,
 					 id_priv->id.port_num, &multicast->rec,
+					 ndev, gid_type,
 					 &event.param.ud.ah_attr);
 		event.param.ud.qp_num = 0xFFFFFF;
 		event.param.ud.qkey = be32_to_cpu(multicast->rec.qkey);
+		if (ndev)
+			dev_put(ndev);
 	} else
 		event.event = RDMA_CM_EVENT_MULTICAST_ERROR;
 
@@ -3672,9 +3832,10 @@ static int cma_iboe_join_multicast(struct rdma_id_private *id_priv,
 {
 	struct iboe_mcast_work *work;
 	struct rdma_dev_addr *dev_addr = &id_priv->id.route.addr.dev_addr;
-	int err;
+	int err = 0;
 	struct sockaddr *addr = (struct sockaddr *)&mc->addr;
 	struct net_device *ndev = NULL;
+	enum ib_gid_type gid_type;
 
 	if (cma_zero_addr((struct sockaddr *)&mc->addr))
 		return -EINVAL;
@@ -3704,9 +3865,25 @@ static int cma_iboe_join_multicast(struct rdma_id_private *id_priv,
 	mc->multicast.ib->rec.rate = iboe_get_rate(ndev);
 	mc->multicast.ib->rec.hop_limit = 1;
 	mc->multicast.ib->rec.mtu = iboe_get_mtu(ndev->mtu);
+
+	gid_type = id_priv->cma_dev->default_gid_type[id_priv->id.port_num -
+		   rdma_start_port(id_priv->cma_dev->device)];
+	if (addr->sa_family == AF_INET) {
+		if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
+			err = cma_igmp_send(ndev, &mc->multicast.ib->rec.mgid,
+					    true);
+		if (!err) {
+			mc->igmp_joined = true;
+			mc->multicast.ib->rec.hop_limit = IPV6_DEFAULT_HOPLIMIT;
+		}
+	} else {
+		if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
+			err = -ENOTSUPP;
+	}
 	dev_put(ndev);
-	if (!mc->multicast.ib->rec.mtu) {
-		err = -EINVAL;
+	if (err || !mc->multicast.ib->rec.mtu) {
+		if (!err)
+			err = -EINVAL;
 		goto out2;
 	}
 	rdma_ip2gid((struct sockaddr *)&id_priv->id.route.addr.src_addr,
@@ -3745,7 +3922,7 @@ int rdma_join_multicast(struct rdma_cm_id *id, struct sockaddr *addr,
 	memcpy(&mc->addr, addr, rdma_addr_size(addr));
 	mc->context = context;
 	mc->id_priv = id_priv;
-
+	mc->igmp_joined = false;
 	spin_lock(&id_priv->lock);
 	list_add(&mc->list, &id_priv->mc_list);
 	spin_unlock(&id_priv->lock);
@@ -3790,9 +3967,25 @@ void rdma_leave_multicast(struct rdma_cm_id *id, struct sockaddr *addr)
 			if (rdma_cap_ib_mcast(id->device, id->port_num)) {
 				ib_sa_free_multicast(mc->multicast.ib);
 				kfree(mc);
-			} else if (rdma_protocol_roce(id->device, id->port_num))
+			} else if (rdma_protocol_roce(id->device, id->port_num)) {
+				if (mc->igmp_joined) {
+					struct rdma_dev_addr *dev_addr =
+						&id->route.addr.dev_addr;
+					struct net_device *ndev = NULL;
+
+					if (dev_addr->bound_dev_if)
+						ndev = dev_get_by_index(&init_net,
+									dev_addr->bound_dev_if);
+					if (ndev) {
+						cma_igmp_send(ndev,
+							      &mc->multicast.ib->rec.mgid,
+							      false);
+						dev_put(ndev);
+					}
+					mc->igmp_joined = false;
+				}
 				kref_put(&mc->mcref, release_mc);
-
+			}
 			return;
 		}
 	}
@@ -3861,12 +4054,27 @@ static void cma_add_one(struct ib_device *device)
 {
 	struct cma_device *cma_dev;
 	struct rdma_id_private *id_priv;
+	unsigned int i;
+	unsigned long supported_gids = 0;
 
 	cma_dev = kmalloc(sizeof *cma_dev, GFP_KERNEL);
 	if (!cma_dev)
 		return;
 
 	cma_dev->device = device;
+	cma_dev->default_gid_type = kcalloc(device->phys_port_cnt,
+					    sizeof(*cma_dev->default_gid_type),
+					    GFP_KERNEL);
+	if (!cma_dev->default_gid_type) {
+		kfree(cma_dev);
+		return;
+	}
+	for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
+		supported_gids = roce_gid_type_mask_support(device, i);
+		WARN_ON(!supported_gids);
+		cma_dev->default_gid_type[i - rdma_start_port(device)] =
+			find_first_bit(&supported_gids, BITS_PER_LONG);
+	}
 
 	init_completion(&cma_dev->comp);
 	atomic_set(&cma_dev->refcount, 1);
@@ -3946,6 +4154,7 @@ static void cma_remove_one(struct ib_device *device, void *client_data)
 	mutex_unlock(&lock);
 
 	cma_process_remove(cma_dev);
+	kfree(cma_dev->default_gid_type);
 	kfree(cma_dev);
 }
 
@@ -4079,6 +4288,7 @@ static int __init cma_init(void)
 
 	if (ibnl_add_client(RDMA_NL_RDMA_CM, RDMA_NL_RDMA_CM_NUM_OPS, cma_cb_table))
 		printk(KERN_WARNING "RDMA CMA: failed to add netlink callback\n");
+	cma_configfs_init();
 
 	return 0;
 
@@ -4093,6 +4303,7 @@ err_wq:
 
 static void __exit cma_cleanup(void)
 {
+	cma_configfs_exit();
 	ibnl_remove_client(RDMA_NL_RDMA_CM);
 	ib_unregister_client(&cma_client);
 	unregister_netdevice_notifier(&cma_nb);
diff --git a/drivers/infiniband/core/cma_configfs.c b/drivers/infiniband/core/cma_configfs.c
new file mode 100644
index 0000000..18b112a
--- /dev/null
+++ b/drivers/infiniband/core/cma_configfs.c
@@ -0,0 +1,321 @@
+/*
+ * Copyright (c) 2015, Mellanox Technologies inc.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/module.h>
+#include <linux/configfs.h>
+#include <rdma/ib_verbs.h>
+#include "core_priv.h"
+
+struct cma_device;
+
+struct cma_dev_group;
+
+struct cma_dev_port_group {
+	unsigned int		port_num;
+	struct cma_dev_group	*cma_dev_group;
+	struct config_group	group;
+};
+
+struct cma_dev_group {
+	char				name[IB_DEVICE_NAME_MAX];
+	struct config_group		device_group;
+	struct config_group		ports_group;
+	struct config_group		*default_dev_group[2];
+	struct config_group		**default_ports_group;
+	struct cma_dev_port_group	*ports;
+};
+
+static struct cma_dev_port_group *to_dev_port_group(struct config_item *item)
+{
+	struct config_group *group;
+
+	if (!item)
+		return NULL;
+
+	group = container_of(item, struct config_group, cg_item);
+	return container_of(group, struct cma_dev_port_group, group);
+}
+
+static bool filter_by_name(struct ib_device *ib_dev, void *cookie)
+{
+	return !strcmp(ib_dev->name, cookie);
+}
+
+static int cma_configfs_params_get(struct config_item *item,
+				   struct cma_device **pcma_dev,
+				   struct cma_dev_port_group **pgroup)
+{
+	struct cma_dev_port_group *group = to_dev_port_group(item);
+	struct cma_device *cma_dev;
+
+	if (!group)
+		return -ENODEV;
+
+	cma_dev = cma_enum_devices_by_ibdev(filter_by_name,
+					    group->cma_dev_group->name);
+	if (!cma_dev)
+		return -ENODEV;
+
+	*pcma_dev = cma_dev;
+	*pgroup = group;
+
+	return 0;
+}
+
+static void cma_configfs_params_put(struct cma_device *cma_dev)
+{
+	cma_deref_dev(cma_dev);
+}
+
+static ssize_t default_roce_mode_show(struct config_item *item,
+				      char *buf)
+{
+	struct cma_device *cma_dev;
+	struct cma_dev_port_group *group;
+	int gid_type;
+	ssize_t ret;
+
+	ret = cma_configfs_params_get(item, &cma_dev, &group);
+	if (ret)
+		return ret;
+
+	gid_type = cma_get_default_gid_type(cma_dev, group->port_num);
+	cma_configfs_params_put(cma_dev);
+
+	if (gid_type < 0)
+		return gid_type;
+
+	return sprintf(buf, "%s\n", ib_cache_gid_type_str(gid_type));
+}
+
+static ssize_t default_roce_mode_store(struct config_item *item,
+				       const char *buf, size_t count)
+{
+	struct cma_device *cma_dev;
+	struct cma_dev_port_group *group;
+	int gid_type = ib_cache_gid_parse_type_str(buf);
+	ssize_t ret;
+
+	if (gid_type < 0)
+		return -EINVAL;
+
+	ret = cma_configfs_params_get(item, &cma_dev, &group);
+	if (ret)
+		return ret;
+
+	ret = cma_set_default_gid_type(cma_dev, group->port_num, gid_type);
+
+	cma_configfs_params_put(cma_dev);
+
+	return !ret ? strnlen(buf, count) : ret;
+}
+
+CONFIGFS_ATTR(, default_roce_mode);
+
+static struct configfs_attribute *cma_configfs_attributes[] = {
+	&attr_default_roce_mode,
+	NULL,
+};
+
+static struct config_item_type cma_port_group_type = {
+	.ct_attrs	= cma_configfs_attributes,
+	.ct_owner	= THIS_MODULE
+};
+
+static int make_cma_ports(struct cma_dev_group *cma_dev_group,
+			  struct cma_device *cma_dev)
+{
+	struct ib_device *ibdev;
+	unsigned int i;
+	unsigned int ports_num;
+	struct cma_dev_port_group *ports;
+	struct config_group **ports_group;
+	int err;
+
+	ibdev = cma_get_ib_dev(cma_dev);
+
+	if (!ibdev)
+		return -ENODEV;
+
+	ports_num = ibdev->phys_port_cnt;
+	ports = kcalloc(ports_num, sizeof(*cma_dev_group->ports),
+			GFP_KERNEL);
+	ports_group = kcalloc(ports_num + 1, sizeof(*ports_group), GFP_KERNEL);
+
+	if (!ports || !ports_group) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	for (i = 0; i < ports_num; i++) {
+		char port_str[10];
+
+		ports[i].port_num = i + 1;
+		snprintf(port_str, sizeof(port_str), "%u", i + 1);
+		ports[i].cma_dev_group = cma_dev_group;
+		config_group_init_type_name(&ports[i].group,
+					    port_str,
+					    &cma_port_group_type);
+		ports_group[i] = &ports[i].group;
+	}
+	ports_group[i] = NULL;
+	cma_dev_group->default_ports_group = ports_group;
+	cma_dev_group->ports = ports;
+
+	return 0;
+free:
+	kfree(ports);
+	kfree(ports_group);
+	cma_dev_group->ports = NULL;
+	cma_dev_group->default_ports_group = NULL;
+	return err;
+}
+
+static void release_cma_dev(struct config_item  *item)
+{
+	struct config_group *group = container_of(item, struct config_group,
+						  cg_item);
+	struct cma_dev_group *cma_dev_group = container_of(group,
+							   struct cma_dev_group,
+							   device_group);
+
+	kfree(cma_dev_group);
+};
+
+static void release_cma_ports_group(struct config_item  *item)
+{
+	struct config_group *group = container_of(item, struct config_group,
+						  cg_item);
+	struct cma_dev_group *cma_dev_group = container_of(group,
+							   struct cma_dev_group,
+							   ports_group);
+
+	kfree(cma_dev_group->ports);
+	kfree(cma_dev_group->default_ports_group);
+	cma_dev_group->ports = NULL;
+	cma_dev_group->default_ports_group = NULL;
+};
+
+static struct configfs_item_operations cma_ports_item_ops = {
+	.release = release_cma_ports_group
+};
+
+static struct config_item_type cma_ports_group_type = {
+	.ct_item_ops	= &cma_ports_item_ops,
+	.ct_owner	= THIS_MODULE
+};
+
+static struct configfs_item_operations cma_device_item_ops = {
+	.release = release_cma_dev
+};
+
+static struct config_item_type cma_device_group_type = {
+	.ct_item_ops	= &cma_device_item_ops,
+	.ct_owner	= THIS_MODULE
+};
+
+static struct config_group *make_cma_dev(struct config_group *group,
+					 const char *name)
+{
+	int err = -ENODEV;
+	struct cma_device *cma_dev = cma_enum_devices_by_ibdev(filter_by_name,
+							       (void *)name);
+	struct cma_dev_group *cma_dev_group = NULL;
+
+	if (!cma_dev)
+		goto fail;
+
+	cma_dev_group = kzalloc(sizeof(*cma_dev_group), GFP_KERNEL);
+
+	if (!cma_dev_group) {
+		err = -ENOMEM;
+		goto fail;
+	}
+
+	strncpy(cma_dev_group->name, name, sizeof(cma_dev_group->name));
+
+	err = make_cma_ports(cma_dev_group, cma_dev);
+	if (err)
+		goto fail;
+
+	cma_dev_group->ports_group.default_groups =
+		cma_dev_group->default_ports_group;
+	config_group_init_type_name(&cma_dev_group->ports_group, "ports",
+				    &cma_ports_group_type);
+
+	cma_dev_group->device_group.default_groups
+		= cma_dev_group->default_dev_group;
+	cma_dev_group->default_dev_group[0] = &cma_dev_group->ports_group;
+	cma_dev_group->default_dev_group[1] = NULL;
+
+	config_group_init_type_name(&cma_dev_group->device_group, name,
+				    &cma_device_group_type);
+
+	cma_deref_dev(cma_dev);
+	return &cma_dev_group->device_group;
+
+fail:
+	if (cma_dev)
+		cma_deref_dev(cma_dev);
+	kfree(cma_dev_group);
+	return ERR_PTR(err);
+}
+
+static struct configfs_group_operations cma_subsys_group_ops = {
+	.make_group	= make_cma_dev,
+};
+
+static struct config_item_type cma_subsys_type = {
+	.ct_group_ops	= &cma_subsys_group_ops,
+	.ct_owner	= THIS_MODULE,
+};
+
+static struct configfs_subsystem cma_subsys = {
+	.su_group	= {
+		.cg_item	= {
+			.ci_namebuf	= "rdma_cm",
+			.ci_type	= &cma_subsys_type,
+		},
+	},
+};
+
+int __init cma_configfs_init(void)
+{
+	config_group_init(&cma_subsys.su_group);
+	mutex_init(&cma_subsys.su_mutex);
+	return configfs_register_subsystem(&cma_subsys);
+}
+
+void __exit cma_configfs_exit(void)
+{
+	configfs_unregister_subsystem(&cma_subsys);
+}
diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h
index 5cf6eb7..eab3221 100644
--- a/drivers/infiniband/core/core_priv.h
+++ b/drivers/infiniband/core/core_priv.h
@@ -38,6 +38,32 @@
 
 #include <rdma/ib_verbs.h>
 
+#if IS_ENABLED(CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS)
+int cma_configfs_init(void);
+void cma_configfs_exit(void);
+#else
+static inline int cma_configfs_init(void)
+{
+	return 0;
+}
+
+static inline void cma_configfs_exit(void)
+{
+}
+#endif
+struct cma_device;
+void cma_ref_dev(struct cma_device *cma_dev);
+void cma_deref_dev(struct cma_device *cma_dev);
+typedef bool (*cma_device_filter)(struct ib_device *, void *);
+struct cma_device *cma_enum_devices_by_ibdev(cma_device_filter	filter,
+					     void		*cookie);
+int cma_get_default_gid_type(struct cma_device *cma_dev,
+			     unsigned int port);
+int cma_set_default_gid_type(struct cma_device *cma_dev,
+			     unsigned int port,
+			     enum ib_gid_type default_gid_type);
+struct ib_device *cma_get_ib_dev(struct cma_device *cma_dev);
+
 int  ib_device_register_sysfs(struct ib_device *device,
 			      int (*port_callback)(struct ib_device *,
 						   u8, struct kobject *));
@@ -70,8 +96,13 @@ enum ib_cache_gid_default_mode {
 	IB_CACHE_GID_DEFAULT_MODE_DELETE
 };
 
+int ib_cache_gid_parse_type_str(const char *buf);
+
+const char *ib_cache_gid_type_str(enum ib_gid_type gid_type);
+
 void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port,
 				  struct net_device *ndev,
+				  unsigned long gid_type_mask,
 				  enum ib_cache_gid_default_mode mode);
 
 int ib_cache_gid_add(struct ib_device *ib_dev, u8 port,
@@ -87,9 +118,23 @@ int roce_gid_mgmt_init(void);
 void roce_gid_mgmt_cleanup(void);
 
 int roce_rescan_device(struct ib_device *ib_dev);
+unsigned long roce_gid_type_mask_support(struct ib_device *ib_dev, u8 port);
 
 int ib_cache_setup_one(struct ib_device *device);
 void ib_cache_cleanup_one(struct ib_device *device);
 void ib_cache_release_one(struct ib_device *device);
 
+static inline bool rdma_is_upper_dev_rcu(struct net_device *dev,
+					 struct net_device *upper)
+{
+	struct net_device *_upper = NULL;
+	struct list_head *iter;
+
+	netdev_for_each_all_upper_dev_rcu(dev, _upper, iter)
+		if (_upper == upper)
+			break;
+
+	return _upper == upper;
+}
+
 #endif /* _CORE_PRIV_H */
diff --git a/drivers/infiniband/core/cq.c b/drivers/infiniband/core/cq.c
new file mode 100644
index 0000000..a754fc7
--- /dev/null
+++ b/drivers/infiniband/core/cq.c
@@ -0,0 +1,209 @@
+/*
+ * Copyright (c) 2015 HGST, a Western Digital Company.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <rdma/ib_verbs.h>
+
+/* # of WCs to poll for with a single call to ib_poll_cq */
+#define IB_POLL_BATCH			16
+
+/* # of WCs to iterate over before yielding */
+#define IB_POLL_BUDGET_IRQ		256
+#define IB_POLL_BUDGET_WORKQUEUE	65536
+
+#define IB_POLL_FLAGS \
+	(IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS)
+
+static int __ib_process_cq(struct ib_cq *cq, int budget)
+{
+	int i, n, completed = 0;
+
+	while ((n = ib_poll_cq(cq, IB_POLL_BATCH, cq->wc)) > 0) {
+		for (i = 0; i < n; i++) {
+			struct ib_wc *wc = &cq->wc[i];
+
+			if (wc->wr_cqe)
+				wc->wr_cqe->done(cq, wc);
+			else
+				WARN_ON_ONCE(wc->status == IB_WC_SUCCESS);
+		}
+
+		completed += n;
+
+		if (n != IB_POLL_BATCH ||
+		    (budget != -1 && completed >= budget))
+			break;
+	}
+
+	return completed;
+}
+
+/**
+ * ib_process_direct_cq - process a CQ in caller context
+ * @cq:		CQ to process
+ * @budget:	number of CQEs to poll for
+ *
+ * This function is used to process all outstanding CQ entries on a
+ * %IB_POLL_DIRECT CQ.  It does not offload CQ processing to a different
+ * context and does not ask for completion interrupts from the HCA.
+ *
+ * Note: for compatibility reasons -1 can be passed in %budget for unlimited
+ * polling.  Do not use this feature in new code, it will be removed soon.
+ */
+int ib_process_cq_direct(struct ib_cq *cq, int budget)
+{
+	WARN_ON_ONCE(cq->poll_ctx != IB_POLL_DIRECT);
+
+	return __ib_process_cq(cq, budget);
+}
+EXPORT_SYMBOL(ib_process_cq_direct);
+
+static void ib_cq_completion_direct(struct ib_cq *cq, void *private)
+{
+	WARN_ONCE(1, "got unsolicited completion for CQ 0x%p\n", cq);
+}
+
+static int ib_poll_handler(struct irq_poll *iop, int budget)
+{
+	struct ib_cq *cq = container_of(iop, struct ib_cq, iop);
+	int completed;
+
+	completed = __ib_process_cq(cq, budget);
+	if (completed < budget) {
+		irq_poll_complete(&cq->iop);
+		if (ib_req_notify_cq(cq, IB_POLL_FLAGS) > 0)
+			irq_poll_sched(&cq->iop);
+	}
+
+	return completed;
+}
+
+static void ib_cq_completion_softirq(struct ib_cq *cq, void *private)
+{
+	irq_poll_sched(&cq->iop);
+}
+
+static void ib_cq_poll_work(struct work_struct *work)
+{
+	struct ib_cq *cq = container_of(work, struct ib_cq, work);
+	int completed;
+
+	completed = __ib_process_cq(cq, IB_POLL_BUDGET_WORKQUEUE);
+	if (completed >= IB_POLL_BUDGET_WORKQUEUE ||
+	    ib_req_notify_cq(cq, IB_POLL_FLAGS) > 0)
+		queue_work(ib_comp_wq, &cq->work);
+}
+
+static void ib_cq_completion_workqueue(struct ib_cq *cq, void *private)
+{
+	queue_work(ib_comp_wq, &cq->work);
+}
+
+/**
+ * ib_alloc_cq - allocate a completion queue
+ * @dev:		device to allocate the CQ for
+ * @private:		driver private data, accessible from cq->cq_context
+ * @nr_cqe:		number of CQEs to allocate
+ * @comp_vector:	HCA completion vectors for this CQ
+ * @poll_ctx:		context to poll the CQ from.
+ *
+ * This is the proper interface to allocate a CQ for in-kernel users. A
+ * CQ allocated with this interface will automatically be polled from the
+ * specified context.  The ULP needs must use wr->wr_cqe instead of wr->wr_id
+ * to use this CQ abstraction.
+ */
+struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private,
+		int nr_cqe, int comp_vector, enum ib_poll_context poll_ctx)
+{
+	struct ib_cq_init_attr cq_attr = {
+		.cqe		= nr_cqe,
+		.comp_vector	= comp_vector,
+	};
+	struct ib_cq *cq;
+	int ret = -ENOMEM;
+
+	cq = dev->create_cq(dev, &cq_attr, NULL, NULL);
+	if (IS_ERR(cq))
+		return cq;
+
+	cq->device = dev;
+	cq->uobject = NULL;
+	cq->event_handler = NULL;
+	cq->cq_context = private;
+	cq->poll_ctx = poll_ctx;
+	atomic_set(&cq->usecnt, 0);
+
+	cq->wc = kmalloc_array(IB_POLL_BATCH, sizeof(*cq->wc), GFP_KERNEL);
+	if (!cq->wc)
+		goto out_destroy_cq;
+
+	switch (cq->poll_ctx) {
+	case IB_POLL_DIRECT:
+		cq->comp_handler = ib_cq_completion_direct;
+		break;
+	case IB_POLL_SOFTIRQ:
+		cq->comp_handler = ib_cq_completion_softirq;
+
+		irq_poll_init(&cq->iop, IB_POLL_BUDGET_IRQ, ib_poll_handler);
+		ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
+		break;
+	case IB_POLL_WORKQUEUE:
+		cq->comp_handler = ib_cq_completion_workqueue;
+		INIT_WORK(&cq->work, ib_cq_poll_work);
+		ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
+		break;
+	default:
+		ret = -EINVAL;
+		goto out_free_wc;
+	}
+
+	return cq;
+
+out_free_wc:
+	kfree(cq->wc);
+out_destroy_cq:
+	cq->device->destroy_cq(cq);
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL(ib_alloc_cq);
+
+/**
+ * ib_free_cq - free a completion queue
+ * @cq:		completion queue to free.
+ */
+void ib_free_cq(struct ib_cq *cq)
+{
+	int ret;
+
+	if (WARN_ON_ONCE(atomic_read(&cq->usecnt)))
+		return;
+
+	switch (cq->poll_ctx) {
+	case IB_POLL_DIRECT:
+		break;
+	case IB_POLL_SOFTIRQ:
+		irq_poll_disable(&cq->iop);
+		break;
+	case IB_POLL_WORKQUEUE:
+		flush_work(&cq->work);
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+
+	kfree(cq->wc);
+	ret = cq->device->destroy_cq(cq);
+	WARN_ON_ONCE(ret);
+}
+EXPORT_SYMBOL(ib_free_cq);
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 179e813..00da80e 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -58,6 +58,7 @@ struct ib_client_data {
 	bool		  going_down;
 };
 
+struct workqueue_struct *ib_comp_wq;
 struct workqueue_struct *ib_wq;
 EXPORT_SYMBOL_GPL(ib_wq);
 
@@ -325,6 +326,7 @@ int ib_register_device(struct ib_device *device,
 {
 	int ret;
 	struct ib_client *client;
+	struct ib_udata uhw = {.outlen = 0, .inlen = 0};
 
 	mutex_lock(&device_mutex);
 
@@ -352,6 +354,13 @@ int ib_register_device(struct ib_device *device,
 		goto out;
 	}
 
+	memset(&device->attrs, 0, sizeof(device->attrs));
+	ret = device->query_device(device, &device->attrs, &uhw);
+	if (ret) {
+		printk(KERN_WARNING "Couldn't query the device attributes\n");
+		goto out;
+	}
+
 	ret = ib_device_register_sysfs(device, port_callback);
 	if (ret) {
 		printk(KERN_WARNING "Couldn't register device %s with driver model\n",
@@ -628,25 +637,6 @@ void ib_dispatch_event(struct ib_event *event)
 EXPORT_SYMBOL(ib_dispatch_event);
 
 /**
- * ib_query_device - Query IB device attributes
- * @device:Device to query
- * @device_attr:Device attributes
- *
- * ib_query_device() returns the attributes of a device through the
- * @device_attr pointer.
- */
-int ib_query_device(struct ib_device *device,
-		    struct ib_device_attr *device_attr)
-{
-	struct ib_udata uhw = {.outlen = 0, .inlen = 0};
-
-	memset(device_attr, 0, sizeof(*device_attr));
-
-	return device->query_device(device, device_attr, &uhw);
-}
-EXPORT_SYMBOL(ib_query_device);
-
-/**
  * ib_query_port - Query IB port attributes
  * @device:Device to query
  * @port_num:Port number to query
@@ -825,26 +815,31 @@ EXPORT_SYMBOL(ib_modify_port);
  *   a specified GID value occurs.
  * @device: The device to query.
  * @gid: The GID value to search for.
+ * @gid_type: Type of GID.
  * @ndev: The ndev related to the GID to search for.
  * @port_num: The port number of the device where the GID value was found.
  * @index: The index into the GID table where the GID was found.  This
  *   parameter may be NULL.
  */
 int ib_find_gid(struct ib_device *device, union ib_gid *gid,
-		struct net_device *ndev, u8 *port_num, u16 *index)
+		enum ib_gid_type gid_type, struct net_device *ndev,
+		u8 *port_num, u16 *index)
 {
 	union ib_gid tmp_gid;
 	int ret, port, i;
 
 	for (port = rdma_start_port(device); port <= rdma_end_port(device); ++port) {
 		if (rdma_cap_roce_gid_table(device, port)) {
-			if (!ib_find_cached_gid_by_port(device, gid, port,
+			if (!ib_find_cached_gid_by_port(device, gid, gid_type, port,
 							ndev, index)) {
 				*port_num = port;
 				return 0;
 			}
 		}
 
+		if (gid_type != IB_GID_TYPE_IB)
+			continue;
+
 		for (i = 0; i < device->port_immutable[port].gid_tbl_len; ++i) {
 			ret = ib_query_gid(device, port, i, &tmp_gid, NULL);
 			if (ret)
@@ -954,10 +949,18 @@ static int __init ib_core_init(void)
 	if (!ib_wq)
 		return -ENOMEM;
 
+	ib_comp_wq = alloc_workqueue("ib-comp-wq",
+			WQ_UNBOUND | WQ_HIGHPRI | WQ_MEM_RECLAIM,
+			WQ_UNBOUND_MAX_ACTIVE);
+	if (!ib_comp_wq) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
 	ret = class_register(&ib_class);
 	if (ret) {
 		printk(KERN_WARNING "Couldn't create InfiniBand device class\n");
-		goto err;
+		goto err_comp;
 	}
 
 	ret = ibnl_init();
@@ -972,7 +975,8 @@ static int __init ib_core_init(void)
 
 err_sysfs:
 	class_unregister(&ib_class);
-
+err_comp:
+	destroy_workqueue(ib_comp_wq);
 err:
 	destroy_workqueue(ib_wq);
 	return ret;
@@ -983,6 +987,7 @@ static void __exit ib_core_cleanup(void)
 	ib_cache_cleanup();
 	ibnl_cleanup();
 	class_unregister(&ib_class);
+	destroy_workqueue(ib_comp_wq);
 	/* Make sure that any pending umem accounting work is done. */
 	destroy_workqueue(ib_wq);
 }
diff --git a/drivers/infiniband/core/fmr_pool.c b/drivers/infiniband/core/fmr_pool.c
index 9f5ad7c..6ac3683 100644
--- a/drivers/infiniband/core/fmr_pool.c
+++ b/drivers/infiniband/core/fmr_pool.c
@@ -212,7 +212,6 @@ struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd             *pd,
 {
 	struct ib_device   *device;
 	struct ib_fmr_pool *pool;
-	struct ib_device_attr *attr;
 	int i;
 	int ret;
 	int max_remaps;
@@ -228,25 +227,10 @@ struct ib_fmr_pool *ib_create_fmr_pool(struct ib_pd             *pd,
 		return ERR_PTR(-ENOSYS);
 	}
 
-	attr = kmalloc(sizeof *attr, GFP_KERNEL);
-	if (!attr) {
-		printk(KERN_WARNING PFX "couldn't allocate device attr struct\n");
-		return ERR_PTR(-ENOMEM);
-	}
-
-	ret = ib_query_device(device, attr);
-	if (ret) {
-		printk(KERN_WARNING PFX "couldn't query device: %d\n", ret);
-		kfree(attr);
-		return ERR_PTR(ret);
-	}
-
-	if (!attr->max_map_per_fmr)
+	if (!device->attrs.max_map_per_fmr)
 		max_remaps = IB_FMR_MAX_REMAPS;
 	else
-		max_remaps = attr->max_map_per_fmr;
-
-	kfree(attr);
+		max_remaps = device->attrs.max_map_per_fmr;
 
 	pool = kmalloc(sizeof *pool, GFP_KERNEL);
 	if (!pool) {
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 2281de1..9fa5bf3 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -84,6 +84,9 @@ static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req,
 			      u8 mgmt_class);
 static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req,
 			   struct ib_mad_agent_private *agent_priv);
+static bool ib_mad_send_error(struct ib_mad_port_private *port_priv,
+			      struct ib_wc *wc);
+static void ib_mad_send_done(struct ib_cq *cq, struct ib_wc *wc);
 
 /*
  * Returns a ib_mad_port_private structure or NULL for a device/port
@@ -681,7 +684,7 @@ static void snoop_recv(struct ib_mad_qp_info *qp_info,
 
 		atomic_inc(&mad_snoop_priv->refcount);
 		spin_unlock_irqrestore(&qp_info->snoop_lock, flags);
-		mad_snoop_priv->agent.recv_handler(&mad_snoop_priv->agent,
+		mad_snoop_priv->agent.recv_handler(&mad_snoop_priv->agent, NULL,
 						   mad_recv_wc);
 		deref_snoop_agent(mad_snoop_priv);
 		spin_lock_irqsave(&qp_info->snoop_lock, flags);
@@ -689,12 +692,11 @@ static void snoop_recv(struct ib_mad_qp_info *qp_info,
 	spin_unlock_irqrestore(&qp_info->snoop_lock, flags);
 }
 
-static void build_smp_wc(struct ib_qp *qp,
-			 u64 wr_id, u16 slid, u16 pkey_index, u8 port_num,
-			 struct ib_wc *wc)
+static void build_smp_wc(struct ib_qp *qp, struct ib_cqe *cqe, u16 slid,
+		u16 pkey_index, u8 port_num, struct ib_wc *wc)
 {
 	memset(wc, 0, sizeof *wc);
-	wc->wr_id = wr_id;
+	wc->wr_cqe = cqe;
 	wc->status = IB_WC_SUCCESS;
 	wc->opcode = IB_WC_RECV;
 	wc->pkey_index = pkey_index;
@@ -832,7 +834,7 @@ static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv,
 	}
 
 	build_smp_wc(mad_agent_priv->agent.qp,
-		     send_wr->wr.wr_id, drslid,
+		     send_wr->wr.wr_cqe, drslid,
 		     send_wr->pkey_index,
 		     send_wr->port_num, &mad_wc);
 
@@ -1039,7 +1041,9 @@ struct ib_mad_send_buf * ib_create_send_mad(struct ib_mad_agent *mad_agent,
 
 	mad_send_wr->sg_list[1].lkey = mad_agent->qp->pd->local_dma_lkey;
 
-	mad_send_wr->send_wr.wr.wr_id = (unsigned long) mad_send_wr;
+	mad_send_wr->mad_list.cqe.done = ib_mad_send_done;
+
+	mad_send_wr->send_wr.wr.wr_cqe = &mad_send_wr->mad_list.cqe;
 	mad_send_wr->send_wr.wr.sg_list = mad_send_wr->sg_list;
 	mad_send_wr->send_wr.wr.num_sge = 2;
 	mad_send_wr->send_wr.wr.opcode = IB_WR_SEND;
@@ -1151,8 +1155,9 @@ int ib_send_mad(struct ib_mad_send_wr_private *mad_send_wr)
 
 	/* Set WR ID to find mad_send_wr upon completion */
 	qp_info = mad_send_wr->mad_agent_priv->qp_info;
-	mad_send_wr->send_wr.wr.wr_id = (unsigned long)&mad_send_wr->mad_list;
 	mad_send_wr->mad_list.mad_queue = &qp_info->send_queue;
+	mad_send_wr->mad_list.cqe.done = ib_mad_send_done;
+	mad_send_wr->send_wr.wr.wr_cqe = &mad_send_wr->mad_list.cqe;
 
 	mad_agent = mad_send_wr->send_buf.mad_agent;
 	sge = mad_send_wr->sg_list;
@@ -1982,9 +1987,9 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv,
 				/* user rmpp is in effect
 				 * and this is an active RMPP MAD
 				 */
-				mad_recv_wc->wc->wr_id = 0;
-				mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
-								   mad_recv_wc);
+				mad_agent_priv->agent.recv_handler(
+						&mad_agent_priv->agent, NULL,
+						mad_recv_wc);
 				atomic_dec(&mad_agent_priv->refcount);
 			} else {
 				/* not user rmpp, revert to normal behavior and
@@ -1998,9 +2003,10 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv,
 			spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
 
 			/* Defined behavior is to complete response before request */
-			mad_recv_wc->wc->wr_id = (unsigned long) &mad_send_wr->send_buf;
-			mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
-							   mad_recv_wc);
+			mad_agent_priv->agent.recv_handler(
+					&mad_agent_priv->agent,
+					&mad_send_wr->send_buf,
+					mad_recv_wc);
 			atomic_dec(&mad_agent_priv->refcount);
 
 			mad_send_wc.status = IB_WC_SUCCESS;
@@ -2009,7 +2015,7 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv,
 			ib_mad_complete_send_wr(mad_send_wr, &mad_send_wc);
 		}
 	} else {
-		mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
+		mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent, NULL,
 						   mad_recv_wc);
 		deref_mad_agent(mad_agent_priv);
 	}
@@ -2172,13 +2178,14 @@ handle_smi(struct ib_mad_port_private *port_priv,
 	return handle_ib_smi(port_priv, qp_info, wc, port_num, recv, response);
 }
 
-static void ib_mad_recv_done_handler(struct ib_mad_port_private *port_priv,
-				     struct ib_wc *wc)
+static void ib_mad_recv_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct ib_mad_port_private *port_priv = cq->cq_context;
+	struct ib_mad_list_head *mad_list =
+		container_of(wc->wr_cqe, struct ib_mad_list_head, cqe);
 	struct ib_mad_qp_info *qp_info;
 	struct ib_mad_private_header *mad_priv_hdr;
 	struct ib_mad_private *recv, *response = NULL;
-	struct ib_mad_list_head *mad_list;
 	struct ib_mad_agent_private *mad_agent;
 	int port_num;
 	int ret = IB_MAD_RESULT_SUCCESS;
@@ -2186,7 +2193,17 @@ static void ib_mad_recv_done_handler(struct ib_mad_port_private *port_priv,
 	u16 resp_mad_pkey_index = 0;
 	bool opa;
 
-	mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id;
+	if (list_empty_careful(&port_priv->port_list))
+		return;
+
+	if (wc->status != IB_WC_SUCCESS) {
+		/*
+		 * Receive errors indicate that the QP has entered the error
+		 * state - error handling/shutdown code will cleanup
+		 */
+		return;
+	}
+
 	qp_info = mad_list->mad_queue->qp_info;
 	dequeue_mad(mad_list);
 
@@ -2227,7 +2244,7 @@ static void ib_mad_recv_done_handler(struct ib_mad_port_private *port_priv,
 	response = alloc_mad_private(mad_size, GFP_KERNEL);
 	if (!response) {
 		dev_err(&port_priv->device->dev,
-			"ib_mad_recv_done_handler no memory for response buffer\n");
+			"%s: no memory for response buffer\n", __func__);
 		goto out;
 	}
 
@@ -2413,11 +2430,12 @@ done:
 	spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
 }
 
-static void ib_mad_send_done_handler(struct ib_mad_port_private *port_priv,
-				     struct ib_wc *wc)
+static void ib_mad_send_done(struct ib_cq *cq, struct ib_wc *wc)
 {
+	struct ib_mad_port_private *port_priv = cq->cq_context;
+	struct ib_mad_list_head *mad_list =
+		container_of(wc->wr_cqe, struct ib_mad_list_head, cqe);
 	struct ib_mad_send_wr_private	*mad_send_wr, *queued_send_wr;
-	struct ib_mad_list_head		*mad_list;
 	struct ib_mad_qp_info		*qp_info;
 	struct ib_mad_queue		*send_queue;
 	struct ib_send_wr		*bad_send_wr;
@@ -2425,7 +2443,14 @@ static void ib_mad_send_done_handler(struct ib_mad_port_private *port_priv,
 	unsigned long flags;
 	int ret;
 
-	mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id;
+	if (list_empty_careful(&port_priv->port_list))
+		return;
+
+	if (wc->status != IB_WC_SUCCESS) {
+		if (!ib_mad_send_error(port_priv, wc))
+			return;
+	}
+
 	mad_send_wr = container_of(mad_list, struct ib_mad_send_wr_private,
 				   mad_list);
 	send_queue = mad_list->mad_queue;
@@ -2490,24 +2515,15 @@ static void mark_sends_for_retry(struct ib_mad_qp_info *qp_info)
 	spin_unlock_irqrestore(&qp_info->send_queue.lock, flags);
 }
 
-static void mad_error_handler(struct ib_mad_port_private *port_priv,
-			      struct ib_wc *wc)
+static bool ib_mad_send_error(struct ib_mad_port_private *port_priv,
+		struct ib_wc *wc)
 {
-	struct ib_mad_list_head *mad_list;
-	struct ib_mad_qp_info *qp_info;
+	struct ib_mad_list_head *mad_list =
+		container_of(wc->wr_cqe, struct ib_mad_list_head, cqe);
+	struct ib_mad_qp_info *qp_info = mad_list->mad_queue->qp_info;
 	struct ib_mad_send_wr_private *mad_send_wr;
 	int ret;
 
-	/* Determine if failure was a send or receive */
-	mad_list = (struct ib_mad_list_head *)(unsigned long)wc->wr_id;
-	qp_info = mad_list->mad_queue->qp_info;
-	if (mad_list->mad_queue == &qp_info->recv_queue)
-		/*
-		 * Receive errors indicate that the QP has entered the error
-		 * state - error handling/shutdown code will cleanup
-		 */
-		return;
-
 	/*
 	 * Send errors will transition the QP to SQE - move
 	 * QP to RTS and repost flushed work requests
@@ -2522,10 +2538,9 @@ static void mad_error_handler(struct ib_mad_port_private *port_priv,
 			mad_send_wr->retry = 0;
 			ret = ib_post_send(qp_info->qp, &mad_send_wr->send_wr.wr,
 					&bad_send_wr);
-			if (ret)
-				ib_mad_send_done_handler(port_priv, wc);
-		} else
-			ib_mad_send_done_handler(port_priv, wc);
+			if (!ret)
+				return false;
+		}
 	} else {
 		struct ib_qp_attr *attr;
 
@@ -2539,42 +2554,14 @@ static void mad_error_handler(struct ib_mad_port_private *port_priv,
 			kfree(attr);
 			if (ret)
 				dev_err(&port_priv->device->dev,
-					"mad_error_handler - ib_modify_qp to RTS : %d\n",
-					ret);
+					"%s - ib_modify_qp to RTS: %d\n",
+					__func__, ret);
 			else
 				mark_sends_for_retry(qp_info);
 		}
-		ib_mad_send_done_handler(port_priv, wc);
 	}
-}
 
-/*
- * IB MAD completion callback
- */
-static void ib_mad_completion_handler(struct work_struct *work)
-{
-	struct ib_mad_port_private *port_priv;
-	struct ib_wc wc;
-
-	port_priv = container_of(work, struct ib_mad_port_private, work);
-	ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP);
-
-	while (ib_poll_cq(port_priv->cq, 1, &wc) == 1) {
-		if (wc.status == IB_WC_SUCCESS) {
-			switch (wc.opcode) {
-			case IB_WC_SEND:
-				ib_mad_send_done_handler(port_priv, &wc);
-				break;
-			case IB_WC_RECV:
-				ib_mad_recv_done_handler(port_priv, &wc);
-				break;
-			default:
-				BUG_ON(1);
-				break;
-			}
-		} else
-			mad_error_handler(port_priv, &wc);
-	}
+	return true;
 }
 
 static void cancel_mads(struct ib_mad_agent_private *mad_agent_priv)
@@ -2716,7 +2703,7 @@ static void local_completions(struct work_struct *work)
 			 * before request
 			 */
 			build_smp_wc(recv_mad_agent->agent.qp,
-				     (unsigned long) local->mad_send_wr,
+				     local->mad_send_wr->send_wr.wr.wr_cqe,
 				     be16_to_cpu(IB_LID_PERMISSIVE),
 				     local->mad_send_wr->send_wr.pkey_index,
 				     recv_mad_agent->agent.port_num, &wc);
@@ -2744,6 +2731,7 @@ static void local_completions(struct work_struct *work)
 					   IB_MAD_SNOOP_RECVS);
 			recv_mad_agent->agent.recv_handler(
 						&recv_mad_agent->agent,
+						&local->mad_send_wr->send_buf,
 						&local->mad_priv->header.recv_wc);
 			spin_lock_irqsave(&recv_mad_agent->lock, flags);
 			atomic_dec(&recv_mad_agent->refcount);
@@ -2855,17 +2843,6 @@ static void timeout_sends(struct work_struct *work)
 	spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
 }
 
-static void ib_mad_thread_completion_handler(struct ib_cq *cq, void *arg)
-{
-	struct ib_mad_port_private *port_priv = cq->cq_context;
-	unsigned long flags;
-
-	spin_lock_irqsave(&ib_mad_port_list_lock, flags);
-	if (!list_empty(&port_priv->port_list))
-		queue_work(port_priv->wq, &port_priv->work);
-	spin_unlock_irqrestore(&ib_mad_port_list_lock, flags);
-}
-
 /*
  * Allocate receive MADs and post receive WRs for them
  */
@@ -2913,8 +2890,9 @@ static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info,
 			break;
 		}
 		mad_priv->header.mapping = sg_list.addr;
-		recv_wr.wr_id = (unsigned long)&mad_priv->header.mad_list;
 		mad_priv->header.mad_list.mad_queue = recv_queue;
+		mad_priv->header.mad_list.cqe.done = ib_mad_recv_done;
+		recv_wr.wr_cqe = &mad_priv->header.mad_list.cqe;
 
 		/* Post receive WR */
 		spin_lock_irqsave(&recv_queue->lock, flags);
@@ -3151,7 +3129,6 @@ static int ib_mad_port_open(struct ib_device *device,
 	unsigned long flags;
 	char name[sizeof "ib_mad123"];
 	int has_smi;
-	struct ib_cq_init_attr cq_attr = {};
 
 	if (WARN_ON(rdma_max_mad_size(device, port_num) < IB_MGMT_MAD_SIZE))
 		return -EFAULT;
@@ -3179,10 +3156,8 @@ static int ib_mad_port_open(struct ib_device *device,
 	if (has_smi)
 		cq_size *= 2;
 
-	cq_attr.cqe = cq_size;
-	port_priv->cq = ib_create_cq(port_priv->device,
-				     ib_mad_thread_completion_handler,
-				     NULL, port_priv, &cq_attr);
+	port_priv->cq = ib_alloc_cq(port_priv->device, port_priv, cq_size, 0,
+			IB_POLL_WORKQUEUE);
 	if (IS_ERR(port_priv->cq)) {
 		dev_err(&device->dev, "Couldn't create ib_mad CQ\n");
 		ret = PTR_ERR(port_priv->cq);
@@ -3211,7 +3186,6 @@ static int ib_mad_port_open(struct ib_device *device,
 		ret = -ENOMEM;
 		goto error8;
 	}
-	INIT_WORK(&port_priv->work, ib_mad_completion_handler);
 
 	spin_lock_irqsave(&ib_mad_port_list_lock, flags);
 	list_add_tail(&port_priv->port_list, &ib_mad_port_list);
@@ -3238,7 +3212,7 @@ error7:
 error6:
 	ib_dealloc_pd(port_priv->pd);
 error4:
-	ib_destroy_cq(port_priv->cq);
+	ib_free_cq(port_priv->cq);
 	cleanup_recv_queue(&port_priv->qp_info[1]);
 	cleanup_recv_queue(&port_priv->qp_info[0]);
 error3:
@@ -3271,7 +3245,7 @@ static int ib_mad_port_close(struct ib_device *device, int port_num)
 	destroy_mad_qp(&port_priv->qp_info[1]);
 	destroy_mad_qp(&port_priv->qp_info[0]);
 	ib_dealloc_pd(port_priv->pd);
-	ib_destroy_cq(port_priv->cq);
+	ib_free_cq(port_priv->cq);
 	cleanup_recv_queue(&port_priv->qp_info[1]);
 	cleanup_recv_queue(&port_priv->qp_info[0]);
 	/* XXX: Handle deallocation of MAD registration tables */
diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h
index 990698a..28669f6 100644
--- a/drivers/infiniband/core/mad_priv.h
+++ b/drivers/infiniband/core/mad_priv.h
@@ -64,6 +64,7 @@
 
 struct ib_mad_list_head {
 	struct list_head list;
+	struct ib_cqe cqe;
 	struct ib_mad_queue *mad_queue;
 };
 
@@ -204,7 +205,6 @@ struct ib_mad_port_private {
 	struct ib_mad_mgmt_version_table version[MAX_MGMT_VERSION];
 	struct list_head agent_list;
 	struct workqueue_struct *wq;
-	struct work_struct work;
 	struct ib_mad_qp_info qp_info[IB_MAD_QPS_CORE];
 };
 
diff --git a/drivers/infiniband/core/multicast.c b/drivers/infiniband/core/multicast.c
index bb6685f..250937c 100644
--- a/drivers/infiniband/core/multicast.c
+++ b/drivers/infiniband/core/multicast.c
@@ -723,14 +723,27 @@ EXPORT_SYMBOL(ib_sa_get_mcmember_rec);
 
 int ib_init_ah_from_mcmember(struct ib_device *device, u8 port_num,
 			     struct ib_sa_mcmember_rec *rec,
+			     struct net_device *ndev,
+			     enum ib_gid_type gid_type,
 			     struct ib_ah_attr *ah_attr)
 {
 	int ret;
 	u16 gid_index;
 	u8 p;
 
-	ret = ib_find_cached_gid(device, &rec->port_gid,
-				 NULL, &p, &gid_index);
+	if (rdma_protocol_roce(device, port_num)) {
+		ret = ib_find_cached_gid_by_port(device, &rec->port_gid,
+						 gid_type, port_num,
+						 ndev,
+						 &gid_index);
+	} else if (rdma_protocol_ib(device, port_num)) {
+		ret = ib_find_cached_gid(device, &rec->port_gid,
+					 IB_GID_TYPE_IB, NULL, &p,
+					 &gid_index);
+	} else {
+		ret = -EINVAL;
+	}
+
 	if (ret)
 		return ret;
 
diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c
index 178f984..06556c3 100644
--- a/drivers/infiniband/core/roce_gid_mgmt.c
+++ b/drivers/infiniband/core/roce_gid_mgmt.c
@@ -67,17 +67,53 @@ struct netdev_event_work {
 	struct netdev_event_work_cmd	cmds[ROCE_NETDEV_CALLBACK_SZ];
 };
 
+static const struct {
+	bool (*is_supported)(const struct ib_device *device, u8 port_num);
+	enum ib_gid_type gid_type;
+} PORT_CAP_TO_GID_TYPE[] = {
+	{rdma_protocol_roce_eth_encap, IB_GID_TYPE_ROCE},
+	{rdma_protocol_roce_udp_encap, IB_GID_TYPE_ROCE_UDP_ENCAP},
+};
+
+#define CAP_TO_GID_TABLE_SIZE	ARRAY_SIZE(PORT_CAP_TO_GID_TYPE)
+
+unsigned long roce_gid_type_mask_support(struct ib_device *ib_dev, u8 port)
+{
+	int i;
+	unsigned int ret_flags = 0;
+
+	if (!rdma_protocol_roce(ib_dev, port))
+		return 1UL << IB_GID_TYPE_IB;
+
+	for (i = 0; i < CAP_TO_GID_TABLE_SIZE; i++)
+		if (PORT_CAP_TO_GID_TYPE[i].is_supported(ib_dev, port))
+			ret_flags |= 1UL << PORT_CAP_TO_GID_TYPE[i].gid_type;
+
+	return ret_flags;
+}
+EXPORT_SYMBOL(roce_gid_type_mask_support);
+
 static void update_gid(enum gid_op_type gid_op, struct ib_device *ib_dev,
 		       u8 port, union ib_gid *gid,
 		       struct ib_gid_attr *gid_attr)
 {
-	switch (gid_op) {
-	case GID_ADD:
-		ib_cache_gid_add(ib_dev, port, gid, gid_attr);
-		break;
-	case GID_DEL:
-		ib_cache_gid_del(ib_dev, port, gid, gid_attr);
-		break;
+	int i;
+	unsigned long gid_type_mask = roce_gid_type_mask_support(ib_dev, port);
+
+	for (i = 0; i < IB_GID_TYPE_SIZE; i++) {
+		if ((1UL << i) & gid_type_mask) {
+			gid_attr->gid_type = i;
+			switch (gid_op) {
+			case GID_ADD:
+				ib_cache_gid_add(ib_dev, port,
+						 gid, gid_attr);
+				break;
+			case GID_DEL:
+				ib_cache_gid_del(ib_dev, port,
+						 gid, gid_attr);
+				break;
+			}
+		}
 	}
 }
 
@@ -103,18 +139,6 @@ static enum bonding_slave_state is_eth_active_slave_of_bonding_rcu(struct net_de
 	return BONDING_SLAVE_STATE_NA;
 }
 
-static bool is_upper_dev_rcu(struct net_device *dev, struct net_device *upper)
-{
-	struct net_device *_upper = NULL;
-	struct list_head *iter;
-
-	netdev_for_each_all_upper_dev_rcu(dev, _upper, iter)
-		if (_upper == upper)
-			break;
-
-	return _upper == upper;
-}
-
 #define REQUIRED_BOND_STATES		(BONDING_SLAVE_STATE_ACTIVE |	\
 					 BONDING_SLAVE_STATE_NA)
 static int is_eth_port_of_netdev(struct ib_device *ib_dev, u8 port,
@@ -132,7 +156,7 @@ static int is_eth_port_of_netdev(struct ib_device *ib_dev, u8 port,
 	if (!real_dev)
 		real_dev = event_ndev;
 
-	res = ((is_upper_dev_rcu(rdma_ndev, event_ndev) &&
+	res = ((rdma_is_upper_dev_rcu(rdma_ndev, event_ndev) &&
 	       (is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) &
 		REQUIRED_BOND_STATES)) ||
 	       real_dev == rdma_ndev);
@@ -178,7 +202,7 @@ static int upper_device_filter(struct ib_device *ib_dev, u8 port,
 		return 1;
 
 	rcu_read_lock();
-	res = is_upper_dev_rcu(rdma_ndev, event_ndev);
+	res = rdma_is_upper_dev_rcu(rdma_ndev, event_ndev);
 	rcu_read_unlock();
 
 	return res;
@@ -203,10 +227,12 @@ static void enum_netdev_default_gids(struct ib_device *ib_dev,
 				     u8 port, struct net_device *event_ndev,
 				     struct net_device *rdma_ndev)
 {
+	unsigned long gid_type_mask;
+
 	rcu_read_lock();
 	if (!rdma_ndev ||
 	    ((rdma_ndev != event_ndev &&
-	      !is_upper_dev_rcu(rdma_ndev, event_ndev)) ||
+	      !rdma_is_upper_dev_rcu(rdma_ndev, event_ndev)) ||
 	     is_eth_active_slave_of_bonding_rcu(rdma_ndev,
 						netdev_master_upper_dev_get_rcu(rdma_ndev)) ==
 	     BONDING_SLAVE_STATE_INACTIVE)) {
@@ -215,7 +241,9 @@ static void enum_netdev_default_gids(struct ib_device *ib_dev,
 	}
 	rcu_read_unlock();
 
-	ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev,
+	gid_type_mask = roce_gid_type_mask_support(ib_dev, port);
+
+	ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev, gid_type_mask,
 				     IB_CACHE_GID_DEFAULT_MODE_SET);
 }
 
@@ -234,12 +262,17 @@ static void bond_delete_netdev_default_gids(struct ib_device *ib_dev,
 
 	rcu_read_lock();
 
-	if (is_upper_dev_rcu(rdma_ndev, event_ndev) &&
+	if (rdma_is_upper_dev_rcu(rdma_ndev, event_ndev) &&
 	    is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) ==
 	    BONDING_SLAVE_STATE_INACTIVE) {
+		unsigned long gid_type_mask;
+
 		rcu_read_unlock();
 
+		gid_type_mask = roce_gid_type_mask_support(ib_dev, port);
+
 		ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev,
+					     gid_type_mask,
 					     IB_CACHE_GID_DEFAULT_MODE_DELETE);
 	} else {
 		rcu_read_unlock();
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index a95a32b..f334090 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -49,7 +49,9 @@
 #include <net/netlink.h>
 #include <uapi/rdma/ib_user_sa.h>
 #include <rdma/ib_marshall.h>
+#include <rdma/ib_addr.h>
 #include "sa.h"
+#include "core_priv.h"
 
 MODULE_AUTHOR("Roland Dreier");
 MODULE_DESCRIPTION("InfiniBand subnet administration query support");
@@ -715,7 +717,9 @@ static int ib_nl_handle_set_timeout(struct sk_buff *skb,
 	struct nlattr *tb[LS_NLA_TYPE_MAX];
 	int ret;
 
-	if (!netlink_capable(skb, CAP_NET_ADMIN))
+	if (!(nlh->nlmsg_flags & NLM_F_REQUEST) ||
+	    !(NETLINK_CB(skb).sk) ||
+	    !netlink_capable(skb, CAP_NET_ADMIN))
 		return -EPERM;
 
 	ret = nla_parse(tb, LS_NLA_TYPE_MAX - 1, nlmsg_data(nlh),
@@ -789,7 +793,9 @@ static int ib_nl_handle_resolve_resp(struct sk_buff *skb,
 	int found = 0;
 	int ret;
 
-	if (!netlink_capable(skb, CAP_NET_ADMIN))
+	if ((nlh->nlmsg_flags & NLM_F_REQUEST) ||
+	    !(NETLINK_CB(skb).sk) ||
+	    !netlink_capable(skb, CAP_NET_ADMIN))
 		return -EPERM;
 
 	spin_lock_irqsave(&ib_nl_request_lock, flags);
@@ -996,7 +1002,8 @@ int ib_init_ah_from_path(struct ib_device *device, u8 port_num,
 {
 	int ret;
 	u16 gid_index;
-	int force_grh;
+	int use_roce;
+	struct net_device *ndev = NULL;
 
 	memset(ah_attr, 0, sizeof *ah_attr);
 	ah_attr->dlid = be16_to_cpu(rec->dlid);
@@ -1006,16 +1013,71 @@ int ib_init_ah_from_path(struct ib_device *device, u8 port_num,
 	ah_attr->port_num = port_num;
 	ah_attr->static_rate = rec->rate;
 
-	force_grh = rdma_cap_eth_ah(device, port_num);
+	use_roce = rdma_cap_eth_ah(device, port_num);
+
+	if (use_roce) {
+		struct net_device *idev;
+		struct net_device *resolved_dev;
+		struct rdma_dev_addr dev_addr = {.bound_dev_if = rec->ifindex,
+						 .net = rec->net ? rec->net :
+							 &init_net};
+		union {
+			struct sockaddr     _sockaddr;
+			struct sockaddr_in  _sockaddr_in;
+			struct sockaddr_in6 _sockaddr_in6;
+		} sgid_addr, dgid_addr;
+
+		if (!device->get_netdev)
+			return -EOPNOTSUPP;
+
+		rdma_gid2ip(&sgid_addr._sockaddr, &rec->sgid);
+		rdma_gid2ip(&dgid_addr._sockaddr, &rec->dgid);
+
+		/* validate the route */
+		ret = rdma_resolve_ip_route(&sgid_addr._sockaddr,
+					    &dgid_addr._sockaddr, &dev_addr);
+		if (ret)
+			return ret;
 
-	if (rec->hop_limit > 1 || force_grh) {
-		struct net_device *ndev = ib_get_ndev_from_path(rec);
+		if ((dev_addr.network == RDMA_NETWORK_IPV4 ||
+		     dev_addr.network == RDMA_NETWORK_IPV6) &&
+		    rec->gid_type != IB_GID_TYPE_ROCE_UDP_ENCAP)
+			return -EINVAL;
+
+		idev = device->get_netdev(device, port_num);
+		if (!idev)
+			return -ENODEV;
+
+		resolved_dev = dev_get_by_index(dev_addr.net,
+						dev_addr.bound_dev_if);
+		if (resolved_dev->flags & IFF_LOOPBACK) {
+			dev_put(resolved_dev);
+			resolved_dev = idev;
+			dev_hold(resolved_dev);
+		}
+		ndev = ib_get_ndev_from_path(rec);
+		rcu_read_lock();
+		if ((ndev && ndev != resolved_dev) ||
+		    (resolved_dev != idev &&
+		     !rdma_is_upper_dev_rcu(idev, resolved_dev)))
+			ret = -EHOSTUNREACH;
+		rcu_read_unlock();
+		dev_put(idev);
+		dev_put(resolved_dev);
+		if (ret) {
+			if (ndev)
+				dev_put(ndev);
+			return ret;
+		}
+	}
 
+	if (rec->hop_limit > 1 || use_roce) {
 		ah_attr->ah_flags = IB_AH_GRH;
 		ah_attr->grh.dgid = rec->dgid;
 
-		ret = ib_find_cached_gid(device, &rec->sgid, ndev, &port_num,
-					 &gid_index);
+		ret = ib_find_cached_gid_by_port(device, &rec->sgid,
+						 rec->gid_type, port_num, ndev,
+						 &gid_index);
 		if (ret) {
 			if (ndev)
 				dev_put(ndev);
@@ -1029,9 +1091,10 @@ int ib_init_ah_from_path(struct ib_device *device, u8 port_num,
 		if (ndev)
 			dev_put(ndev);
 	}
-	if (force_grh) {
+
+	if (use_roce)
 		memcpy(ah_attr->dmac, rec->dmac, ETH_ALEN);
-	}
+
 	return 0;
 }
 EXPORT_SYMBOL(ib_init_ah_from_path);
@@ -1157,6 +1220,7 @@ static void ib_sa_path_rec_callback(struct ib_sa_query *sa_query,
 			  mad->data, &rec);
 		rec.net = NULL;
 		rec.ifindex = 0;
+		rec.gid_type = IB_GID_TYPE_IB;
 		memset(rec.dmac, 0, ETH_ALEN);
 		query->callback(status, &rec, query->context);
 	} else
@@ -1609,14 +1673,15 @@ static void send_handler(struct ib_mad_agent *agent,
 }
 
 static void recv_handler(struct ib_mad_agent *mad_agent,
+			 struct ib_mad_send_buf *send_buf,
 			 struct ib_mad_recv_wc *mad_recv_wc)
 {
 	struct ib_sa_query *query;
-	struct ib_mad_send_buf *mad_buf;
 
-	mad_buf = (void *) (unsigned long) mad_recv_wc->wc->wr_id;
-	query = mad_buf->context[0];
+	if (!send_buf)
+		return;
 
+	query = send_buf->context[0];
 	if (query->callback) {
 		if (mad_recv_wc->wc->status == IB_WC_SUCCESS)
 			query->callback(query,
diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index b1f37d4..3de9351 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -37,15 +37,27 @@
 #include <linux/slab.h>
 #include <linux/stat.h>
 #include <linux/string.h>
+#include <linux/netdevice.h>
 
 #include <rdma/ib_mad.h>
+#include <rdma/ib_pma.h>
 
+struct ib_port;
+
+struct gid_attr_group {
+	struct ib_port		*port;
+	struct kobject		kobj;
+	struct attribute_group	ndev;
+	struct attribute_group	type;
+};
 struct ib_port {
 	struct kobject         kobj;
 	struct ib_device      *ibdev;
+	struct gid_attr_group *gid_attr_group;
 	struct attribute_group gid_group;
 	struct attribute_group pkey_group;
 	u8                     port_num;
+	struct attribute_group *pma_table;
 };
 
 struct port_attribute {
@@ -65,6 +77,7 @@ struct port_table_attribute {
 	struct port_attribute	attr;
 	char			name[8];
 	int			index;
+	__be16			attr_id;
 };
 
 static ssize_t port_attr_show(struct kobject *kobj,
@@ -84,6 +97,24 @@ static const struct sysfs_ops port_sysfs_ops = {
 	.show = port_attr_show
 };
 
+static ssize_t gid_attr_show(struct kobject *kobj,
+			     struct attribute *attr, char *buf)
+{
+	struct port_attribute *port_attr =
+		container_of(attr, struct port_attribute, attr);
+	struct ib_port *p = container_of(kobj, struct gid_attr_group,
+					 kobj)->port;
+
+	if (!port_attr->show)
+		return -EIO;
+
+	return port_attr->show(p, port_attr, buf);
+}
+
+static const struct sysfs_ops gid_attr_sysfs_ops = {
+	.show = gid_attr_show
+};
+
 static ssize_t state_show(struct ib_port *p, struct port_attribute *unused,
 			  char *buf)
 {
@@ -281,6 +312,46 @@ static struct attribute *port_default_attrs[] = {
 	NULL
 };
 
+static size_t print_ndev(struct ib_gid_attr *gid_attr, char *buf)
+{
+	if (!gid_attr->ndev)
+		return -EINVAL;
+
+	return sprintf(buf, "%s\n", gid_attr->ndev->name);
+}
+
+static size_t print_gid_type(struct ib_gid_attr *gid_attr, char *buf)
+{
+	return sprintf(buf, "%s\n", ib_cache_gid_type_str(gid_attr->gid_type));
+}
+
+static ssize_t _show_port_gid_attr(struct ib_port *p,
+				   struct port_attribute *attr,
+				   char *buf,
+				   size_t (*print)(struct ib_gid_attr *gid_attr,
+						   char *buf))
+{
+	struct port_table_attribute *tab_attr =
+		container_of(attr, struct port_table_attribute, attr);
+	union ib_gid gid;
+	struct ib_gid_attr gid_attr = {};
+	ssize_t ret;
+	va_list args;
+
+	ret = ib_query_gid(p->ibdev, p->port_num, tab_attr->index, &gid,
+			   &gid_attr);
+	if (ret)
+		goto err;
+
+	ret = print(&gid_attr, buf);
+
+err:
+	if (gid_attr.ndev)
+		dev_put(gid_attr.ndev);
+	va_end(args);
+	return ret;
+}
+
 static ssize_t show_port_gid(struct ib_port *p, struct port_attribute *attr,
 			     char *buf)
 {
@@ -296,6 +367,19 @@ static ssize_t show_port_gid(struct ib_port *p, struct port_attribute *attr,
 	return sprintf(buf, "%pI6\n", gid.raw);
 }
 
+static ssize_t show_port_gid_attr_ndev(struct ib_port *p,
+				       struct port_attribute *attr, char *buf)
+{
+	return _show_port_gid_attr(p, attr, buf, print_ndev);
+}
+
+static ssize_t show_port_gid_attr_gid_type(struct ib_port *p,
+					   struct port_attribute *attr,
+					   char *buf)
+{
+	return _show_port_gid_attr(p, attr, buf, print_gid_type);
+}
+
 static ssize_t show_port_pkey(struct ib_port *p, struct port_attribute *attr,
 			      char *buf)
 {
@@ -314,24 +398,32 @@ static ssize_t show_port_pkey(struct ib_port *p, struct port_attribute *attr,
 #define PORT_PMA_ATTR(_name, _counter, _width, _offset)			\
 struct port_table_attribute port_pma_attr_##_name = {			\
 	.attr  = __ATTR(_name, S_IRUGO, show_pma_counter, NULL),	\
-	.index = (_offset) | ((_width) << 16) | ((_counter) << 24)	\
+	.index = (_offset) | ((_width) << 16) | ((_counter) << 24),	\
+	.attr_id = IB_PMA_PORT_COUNTERS ,				\
 }
 
-static ssize_t show_pma_counter(struct ib_port *p, struct port_attribute *attr,
-				char *buf)
+#define PORT_PMA_ATTR_EXT(_name, _width, _offset)			\
+struct port_table_attribute port_pma_attr_ext_##_name = {		\
+	.attr  = __ATTR(_name, S_IRUGO, show_pma_counter, NULL),	\
+	.index = (_offset) | ((_width) << 16),				\
+	.attr_id = IB_PMA_PORT_COUNTERS_EXT ,				\
+}
+
+/*
+ * Get a Perfmgmt MAD block of data.
+ * Returns error code or the number of bytes retrieved.
+ */
+static int get_perf_mad(struct ib_device *dev, int port_num, __be16 attr,
+		void *data, int offset, size_t size)
 {
-	struct port_table_attribute *tab_attr =
-		container_of(attr, struct port_table_attribute, attr);
-	int offset = tab_attr->index & 0xffff;
-	int width  = (tab_attr->index >> 16) & 0xff;
-	struct ib_mad *in_mad  = NULL;
-	struct ib_mad *out_mad = NULL;
+	struct ib_mad *in_mad;
+	struct ib_mad *out_mad;
 	size_t mad_size = sizeof(*out_mad);
 	u16 out_mad_pkey_index = 0;
 	ssize_t ret;
 
-	if (!p->ibdev->process_mad)
-		return sprintf(buf, "N/A (no PMA)\n");
+	if (!dev->process_mad)
+		return -ENOSYS;
 
 	in_mad  = kzalloc(sizeof *in_mad, GFP_KERNEL);
 	out_mad = kmalloc(sizeof *out_mad, GFP_KERNEL);
@@ -344,12 +436,13 @@ static ssize_t show_pma_counter(struct ib_port *p, struct port_attribute *attr,
 	in_mad->mad_hdr.mgmt_class    = IB_MGMT_CLASS_PERF_MGMT;
 	in_mad->mad_hdr.class_version = 1;
 	in_mad->mad_hdr.method        = IB_MGMT_METHOD_GET;
-	in_mad->mad_hdr.attr_id       = cpu_to_be16(0x12); /* PortCounters */
+	in_mad->mad_hdr.attr_id       = attr;
 
-	in_mad->data[41] = p->port_num;	/* PortSelect field */
+	if (attr != IB_PMA_CLASS_PORT_INFO)
+		in_mad->data[41] = port_num;	/* PortSelect field */
 
-	if ((p->ibdev->process_mad(p->ibdev, IB_MAD_IGNORE_MKEY,
-		 p->port_num, NULL, NULL,
+	if ((dev->process_mad(dev, IB_MAD_IGNORE_MKEY,
+		 port_num, NULL, NULL,
 		 (const struct ib_mad_hdr *)in_mad, mad_size,
 		 (struct ib_mad_hdr *)out_mad, &mad_size,
 		 &out_mad_pkey_index) &
@@ -358,31 +451,54 @@ static ssize_t show_pma_counter(struct ib_port *p, struct port_attribute *attr,
 		ret = -EINVAL;
 		goto out;
 	}
+	memcpy(data, out_mad->data + offset, size);
+	ret = size;
+out:
+	kfree(in_mad);
+	kfree(out_mad);
+	return ret;
+}
+
+static ssize_t show_pma_counter(struct ib_port *p, struct port_attribute *attr,
+				char *buf)
+{
+	struct port_table_attribute *tab_attr =
+		container_of(attr, struct port_table_attribute, attr);
+	int offset = tab_attr->index & 0xffff;
+	int width  = (tab_attr->index >> 16) & 0xff;
+	ssize_t ret;
+	u8 data[8];
+
+	ret = get_perf_mad(p->ibdev, p->port_num, tab_attr->attr_id, &data,
+			40 + offset / 8, sizeof(data));
+	if (ret < 0)
+		return sprintf(buf, "N/A (no PMA)\n");
 
 	switch (width) {
 	case 4:
-		ret = sprintf(buf, "%u\n", (out_mad->data[40 + offset / 8] >>
+		ret = sprintf(buf, "%u\n", (*data >>
 					    (4 - (offset % 8))) & 0xf);
 		break;
 	case 8:
-		ret = sprintf(buf, "%u\n", out_mad->data[40 + offset / 8]);
+		ret = sprintf(buf, "%u\n", *data);
 		break;
 	case 16:
 		ret = sprintf(buf, "%u\n",
-			      be16_to_cpup((__be16 *)(out_mad->data + 40 + offset / 8)));
+			      be16_to_cpup((__be16 *)data));
 		break;
 	case 32:
 		ret = sprintf(buf, "%u\n",
-			      be32_to_cpup((__be32 *)(out_mad->data + 40 + offset / 8)));
+			      be32_to_cpup((__be32 *)data));
+		break;
+	case 64:
+		ret = sprintf(buf, "%llu\n",
+				be64_to_cpup((__be64 *)data));
 		break;
+
 	default:
 		ret = 0;
 	}
 
-out:
-	kfree(in_mad);
-	kfree(out_mad);
-
 	return ret;
 }
 
@@ -403,6 +519,18 @@ static PORT_PMA_ATTR(port_rcv_data		    , 13, 32, 224);
 static PORT_PMA_ATTR(port_xmit_packets		    , 14, 32, 256);
 static PORT_PMA_ATTR(port_rcv_packets		    , 15, 32, 288);
 
+/*
+ * Counters added by extended set
+ */
+static PORT_PMA_ATTR_EXT(port_xmit_data		    , 64,  64);
+static PORT_PMA_ATTR_EXT(port_rcv_data		    , 64, 128);
+static PORT_PMA_ATTR_EXT(port_xmit_packets	    , 64, 192);
+static PORT_PMA_ATTR_EXT(port_rcv_packets	    , 64, 256);
+static PORT_PMA_ATTR_EXT(unicast_xmit_packets	    , 64, 320);
+static PORT_PMA_ATTR_EXT(unicast_rcv_packets	    , 64, 384);
+static PORT_PMA_ATTR_EXT(multicast_xmit_packets	    , 64, 448);
+static PORT_PMA_ATTR_EXT(multicast_rcv_packets	    , 64, 512);
+
 static struct attribute *pma_attrs[] = {
 	&port_pma_attr_symbol_error.attr.attr,
 	&port_pma_attr_link_error_recovery.attr.attr,
@@ -423,11 +551,65 @@ static struct attribute *pma_attrs[] = {
 	NULL
 };
 
+static struct attribute *pma_attrs_ext[] = {
+	&port_pma_attr_symbol_error.attr.attr,
+	&port_pma_attr_link_error_recovery.attr.attr,
+	&port_pma_attr_link_downed.attr.attr,
+	&port_pma_attr_port_rcv_errors.attr.attr,
+	&port_pma_attr_port_rcv_remote_physical_errors.attr.attr,
+	&port_pma_attr_port_rcv_switch_relay_errors.attr.attr,
+	&port_pma_attr_port_xmit_discards.attr.attr,
+	&port_pma_attr_port_xmit_constraint_errors.attr.attr,
+	&port_pma_attr_port_rcv_constraint_errors.attr.attr,
+	&port_pma_attr_local_link_integrity_errors.attr.attr,
+	&port_pma_attr_excessive_buffer_overrun_errors.attr.attr,
+	&port_pma_attr_VL15_dropped.attr.attr,
+	&port_pma_attr_ext_port_xmit_data.attr.attr,
+	&port_pma_attr_ext_port_rcv_data.attr.attr,
+	&port_pma_attr_ext_port_xmit_packets.attr.attr,
+	&port_pma_attr_ext_port_rcv_packets.attr.attr,
+	&port_pma_attr_ext_unicast_rcv_packets.attr.attr,
+	&port_pma_attr_ext_unicast_xmit_packets.attr.attr,
+	&port_pma_attr_ext_multicast_rcv_packets.attr.attr,
+	&port_pma_attr_ext_multicast_xmit_packets.attr.attr,
+	NULL
+};
+
+static struct attribute *pma_attrs_noietf[] = {
+	&port_pma_attr_symbol_error.attr.attr,
+	&port_pma_attr_link_error_recovery.attr.attr,
+	&port_pma_attr_link_downed.attr.attr,
+	&port_pma_attr_port_rcv_errors.attr.attr,
+	&port_pma_attr_port_rcv_remote_physical_errors.attr.attr,
+	&port_pma_attr_port_rcv_switch_relay_errors.attr.attr,
+	&port_pma_attr_port_xmit_discards.attr.attr,
+	&port_pma_attr_port_xmit_constraint_errors.attr.attr,
+	&port_pma_attr_port_rcv_constraint_errors.attr.attr,
+	&port_pma_attr_local_link_integrity_errors.attr.attr,
+	&port_pma_attr_excessive_buffer_overrun_errors.attr.attr,
+	&port_pma_attr_VL15_dropped.attr.attr,
+	&port_pma_attr_ext_port_xmit_data.attr.attr,
+	&port_pma_attr_ext_port_rcv_data.attr.attr,
+	&port_pma_attr_ext_port_xmit_packets.attr.attr,
+	&port_pma_attr_ext_port_rcv_packets.attr.attr,
+	NULL
+};
+
 static struct attribute_group pma_group = {
 	.name  = "counters",
 	.attrs  = pma_attrs
 };
 
+static struct attribute_group pma_group_ext = {
+	.name  = "counters",
+	.attrs  = pma_attrs_ext
+};
+
+static struct attribute_group pma_group_noietf = {
+	.name  = "counters",
+	.attrs  = pma_attrs_noietf
+};
+
 static void ib_port_release(struct kobject *kobj)
 {
 	struct ib_port *p = container_of(kobj, struct ib_port, kobj);
@@ -451,12 +633,41 @@ static void ib_port_release(struct kobject *kobj)
 	kfree(p);
 }
 
+static void ib_port_gid_attr_release(struct kobject *kobj)
+{
+	struct gid_attr_group *g = container_of(kobj, struct gid_attr_group,
+						kobj);
+	struct attribute *a;
+	int i;
+
+	if (g->ndev.attrs) {
+		for (i = 0; (a = g->ndev.attrs[i]); ++i)
+			kfree(a);
+
+		kfree(g->ndev.attrs);
+	}
+
+	if (g->type.attrs) {
+		for (i = 0; (a = g->type.attrs[i]); ++i)
+			kfree(a);
+
+		kfree(g->type.attrs);
+	}
+
+	kfree(g);
+}
+
 static struct kobj_type port_type = {
 	.release       = ib_port_release,
 	.sysfs_ops     = &port_sysfs_ops,
 	.default_attrs = port_default_attrs
 };
 
+static struct kobj_type gid_attr_type = {
+	.sysfs_ops      = &gid_attr_sysfs_ops,
+	.release        = ib_port_gid_attr_release
+};
+
 static struct attribute **
 alloc_group_attrs(ssize_t (*show)(struct ib_port *,
 				  struct port_attribute *, char *buf),
@@ -500,6 +711,31 @@ err:
 	return NULL;
 }
 
+/*
+ * Figure out which counter table to use depending on
+ * the device capabilities.
+ */
+static struct attribute_group *get_counter_table(struct ib_device *dev,
+						 int port_num)
+{
+	struct ib_class_port_info cpi;
+
+	if (get_perf_mad(dev, port_num, IB_PMA_CLASS_PORT_INFO,
+				&cpi, 40, sizeof(cpi)) >= 0) {
+
+		if (cpi.capability_mask && IB_PMA_CLASS_CAP_EXT_WIDTH)
+			/* We have extended counters */
+			return &pma_group_ext;
+
+		if (cpi.capability_mask && IB_PMA_CLASS_CAP_EXT_WIDTH_NOIETF)
+			/* But not the IETF ones */
+			return &pma_group_noietf;
+	}
+
+	/* Fall back to normal counters */
+	return &pma_group;
+}
+
 static int add_port(struct ib_device *device, int port_num,
 		    int (*port_callback)(struct ib_device *,
 					 u8, struct kobject *))
@@ -528,9 +764,24 @@ static int add_port(struct ib_device *device, int port_num,
 		return ret;
 	}
 
-	ret = sysfs_create_group(&p->kobj, &pma_group);
-	if (ret)
+	p->gid_attr_group = kzalloc(sizeof(*p->gid_attr_group), GFP_KERNEL);
+	if (!p->gid_attr_group) {
+		ret = -ENOMEM;
 		goto err_put;
+	}
+
+	p->gid_attr_group->port = p;
+	ret = kobject_init_and_add(&p->gid_attr_group->kobj, &gid_attr_type,
+				   &p->kobj, "gid_attrs");
+	if (ret) {
+		kfree(p->gid_attr_group);
+		goto err_put;
+	}
+
+	p->pma_table = get_counter_table(device, port_num);
+	ret = sysfs_create_group(&p->kobj, p->pma_table);
+	if (ret)
+		goto err_put_gid_attrs;
 
 	p->gid_group.name  = "gids";
 	p->gid_group.attrs = alloc_group_attrs(show_port_gid, attr.gid_tbl_len);
@@ -543,12 +794,38 @@ static int add_port(struct ib_device *device, int port_num,
 	if (ret)
 		goto err_free_gid;
 
+	p->gid_attr_group->ndev.name = "ndevs";
+	p->gid_attr_group->ndev.attrs = alloc_group_attrs(show_port_gid_attr_ndev,
+							  attr.gid_tbl_len);
+	if (!p->gid_attr_group->ndev.attrs) {
+		ret = -ENOMEM;
+		goto err_remove_gid;
+	}
+
+	ret = sysfs_create_group(&p->gid_attr_group->kobj,
+				 &p->gid_attr_group->ndev);
+	if (ret)
+		goto err_free_gid_ndev;
+
+	p->gid_attr_group->type.name = "types";
+	p->gid_attr_group->type.attrs = alloc_group_attrs(show_port_gid_attr_gid_type,
+							  attr.gid_tbl_len);
+	if (!p->gid_attr_group->type.attrs) {
+		ret = -ENOMEM;
+		goto err_remove_gid_ndev;
+	}
+
+	ret = sysfs_create_group(&p->gid_attr_group->kobj,
+				 &p->gid_attr_group->type);
+	if (ret)
+		goto err_free_gid_type;
+
 	p->pkey_group.name  = "pkeys";
 	p->pkey_group.attrs = alloc_group_attrs(show_port_pkey,
 						attr.pkey_tbl_len);
 	if (!p->pkey_group.attrs) {
 		ret = -ENOMEM;
-		goto err_remove_gid;
+		goto err_remove_gid_type;
 	}
 
 	ret = sysfs_create_group(&p->kobj, &p->pkey_group);
@@ -576,6 +853,28 @@ err_free_pkey:
 	kfree(p->pkey_group.attrs);
 	p->pkey_group.attrs = NULL;
 
+err_remove_gid_type:
+	sysfs_remove_group(&p->gid_attr_group->kobj,
+			   &p->gid_attr_group->type);
+
+err_free_gid_type:
+	for (i = 0; i < attr.gid_tbl_len; ++i)
+		kfree(p->gid_attr_group->type.attrs[i]);
+
+	kfree(p->gid_attr_group->type.attrs);
+	p->gid_attr_group->type.attrs = NULL;
+
+err_remove_gid_ndev:
+	sysfs_remove_group(&p->gid_attr_group->kobj,
+			   &p->gid_attr_group->ndev);
+
+err_free_gid_ndev:
+	for (i = 0; i < attr.gid_tbl_len; ++i)
+		kfree(p->gid_attr_group->ndev.attrs[i]);
+
+	kfree(p->gid_attr_group->ndev.attrs);
+	p->gid_attr_group->ndev.attrs = NULL;
+
 err_remove_gid:
 	sysfs_remove_group(&p->kobj, &p->gid_group);
 
@@ -587,7 +886,10 @@ err_free_gid:
 	p->gid_group.attrs = NULL;
 
 err_remove_pma:
-	sysfs_remove_group(&p->kobj, &pma_group);
+	sysfs_remove_group(&p->kobj, p->pma_table);
+
+err_put_gid_attrs:
+	kobject_put(&p->gid_attr_group->kobj);
 
 err_put:
 	kobject_put(&p->kobj);
@@ -614,18 +916,12 @@ static ssize_t show_sys_image_guid(struct device *device,
 				   struct device_attribute *dev_attr, char *buf)
 {
 	struct ib_device *dev = container_of(device, struct ib_device, dev);
-	struct ib_device_attr attr;
-	ssize_t ret;
-
-	ret = ib_query_device(dev, &attr);
-	if (ret)
-		return ret;
 
 	return sprintf(buf, "%04x:%04x:%04x:%04x\n",
-		       be16_to_cpu(((__be16 *) &attr.sys_image_guid)[0]),
-		       be16_to_cpu(((__be16 *) &attr.sys_image_guid)[1]),
-		       be16_to_cpu(((__be16 *) &attr.sys_image_guid)[2]),
-		       be16_to_cpu(((__be16 *) &attr.sys_image_guid)[3]));
+		       be16_to_cpu(((__be16 *) &dev->attrs.sys_image_guid)[0]),
+		       be16_to_cpu(((__be16 *) &dev->attrs.sys_image_guid)[1]),
+		       be16_to_cpu(((__be16 *) &dev->attrs.sys_image_guid)[2]),
+		       be16_to_cpu(((__be16 *) &dev->attrs.sys_image_guid)[3]));
 }
 
 static ssize_t show_node_guid(struct device *device,
@@ -800,9 +1096,14 @@ static void free_port_list_attributes(struct ib_device *device)
 	list_for_each_entry_safe(p, t, &device->port_list, entry) {
 		struct ib_port *port = container_of(p, struct ib_port, kobj);
 		list_del(&p->entry);
-		sysfs_remove_group(p, &pma_group);
+		sysfs_remove_group(p, port->pma_table);
 		sysfs_remove_group(p, &port->pkey_group);
 		sysfs_remove_group(p, &port->gid_group);
+		sysfs_remove_group(&port->gid_attr_group->kobj,
+				   &port->gid_attr_group->ndev);
+		sysfs_remove_group(&port->gid_attr_group->kobj,
+				   &port->gid_attr_group->type);
+		kobject_put(&port->gid_attr_group->kobj);
 		kobject_put(p);
 	}
 
diff --git a/drivers/infiniband/core/ud_header.c b/drivers/infiniband/core/ud_header.c
index 72feee6..19837d2 100644
--- a/drivers/infiniband/core/ud_header.c
+++ b/drivers/infiniband/core/ud_header.c
@@ -35,6 +35,7 @@
 #include <linux/string.h>
 #include <linux/export.h>
 #include <linux/if_ether.h>
+#include <linux/ip.h>
 
 #include <rdma/ib_pack.h>
 
@@ -116,6 +117,72 @@ static const struct ib_field vlan_table[]  = {
 	  .size_bits    = 16 }
 };
 
+static const struct ib_field ip4_table[]  = {
+	{ STRUCT_FIELD(ip4, ver),
+	  .offset_words = 0,
+	  .offset_bits  = 0,
+	  .size_bits    = 4 },
+	{ STRUCT_FIELD(ip4, hdr_len),
+	  .offset_words = 0,
+	  .offset_bits  = 4,
+	  .size_bits    = 4 },
+	{ STRUCT_FIELD(ip4, tos),
+	  .offset_words = 0,
+	  .offset_bits  = 8,
+	  .size_bits    = 8 },
+	{ STRUCT_FIELD(ip4, tot_len),
+	  .offset_words = 0,
+	  .offset_bits  = 16,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(ip4, id),
+	  .offset_words = 1,
+	  .offset_bits  = 0,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(ip4, frag_off),
+	  .offset_words = 1,
+	  .offset_bits  = 16,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(ip4, ttl),
+	  .offset_words = 2,
+	  .offset_bits  = 0,
+	  .size_bits    = 8 },
+	{ STRUCT_FIELD(ip4, protocol),
+	  .offset_words = 2,
+	  .offset_bits  = 8,
+	  .size_bits    = 8 },
+	{ STRUCT_FIELD(ip4, check),
+	  .offset_words = 2,
+	  .offset_bits  = 16,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(ip4, saddr),
+	  .offset_words = 3,
+	  .offset_bits  = 0,
+	  .size_bits    = 32 },
+	{ STRUCT_FIELD(ip4, daddr),
+	  .offset_words = 4,
+	  .offset_bits  = 0,
+	  .size_bits    = 32 }
+};
+
+static const struct ib_field udp_table[]  = {
+	{ STRUCT_FIELD(udp, sport),
+	  .offset_words = 0,
+	  .offset_bits  = 0,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(udp, dport),
+	  .offset_words = 0,
+	  .offset_bits  = 16,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(udp, length),
+	  .offset_words = 1,
+	  .offset_bits  = 0,
+	  .size_bits    = 16 },
+	{ STRUCT_FIELD(udp, csum),
+	  .offset_words = 1,
+	  .offset_bits  = 16,
+	  .size_bits    = 16 }
+};
+
 static const struct ib_field grh_table[]  = {
 	{ STRUCT_FIELD(grh, ip_version),
 	  .offset_words = 0,
@@ -213,26 +280,57 @@ static const struct ib_field deth_table[] = {
 	  .size_bits    = 24 }
 };
 
+__sum16 ib_ud_ip4_csum(struct ib_ud_header *header)
+{
+	struct iphdr iph;
+
+	iph.ihl		= 5;
+	iph.version	= 4;
+	iph.tos		= header->ip4.tos;
+	iph.tot_len	= header->ip4.tot_len;
+	iph.id		= header->ip4.id;
+	iph.frag_off	= header->ip4.frag_off;
+	iph.ttl		= header->ip4.ttl;
+	iph.protocol	= header->ip4.protocol;
+	iph.check	= 0;
+	iph.saddr	= header->ip4.saddr;
+	iph.daddr	= header->ip4.daddr;
+
+	return ip_fast_csum((u8 *)&iph, iph.ihl);
+}
+EXPORT_SYMBOL(ib_ud_ip4_csum);
+
 /**
  * ib_ud_header_init - Initialize UD header structure
  * @payload_bytes:Length of packet payload
  * @lrh_present: specify if LRH is present
  * @eth_present: specify if Eth header is present
  * @vlan_present: packet is tagged vlan
- * @grh_present:GRH flag (if non-zero, GRH will be included)
+ * @grh_present: GRH flag (if non-zero, GRH will be included)
+ * @ip_version: if non-zero, IP header, V4 or V6, will be included
+ * @udp_present :if non-zero, UDP header will be included
  * @immediate_present: specify if immediate data is present
  * @header:Structure to initialize
  */
-void ib_ud_header_init(int     		    payload_bytes,
-		       int		    lrh_present,
-		       int		    eth_present,
-		       int		    vlan_present,
-		       int    		    grh_present,
-		       int		    immediate_present,
-		       struct ib_ud_header *header)
+int ib_ud_header_init(int     payload_bytes,
+		      int    lrh_present,
+		      int    eth_present,
+		      int    vlan_present,
+		      int    grh_present,
+		      int    ip_version,
+		      int    udp_present,
+		      int    immediate_present,
+		      struct ib_ud_header *header)
 {
+	grh_present = grh_present && !ip_version;
 	memset(header, 0, sizeof *header);
 
+	/*
+	 * UDP header without IP header doesn't make sense
+	 */
+	if (udp_present && ip_version != 4 && ip_version != 6)
+		return -EINVAL;
+
 	if (lrh_present) {
 		u16 packet_length;
 
@@ -252,7 +350,7 @@ void ib_ud_header_init(int     		    payload_bytes,
 	if (vlan_present)
 		header->eth.type = cpu_to_be16(ETH_P_8021Q);
 
-	if (grh_present) {
+	if (ip_version == 6 || grh_present) {
 		header->grh.ip_version      = 6;
 		header->grh.payload_length  =
 			cpu_to_be16((IB_BTH_BYTES     +
@@ -260,8 +358,30 @@ void ib_ud_header_init(int     		    payload_bytes,
 				     payload_bytes    +
 				     4                + /* ICRC     */
 				     3) & ~3);          /* round up */
-		header->grh.next_header     = 0x1b;
+		header->grh.next_header     = udp_present ? IPPROTO_UDP : 0x1b;
+	}
+
+	if (ip_version == 4) {
+		int udp_bytes = udp_present ? IB_UDP_BYTES : 0;
+
+		header->ip4.ver = 4; /* version 4 */
+		header->ip4.hdr_len = 5; /* 5 words */
+		header->ip4.tot_len =
+			cpu_to_be16(IB_IP4_BYTES   +
+				     udp_bytes     +
+				     IB_BTH_BYTES  +
+				     IB_DETH_BYTES +
+				     payload_bytes +
+				     4);     /* ICRC     */
+		header->ip4.protocol = IPPROTO_UDP;
 	}
+	if (udp_present && ip_version)
+		header->udp.length =
+			cpu_to_be16(IB_UDP_BYTES   +
+				     IB_BTH_BYTES  +
+				     IB_DETH_BYTES +
+				     payload_bytes +
+				     4);     /* ICRC     */
 
 	if (immediate_present)
 		header->bth.opcode           = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE;
@@ -273,8 +393,11 @@ void ib_ud_header_init(int     		    payload_bytes,
 	header->lrh_present = lrh_present;
 	header->eth_present = eth_present;
 	header->vlan_present = vlan_present;
-	header->grh_present = grh_present;
+	header->grh_present = grh_present || (ip_version == 6);
+	header->ipv4_present = ip_version == 4;
+	header->udp_present = udp_present;
 	header->immediate_present = immediate_present;
+	return 0;
 }
 EXPORT_SYMBOL(ib_ud_header_init);
 
@@ -311,6 +434,16 @@ int ib_ud_header_pack(struct ib_ud_header *header,
 			&header->grh, buf + len);
 		len += IB_GRH_BYTES;
 	}
+	if (header->ipv4_present) {
+		ib_pack(ip4_table, ARRAY_SIZE(ip4_table),
+			&header->ip4, buf + len);
+		len += IB_IP4_BYTES;
+	}
+	if (header->udp_present) {
+		ib_pack(udp_table, ARRAY_SIZE(udp_table),
+			&header->udp, buf + len);
+		len += IB_UDP_BYTES;
+	}
 
 	ib_pack(bth_table, ARRAY_SIZE(bth_table),
 		&header->bth, buf + len);
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index 40becdb..e69bf26 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -232,7 +232,7 @@ static void ib_umem_notifier_invalidate_range_end(struct mmu_notifier *mn,
 	ib_ucontext_notifier_end_account(context);
 }
 
-static struct mmu_notifier_ops ib_umem_notifiers = {
+static const struct mmu_notifier_ops ib_umem_notifiers = {
 	.release                    = ib_umem_notifier_release,
 	.invalidate_page            = ib_umem_notifier_invalidate_page,
 	.invalidate_range_start     = ib_umem_notifier_invalidate_range_start,
diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c
index 57f281f..415a318 100644
--- a/drivers/infiniband/core/user_mad.c
+++ b/drivers/infiniband/core/user_mad.c
@@ -210,6 +210,7 @@ static void send_handler(struct ib_mad_agent *agent,
 }
 
 static void recv_handler(struct ib_mad_agent *agent,
+			 struct ib_mad_send_buf *send_buf,
 			 struct ib_mad_recv_wc *mad_recv_wc)
 {
 	struct ib_umad_file *file = agent->context;
diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h
index 94bbd8c..612ccfd 100644
--- a/drivers/infiniband/core/uverbs.h
+++ b/drivers/infiniband/core/uverbs.h
@@ -204,6 +204,8 @@ void ib_uverbs_event_handler(struct ib_event_handler *handler,
 			     struct ib_event *event);
 void ib_uverbs_dealloc_xrcd(struct ib_uverbs_device *dev, struct ib_xrcd *xrcd);
 
+int uverbs_dealloc_mw(struct ib_mw *mw);
+
 struct ib_uverbs_flow_spec {
 	union {
 		union {
diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 1c02dea..6ffc9c4 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -291,9 +291,6 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file,
 	struct ib_uverbs_get_context      cmd;
 	struct ib_uverbs_get_context_resp resp;
 	struct ib_udata                   udata;
-#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
-	struct ib_device_attr		  dev_attr;
-#endif
 	struct ib_ucontext		 *ucontext;
 	struct file			 *filp;
 	int ret;
@@ -342,10 +339,7 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file,
 	ucontext->odp_mrs_count = 0;
 	INIT_LIST_HEAD(&ucontext->no_private_counters);
 
-	ret = ib_query_device(ib_dev, &dev_attr);
-	if (ret)
-		goto err_free;
-	if (!(dev_attr.device_cap_flags & IB_DEVICE_ON_DEMAND_PAGING))
+	if (!(ib_dev->attrs.device_cap_flags & IB_DEVICE_ON_DEMAND_PAGING))
 		ucontext->invalidate_range = NULL;
 
 #endif
@@ -447,8 +441,6 @@ ssize_t ib_uverbs_query_device(struct ib_uverbs_file *file,
 {
 	struct ib_uverbs_query_device      cmd;
 	struct ib_uverbs_query_device_resp resp;
-	struct ib_device_attr              attr;
-	int                                ret;
 
 	if (out_len < sizeof resp)
 		return -ENOSPC;
@@ -456,12 +448,8 @@ ssize_t ib_uverbs_query_device(struct ib_uverbs_file *file,
 	if (copy_from_user(&cmd, buf, sizeof cmd))
 		return -EFAULT;
 
-	ret = ib_query_device(ib_dev, &attr);
-	if (ret)
-		return ret;
-
 	memset(&resp, 0, sizeof resp);
-	copy_query_dev_fields(file, ib_dev, &resp, &attr);
+	copy_query_dev_fields(file, ib_dev, &resp, &ib_dev->attrs);
 
 	if (copy_to_user((void __user *) (unsigned long) cmd.response,
 			 &resp, sizeof resp))
@@ -986,11 +974,8 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file,
 	}
 
 	if (cmd.access_flags & IB_ACCESS_ON_DEMAND) {
-		struct ib_device_attr attr;
-
-		ret = ib_query_device(pd->device, &attr);
-		if (ret || !(attr.device_cap_flags &
-				IB_DEVICE_ON_DEMAND_PAGING)) {
+		if (!(pd->device->attrs.device_cap_flags &
+		      IB_DEVICE_ON_DEMAND_PAGING)) {
 			pr_debug("ODP support not available\n");
 			ret = -EINVAL;
 			goto err_put;
@@ -1008,7 +993,6 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file,
 	mr->pd      = pd;
 	mr->uobject = uobj;
 	atomic_inc(&pd->usecnt);
-	atomic_set(&mr->usecnt, 0);
 
 	uobj->object = mr;
 	ret = idr_add_uobj(&ib_uverbs_mr_idr, uobj);
@@ -1106,11 +1090,6 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file,
 		}
 	}
 
-	if (atomic_read(&mr->usecnt)) {
-		ret = -EBUSY;
-		goto put_uobj_pd;
-	}
-
 	old_pd = mr->pd;
 	ret = mr->device->rereg_user_mr(mr, cmd.flags, cmd.start,
 					cmd.length, cmd.hca_va,
@@ -1258,7 +1237,7 @@ err_copy:
 	idr_remove_uobj(&ib_uverbs_mw_idr, uobj);
 
 err_unalloc:
-	ib_dealloc_mw(mw);
+	uverbs_dealloc_mw(mw);
 
 err_put:
 	put_pd_read(pd);
@@ -1287,7 +1266,7 @@ ssize_t ib_uverbs_dealloc_mw(struct ib_uverbs_file *file,
 
 	mw = uobj->object;
 
-	ret = ib_dealloc_mw(mw);
+	ret = uverbs_dealloc_mw(mw);
 	if (!ret)
 		uobj->live = 0;
 
@@ -1845,7 +1824,10 @@ static int create_qp(struct ib_uverbs_file *file,
 		      sizeof(cmd->create_flags))
 		attr.create_flags = cmd->create_flags;
 
-	if (attr.create_flags & ~IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) {
+	if (attr.create_flags & ~(IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK |
+				IB_QP_CREATE_CROSS_CHANNEL |
+				IB_QP_CREATE_MANAGED_SEND |
+				IB_QP_CREATE_MANAGED_RECV)) {
 		ret = -EINVAL;
 		goto err_put;
 	}
diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
index e3ef288..39680ae 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -133,6 +133,17 @@ static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file,
 static void ib_uverbs_add_one(struct ib_device *device);
 static void ib_uverbs_remove_one(struct ib_device *device, void *client_data);
 
+int uverbs_dealloc_mw(struct ib_mw *mw)
+{
+	struct ib_pd *pd = mw->pd;
+	int ret;
+
+	ret = mw->device->dealloc_mw(mw);
+	if (!ret)
+		atomic_dec(&pd->usecnt);
+	return ret;
+}
+
 static void ib_uverbs_release_dev(struct kobject *kobj)
 {
 	struct ib_uverbs_device *dev =
@@ -224,7 +235,7 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file,
 		struct ib_mw *mw = uobj->object;
 
 		idr_remove_uobj(&ib_uverbs_mw_idr, uobj);
-		ib_dealloc_mw(mw);
+		uverbs_dealloc_mw(mw);
 		kfree(uobj);
 	}
 
diff --git a/drivers/infiniband/core/uverbs_marshall.c b/drivers/infiniband/core/uverbs_marshall.c
index 7d2f14c..af020f8 100644
--- a/drivers/infiniband/core/uverbs_marshall.c
+++ b/drivers/infiniband/core/uverbs_marshall.c
@@ -144,5 +144,6 @@ void ib_copy_path_rec_from_user(struct ib_sa_path_rec *dst,
 	memset(dst->dmac, 0, sizeof(dst->dmac));
 	dst->net = NULL;
 	dst->ifindex = 0;
+	dst->gid_type = IB_GID_TYPE_IB;
 }
 EXPORT_SYMBOL(ib_copy_path_rec_from_user);
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 545906d..5af6d02 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -229,12 +229,6 @@ EXPORT_SYMBOL(rdma_port_get_link_layer);
 struct ib_pd *ib_alloc_pd(struct ib_device *device)
 {
 	struct ib_pd *pd;
-	struct ib_device_attr devattr;
-	int rc;
-
-	rc = ib_query_device(device, &devattr);
-	if (rc)
-		return ERR_PTR(rc);
 
 	pd = device->alloc_pd(device, NULL, NULL);
 	if (IS_ERR(pd))
@@ -245,7 +239,7 @@ struct ib_pd *ib_alloc_pd(struct ib_device *device)
 	pd->local_mr = NULL;
 	atomic_set(&pd->usecnt, 0);
 
-	if (devattr.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY)
+	if (device->attrs.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY)
 		pd->local_dma_lkey = device->local_dma_lkey;
 	else {
 		struct ib_mr *mr;
@@ -311,8 +305,61 @@ struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr)
 }
 EXPORT_SYMBOL(ib_create_ah);
 
+static int ib_get_header_version(const union rdma_network_hdr *hdr)
+{
+	const struct iphdr *ip4h = (struct iphdr *)&hdr->roce4grh;
+	struct iphdr ip4h_checked;
+	const struct ipv6hdr *ip6h = (struct ipv6hdr *)&hdr->ibgrh;
+
+	/* If it's IPv6, the version must be 6, otherwise, the first
+	 * 20 bytes (before the IPv4 header) are garbled.
+	 */
+	if (ip6h->version != 6)
+		return (ip4h->version == 4) ? 4 : 0;
+	/* version may be 6 or 4 because the first 20 bytes could be garbled */
+
+	/* RoCE v2 requires no options, thus header length
+	 * must be 5 words
+	 */
+	if (ip4h->ihl != 5)
+		return 6;
+
+	/* Verify checksum.
+	 * We can't write on scattered buffers so we need to copy to
+	 * temp buffer.
+	 */
+	memcpy(&ip4h_checked, ip4h, sizeof(ip4h_checked));
+	ip4h_checked.check = 0;
+	ip4h_checked.check = ip_fast_csum((u8 *)&ip4h_checked, 5);
+	/* if IPv4 header checksum is OK, believe it */
+	if (ip4h->check == ip4h_checked.check)
+		return 4;
+	return 6;
+}
+
+static enum rdma_network_type ib_get_net_type_by_grh(struct ib_device *device,
+						     u8 port_num,
+						     const struct ib_grh *grh)
+{
+	int grh_version;
+
+	if (rdma_protocol_ib(device, port_num))
+		return RDMA_NETWORK_IB;
+
+	grh_version = ib_get_header_version((union rdma_network_hdr *)grh);
+
+	if (grh_version == 4)
+		return RDMA_NETWORK_IPV4;
+
+	if (grh->next_hdr == IPPROTO_UDP)
+		return RDMA_NETWORK_IPV6;
+
+	return RDMA_NETWORK_ROCE_V1;
+}
+
 struct find_gid_index_context {
 	u16 vlan_id;
+	enum ib_gid_type gid_type;
 };
 
 static bool find_gid_index(const union ib_gid *gid,
@@ -322,6 +369,9 @@ static bool find_gid_index(const union ib_gid *gid,
 	struct find_gid_index_context *ctx =
 		(struct find_gid_index_context *)context;
 
+	if (ctx->gid_type != gid_attr->gid_type)
+		return false;
+
 	if ((!!(ctx->vlan_id != 0xffff) == !is_vlan_dev(gid_attr->ndev)) ||
 	    (is_vlan_dev(gid_attr->ndev) &&
 	     vlan_dev_vlan_id(gid_attr->ndev) != ctx->vlan_id))
@@ -332,14 +382,49 @@ static bool find_gid_index(const union ib_gid *gid,
 
 static int get_sgid_index_from_eth(struct ib_device *device, u8 port_num,
 				   u16 vlan_id, const union ib_gid *sgid,
+				   enum ib_gid_type gid_type,
 				   u16 *gid_index)
 {
-	struct find_gid_index_context context = {.vlan_id = vlan_id};
+	struct find_gid_index_context context = {.vlan_id = vlan_id,
+						 .gid_type = gid_type};
 
 	return ib_find_gid_by_filter(device, sgid, port_num, find_gid_index,
 				     &context, gid_index);
 }
 
+static int get_gids_from_rdma_hdr(union rdma_network_hdr *hdr,
+				  enum rdma_network_type net_type,
+				  union ib_gid *sgid, union ib_gid *dgid)
+{
+	struct sockaddr_in  src_in;
+	struct sockaddr_in  dst_in;
+	__be32 src_saddr, dst_saddr;
+
+	if (!sgid || !dgid)
+		return -EINVAL;
+
+	if (net_type == RDMA_NETWORK_IPV4) {
+		memcpy(&src_in.sin_addr.s_addr,
+		       &hdr->roce4grh.saddr, 4);
+		memcpy(&dst_in.sin_addr.s_addr,
+		       &hdr->roce4grh.daddr, 4);
+		src_saddr = src_in.sin_addr.s_addr;
+		dst_saddr = dst_in.sin_addr.s_addr;
+		ipv6_addr_set_v4mapped(src_saddr,
+				       (struct in6_addr *)sgid);
+		ipv6_addr_set_v4mapped(dst_saddr,
+				       (struct in6_addr *)dgid);
+		return 0;
+	} else if (net_type == RDMA_NETWORK_IPV6 ||
+		   net_type == RDMA_NETWORK_IB) {
+		*dgid = hdr->ibgrh.dgid;
+		*sgid = hdr->ibgrh.sgid;
+		return 0;
+	} else {
+		return -EINVAL;
+	}
+}
+
 int ib_init_ah_from_wc(struct ib_device *device, u8 port_num,
 		       const struct ib_wc *wc, const struct ib_grh *grh,
 		       struct ib_ah_attr *ah_attr)
@@ -347,33 +432,72 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num,
 	u32 flow_class;
 	u16 gid_index;
 	int ret;
+	enum rdma_network_type net_type = RDMA_NETWORK_IB;
+	enum ib_gid_type gid_type = IB_GID_TYPE_IB;
+	int hoplimit = 0xff;
+	union ib_gid dgid;
+	union ib_gid sgid;
 
 	memset(ah_attr, 0, sizeof *ah_attr);
 	if (rdma_cap_eth_ah(device, port_num)) {
+		if (wc->wc_flags & IB_WC_WITH_NETWORK_HDR_TYPE)
+			net_type = wc->network_hdr_type;
+		else
+			net_type = ib_get_net_type_by_grh(device, port_num, grh);
+		gid_type = ib_network_to_gid_type(net_type);
+	}
+	ret = get_gids_from_rdma_hdr((union rdma_network_hdr *)grh, net_type,
+				     &sgid, &dgid);
+	if (ret)
+		return ret;
+
+	if (rdma_protocol_roce(device, port_num)) {
+		int if_index = 0;
 		u16 vlan_id = wc->wc_flags & IB_WC_WITH_VLAN ?
 				wc->vlan_id : 0xffff;
+		struct net_device *idev;
+		struct net_device *resolved_dev;
 
 		if (!(wc->wc_flags & IB_WC_GRH))
 			return -EPROTOTYPE;
 
-		if (!(wc->wc_flags & IB_WC_WITH_SMAC) ||
-		    !(wc->wc_flags & IB_WC_WITH_VLAN)) {
-			ret = rdma_addr_find_dmac_by_grh(&grh->dgid, &grh->sgid,
-							 ah_attr->dmac,
-							 wc->wc_flags & IB_WC_WITH_VLAN ?
-							 NULL : &vlan_id,
-							 0);
-			if (ret)
-				return ret;
+		if (!device->get_netdev)
+			return -EOPNOTSUPP;
+
+		idev = device->get_netdev(device, port_num);
+		if (!idev)
+			return -ENODEV;
+
+		ret = rdma_addr_find_l2_eth_by_grh(&dgid, &sgid,
+						   ah_attr->dmac,
+						   wc->wc_flags & IB_WC_WITH_VLAN ?
+						   NULL : &vlan_id,
+						   &if_index, &hoplimit);
+		if (ret) {
+			dev_put(idev);
+			return ret;
 		}
 
-		ret = get_sgid_index_from_eth(device, port_num, vlan_id,
-					      &grh->dgid, &gid_index);
+		resolved_dev = dev_get_by_index(&init_net, if_index);
+		if (resolved_dev->flags & IFF_LOOPBACK) {
+			dev_put(resolved_dev);
+			resolved_dev = idev;
+			dev_hold(resolved_dev);
+		}
+		rcu_read_lock();
+		if (resolved_dev != idev && !rdma_is_upper_dev_rcu(idev,
+								   resolved_dev))
+			ret = -EHOSTUNREACH;
+		rcu_read_unlock();
+		dev_put(idev);
+		dev_put(resolved_dev);
 		if (ret)
 			return ret;
 
-		if (wc->wc_flags & IB_WC_WITH_SMAC)
-			memcpy(ah_attr->dmac, wc->smac, ETH_ALEN);
+		ret = get_sgid_index_from_eth(device, port_num, vlan_id,
+					      &dgid, gid_type, &gid_index);
+		if (ret)
+			return ret;
 	}
 
 	ah_attr->dlid = wc->slid;
@@ -383,10 +507,11 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num,
 
 	if (wc->wc_flags & IB_WC_GRH) {
 		ah_attr->ah_flags = IB_AH_GRH;
-		ah_attr->grh.dgid = grh->sgid;
+		ah_attr->grh.dgid = sgid;
 
 		if (!rdma_cap_eth_ah(device, port_num)) {
-			ret = ib_find_cached_gid_by_port(device, &grh->dgid,
+			ret = ib_find_cached_gid_by_port(device, &dgid,
+							 IB_GID_TYPE_IB,
 							 port_num, NULL,
 							 &gid_index);
 			if (ret)
@@ -396,7 +521,7 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num,
 		ah_attr->grh.sgid_index = (u8) gid_index;
 		flow_class = be32_to_cpu(grh->version_tclass_flow);
 		ah_attr->grh.flow_label = flow_class & 0xFFFFF;
-		ah_attr->grh.hop_limit = 0xFF;
+		ah_attr->grh.hop_limit = hoplimit;
 		ah_attr->grh.traffic_class = (flow_class >> 20) & 0xFF;
 	}
 	return 0;
@@ -1014,6 +1139,7 @@ int ib_resolve_eth_dmac(struct ib_qp *qp,
 			union ib_gid		sgid;
 			struct ib_gid_attr	sgid_attr;
 			int			ifindex;
+			int			hop_limit;
 
 			ret = ib_query_gid(qp->device,
 					   qp_attr->ah_attr.port_num,
@@ -1028,12 +1154,14 @@ int ib_resolve_eth_dmac(struct ib_qp *qp,
 
 			ifindex = sgid_attr.ndev->ifindex;
 
-			ret = rdma_addr_find_dmac_by_grh(&sgid,
-							 &qp_attr->ah_attr.grh.dgid,
-							 qp_attr->ah_attr.dmac,
-							 NULL, ifindex);
+			ret = rdma_addr_find_l2_eth_by_grh(&sgid,
+							   &qp_attr->ah_attr.grh.dgid,
+							   qp_attr->ah_attr.dmac,
+							   NULL, &ifindex, &hop_limit);
 
 			dev_put(sgid_attr.ndev);
+
+			qp_attr->ah_attr.grh.hop_limit = hop_limit;
 		}
 	}
 out:
@@ -1215,29 +1343,17 @@ struct ib_mr *ib_get_dma_mr(struct ib_pd *pd, int mr_access_flags)
 		mr->pd      = pd;
 		mr->uobject = NULL;
 		atomic_inc(&pd->usecnt);
-		atomic_set(&mr->usecnt, 0);
 	}
 
 	return mr;
 }
 EXPORT_SYMBOL(ib_get_dma_mr);
 
-int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr)
-{
-	return mr->device->query_mr ?
-		mr->device->query_mr(mr, mr_attr) : -ENOSYS;
-}
-EXPORT_SYMBOL(ib_query_mr);
-
 int ib_dereg_mr(struct ib_mr *mr)
 {
-	struct ib_pd *pd;
+	struct ib_pd *pd = mr->pd;
 	int ret;
 
-	if (atomic_read(&mr->usecnt))
-		return -EBUSY;
-
-	pd = mr->pd;
 	ret = mr->device->dereg_mr(mr);
 	if (!ret)
 		atomic_dec(&pd->usecnt);
@@ -1273,49 +1389,12 @@ struct ib_mr *ib_alloc_mr(struct ib_pd *pd,
 		mr->pd      = pd;
 		mr->uobject = NULL;
 		atomic_inc(&pd->usecnt);
-		atomic_set(&mr->usecnt, 0);
 	}
 
 	return mr;
 }
 EXPORT_SYMBOL(ib_alloc_mr);
 
-/* Memory windows */
-
-struct ib_mw *ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type)
-{
-	struct ib_mw *mw;
-
-	if (!pd->device->alloc_mw)
-		return ERR_PTR(-ENOSYS);
-
-	mw = pd->device->alloc_mw(pd, type);
-	if (!IS_ERR(mw)) {
-		mw->device  = pd->device;
-		mw->pd      = pd;
-		mw->uobject = NULL;
-		mw->type    = type;
-		atomic_inc(&pd->usecnt);
-	}
-
-	return mw;
-}
-EXPORT_SYMBOL(ib_alloc_mw);
-
-int ib_dealloc_mw(struct ib_mw *mw)
-{
-	struct ib_pd *pd;
-	int ret;
-
-	pd = mw->pd;
-	ret = mw->device->dealloc_mw(mw);
-	if (!ret)
-		atomic_dec(&pd->usecnt);
-
-	return ret;
-}
-EXPORT_SYMBOL(ib_dealloc_mw);
-
 /* "Fast" memory regions */
 
 struct ib_fmr *ib_alloc_fmr(struct ib_pd *pd,
@@ -1530,7 +1609,7 @@ int ib_sg_to_pages(struct ib_mr *mr,
 		   int (*set_page)(struct ib_mr *, u64))
 {
 	struct scatterlist *sg;
-	u64 last_end_dma_addr = 0, last_page_addr = 0;
+	u64 last_end_dma_addr = 0;
 	unsigned int last_page_off = 0;
 	u64 page_mask = ~((u64)mr->page_size - 1);
 	int i, ret;
@@ -1572,7 +1651,6 @@ next_page:
 
 		mr->length += dma_len;
 		last_end_dma_addr = end_dma_addr;
-		last_page_addr = end_dma_addr & page_mask;
 		last_page_off = end_dma_addr & ~page_mask;
 	}