Merge commit 'v2.6.28-rc9' into next

15 files changed, 162 insertions, 83 deletions

diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
index d85af1b67027..eb36a81dd09b 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -358,8 +358,6 @@ static int ib_uverbs_event_close(struct inode *inode, struct file *filp)
 	}
 	spin_unlock_irq(&file->lock);
 
-	ib_uverbs_event_fasync(-1, filp, 0);
-
 	if (file->is_async) {
 		ib_unregister_event_handler(&file->uverbs_file->event_handler);
 		kref_put(&file->uverbs_file->ref, ib_uverbs_release_file);
diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c
index ecff98043589..160ef482712d 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_provider.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c
@@ -1102,9 +1102,7 @@ static u64 fw_vers_string_to_u64(struct iwch_dev *iwch_dev)
 	char *cp, *next;
 	unsigned fw_maj, fw_min, fw_mic;
 
-	rtnl_lock();
 	lldev->ethtool_ops->get_drvinfo(lldev, &info);
-	rtnl_unlock();
 
 	next = info.fw_version + 1;
 	cp = strsep(&next, ".");
@@ -1192,9 +1190,7 @@ static ssize_t show_fw_ver(struct device *dev, struct device_attribute *attr, ch
 	struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev;
 
 	PDBG("%s dev 0x%p\n", __func__, dev);
-	rtnl_lock();
 	lldev->ethtool_ops->get_drvinfo(lldev, &info);
-	rtnl_unlock();
 	return sprintf(buf, "%s\n", info.fw_version);
 }
 
@@ -1207,9 +1203,7 @@ static ssize_t show_hca(struct device *dev, struct device_attribute *attr,
 	struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev;
 
 	PDBG("%s dev 0x%p\n", __func__, dev);
-	rtnl_lock();
 	lldev->ethtool_ops->get_drvinfo(lldev, &info);
-	rtnl_unlock();
 	return sprintf(buf, "%s\n", info.driver);
 }
 
diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c
index 3e4585c2318a..19661b2f0406 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_qp.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c
@@ -745,7 +745,6 @@ int iwch_post_zb_read(struct iwch_qp *qhp)
 	wqe->read.rdmaop = T3_READ_REQ;
 	wqe->read.reserved[0] = 0;
 	wqe->read.reserved[1] = 0;
-	wqe->read.reserved[2] = 0;
 	wqe->read.rem_stag = cpu_to_be32(1);
 	wqe->read.rem_to = cpu_to_be64(1);
 	wqe->read.local_stag = cpu_to_be32(1);
diff --git a/drivers/infiniband/hw/ehca/ehca_classes.h b/drivers/infiniband/hw/ehca/ehca_classes.h
index 4df887af66a5..7fc35cf0cddf 100644
--- a/drivers/infiniband/hw/ehca/ehca_classes.h
+++ b/drivers/infiniband/hw/ehca/ehca_classes.h
@@ -163,7 +163,8 @@ struct ehca_mod_qp_parm {
 /* struct for tracking if cqes have been reported to the application */
 struct ehca_qmap_entry {
 	u16 app_wr_id;
-	u16 reported;
+	u8 reported;
+	u8 cqe_req;
 };
 
 struct ehca_queue_map {
@@ -171,6 +172,7 @@ struct ehca_queue_map {
 	unsigned int entries;
 	unsigned int tail;
 	unsigned int left_to_poll;
+	unsigned int next_wqe_idx;   /* Idx to first wqe to be flushed */
 };
 
 struct ehca_qp {
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index cb55be04442c..757035ea246f 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -359,36 +359,48 @@ static void notify_port_conf_change(struct ehca_shca *shca, int port_num)
 	*old_attr = new_attr;
 }
 
+/* replay modify_qp for sqps -- return 0 if all is well, 1 if AQP1 destroyed */
+static int replay_modify_qp(struct ehca_sport *sport)
+{
+	int aqp1_destroyed;
+	unsigned long flags;
+
+	spin_lock_irqsave(&sport->mod_sqp_lock, flags);
+
+	aqp1_destroyed = !sport->ibqp_sqp[IB_QPT_GSI];
+
+	if (sport->ibqp_sqp[IB_QPT_SMI])
+		ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_SMI]);
+	if (!aqp1_destroyed)
+		ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_GSI]);
+
+	spin_unlock_irqrestore(&sport->mod_sqp_lock, flags);
+
+	return aqp1_destroyed;
+}
+
 static void parse_ec(struct ehca_shca *shca, u64 eqe)
 {
 	u8 ec   = EHCA_BMASK_GET(NEQE_EVENT_CODE, eqe);
 	u8 port = EHCA_BMASK_GET(NEQE_PORT_NUMBER, eqe);
 	u8 spec_event;
 	struct ehca_sport *sport = &shca->sport[port - 1];
-	unsigned long flags;
 
 	switch (ec) {
 	case 0x30: /* port availability change */
 		if (EHCA_BMASK_GET(NEQE_PORT_AVAILABILITY, eqe)) {
-			int suppress_event;
-			/* replay modify_qp for sqps */
-			spin_lock_irqsave(&sport->mod_sqp_lock, flags);
-			suppress_event = !sport->ibqp_sqp[IB_QPT_GSI];
-			if (sport->ibqp_sqp[IB_QPT_SMI])
-				ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_SMI]);
-			if (!suppress_event)
-				ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_GSI]);
-			spin_unlock_irqrestore(&sport->mod_sqp_lock, flags);
-
-			/* AQP1 was destroyed, ignore this event */
-			if (suppress_event)
-				break;
+			/* only replay modify_qp calls in autodetect mode;
+			 * if AQP1 was destroyed, the port is already down
+			 * again and we can drop the event.
+			 */
+			if (ehca_nr_ports < 0)
+				if (replay_modify_qp(sport))
+					break;
 
 			sport->port_state = IB_PORT_ACTIVE;
 			dispatch_port_event(shca, port, IB_EVENT_PORT_ACTIVE,
 					    "is active");
-			ehca_query_sma_attr(shca, port,
-					    &sport->saved_attr);
+			ehca_query_sma_attr(shca, port, &sport->saved_attr);
 		} else {
 			sport->port_state = IB_PORT_DOWN;
 			dispatch_port_event(shca, port, IB_EVENT_PORT_ERR,
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c
index bb02a86aa526..bec7e0249358 100644
--- a/drivers/infiniband/hw/ehca/ehca_main.c
+++ b/drivers/infiniband/hw/ehca/ehca_main.c
@@ -994,8 +994,7 @@ static int ehca_mem_notifier(struct notifier_block *nb,
 			if (printk_timed_ratelimit(&ehca_dmem_warn_time,
 						   30 * 1000))
 				ehca_gen_err("DMEM operations are not allowed"
-					     "as long as an ehca adapter is"
-					     "attached to the LPAR");
+					     "in conjunction with eHCA");
 			return NOTIFY_BAD;
 		}
 	}
diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c
index 4d54b9f64567..cadbf0cdd910 100644
--- a/drivers/infiniband/hw/ehca/ehca_qp.c
+++ b/drivers/infiniband/hw/ehca/ehca_qp.c
@@ -435,9 +435,13 @@ static void reset_queue_map(struct ehca_queue_map *qmap)
 {
 	int i;
 
-	qmap->tail = 0;
-	for (i = 0; i < qmap->entries; i++)
+	qmap->tail = qmap->entries - 1;
+	qmap->left_to_poll = 0;
+	qmap->next_wqe_idx = 0;
+	for (i = 0; i < qmap->entries; i++) {
 		qmap->map[i].reported = 1;
+		qmap->map[i].cqe_req = 0;
+	}
 }
 
 /*
@@ -860,6 +864,11 @@ static struct ehca_qp *internal_create_qp(
 	if (qp_type == IB_QPT_GSI) {
 		h_ret = ehca_define_sqp(shca, my_qp, init_attr);
 		if (h_ret != H_SUCCESS) {
+			kfree(my_qp->mod_qp_parm);
+			my_qp->mod_qp_parm = NULL;
+			/* the QP pointer is no longer valid */
+			shca->sport[init_attr->port_num - 1].ibqp_sqp[qp_type] =
+				NULL;
 			ret = ehca2ib_return_code(h_ret);
 			goto create_qp_exit6;
 		}
@@ -1116,6 +1125,7 @@ static int calc_left_cqes(u64 wqe_p, struct ipz_queue *ipz_queue,
 	void *wqe_v;
 	u64 q_ofs;
 	u32 wqe_idx;
+	unsigned int tail_idx;
 
 	/* convert real to abs address */
 	wqe_p = wqe_p & (~(1UL << 63));
@@ -1128,12 +1138,17 @@ static int calc_left_cqes(u64 wqe_p, struct ipz_queue *ipz_queue,
 		return -EFAULT;
 	}
 
+	tail_idx = (qmap->tail + 1) % qmap->entries;
 	wqe_idx = q_ofs / ipz_queue->qe_size;
-	if (wqe_idx < qmap->tail)
-		qmap->left_to_poll = (qmap->entries - qmap->tail) + wqe_idx;
-	else
-		qmap->left_to_poll = wqe_idx - qmap->tail;
 
+	/* check all processed wqes, whether a cqe is requested or not */
+	while (tail_idx != wqe_idx) {
+		if (qmap->map[tail_idx].cqe_req)
+			qmap->left_to_poll++;
+		tail_idx = (tail_idx + 1) % qmap->entries;
+	}
+	/* save index in queue, where we have to start flushing */
+	qmap->next_wqe_idx = wqe_idx;
 	return 0;
 }
 
@@ -1180,10 +1195,14 @@ static int check_for_left_cqes(struct ehca_qp *my_qp, struct ehca_shca *shca)
 	} else {
 		spin_lock_irqsave(&my_qp->send_cq->spinlock, flags);
 		my_qp->sq_map.left_to_poll = 0;
+		my_qp->sq_map.next_wqe_idx = (my_qp->sq_map.tail + 1) %
+						my_qp->sq_map.entries;
 		spin_unlock_irqrestore(&my_qp->send_cq->spinlock, flags);
 
 		spin_lock_irqsave(&my_qp->recv_cq->spinlock, flags);
 		my_qp->rq_map.left_to_poll = 0;
+		my_qp->rq_map.next_wqe_idx = (my_qp->rq_map.tail + 1) %
+						my_qp->rq_map.entries;
 		spin_unlock_irqrestore(&my_qp->recv_cq->spinlock, flags);
 	}
 
diff --git a/drivers/infiniband/hw/ehca/ehca_reqs.c b/drivers/infiniband/hw/ehca/ehca_reqs.c
index 64928079eafa..00a648f4316c 100644
--- a/drivers/infiniband/hw/ehca/ehca_reqs.c
+++ b/drivers/infiniband/hw/ehca/ehca_reqs.c
@@ -179,6 +179,7 @@ static inline int ehca_write_swqe(struct ehca_qp *qp,
 
 	qmap_entry->app_wr_id = get_app_wr_id(send_wr->wr_id);
 	qmap_entry->reported = 0;
+	qmap_entry->cqe_req = 0;
 
 	switch (send_wr->opcode) {
 	case IB_WR_SEND:
@@ -203,8 +204,10 @@ static inline int ehca_write_swqe(struct ehca_qp *qp,
 
 	if ((send_wr->send_flags & IB_SEND_SIGNALED ||
 	    qp->init_attr.sq_sig_type == IB_SIGNAL_ALL_WR)
-	    && !hidden)
+	    && !hidden) {
 		wqe_p->wr_flag |= WQE_WRFLAG_REQ_SIGNAL_COM;
+		qmap_entry->cqe_req = 1;
+	}
 
 	if (send_wr->opcode == IB_WR_SEND_WITH_IMM ||
 	    send_wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) {
@@ -569,6 +572,7 @@ static int internal_post_recv(struct ehca_qp *my_qp,
 		qmap_entry = &my_qp->rq_map.map[rq_map_idx];
 		qmap_entry->app_wr_id = get_app_wr_id(cur_recv_wr->wr_id);
 		qmap_entry->reported = 0;
+		qmap_entry->cqe_req = 1;
 
 		wqe_cnt++;
 	} /* eof for cur_recv_wr */
@@ -706,27 +710,34 @@ repoll:
 		goto repoll;
 	wc->qp = &my_qp->ib_qp;
 
+	qmap_tail_idx = get_app_wr_id(cqe->work_request_id);
+	if (!(cqe->w_completion_flags & WC_SEND_RECEIVE_BIT))
+		/* We got a send completion. */
+		qmap = &my_qp->sq_map;
+	else
+		/* We got a receive completion. */
+		qmap = &my_qp->rq_map;
+
+	/* advance the tail pointer */
+	qmap->tail = qmap_tail_idx;
+
 	if (is_error) {
 		/*
 		 * set left_to_poll to 0 because in error state, we will not
 		 * get any additional CQEs
 		 */
-		ehca_add_to_err_list(my_qp, 1);
+		my_qp->sq_map.next_wqe_idx = (my_qp->sq_map.tail + 1) %
+						my_qp->sq_map.entries;
 		my_qp->sq_map.left_to_poll = 0;
+		ehca_add_to_err_list(my_qp, 1);
 
+		my_qp->rq_map.next_wqe_idx = (my_qp->rq_map.tail + 1) %
+						my_qp->rq_map.entries;
+		my_qp->rq_map.left_to_poll = 0;
 		if (HAS_RQ(my_qp))
 			ehca_add_to_err_list(my_qp, 0);
-		my_qp->rq_map.left_to_poll = 0;
 	}
 
-	qmap_tail_idx = get_app_wr_id(cqe->work_request_id);
-	if (!(cqe->w_completion_flags & WC_SEND_RECEIVE_BIT))
-		/* We got a send completion. */
-		qmap = &my_qp->sq_map;
-	else
-		/* We got a receive completion. */
-		qmap = &my_qp->rq_map;
-
 	qmap_entry = &qmap->map[qmap_tail_idx];
 	if (qmap_entry->reported) {
 		ehca_warn(cq->device, "Double cqe on qp_num=%#x",
@@ -738,10 +749,6 @@ repoll:
 	wc->wr_id = replace_wr_id(cqe->work_request_id, qmap_entry->app_wr_id);
 	qmap_entry->reported = 1;
 
-	/* this is a proper completion, we need to advance the tail pointer */
-	if (++qmap->tail == qmap->entries)
-		qmap->tail = 0;
-
 	/* if left_to_poll is decremented to 0, add the QP to the error list */
 	if (qmap->left_to_poll > 0) {
 		qmap->left_to_poll--;
@@ -805,13 +812,14 @@ static int generate_flush_cqes(struct ehca_qp *my_qp, struct ib_cq *cq,
 	else
 		qmap = &my_qp->rq_map;
 
-	qmap_entry = &qmap->map[qmap->tail];
+	qmap_entry = &qmap->map[qmap->next_wqe_idx];
 
 	while ((nr < num_entries) && (qmap_entry->reported == 0)) {
 		/* generate flush CQE */
+
 		memset(wc, 0, sizeof(*wc));
 
-		offset = qmap->tail * ipz_queue->qe_size;
+		offset = qmap->next_wqe_idx * ipz_queue->qe_size;
 		wqe = (struct ehca_wqe *)ipz_qeit_calc(ipz_queue, offset);
 		if (!wqe) {
 			ehca_err(cq->device, "Invalid wqe offset=%#lx on "
@@ -850,11 +858,12 @@ static int generate_flush_cqes(struct ehca_qp *my_qp, struct ib_cq *cq,
 
 		wc->qp = &my_qp->ib_qp;
 
-		/* mark as reported and advance tail pointer */
+		/* mark as reported and advance next_wqe pointer */
 		qmap_entry->reported = 1;
-		if (++qmap->tail == qmap->entries)
-			qmap->tail = 0;
-		qmap_entry = &qmap->map[qmap->tail];
+		qmap->next_wqe_idx++;
+		if (qmap->next_wqe_idx == qmap->entries)
+			qmap->next_wqe_idx = 0;
+		qmap_entry = &qmap->map[qmap->next_wqe_idx];
 
 		wc++; nr++;
 	}
diff --git a/drivers/infiniband/hw/ipath/ipath_ruc.c b/drivers/infiniband/hw/ipath/ipath_ruc.c
index fc0f6d9e6030..2296832f94da 100644
--- a/drivers/infiniband/hw/ipath/ipath_ruc.c
+++ b/drivers/infiniband/hw/ipath/ipath_ruc.c
@@ -156,7 +156,7 @@ bail:
 /**
  * ipath_get_rwqe - copy the next RWQE into the QP's RWQE
  * @qp: the QP
- * @wr_id_only: update wr_id only, not SGEs
+ * @wr_id_only: update qp->r_wr_id only, not qp->r_sge
  *
  * Return 0 if no RWQE is available, otherwise return 1.
  *
@@ -173,8 +173,6 @@ int ipath_get_rwqe(struct ipath_qp *qp, int wr_id_only)
 	u32 tail;
 	int ret;
 
-	qp->r_sge.sg_list = qp->r_sg_list;
-
 	if (qp->ibqp.srq) {
 		srq = to_isrq(qp->ibqp.srq);
 		handler = srq->ibsrq.event_handler;
@@ -206,8 +204,10 @@ int ipath_get_rwqe(struct ipath_qp *qp, int wr_id_only)
 		wqe = get_rwqe_ptr(rq, tail);
 		if (++tail >= rq->size)
 			tail = 0;
-	} while (!wr_id_only && !ipath_init_sge(qp, wqe, &qp->r_len,
-						&qp->r_sge));
+		if (wr_id_only)
+			break;
+		qp->r_sge.sg_list = qp->r_sg_list;
+	} while (!ipath_init_sge(qp, wqe, &qp->r_len, &qp->r_sge));
 	qp->r_wr_id = wqe->wr_id;
 	wq->tail = tail;
 
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index d0866a3636e2..18308494a195 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -343,6 +343,7 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 {
 	struct mlx4_ib_dev *dev = to_mdev(ibcq->device);
 	struct mlx4_ib_cq *cq = to_mcq(ibcq);
+	struct mlx4_mtt mtt;
 	int outst_cqe;
 	int err;
 
@@ -376,10 +377,13 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 			goto out;
 	}
 
+	mtt = cq->buf.mtt;
+
 	err = mlx4_cq_resize(dev->dev, &cq->mcq, entries, &cq->resize_buf->buf.mtt);
 	if (err)
 		goto err_buf;
 
+	mlx4_mtt_cleanup(dev->dev, &mtt);
 	if (ibcq->uobject) {
 		cq->buf      = cq->resize_buf->buf;
 		cq->ibcq.cqe = cq->resize_buf->cqe;
@@ -406,6 +410,7 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 	goto out;
 
 err_buf:
+	mlx4_mtt_cleanup(dev->dev, &cq->resize_buf->buf.mtt);
 	if (!ibcq->uobject)
 		mlx4_ib_free_cq_buf(dev, &cq->resize_buf->buf,
 				    cq->resize_buf->cqe);
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index 87f5c5a87b98..8e4d26d56a95 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -205,6 +205,7 @@ struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
 		goto err_mr;
 
 	mr->ibmr.rkey = mr->ibmr.lkey = mr->mmr.key;
+	mr->umem = NULL;
 
 	return &mr->ibmr;
 
diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c
index a2b04d62b1a4..aa1dc41f04c8 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -95,6 +95,10 @@ unsigned int wqm_quanta = 0x10000;
 module_param(wqm_quanta, int, 0644);
 MODULE_PARM_DESC(wqm_quanta, "WQM quanta");
 
+static unsigned int limit_maxrdreqsz;
+module_param(limit_maxrdreqsz, bool, 0644);
+MODULE_PARM_DESC(limit_maxrdreqsz, "Limit max read request size to 256 Bytes");
+
 LIST_HEAD(nes_adapter_list);
 static LIST_HEAD(nes_dev_list);
 
@@ -588,6 +592,18 @@ static int __devinit nes_probe(struct pci_dev *pcidev, const struct pci_device_i
 						nesdev->nesadapter->port_count;
 	}
 
+	if ((limit_maxrdreqsz ||
+	     ((nesdev->nesadapter->phy_type[0] == NES_PHY_TYPE_GLADIUS) &&
+	      (hw_rev == NE020_REV1))) &&
+	    (pcie_get_readrq(pcidev) > 256)) {
+		if (pcie_set_readrq(pcidev, 256))
+			printk(KERN_ERR PFX "Unable to set max read request"
+				" to 256 bytes\n");
+		else
+			nes_debug(NES_DBG_INIT, "Max read request size set"
+				" to 256 bytes\n");
+	}
+
 	tasklet_init(&nesdev->dpc_tasklet, nes_dpc, (unsigned long)nesdev);
 
 	/* bring up the Control QP */
diff --git a/drivers/infiniband/hw/nes/nes_hw.h b/drivers/infiniband/hw/nes/nes_hw.h
index 610b9d859597..bc0b4de04450 100644
--- a/drivers/infiniband/hw/nes/nes_hw.h
+++ b/drivers/infiniband/hw/nes/nes_hw.h
@@ -40,6 +40,7 @@
 #define NES_PHY_TYPE_ARGUS     4
 #define NES_PHY_TYPE_PUMA_1G   5
 #define NES_PHY_TYPE_PUMA_10G  6
+#define NES_PHY_TYPE_GLADIUS   7
 
 #define NES_MULTICAST_PF_MAX 8
 
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index 932e56fcf774..d36c9a0bf1bb 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -220,14 +220,14 @@ static int nes_bind_mw(struct ib_qp *ibqp, struct ib_mw *ibmw,
 	if (nesqp->ibqp_state > IB_QPS_RTS)
 		return -EINVAL;
 
-		spin_lock_irqsave(&nesqp->lock, flags);
+	spin_lock_irqsave(&nesqp->lock, flags);
 
 	head = nesqp->hwqp.sq_head;
 	qsize = nesqp->hwqp.sq_tail;
 
 	/* Check for SQ overflow */
 	if (((head + (2 * qsize) - nesqp->hwqp.sq_tail) % qsize) == (qsize - 1)) {
-			spin_unlock_irqrestore(&nesqp->lock, flags);
+		spin_unlock_irqrestore(&nesqp->lock, flags);
 		return -EINVAL;
 	}
 
@@ -269,7 +269,7 @@ static int nes_bind_mw(struct ib_qp *ibqp, struct ib_mw *ibmw,
 	nes_write32(nesdev->regs+NES_WQE_ALLOC,
 			(1 << 24) | 0x00800000 | nesqp->hwqp.qp_id);
 
-		spin_unlock_irqrestore(&nesqp->lock, flags);
+	spin_unlock_irqrestore(&nesqp->lock, flags);
 
 	return 0;
 }
@@ -349,7 +349,7 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd,
 			if (nesfmr->nesmr.pbls_used > nesadapter->free_4kpbl) {
 				spin_unlock_irqrestore(&nesadapter->pbl_lock, flags);
 				ret = -ENOMEM;
-				goto failed_vpbl_alloc;
+				goto failed_vpbl_avail;
 			} else {
 				nesadapter->free_4kpbl -= nesfmr->nesmr.pbls_used;
 			}
@@ -357,7 +357,7 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd,
 			if (nesfmr->nesmr.pbls_used > nesadapter->free_256pbl) {
 				spin_unlock_irqrestore(&nesadapter->pbl_lock, flags);
 				ret = -ENOMEM;
-				goto failed_vpbl_alloc;
+				goto failed_vpbl_avail;
 			} else {
 				nesadapter->free_256pbl -= nesfmr->nesmr.pbls_used;
 			}
@@ -391,14 +391,14 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd,
 			goto failed_vpbl_alloc;
 		}
 
-		nesfmr->root_vpbl.leaf_vpbl = kzalloc(sizeof(*nesfmr->root_vpbl.leaf_vpbl)*1024, GFP_KERNEL);
+		nesfmr->leaf_pbl_cnt = nesfmr->nesmr.pbls_used-1;
+		nesfmr->root_vpbl.leaf_vpbl = kzalloc(sizeof(*nesfmr->root_vpbl.leaf_vpbl)*1024, GFP_ATOMIC);
 		if (!nesfmr->root_vpbl.leaf_vpbl) {
 			spin_unlock_irqrestore(&nesadapter->pbl_lock, flags);
 			ret = -ENOMEM;
 			goto failed_leaf_vpbl_alloc;
 		}
 
-		nesfmr->leaf_pbl_cnt = nesfmr->nesmr.pbls_used-1;
 		nes_debug(NES_DBG_MR, "two level pbl, root_vpbl.pbl_vbase=%p"
 				" leaf_pbl_cnt=%d root_vpbl.leaf_vpbl=%p\n",
 				nesfmr->root_vpbl.pbl_vbase, nesfmr->leaf_pbl_cnt, nesfmr->root_vpbl.leaf_vpbl);
@@ -519,6 +519,16 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd,
 				nesfmr->root_vpbl.pbl_pbase);
 
 	failed_vpbl_alloc:
+	if (nesfmr->nesmr.pbls_used != 0) {
+		spin_lock_irqsave(&nesadapter->pbl_lock, flags);
+		if (nesfmr->nesmr.pbl_4k)
+			nesadapter->free_4kpbl += nesfmr->nesmr.pbls_used;
+		else
+			nesadapter->free_256pbl += nesfmr->nesmr.pbls_used;
+		spin_unlock_irqrestore(&nesadapter->pbl_lock, flags);
+	}
+
+failed_vpbl_avail:
 	kfree(nesfmr);
 
 	failed_fmr_alloc:
@@ -534,18 +544,14 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd,
  */
 static int nes_dealloc_fmr(struct ib_fmr *ibfmr)
 {
+	unsigned long flags;
 	struct nes_mr *nesmr = to_nesmr_from_ibfmr(ibfmr);
 	struct nes_fmr *nesfmr = to_nesfmr(nesmr);
 	struct nes_vnic *nesvnic = to_nesvnic(ibfmr->device);
 	struct nes_device *nesdev = nesvnic->nesdev;
-	struct nes_mr temp_nesmr = *nesmr;
+	struct nes_adapter *nesadapter = nesdev->nesadapter;
 	int i = 0;
 
-	temp_nesmr.ibmw.device = ibfmr->device;
-	temp_nesmr.ibmw.pd = ibfmr->pd;
-	temp_nesmr.ibmw.rkey = ibfmr->rkey;
-	temp_nesmr.ibmw.uobject = NULL;
-
 	/* free the resources */
 	if (nesfmr->leaf_pbl_cnt == 0) {
 		/* single PBL case */
@@ -561,8 +567,24 @@ static int nes_dealloc_fmr(struct ib_fmr *ibfmr)
 		pci_free_consistent(nesdev->pcidev, 8192, nesfmr->root_vpbl.pbl_vbase,
 				nesfmr->root_vpbl.pbl_pbase);
 	}
+	nesmr->ibmw.device = ibfmr->device;
+	nesmr->ibmw.pd = ibfmr->pd;
+	nesmr->ibmw.rkey = ibfmr->rkey;
+	nesmr->ibmw.uobject = NULL;
 
-	return nes_dealloc_mw(&temp_nesmr.ibmw);
+	if (nesfmr->nesmr.pbls_used != 0) {
+		spin_lock_irqsave(&nesadapter->pbl_lock, flags);
+		if (nesfmr->nesmr.pbl_4k) {
+			nesadapter->free_4kpbl += nesfmr->nesmr.pbls_used;
+			WARN_ON(nesadapter->free_4kpbl > nesadapter->max_4kpbl);
+		} else {
+			nesadapter->free_256pbl += nesfmr->nesmr.pbls_used;
+			WARN_ON(nesadapter->free_256pbl > nesadapter->max_256pbl);
+		}
+		spin_unlock_irqrestore(&nesadapter->pbl_lock, flags);
+	}
+
+	return nes_dealloc_mw(&nesmr->ibmw);
 }
 
 
@@ -1595,7 +1617,7 @@ static struct ib_cq *nes_create_cq(struct ib_device *ibdev, int entries,
 		nes_ucontext->mcrqf = req.mcrqf;
 		if (nes_ucontext->mcrqf) {
 			if (nes_ucontext->mcrqf & 0x80000000)
-				nescq->hw_cq.cq_number = nesvnic->nic.qp_id + 12 + (nes_ucontext->mcrqf & 0xf) - 1;
+				nescq->hw_cq.cq_number = nesvnic->nic.qp_id + 28 + 2 * ((nes_ucontext->mcrqf & 0xf) - 1);
 			else if (nes_ucontext->mcrqf & 0x40000000)
 				nescq->hw_cq.cq_number = nes_ucontext->mcrqf & 0xffff;
 			else
@@ -3212,7 +3234,7 @@ static int nes_post_send(struct ib_qp *ibqp, struct ib_send_wr *ib_wr,
 	if (nesqp->ibqp_state > IB_QPS_RTS)
 		return -EINVAL;
 
-		spin_lock_irqsave(&nesqp->lock, flags);
+	spin_lock_irqsave(&nesqp->lock, flags);
 
 	head = nesqp->hwqp.sq_head;
 
@@ -3337,7 +3359,7 @@ static int nes_post_send(struct ib_qp *ibqp, struct ib_send_wr *ib_wr,
 				(counter << 24) | 0x00800000 | nesqp->hwqp.qp_id);
 	}
 
-		spin_unlock_irqrestore(&nesqp->lock, flags);
+	spin_unlock_irqrestore(&nesqp->lock, flags);
 
 	if (err)
 		*bad_wr = ib_wr;
@@ -3368,7 +3390,7 @@ static int nes_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *ib_wr,
 	if (nesqp->ibqp_state > IB_QPS_RTS)
 		return -EINVAL;
 
-		spin_lock_irqsave(&nesqp->lock, flags);
+	spin_lock_irqsave(&nesqp->lock, flags);
 
 	head = nesqp->hwqp.rq_head;
 
@@ -3421,7 +3443,7 @@ static int nes_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *ib_wr,
 		nes_write32(nesdev->regs+NES_WQE_ALLOC, (counter<<24) | nesqp->hwqp.qp_id);
 	}
 
-		spin_unlock_irqrestore(&nesqp->lock, flags);
+	spin_unlock_irqrestore(&nesqp->lock, flags);
 
 	if (err)
 		*bad_wr = ib_wr;
@@ -3453,7 +3475,7 @@ static int nes_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry)
 
 	nes_debug(NES_DBG_CQ, "\n");
 
-		spin_lock_irqsave(&nescq->lock, flags);
+	spin_lock_irqsave(&nescq->lock, flags);
 
 	head = nescq->hw_cq.cq_head;
 	cq_size = nescq->hw_cq.cq_size;
@@ -3562,7 +3584,7 @@ static int nes_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry)
 	nes_debug(NES_DBG_CQ, "Reporting %u completions for CQ%u.\n",
 			cqe_count, nescq->hw_cq.cq_number);
 
-		spin_unlock_irqrestore(&nescq->lock, flags);
+	spin_unlock_irqrestore(&nescq->lock, flags);
 
 	return cqe_count;
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index fddded7900d1..85257f6b9576 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -106,12 +106,13 @@ int ipoib_open(struct net_device *dev)
 
 	ipoib_dbg(priv, "bringing up interface\n");
 
-	napi_enable(&priv->napi);
 	set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
 
 	if (ipoib_pkey_dev_delay_open(dev))
 		return 0;
 
+	napi_enable(&priv->napi);
+
 	if (ipoib_ib_dev_open(dev)) {
 		napi_disable(&priv->napi);
 		return -EINVAL;
@@ -546,6 +547,7 @@ static int path_rec_start(struct net_device *dev,
 	if (path->query_id < 0) {
 		ipoib_warn(priv, "ib_sa_path_rec_get failed: %d\n", path->query_id);
 		path->query = NULL;
+		complete(&path->done);
 		return path->query_id;
 	}
 
@@ -662,7 +664,7 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev,
 			skb_push(skb, sizeof *phdr);
 			__skb_queue_tail(&path->queue, skb);
 
-			if (path_rec_start(dev, path)) {
+			if (!path->query && path_rec_start(dev, path)) {
 				spin_unlock_irqrestore(&priv->lock, flags);
 				path_free(dev, path);
 				return;