10 files changed, 266 insertions, 163 deletions

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 474aa214ab57..0b8a79d53a00 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -336,6 +336,8 @@ static inline void ipoib_unregister_debugfs(void) { }
 extern int ipoib_sendq_size;
 extern int ipoib_recvq_size;
 
+extern struct ib_sa_client ipoib_sa_client;
+
 #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG
 extern int ipoib_debug_level;
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 5033666b1481..f426a69d9a43 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -169,117 +169,129 @@ static int ipoib_ib_post_receives(struct net_device *dev)
 	return 0;
 }
 
-static void ipoib_ib_handle_wc(struct net_device *dev,
-			       struct ib_wc *wc)
+static void ipoib_ib_handle_rx_wc(struct net_device *dev, struct ib_wc *wc)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
-	unsigned int wr_id = wc->wr_id;
+	unsigned int wr_id = wc->wr_id & ~IPOIB_OP_RECV;
+	struct sk_buff *skb;
+	dma_addr_t addr;
 
-	ipoib_dbg_data(priv, "called: id %d, op %d, status: %d\n",
+	ipoib_dbg_data(priv, "recv completion: id %d, op %d, status: %d\n",
 		       wr_id, wc->opcode, wc->status);
 
-	if (wr_id & IPOIB_OP_RECV) {
-		wr_id &= ~IPOIB_OP_RECV;
-
-		if (wr_id < ipoib_recvq_size) {
-			struct sk_buff *skb  = priv->rx_ring[wr_id].skb;
-			dma_addr_t      addr = priv->rx_ring[wr_id].mapping;
-
-			if (unlikely(wc->status != IB_WC_SUCCESS)) {
-				if (wc->status != IB_WC_WR_FLUSH_ERR)
-					ipoib_warn(priv, "failed recv event "
-						   "(status=%d, wrid=%d vend_err %x)\n",
-						   wc->status, wr_id, wc->vendor_err);
-				dma_unmap_single(priv->ca->dma_device, addr,
-						 IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
-				dev_kfree_skb_any(skb);
-				priv->rx_ring[wr_id].skb = NULL;
-				return;
-			}
+	if (unlikely(wr_id >= ipoib_recvq_size)) {
+		ipoib_warn(priv, "recv completion event with wrid %d (> %d)\n",
+			   wr_id, ipoib_recvq_size);
+		return;
+	}
 
-			/*
-			 * If we can't allocate a new RX buffer, dump
-			 * this packet and reuse the old buffer.
-			 */
-			if (unlikely(ipoib_alloc_rx_skb(dev, wr_id))) {
-				++priv->stats.rx_dropped;
-				goto repost;
-			}
+	skb  = priv->rx_ring[wr_id].skb;
+	addr = priv->rx_ring[wr_id].mapping;
 
-			ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n",
-				       wc->byte_len, wc->slid);
+	if (unlikely(wc->status != IB_WC_SUCCESS)) {
+		if (wc->status != IB_WC_WR_FLUSH_ERR)
+			ipoib_warn(priv, "failed recv event "
+				   "(status=%d, wrid=%d vend_err %x)\n",
+				   wc->status, wr_id, wc->vendor_err);
+		dma_unmap_single(priv->ca->dma_device, addr,
+				 IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
+		dev_kfree_skb_any(skb);
+		priv->rx_ring[wr_id].skb = NULL;
+		return;
+	}
 
-			dma_unmap_single(priv->ca->dma_device, addr,
-					 IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
+	/*
+	 * If we can't allocate a new RX buffer, dump
+	 * this packet and reuse the old buffer.
+	 */
+	if (unlikely(ipoib_alloc_rx_skb(dev, wr_id))) {
+		++priv->stats.rx_dropped;
+		goto repost;
+	}
 
-			skb_put(skb, wc->byte_len);
-			skb_pull(skb, IB_GRH_BYTES);
+	ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n",
+		       wc->byte_len, wc->slid);
 
-			if (wc->slid != priv->local_lid ||
-			    wc->src_qp != priv->qp->qp_num) {
-				skb->protocol = ((struct ipoib_header *) skb->data)->proto;
-				skb->mac.raw = skb->data;
-				skb_pull(skb, IPOIB_ENCAP_LEN);
+	dma_unmap_single(priv->ca->dma_device, addr,
+			 IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
 
-				dev->last_rx = jiffies;
-				++priv->stats.rx_packets;
-				priv->stats.rx_bytes += skb->len;
+	skb_put(skb, wc->byte_len);
+	skb_pull(skb, IB_GRH_BYTES);
 
-				skb->dev = dev;
-				/* XXX get correct PACKET_ type here */
-				skb->pkt_type = PACKET_HOST;
-				netif_rx_ni(skb);
-			} else {
-				ipoib_dbg_data(priv, "dropping loopback packet\n");
-				dev_kfree_skb_any(skb);
-			}
+	if (wc->slid != priv->local_lid ||
+	    wc->src_qp != priv->qp->qp_num) {
+		skb->protocol = ((struct ipoib_header *) skb->data)->proto;
+		skb->mac.raw = skb->data;
+		skb_pull(skb, IPOIB_ENCAP_LEN);
 
-		repost:
-			if (unlikely(ipoib_ib_post_receive(dev, wr_id)))
-				ipoib_warn(priv, "ipoib_ib_post_receive failed "
-					   "for buf %d\n", wr_id);
-		} else
-			ipoib_warn(priv, "completion event with wrid %d\n",
-				   wr_id);
+		dev->last_rx = jiffies;
+		++priv->stats.rx_packets;
+		priv->stats.rx_bytes += skb->len;
 
+		skb->dev = dev;
+		/* XXX get correct PACKET_ type here */
+		skb->pkt_type = PACKET_HOST;
+		netif_rx_ni(skb);
 	} else {
-		struct ipoib_tx_buf *tx_req;
-		unsigned long flags;
+		ipoib_dbg_data(priv, "dropping loopback packet\n");
+		dev_kfree_skb_any(skb);
+	}
 
-		if (wr_id >= ipoib_sendq_size) {
-			ipoib_warn(priv, "completion event with wrid %d (> %d)\n",
-				   wr_id, ipoib_sendq_size);
-			return;
-		}
+repost:
+	if (unlikely(ipoib_ib_post_receive(dev, wr_id)))
+		ipoib_warn(priv, "ipoib_ib_post_receive failed "
+			   "for buf %d\n", wr_id);
+}
 
-		ipoib_dbg_data(priv, "send complete, wrid %d\n", wr_id);
+static void ipoib_ib_handle_tx_wc(struct net_device *dev, struct ib_wc *wc)
+{
+	struct ipoib_dev_priv *priv = netdev_priv(dev);
+	unsigned int wr_id = wc->wr_id;
+	struct ipoib_tx_buf *tx_req;
+	unsigned long flags;
 
-		tx_req = &priv->tx_ring[wr_id];
+	ipoib_dbg_data(priv, "send completion: id %d, op %d, status: %d\n",
+		       wr_id, wc->opcode, wc->status);
 
-		dma_unmap_single(priv->ca->dma_device,
-				 pci_unmap_addr(tx_req, mapping),
-				 tx_req->skb->len,
-				 DMA_TO_DEVICE);
+	if (unlikely(wr_id >= ipoib_sendq_size)) {
+		ipoib_warn(priv, "send completion event with wrid %d (> %d)\n",
+			   wr_id, ipoib_sendq_size);
+		return;
+	}
 
-		++priv->stats.tx_packets;
-		priv->stats.tx_bytes += tx_req->skb->len;
+	tx_req = &priv->tx_ring[wr_id];
 
-		dev_kfree_skb_any(tx_req->skb);
+	dma_unmap_single(priv->ca->dma_device,
+			 pci_unmap_addr(tx_req, mapping),
+			 tx_req->skb->len,
+			 DMA_TO_DEVICE);
 
-		spin_lock_irqsave(&priv->tx_lock, flags);
-		++priv->tx_tail;
-		if (netif_queue_stopped(dev) &&
-		    test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags) &&
-		    priv->tx_head - priv->tx_tail <= ipoib_sendq_size >> 1)
-			netif_wake_queue(dev);
-		spin_unlock_irqrestore(&priv->tx_lock, flags);
+	++priv->stats.tx_packets;
+	priv->stats.tx_bytes += tx_req->skb->len;
 
-		if (wc->status != IB_WC_SUCCESS &&
-		    wc->status != IB_WC_WR_FLUSH_ERR)
-			ipoib_warn(priv, "failed send event "
-				   "(status=%d, wrid=%d vend_err %x)\n",
-				   wc->status, wr_id, wc->vendor_err);
-	}
+	dev_kfree_skb_any(tx_req->skb);
+
+	spin_lock_irqsave(&priv->tx_lock, flags);
+	++priv->tx_tail;
+	if (netif_queue_stopped(dev) &&
+	    test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags) &&
+	    priv->tx_head - priv->tx_tail <= ipoib_sendq_size >> 1)
+		netif_wake_queue(dev);
+	spin_unlock_irqrestore(&priv->tx_lock, flags);
+
+	if (wc->status != IB_WC_SUCCESS &&
+	    wc->status != IB_WC_WR_FLUSH_ERR)
+		ipoib_warn(priv, "failed send event "
+			   "(status=%d, wrid=%d vend_err %x)\n",
+			   wc->status, wr_id, wc->vendor_err);
+}
+
+static void ipoib_ib_handle_wc(struct net_device *dev, struct ib_wc *wc)
+{
+	if (wc->wr_id & IPOIB_OP_RECV)
+		ipoib_ib_handle_rx_wc(dev, wc);
+	else
+		ipoib_ib_handle_tx_wc(dev, wc);
 }
 
 void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr)
@@ -320,7 +332,7 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb,
 	struct ipoib_tx_buf *tx_req;
 	dma_addr_t addr;
 
-	if (skb->len > dev->mtu + INFINIBAND_ALEN) {
+	if (unlikely(skb->len > dev->mtu + INFINIBAND_ALEN)) {
 		ipoib_warn(priv, "packet len %d (> %d) too long to send, dropping\n",
 			   skb->len, dev->mtu + INFINIBAND_ALEN);
 		++priv->stats.tx_dropped;
@@ -619,8 +631,10 @@ void ipoib_ib_dev_flush(void *_dev)
 	 * The device could have been brought down between the start and when
 	 * we get here, don't bring it back up if it's not configured up
 	 */
-	if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
+	if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) {
 		ipoib_ib_dev_up(dev);
+		ipoib_mcast_restart_task(dev);
+	}
 
 	mutex_lock(&priv->vlan_mutex);
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index cf71d2a5515c..1eaf00e9862c 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -40,7 +40,6 @@
 
 #include <linux/init.h>
 #include <linux/slab.h>
-#include <linux/vmalloc.h>
 #include <linux/kernel.h>
 
 #include <linux/if_arp.h>	/* For ARPHRD_xxx */
@@ -82,6 +81,8 @@ static const u8 ipv4_bcast_addr[] = {
 
 struct workqueue_struct *ipoib_workqueue;
 
+struct ib_sa_client ipoib_sa_client;
+
 static void ipoib_add_one(struct ib_device *device);
 static void ipoib_remove_one(struct ib_device *device);
 
@@ -336,7 +337,8 @@ void ipoib_flush_paths(struct net_device *dev)
 	struct ipoib_path *path, *tp;
 	LIST_HEAD(remove_list);
 
-	spin_lock_irq(&priv->lock);
+	spin_lock_irq(&priv->tx_lock);
+	spin_lock(&priv->lock);
 
 	list_splice(&priv->path_list, &remove_list);
 	INIT_LIST_HEAD(&priv->path_list);
@@ -347,12 +349,15 @@ void ipoib_flush_paths(struct net_device *dev)
 	list_for_each_entry_safe(path, tp, &remove_list, list) {
 		if (path->query)
 			ib_sa_cancel_query(path->query_id, path->query);
-		spin_unlock_irq(&priv->lock);
+		spin_unlock(&priv->lock);
+		spin_unlock_irq(&priv->tx_lock);
 		wait_for_completion(&path->done);
 		path_free(dev, path);
-		spin_lock_irq(&priv->lock);
+		spin_lock_irq(&priv->tx_lock);
+		spin_lock(&priv->lock);
 	}
-	spin_unlock_irq(&priv->lock);
+	spin_unlock(&priv->lock);
+	spin_unlock_irq(&priv->tx_lock);
 }
 
 static void path_rec_completion(int status,
@@ -459,7 +464,7 @@ static int path_rec_start(struct net_device *dev,
 	init_completion(&path->done);
 
 	path->query_id =
-		ib_sa_path_rec_get(priv->ca, priv->port,
+		ib_sa_path_rec_get(&ipoib_sa_client, priv->ca, priv->port,
 				   &path->pathrec,
 				   IB_SA_PATH_REC_DGID		|
 				   IB_SA_PATH_REC_SGID		|
@@ -615,7 +620,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct ipoib_neigh *neigh;
 	unsigned long flags;
 
-	if (!spin_trylock_irqsave(&priv->tx_lock, flags))
+	if (unlikely(!spin_trylock_irqsave(&priv->tx_lock, flags)))
 		return NETDEV_TX_LOCKED;
 
 	/*
@@ -628,7 +633,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		return NETDEV_TX_BUSY;
 	}
 
-	if (skb->dst && skb->dst->neighbour) {
+	if (likely(skb->dst && skb->dst->neighbour)) {
 		if (unlikely(!*to_ipoib_neigh(skb->dst->neighbour))) {
 			ipoib_path_lookup(skb, dev);
 			goto out;
@@ -1107,13 +1112,16 @@ static void ipoib_add_one(struct ib_device *device)
 	struct ipoib_dev_priv *priv;
 	int s, e, p;
 
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
 	dev_list = kmalloc(sizeof *dev_list, GFP_KERNEL);
 	if (!dev_list)
 		return;
 
 	INIT_LIST_HEAD(dev_list);
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		s = 0;
 		e = 0;
 	} else {
@@ -1137,6 +1145,9 @@ static void ipoib_remove_one(struct ib_device *device)
 	struct ipoib_dev_priv *priv, *tmp;
 	struct list_head *dev_list;
 
+	if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB)
+		return;
+
 	dev_list = ib_get_client_data(device, &ipoib_client);
 
 	list_for_each_entry_safe(priv, tmp, dev_list, list) {
@@ -1181,13 +1192,16 @@ static int __init ipoib_init_module(void)
 		goto err_fs;
 	}
 
+	ib_sa_register_client(&ipoib_sa_client);
+
 	ret = ib_register_client(&ipoib_client);
 	if (ret)
-		goto err_wq;
+		goto err_sa;
 
 	return 0;
 
-err_wq:
+err_sa:
+	ib_sa_unregister_client(&ipoib_sa_client);
 	destroy_workqueue(ipoib_workqueue);
 
 err_fs:
@@ -1199,6 +1213,7 @@ err_fs:
 static void __exit ipoib_cleanup_module(void)
 {
 	ib_unregister_client(&ipoib_client);
+	ib_sa_unregister_client(&ipoib_sa_client);
 	ipoib_unregister_debugfs();
 	destroy_workqueue(ipoib_workqueue);
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index ec356ce7cdcd..3faa1820f0e9 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -361,7 +361,7 @@ static int ipoib_mcast_sendonly_join(struct ipoib_mcast *mcast)
 
 	init_completion(&mcast->done);
 
-	ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec,
+	ret = ib_sa_mcmember_rec_set(&ipoib_sa_client, priv->ca, priv->port, &rec,
 				     IB_SA_MCMEMBER_REC_MGID		|
 				     IB_SA_MCMEMBER_REC_PORT_GID	|
 				     IB_SA_MCMEMBER_REC_PKEY		|
@@ -472,22 +472,32 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast,
 
 	if (create) {
 		comp_mask |=
-			IB_SA_MCMEMBER_REC_QKEY		|
-			IB_SA_MCMEMBER_REC_SL		|
-			IB_SA_MCMEMBER_REC_FLOW_LABEL	|
-			IB_SA_MCMEMBER_REC_TRAFFIC_CLASS;
+			IB_SA_MCMEMBER_REC_QKEY			|
+			IB_SA_MCMEMBER_REC_MTU_SELECTOR		|
+			IB_SA_MCMEMBER_REC_MTU			|
+			IB_SA_MCMEMBER_REC_TRAFFIC_CLASS	|
+			IB_SA_MCMEMBER_REC_RATE_SELECTOR	|
+			IB_SA_MCMEMBER_REC_RATE			|
+			IB_SA_MCMEMBER_REC_SL			|
+			IB_SA_MCMEMBER_REC_FLOW_LABEL		|
+			IB_SA_MCMEMBER_REC_HOP_LIMIT;
 
 		rec.qkey	  = priv->broadcast->mcmember.qkey;
+		rec.mtu_selector  = IB_SA_EQ;
+		rec.mtu		  = priv->broadcast->mcmember.mtu;
+		rec.traffic_class = priv->broadcast->mcmember.traffic_class;
+		rec.rate_selector = IB_SA_EQ;
+		rec.rate	  = priv->broadcast->mcmember.rate;
 		rec.sl		  = priv->broadcast->mcmember.sl;
 		rec.flow_label	  = priv->broadcast->mcmember.flow_label;
-		rec.traffic_class = priv->broadcast->mcmember.traffic_class;
+		rec.hop_limit	  = priv->broadcast->mcmember.hop_limit;
 	}
 
 	init_completion(&mcast->done);
 
-	ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec, comp_mask,
-				     mcast->backoff * 1000, GFP_ATOMIC,
-				     ipoib_mcast_join_complete,
+	ret = ib_sa_mcmember_rec_set(&ipoib_sa_client, priv->ca, priv->port,
+				     &rec, comp_mask, mcast->backoff * 1000,
+				     GFP_ATOMIC, ipoib_mcast_join_complete,
 				     mcast, &mcast->query);
 
 	if (ret < 0) {
@@ -528,7 +538,7 @@ void ipoib_mcast_join_task(void *dev_ptr)
 			priv->local_rate = attr.active_speed *
 				ib_width_enum_to_int(attr.active_width);
 		} else
-			ipoib_warn(priv, "ib_query_port failed\n");
+		ipoib_warn(priv, "ib_query_port failed\n");
 	}
 
 	if (!priv->broadcast) {
@@ -681,7 +691,7 @@ static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast)
 	 * Just make one shot at leaving and don't wait for a reply;
 	 * if we fail, too bad.
 	 */
-	ret = ib_sa_mcmember_rec_delete(priv->ca, priv->port, &rec,
+	ret = ib_sa_mcmember_rec_delete(&ipoib_sa_client, priv->ca, priv->port, &rec,
 					IB_SA_MCMEMBER_REC_MGID		|
 					IB_SA_MCMEMBER_REC_PORT_GID	|
 					IB_SA_MCMEMBER_REC_PKEY		|
@@ -795,7 +805,7 @@ void ipoib_mcast_dev_flush(struct net_device *dev)
 	}
 
 	if (priv->broadcast) {
- 		rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree);
+		rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree);
 		list_add_tail(&priv->broadcast->list, &remove_list);
 		priv->broadcast = NULL;
 	}
diff --git a/drivers/infiniband/ulp/iser/Kconfig b/drivers/infiniband/ulp/iser/Kconfig
index fead87d1eff9..365a1b5f19e0 100644
--- a/drivers/infiniband/ulp/iser/Kconfig
+++ b/drivers/infiniband/ulp/iser/Kconfig
@@ -1,6 +1,6 @@
 config INFINIBAND_ISER
 	tristate "ISCSI RDMA Protocol"
-	depends on INFINIBAND && SCSI
+	depends on INFINIBAND && SCSI && INET
 	select SCSI_ISCSI_ATTRS
 	---help---
 	  Support for the ISCSI RDMA Protocol over InfiniBand.  This
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 1437d7ee3b19..2a14fe2e3226 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -141,18 +141,11 @@ iscsi_iser_cmd_init(struct iscsi_cmd_task *ctask)
 
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
 		BUG_ON(ctask->total_length == 0);
-		/* bytes to be sent via RDMA operations */
-		iser_ctask->rdma_data_count = ctask->total_length -
-					 ctask->imm_count -
-					 ctask->unsol_count;
 
-		debug_scsi("cmd [itt %x total %d imm %d unsol_data %d "
-			   "rdma_data %d]\n",
+		debug_scsi("cmd [itt %x total %d imm %d unsol_data %d\n",
 			   ctask->itt, ctask->total_length, ctask->imm_count,
-			   ctask->unsol_count, iser_ctask->rdma_data_count);
-	} else
-		/* bytes to be sent via RDMA operations */
-		iser_ctask->rdma_data_count = ctask->total_length;
+			   ctask->unsol_count);
+	}
 
 	iser_ctask_rdma_init(iser_ctask);
 }
@@ -196,13 +189,10 @@ iscsi_iser_ctask_xmit_unsol_data(struct iscsi_conn *conn,
 {
 	struct iscsi_data  hdr;
 	int error = 0;
-	struct iscsi_iser_cmd_task *iser_ctask = ctask->dd_data;
 
 	/* Send data-out PDUs while there's still unsolicited data to send */
 	while (ctask->unsol_count > 0) {
-		iscsi_prep_unsolicit_data_pdu(ctask, &hdr,
-					      iser_ctask->rdma_data_count);
-
+		iscsi_prep_unsolicit_data_pdu(ctask, &hdr);
 		debug_scsi("Sending data-out: itt 0x%x, data count %d\n",
 			   hdr.itt, ctask->data_count);
 
@@ -555,6 +545,7 @@ static struct scsi_host_template iscsi_iser_sht = {
 	.queuecommand           = iscsi_queuecommand,
 	.can_queue		= ISCSI_XMIT_CMDS_MAX - 1,
 	.sg_tablesize           = ISCSI_ISER_SG_TABLESIZE,
+	.max_sectors		= 1024,
 	.cmd_per_lun            = ISCSI_MAX_CMD_PER_LUN,
 	.eh_abort_handler       = iscsi_eh_abort,
 	.eh_host_reset_handler	= iscsi_eh_host_reset,
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h
index 3350ba690cfe..2cf9ae0def1c 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.h
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.h
@@ -82,8 +82,12 @@
 		       __func__ , ## arg);		\
 	} while (0)
 
+#define SHIFT_4K	12
+#define SIZE_4K	(1UL << SHIFT_4K)
+#define MASK_4K	(~(SIZE_4K-1))
+
 					/* support upto 512KB in one RDMA */
-#define ISCSI_ISER_SG_TABLESIZE         (0x80000 >> PAGE_SHIFT)
+#define ISCSI_ISER_SG_TABLESIZE         (0x80000 >> SHIFT_4K)
 #define ISCSI_ISER_MAX_LUN		256
 #define ISCSI_ISER_MAX_CMD_LEN		16
 
@@ -171,6 +175,7 @@ struct iser_mem_reg {
 	u64  va;
 	u64  len;
 	void *mem_h;
+	int  is_fmr;
 };
 
 struct iser_regd_buf {
@@ -257,7 +262,6 @@ struct iscsi_iser_conn {
 struct iscsi_iser_cmd_task {
 	struct iser_desc             desc;
 	struct iscsi_iser_conn	     *iser_conn;
-	int			     rdma_data_count;/* RDMA bytes           */
 	enum iser_task_status 	     status;
 	int                          command_sent;  /* set if command  sent  */
 	int                          dir[ISER_DIRS_NUM];      /* set if dir use*/
diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c
index 31950a522a1c..d0b03f426581 100644
--- a/drivers/infiniband/ulp/iser/iser_memory.c
+++ b/drivers/infiniband/ulp/iser/iser_memory.c
@@ -42,6 +42,7 @@
 #include "iscsi_iser.h"
 
 #define ISER_KMALLOC_THRESHOLD 0x20000 /* 128K - kmalloc limit */
+
 /**
  * Decrements the reference count for the
  * registered buffer & releases it
@@ -55,7 +56,7 @@ int iser_regd_buff_release(struct iser_regd_buf *regd_buf)
 	if ((atomic_read(&regd_buf->ref_count) == 0) ||
 	    atomic_dec_and_test(&regd_buf->ref_count)) {
 		/* if we used the dma mr, unreg is just NOP */
-		if (regd_buf->reg.rkey != 0)
+		if (regd_buf->reg.is_fmr)
 			iser_unreg_mem(&regd_buf->reg);
 
 		if (regd_buf->dma_addr) {
@@ -90,9 +91,9 @@ void iser_reg_single(struct iser_device *device,
 	BUG_ON(dma_mapping_error(dma_addr));
 
 	regd_buf->reg.lkey = device->mr->lkey;
-	regd_buf->reg.rkey = 0; /* indicate there's no need to unreg */
 	regd_buf->reg.len  = regd_buf->data_size;
 	regd_buf->reg.va   = dma_addr;
+	regd_buf->reg.is_fmr = 0;
 
 	regd_buf->dma_addr  = dma_addr;
 	regd_buf->direction = direction;
@@ -239,7 +240,7 @@ static int iser_sg_to_page_vec(struct iser_data_buf *data,
 	int i;
 
 	/* compute the offset of first element */
-	page_vec->offset = (u64) sg[0].offset;
+	page_vec->offset = (u64) sg[0].offset & ~MASK_4K;
 
 	for (i = 0; i < data->dma_nents; i++) {
 		total_sz += sg_dma_len(&sg[i]);
@@ -247,21 +248,30 @@ static int iser_sg_to_page_vec(struct iser_data_buf *data,
 		first_addr = sg_dma_address(&sg[i]);
 		last_addr  = first_addr + sg_dma_len(&sg[i]);
 
-		start_aligned = !(first_addr & ~PAGE_MASK);
-		end_aligned   = !(last_addr  & ~PAGE_MASK);
+		start_aligned = !(first_addr & ~MASK_4K);
+		end_aligned   = !(last_addr  & ~MASK_4K);
 
 		/* continue to collect page fragments till aligned or SG ends */
 		while (!end_aligned && (i + 1 < data->dma_nents)) {
 			i++;
 			total_sz += sg_dma_len(&sg[i]);
 			last_addr = sg_dma_address(&sg[i]) + sg_dma_len(&sg[i]);
-			end_aligned = !(last_addr  & ~PAGE_MASK);
+			end_aligned = !(last_addr  & ~MASK_4K);
 		}
 
-		first_addr = first_addr & PAGE_MASK;
-
-		for (page = first_addr; page < last_addr; page += PAGE_SIZE)
-			page_vec->pages[cur_page++] = page;
+		/* handle the 1st page in the 1st DMA element */
+		if (cur_page == 0) {
+			page = first_addr & MASK_4K;
+			page_vec->pages[cur_page] = page;
+			cur_page++;
+			page += SIZE_4K;
+		} else
+			page = first_addr;
+
+		for (; page < last_addr; page += SIZE_4K) {
+			page_vec->pages[cur_page] = page;
+			cur_page++;
+		}
 
 	}
 	page_vec->data_size = total_sz;
@@ -269,8 +279,7 @@ static int iser_sg_to_page_vec(struct iser_data_buf *data,
 	return cur_page;
 }
 
-#define MASK_4K			((1UL << 12) - 1) /* 0xFFF */
-#define IS_4K_ALIGNED(addr)	((((unsigned long)addr) & MASK_4K) == 0)
+#define IS_4K_ALIGNED(addr)	((((unsigned long)addr) & ~MASK_4K) == 0)
 
 /**
  * iser_data_buf_aligned_len - Tries to determine the maximal correctly aligned
@@ -320,9 +329,9 @@ static void iser_data_buf_dump(struct iser_data_buf *data)
 	struct scatterlist *sg = (struct scatterlist *)data->buf;
 	int i;
 
-	for (i = 0; i < data->size; i++)
+	for (i = 0; i < data->dma_nents; i++)
 		iser_err("sg[%d] dma_addr:0x%lX page:0x%p "
-			 "off:%d sz:%d dma_len:%d\n",
+			 "off:0x%x sz:0x%x dma_len:0x%x\n",
 			 i, (unsigned long)sg_dma_address(&sg[i]),
 			 sg[i].page, sg[i].offset,
 			 sg[i].length,sg_dma_len(&sg[i]));
@@ -352,7 +361,7 @@ static void iser_page_vec_build(struct iser_data_buf *data,
 
 	page_vec->length = page_vec_len;
 
-	if (page_vec_len * PAGE_SIZE < page_vec->data_size) {
+	if (page_vec_len * SIZE_4K < page_vec->data_size) {
 		iser_err("page_vec too short to hold this SG\n");
 		iser_data_buf_dump(data);
 		iser_dump_page_vec(page_vec);
@@ -370,15 +379,18 @@ int iser_reg_rdma_mem(struct iscsi_iser_cmd_task *iser_ctask,
 		      enum   iser_data_dir        cmd_dir)
 {
 	struct iser_conn     *ib_conn = iser_ctask->iser_conn->ib_conn;
+	struct iser_device   *device = ib_conn->device;
 	struct iser_data_buf *mem = &iser_ctask->data[cmd_dir];
 	struct iser_regd_buf *regd_buf;
 	int aligned_len;
 	int err;
+	int i;
+	struct scatterlist *sg;
 
 	regd_buf = &iser_ctask->rdma_regd[cmd_dir];
 
 	aligned_len = iser_data_buf_aligned_len(mem);
-	if (aligned_len != mem->size) {
+	if (aligned_len != mem->dma_nents) {
 		iser_err("rdma alignment violation %d/%d aligned\n",
 			 aligned_len, mem->size);
 		iser_data_buf_dump(mem);
@@ -389,10 +401,38 @@ int iser_reg_rdma_mem(struct iscsi_iser_cmd_task *iser_ctask,
 		mem = &iser_ctask->data_copy[cmd_dir];
 	}
 
-	iser_page_vec_build(mem, ib_conn->page_vec);
-	err = iser_reg_page_vec(ib_conn, ib_conn->page_vec, &regd_buf->reg);
-	if (err)
-		return err;
+	/* if there a single dma entry, FMR is not needed */
+	if (mem->dma_nents == 1) {
+		sg = (struct scatterlist *)mem->buf;
+
+		regd_buf->reg.lkey = device->mr->lkey;
+		regd_buf->reg.rkey = device->mr->rkey;
+		regd_buf->reg.len  = sg_dma_len(&sg[0]);
+		regd_buf->reg.va   = sg_dma_address(&sg[0]);
+		regd_buf->reg.is_fmr = 0;
+
+		iser_dbg("PHYSICAL Mem.register: lkey: 0x%08X rkey: 0x%08X  "
+			 "va: 0x%08lX sz: %ld]\n",
+			 (unsigned int)regd_buf->reg.lkey,
+			 (unsigned int)regd_buf->reg.rkey,
+			 (unsigned long)regd_buf->reg.va,
+			 (unsigned long)regd_buf->reg.len);
+	} else { /* use FMR for multiple dma entries */
+		iser_page_vec_build(mem, ib_conn->page_vec);
+		err = iser_reg_page_vec(ib_conn, ib_conn->page_vec, &regd_buf->reg);
+		if (err) {
+			iser_data_buf_dump(mem);
+			iser_err("mem->dma_nents = %d (dlength = 0x%x)\n", mem->dma_nents,
+				 ntoh24(iser_ctask->desc.iscsi_header.dlength));
+			iser_err("page_vec: data_size = 0x%x, length = %d, offset = 0x%x\n",
+				 ib_conn->page_vec->data_size, ib_conn->page_vec->length,
+				 ib_conn->page_vec->offset);
+			for (i=0 ; i<ib_conn->page_vec->length ; i++)
+				iser_err("page_vec[%d] = 0x%llx\n", i,
+					 (unsigned long long) ib_conn->page_vec->pages[i]);
+			return err;
+		}
+	}
 
 	/* take a reference on this regd buf such that it will not be released *
 	 * (eg in send dto completion) before we get the scsi response         */
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 72febf1f8ff8..ecdca7fc1e4c 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -88,8 +88,9 @@ static int iser_create_device_ib_res(struct iser_device *device)
 		     iser_cq_tasklet_fn,
 		     (unsigned long)device);
 
-	device->mr = ib_get_dma_mr(device->pd,
-				   IB_ACCESS_LOCAL_WRITE);
+	device->mr = ib_get_dma_mr(device->pd, IB_ACCESS_LOCAL_WRITE |
+				   IB_ACCESS_REMOTE_WRITE |
+				   IB_ACCESS_REMOTE_READ);
 	if (IS_ERR(device->mr))
 		goto dma_mr_err;
 
@@ -150,7 +151,7 @@ static int iser_create_ib_conn_res(struct iser_conn *ib_conn)
 	}
 	ib_conn->page_vec->pages = (u64 *) (ib_conn->page_vec + 1);
 
-	params.page_shift        = PAGE_SHIFT;
+	params.page_shift        = SHIFT_4K;
 	/* when the first/last SG element are not start/end *
 	 * page aligned, the map whould be of N+1 pages     */
 	params.max_pages_per_fmr = ISCSI_ISER_SG_TABLESIZE + 1;
@@ -604,8 +605,9 @@ int iser_reg_page_vec(struct iser_conn     *ib_conn,
 
 	mem_reg->lkey  = mem->fmr->lkey;
 	mem_reg->rkey  = mem->fmr->rkey;
-	mem_reg->len   = page_vec->length * PAGE_SIZE;
+	mem_reg->len   = page_vec->length * SIZE_4K;
 	mem_reg->va    = io_addr;
+	mem_reg->is_fmr = 1;
 	mem_reg->mem_h = (void *)mem;
 
 	mem_reg->va   += page_vec->offset;
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index fd8344cdc0db..44b9e5be6687 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -96,6 +96,8 @@ static struct ib_client srp_client = {
 	.remove = srp_remove_one
 };
 
+static struct ib_sa_client srp_sa_client;
+
 static inline struct srp_target_port *host_to_target(struct Scsi_Host *host)
 {
 	return (struct srp_target_port *) host->hostdata;
@@ -267,7 +269,8 @@ static int srp_lookup_path(struct srp_target_port *target)
 
 	init_completion(&target->done);
 
-	target->path_query_id = ib_sa_path_rec_get(target->srp_host->dev->dev,
+	target->path_query_id = ib_sa_path_rec_get(&srp_sa_client,
+						   target->srp_host->dev->dev,
 						   target->srp_host->port,
 						   &target->path,
 						   IB_SA_PATH_REC_DGID		|
@@ -330,7 +333,7 @@ static int srp_send_req(struct srp_target_port *target)
 	req->priv.req_buf_fmt 	= cpu_to_be16(SRP_BUF_FORMAT_DIRECT |
 					      SRP_BUF_FORMAT_INDIRECT);
 	/*
-	 * In the published SRP specification (draft rev. 16a), the 
+	 * In the published SRP specification (draft rev. 16a), the
 	 * port identifier format is 8 bytes of ID extension followed
 	 * by 8 bytes of GUID.  Older drafts put the two halves in the
 	 * opposite order, so that the GUID comes first.
@@ -1449,12 +1452,28 @@ static ssize_t show_zero_req_lim(struct class_device *cdev, char *buf)
 	return sprintf(buf, "%d\n", target->zero_req_lim);
 }
 
-static CLASS_DEVICE_ATTR(id_ext,	S_IRUGO, show_id_ext,		NULL);
-static CLASS_DEVICE_ATTR(ioc_guid,	S_IRUGO, show_ioc_guid,		NULL);
-static CLASS_DEVICE_ATTR(service_id,	S_IRUGO, show_service_id,	NULL);
-static CLASS_DEVICE_ATTR(pkey,		S_IRUGO, show_pkey,		NULL);
-static CLASS_DEVICE_ATTR(dgid,		S_IRUGO, show_dgid,		NULL);
-static CLASS_DEVICE_ATTR(zero_req_lim,	S_IRUGO, show_zero_req_lim,	NULL);
+static ssize_t show_local_ib_port(struct class_device *cdev, char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(cdev));
+
+	return sprintf(buf, "%d\n", target->srp_host->port);
+}
+
+static ssize_t show_local_ib_device(struct class_device *cdev, char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(cdev));
+
+	return sprintf(buf, "%s\n", target->srp_host->dev->dev->name);
+}
+
+static CLASS_DEVICE_ATTR(id_ext,	  S_IRUGO, show_id_ext,		 NULL);
+static CLASS_DEVICE_ATTR(ioc_guid,	  S_IRUGO, show_ioc_guid,	 NULL);
+static CLASS_DEVICE_ATTR(service_id,	  S_IRUGO, show_service_id,	 NULL);
+static CLASS_DEVICE_ATTR(pkey,		  S_IRUGO, show_pkey,		 NULL);
+static CLASS_DEVICE_ATTR(dgid,		  S_IRUGO, show_dgid,		 NULL);
+static CLASS_DEVICE_ATTR(zero_req_lim,	  S_IRUGO, show_zero_req_lim,	 NULL);
+static CLASS_DEVICE_ATTR(local_ib_port,   S_IRUGO, show_local_ib_port,	 NULL);
+static CLASS_DEVICE_ATTR(local_ib_device, S_IRUGO, show_local_ib_device, NULL);
 
 static struct class_device_attribute *srp_host_attrs[] = {
 	&class_device_attr_id_ext,
@@ -1463,6 +1482,8 @@ static struct class_device_attribute *srp_host_attrs[] = {
 	&class_device_attr_pkey,
 	&class_device_attr_dgid,
 	&class_device_attr_zero_req_lim,
+	&class_device_attr_local_ib_port,
+	&class_device_attr_local_ib_device,
 	NULL
 };
 
@@ -1881,7 +1902,7 @@ static void srp_add_one(struct ib_device *device)
 	if (IS_ERR(srp_dev->fmr_pool))
 		srp_dev->fmr_pool = NULL;
 
-	if (device->node_type == IB_NODE_SWITCH) {
+	if (device->node_type == RDMA_NODE_IB_SWITCH) {
 		s = 0;
 		e = 0;
 	} else {
@@ -1980,9 +2001,12 @@ static int __init srp_init_module(void)
 		return ret;
 	}
 
+	ib_sa_register_client(&srp_sa_client);
+
 	ret = ib_register_client(&srp_client);
 	if (ret) {
 		printk(KERN_ERR PFX "couldn't register IB client\n");
+		ib_sa_unregister_client(&srp_sa_client);
 		class_unregister(&srp_class);
 		return ret;
 	}
@@ -1993,6 +2017,7 @@ static int __init srp_init_module(void)
 static void __exit srp_cleanup_module(void)
 {
 	ib_unregister_client(&srp_client);
+	ib_sa_unregister_client(&srp_sa_client);
 	class_unregister(&srp_class);
 }