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

Pull infiniband update from Roland Dreier:
 "Main batch of InfiniBand/RDMA changes for 3.9:

   - SRP error handling fixes from Bart Van Assche

   - Implementation of memory windows for mlx4 from Shani Michaeli

   - Lots of cxgb4 HW driver fixes from Vipul Pandya

   - Make iSER work for virtual functions, other fixes from Or Gerlitz

   - Fix for bug in qib HW driver from Mike Marciniszyn

   - IPoIB fixes from me, Itai Garbi, Shlomo Pongratz, Yan Burman

   - Various cleanups and warning fixes from Julia Lawall, Paul Bolle,
     Wei Yongjun"

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (41 commits)
  IB/mlx4: Advertise MW support
  IB/mlx4: Support memory window binding
  mlx4: Implement memory windows allocation and deallocation
  mlx4_core: Enable memory windows in {INIT, QUERY}_HCA
  mlx4_core: Disable memory windows for virtual functions
  IPoIB: Free ipoib neigh on path record failure so path rec queries are retried
  IB/srp: Fail I/O requests if the transport is offline
  IB/srp: Avoid endless SCSI error handling loop
  IB/srp: Avoid sending a task management function needlessly
  IB/srp: Track connection state properly
  IB/mlx4: Remove redundant NULL check before kfree
  IB/mlx4: Fix compiler warning about uninitialized 'vlan' variable
  IB/mlx4: Convert is_xxx variables in build_mlx_header() to bool
  IB/iser: Enable iser when FMRs are not supported
  IB/iser: Avoid error prints on EAGAIN registration failures
  IB/iser: Use proper define for the commands per LUN value advertised to SCSI ML
  IB/uverbs: Implement memory windows support in uverbs
  IB/core: Add "type 2" memory windows support
  mlx4_core: Propagate MR deregistration failures to caller
  mlx4_core: Rename MPT-related functions to have mpt_ prefix
  ...

drivers/infiniband/core/uverbs.h

@@ -188,6 +188,8 @@ IB_UVERBS_DECLARE_CMD(alloc_pd);
 IB_UVERBS_DECLARE_CMD(dealloc_pd);
 IB_UVERBS_DECLARE_CMD(reg_mr);
 IB_UVERBS_DECLARE_CMD(dereg_mr);
+IB_UVERBS_DECLARE_CMD(alloc_mw);
+IB_UVERBS_DECLARE_CMD(dealloc_mw);
 IB_UVERBS_DECLARE_CMD(create_comp_channel);
 IB_UVERBS_DECLARE_CMD(create_cq);
 IB_UVERBS_DECLARE_CMD(resize_cq);

drivers/infiniband/core/uverbs_cmd.c

@@ -48,6 +48,7 @@ struct uverbs_lock_class {
 
 static struct uverbs_lock_class pd_lock_class	= { .name = "PD-uobj" };
 static struct uverbs_lock_class mr_lock_class	= { .name = "MR-uobj" };
+static struct uverbs_lock_class mw_lock_class	= { .name = "MW-uobj" };
 static struct uverbs_lock_class cq_lock_class	= { .name = "CQ-uobj" };
 static struct uverbs_lock_class qp_lock_class	= { .name = "QP-uobj" };
 static struct uverbs_lock_class ah_lock_class	= { .name = "AH-uobj" };
@@ -1049,6 +1050,126 @@ ssize_t ib_uverbs_dereg_mr(struct ib_uverbs_file *file,
 	return in_len;
 }
 
+ssize_t ib_uverbs_alloc_mw(struct ib_uverbs_file *file,
+			 const char __user *buf, int in_len,
+			 int out_len)
+{
+	struct ib_uverbs_alloc_mw      cmd;
+	struct ib_uverbs_alloc_mw_resp resp;
+	struct ib_uobject             *uobj;
+	struct ib_pd                  *pd;
+	struct ib_mw                  *mw;
+	int                            ret;
+
+	if (out_len < sizeof(resp))
+		return -ENOSPC;
+
+	if (copy_from_user(&cmd, buf, sizeof(cmd)))
+		return -EFAULT;
+
+	uobj = kmalloc(sizeof(*uobj), GFP_KERNEL);
+	if (!uobj)
+		return -ENOMEM;
+
+	init_uobj(uobj, 0, file->ucontext, &mw_lock_class);
+	down_write(&uobj->mutex);
+
+	pd = idr_read_pd(cmd.pd_handle, file->ucontext);
+	if (!pd) {
+		ret = -EINVAL;
+		goto err_free;
+	}
+
+	mw = pd->device->alloc_mw(pd, cmd.mw_type);
+	if (IS_ERR(mw)) {
+		ret = PTR_ERR(mw);
+		goto err_put;
+	}
+
+	mw->device  = pd->device;
+	mw->pd      = pd;
+	mw->uobject = uobj;
+	atomic_inc(&pd->usecnt);
+
+	uobj->object = mw;
+	ret = idr_add_uobj(&ib_uverbs_mw_idr, uobj);
+	if (ret)
+		goto err_unalloc;
+
+	memset(&resp, 0, sizeof(resp));
+	resp.rkey      = mw->rkey;
+	resp.mw_handle = uobj->id;
+
+	if (copy_to_user((void __user *)(unsigned long)cmd.response,
+			 &resp, sizeof(resp))) {
+		ret = -EFAULT;
+		goto err_copy;
+	}
+
+	put_pd_read(pd);
+
+	mutex_lock(&file->mutex);
+	list_add_tail(&uobj->list, &file->ucontext->mw_list);
+	mutex_unlock(&file->mutex);
+
+	uobj->live = 1;
+
+	up_write(&uobj->mutex);
+
+	return in_len;
+
+err_copy:
+	idr_remove_uobj(&ib_uverbs_mw_idr, uobj);
+
+err_unalloc:
+	ib_dealloc_mw(mw);
+
+err_put:
+	put_pd_read(pd);
+
+err_free:
+	put_uobj_write(uobj);
+	return ret;
+}
+
+ssize_t ib_uverbs_dealloc_mw(struct ib_uverbs_file *file,
+			   const char __user *buf, int in_len,
+			   int out_len)
+{
+	struct ib_uverbs_dealloc_mw cmd;
+	struct ib_mw               *mw;
+	struct ib_uobject	   *uobj;
+	int                         ret = -EINVAL;
+
+	if (copy_from_user(&cmd, buf, sizeof(cmd)))
+		return -EFAULT;
+
+	uobj = idr_write_uobj(&ib_uverbs_mw_idr, cmd.mw_handle, file->ucontext);
+	if (!uobj)
+		return -EINVAL;
+
+	mw = uobj->object;
+
+	ret = ib_dealloc_mw(mw);
+	if (!ret)
+		uobj->live = 0;
+
+	put_uobj_write(uobj);
+
+	if (ret)
+		return ret;
+
+	idr_remove_uobj(&ib_uverbs_mw_idr, uobj);
+
+	mutex_lock(&file->mutex);
+	list_del(&uobj->list);
+	mutex_unlock(&file->mutex);
+
+	put_uobj(uobj);
+
+	return in_len;
+}
+
 ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file,
 				      const char __user *buf, int in_len,
 				      int out_len)

drivers/infiniband/core/uverbs_main.c

@@ -87,6 +87,8 @@ static ssize_t (*uverbs_cmd_table[])(struct ib_uverbs_file *file,
 	[IB_USER_VERBS_CMD_DEALLOC_PD]		= ib_uverbs_dealloc_pd,
 	[IB_USER_VERBS_CMD_REG_MR]		= ib_uverbs_reg_mr,
 	[IB_USER_VERBS_CMD_DEREG_MR]		= ib_uverbs_dereg_mr,
+	[IB_USER_VERBS_CMD_ALLOC_MW]		= ib_uverbs_alloc_mw,
+	[IB_USER_VERBS_CMD_DEALLOC_MW]		= ib_uverbs_dealloc_mw,
 	[IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL] = ib_uverbs_create_comp_channel,
 	[IB_USER_VERBS_CMD_CREATE_CQ]		= ib_uverbs_create_cq,
 	[IB_USER_VERBS_CMD_RESIZE_CQ]		= ib_uverbs_resize_cq,
@@ -201,6 +203,15 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file,
 		kfree(uobj);
 	}
 
+	/* Remove MWs before QPs, in order to support type 2A MWs. */
+	list_for_each_entry_safe(uobj, tmp, &context->mw_list, list) {
+		struct ib_mw *mw = uobj->object;
+
+		idr_remove_uobj(&ib_uverbs_mw_idr, uobj);
+		ib_dealloc_mw(mw);
+		kfree(uobj);
+	}
+
 	list_for_each_entry_safe(uobj, tmp, &context->qp_list, list) {
 		struct ib_qp *qp = uobj->object;
 		struct ib_uqp_object *uqp =
@@ -240,8 +251,6 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file,
 		kfree(uevent);
 	}
 
-	/* XXX Free MWs */
-
 	list_for_each_entry_safe(uobj, tmp, &context->mr_list, list) {
 		struct ib_mr *mr = uobj->object;
 

drivers/infiniband/core/verbs.c

@@ -1099,18 +1099,19 @@ EXPORT_SYMBOL(ib_free_fast_reg_page_list);
 
 /* Memory windows */
 
-struct ib_mw *ib_alloc_mw(struct ib_pd *pd)
+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);
+	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);
 	}
 

drivers/infiniband/hw/amso1100/c2.c

@@ -1238,15 +1238,4 @@ static struct pci_driver c2_pci_driver = {
 	.remove = c2_remove,
 };
 
-static int __init c2_init_module(void)
-{
-	return pci_register_driver(&c2_pci_driver);
-}
-
-static void __exit c2_exit_module(void)
-{
-	pci_unregister_driver(&c2_pci_driver);
-}
-
-module_init(c2_init_module);
-module_exit(c2_exit_module);
+module_pci_driver(c2_pci_driver);

drivers/infiniband/hw/cxgb3/iwch_provider.c

@@ -738,7 +738,7 @@ static struct ib_mr *iwch_get_dma_mr(struct ib_pd *pd, int acc)
 	return ibmr;
 }
 
-static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd)
+static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd, enum ib_mw_type type)
 {
 	struct iwch_dev *rhp;
 	struct iwch_pd *php;
@@ -747,6 +747,9 @@ static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd)
 	u32 stag = 0;
 	int ret;
 
+	if (type != IB_MW_TYPE_1)
+		return ERR_PTR(-EINVAL);
+
 	php = to_iwch_pd(pd);
 	rhp = php->rhp;
 	mhp = kzalloc(sizeof(*mhp), GFP_KERNEL);

drivers/infiniband/hw/cxgb3/iwch_qp.c

@@ -567,18 +567,19 @@ int iwch_bind_mw(struct ib_qp *qp,
 	if (mw_bind->send_flags & IB_SEND_SIGNALED)
 		t3_wr_flags = T3_COMPLETION_FLAG;
 
-	sgl.addr = mw_bind->addr;
-	sgl.lkey = mw_bind->mr->lkey;
-	sgl.length = mw_bind->length;
+	sgl.addr = mw_bind->bind_info.addr;
+	sgl.lkey = mw_bind->bind_info.mr->lkey;
+	sgl.length = mw_bind->bind_info.length;
 	wqe->bind.reserved = 0;
 	wqe->bind.type = TPT_VATO;
 
 	/* TBD: check perms */
-	wqe->bind.perms = iwch_ib_to_tpt_bind_access(mw_bind->mw_access_flags);
-	wqe->bind.mr_stag = cpu_to_be32(mw_bind->mr->lkey);
+	wqe->bind.perms = iwch_ib_to_tpt_bind_access(
+		mw_bind->bind_info.mw_access_flags);
+	wqe->bind.mr_stag = cpu_to_be32(mw_bind->bind_info.mr->lkey);
 	wqe->bind.mw_stag = cpu_to_be32(mw->rkey);
-	wqe->bind.mw_len = cpu_to_be32(mw_bind->length);
-	wqe->bind.mw_va = cpu_to_be64(mw_bind->addr);
+	wqe->bind.mw_len = cpu_to_be32(mw_bind->bind_info.length);
+	wqe->bind.mw_va = cpu_to_be64(mw_bind->bind_info.addr);
 	err = iwch_sgl2pbl_map(rhp, &sgl, 1, &pbl_addr, &page_size);
 	if (err) {
 		spin_unlock_irqrestore(&qhp->lock, flag);

drivers/infiniband/hw/cxgb4/cm.c

@@ -143,14 +143,28 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status);
 static LIST_HEAD(timeout_list);
 static spinlock_t timeout_lock;
 
+static void deref_qp(struct c4iw_ep *ep)
+{
+	c4iw_qp_rem_ref(&ep->com.qp->ibqp);
+	clear_bit(QP_REFERENCED, &ep->com.flags);
+}
+
+static void ref_qp(struct c4iw_ep *ep)
+{
+	set_bit(QP_REFERENCED, &ep->com.flags);
+	c4iw_qp_add_ref(&ep->com.qp->ibqp);
+}
+
 static void start_ep_timer(struct c4iw_ep *ep)
 {
 	PDBG("%s ep %p\n", __func__, ep);
 	if (timer_pending(&ep->timer)) {
-		PDBG("%s stopped / restarted timer ep %p\n", __func__, ep);
-		del_timer_sync(&ep->timer);
-	} else
-		c4iw_get_ep(&ep->com);
+		pr_err("%s timer already started! ep %p\n",
+		       __func__, ep);
+		return;
+	}
+	clear_bit(TIMEOUT, &ep->com.flags);
+	c4iw_get_ep(&ep->com);
 	ep->timer.expires = jiffies + ep_timeout_secs * HZ;
 	ep->timer.data = (unsigned long)ep;
 	ep->timer.function = ep_timeout;
@@ -159,14 +173,10 @@ static void start_ep_timer(struct c4iw_ep *ep)
 
 static void stop_ep_timer(struct c4iw_ep *ep)
 {
-	PDBG("%s ep %p\n", __func__, ep);
-	if (!timer_pending(&ep->timer)) {
-		WARN(1, "%s timer stopped when its not running! "
-		       "ep %p state %u\n", __func__, ep, ep->com.state);
-		return;
-	}
+	PDBG("%s ep %p stopping\n", __func__, ep);
 	del_timer_sync(&ep->timer);
-	c4iw_put_ep(&ep->com);
+	if (!test_and_set_bit(TIMEOUT, &ep->com.flags))
+		c4iw_put_ep(&ep->com);
 }
 
 static int c4iw_l2t_send(struct c4iw_rdev *rdev, struct sk_buff *skb,
@@ -271,11 +281,13 @@ void _c4iw_free_ep(struct kref *kref)
 
 	ep = container_of(kref, struct c4iw_ep, com.kref);
 	PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]);
+	if (test_bit(QP_REFERENCED, &ep->com.flags))
+		deref_qp(ep);
 	if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) {
+		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
 		dst_release(ep->dst);
 		cxgb4_l2t_release(ep->l2t);
-		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 	}
 	kfree(ep);
 }
@@ -687,7 +699,7 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen)
 	memset(mpa, 0, sizeof(*mpa));
 	memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key));
 	mpa->flags = MPA_REJECT;
-	mpa->revision = mpa_rev;
+	mpa->revision = ep->mpa_attr.version;
 	mpa->private_data_size = htons(plen);
 
 	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
@@ -863,7 +875,6 @@ static void close_complete_upcall(struct c4iw_ep *ep)
 		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 		set_bit(CLOSE_UPCALL, &ep->com.history);
 	}
 }
@@ -906,7 +917,6 @@ static void peer_abort_upcall(struct c4iw_ep *ep)
 		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 		set_bit(ABORT_UPCALL, &ep->com.history);
 	}
 }
@@ -946,7 +956,6 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status)
 	if (status < 0) {
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 	}
 }
 
@@ -1291,11 +1300,13 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	if (mpa->revision > mpa_rev) {
 		printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d,"
 		       " Received = %d\n", __func__, mpa_rev, mpa->revision);
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
 
 	if (memcmp(mpa->key, MPA_KEY_REQ, sizeof(mpa->key))) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -1306,6 +1317,7 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	 * Fail if there's too much private data.
 	 */
 	if (plen > MPA_MAX_PRIVATE_DATA) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -1314,6 +1326,7 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	 * If plen does not account for pkt size
 	 */
 	if (ep->mpa_pkt_len > (sizeof(*mpa) + plen)) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -1391,30 +1404,33 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
 	skb_pull(skb, sizeof(*hdr));
 	skb_trim(skb, dlen);
 
-	ep->rcv_seq += dlen;
-	BUG_ON(ep->rcv_seq != (ntohl(hdr->seq) + dlen));
-
 	/* update RX credits */
 	update_rx_credits(ep, dlen);
 
 	switch (state_read(&ep->com)) {
 	case MPA_REQ_SENT:
+		ep->rcv_seq += dlen;
 		process_mpa_reply(ep, skb);
 		break;
 	case MPA_REQ_WAIT:
+		ep->rcv_seq += dlen;
 		process_mpa_request(ep, skb);
 		break;
-	case MPA_REP_SENT:
+	case FPDU_MODE: {
+		struct c4iw_qp_attributes attrs;
+		BUG_ON(!ep->com.qp);
+		if (status)
+			pr_err("%s Unexpected streaming data." \
+			       " qpid %u ep %p state %d tid %u status %d\n",
+			       __func__, ep->com.qp->wq.sq.qid, ep,
+			       state_read(&ep->com), ep->hwtid, status);
+		attrs.next_state = C4IW_QP_STATE_ERROR;
+		c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
+			       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
+		c4iw_ep_disconnect(ep, 1, GFP_KERNEL);
 		break;
+	}
 	default:
-		pr_err("%s Unexpected streaming data." \
-		       " ep %p state %d tid %u status %d\n",
-		       __func__, ep, state_read(&ep->com), ep->hwtid, status);
-
-		/*
-		 * The ep will timeout and inform the ULP of the failure.
-		 * See ep_timeout().
-		 */
 		break;
 	}
 	return 0;
@@ -1437,6 +1453,7 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
 	mutex_lock(&ep->com.mutex);
 	switch (ep->com.state) {
 	case ABORTING:
+		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
 		__state_set(&ep->com, DEAD);
 		release = 1;
 		break;
@@ -1475,11 +1492,11 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 			V_FW_OFLD_CONNECTION_WR_ASTID(atid));
 	req->tcb.cplrxdataack_cplpassacceptrpl =
 			htons(F_FW_OFLD_CONNECTION_WR_CPLRXDATAACK);
-	req->tcb.tx_max = jiffies;
+	req->tcb.tx_max = (__force __be32) jiffies;
 	req->tcb.rcv_adv = htons(1);
 	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
 	wscale = compute_wscale(rcv_win);
-	req->tcb.opt0 = TCAM_BYPASS(1) |
+	req->tcb.opt0 = (__force __be64) (TCAM_BYPASS(1) |
 		(nocong ? NO_CONG(1) : 0) |
 		KEEP_ALIVE(1) |
 		DELACK(1) |
@@ -1490,20 +1507,20 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 		SMAC_SEL(ep->smac_idx) |
 		DSCP(ep->tos) |
 		ULP_MODE(ULP_MODE_TCPDDP) |
-		RCV_BUFSIZ(rcv_win >> 10);
-	req->tcb.opt2 = PACE(1) |
+		RCV_BUFSIZ(rcv_win >> 10));
+	req->tcb.opt2 = (__force __be32) (PACE(1) |
 		TX_QUEUE(ep->com.dev->rdev.lldi.tx_modq[ep->tx_chan]) |
 		RX_CHANNEL(0) |
 		CCTRL_ECN(enable_ecn) |
-		RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
+		RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid));
 	if (enable_tcp_timestamps)
-		req->tcb.opt2 |= TSTAMPS_EN(1);
+		req->tcb.opt2 |= (__force __be32) TSTAMPS_EN(1);
 	if (enable_tcp_sack)
-		req->tcb.opt2 |= SACK_EN(1);
+		req->tcb.opt2 |= (__force __be32) SACK_EN(1);
 	if (wscale && enable_tcp_window_scaling)
-		req->tcb.opt2 |= WND_SCALE_EN(1);
-	req->tcb.opt0 = cpu_to_be64(req->tcb.opt0);
-	req->tcb.opt2 = cpu_to_be32(req->tcb.opt2);
+		req->tcb.opt2 |= (__force __be32) WND_SCALE_EN(1);
+	req->tcb.opt0 = cpu_to_be64((__force u64) req->tcb.opt0);
+	req->tcb.opt2 = cpu_to_be32((__force u32) req->tcb.opt2);
 	set_wr_txq(skb, CPL_PRIORITY_CONTROL, ep->ctrlq_idx);
 	set_bit(ACT_OFLD_CONN, &ep->com.history);
 	c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
@@ -1993,6 +2010,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	init_timer(&child_ep->timer);
 	cxgb4_insert_tid(t, child_ep, hwtid);
+	insert_handle(dev, &dev->hwtid_idr, child_ep, child_ep->hwtid);
 	accept_cr(child_ep, peer_ip, skb, req);
 	set_bit(PASS_ACCEPT_REQ, &child_ep->com.history);
 	goto out;
@@ -2018,7 +2036,6 @@ static int pass_establish(struct c4iw_dev *dev, struct sk_buff *skb)
 	     ntohs(req->tcp_opt));
 
 	set_emss(ep, ntohs(req->tcp_opt));
-	insert_handle(dev, &dev->hwtid_idr, ep, ep->hwtid);
 
 	dst_confirm(ep->dst);
 	state_set(&ep->com, MPA_REQ_WAIT);
@@ -2163,7 +2180,7 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb)
 		break;
 	case MPA_REQ_SENT:
 		stop_ep_timer(ep);
-		if (mpa_rev == 2 && ep->tried_with_mpa_v1)
+		if (mpa_rev == 1 || (mpa_rev == 2 && ep->tried_with_mpa_v1))
 			connect_reply_upcall(ep, -ECONNRESET);
 		else {
 			/*
@@ -2235,9 +2252,8 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb)
 out:
 	if (release)
 		release_ep_resources(ep);
-
-	/* retry with mpa-v1 */
-	if (ep && ep->retry_with_mpa_v1) {
+	else if (ep->retry_with_mpa_v1) {
+		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
 		dst_release(ep->dst);
 		cxgb4_l2t_release(ep->l2t);
@@ -2430,6 +2446,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	cm_id->add_ref(cm_id);
 	ep->com.cm_id = cm_id;
 	ep->com.qp = qp;
+	ref_qp(ep);
 
 	/* bind QP to EP and move to RTS */
 	attrs.mpa_attr = ep->mpa_attr;
@@ -2460,7 +2477,6 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	return 0;
 err1:
 	ep->com.cm_id = NULL;
-	ep->com.qp = NULL;
 	cm_id->rem_ref(cm_id);
 err:
 	c4iw_put_ep(&ep->com);
@@ -2501,6 +2517,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	ep->com.cm_id = cm_id;
 	ep->com.qp = get_qhp(dev, conn_param->qpn);
 	BUG_ON(!ep->com.qp);
+	ref_qp(ep);
 	PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn,
 	     ep->com.qp, cm_id);
 
@@ -2756,7 +2773,8 @@ static void active_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	struct c4iw_ep *ep;
 	int atid = be32_to_cpu(req->tid);
 
-	ep = (struct c4iw_ep *)lookup_atid(dev->rdev.lldi.tids, req->tid);
+	ep = (struct c4iw_ep *)lookup_atid(dev->rdev.lldi.tids,
+					   (__force u32) req->tid);
 	if (!ep)
 		return;
 
@@ -2800,7 +2818,7 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	struct cpl_pass_accept_req *cpl;
 	int ret;
 
-	rpl_skb = (struct sk_buff *)cpu_to_be64(req->cookie);
+	rpl_skb = (struct sk_buff *)(unsigned long)req->cookie;
 	BUG_ON(!rpl_skb);
 	if (req->retval) {
 		PDBG("%s passive open failure %d\n", __func__, req->retval);
@@ -2811,7 +2829,8 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	} else {
 		cpl = (struct cpl_pass_accept_req *)cplhdr(rpl_skb);
 		OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_PASS_ACCEPT_REQ,
-						      htonl(req->tid)));
+					(__force u32) htonl(
+					(__force u32) req->tid)));
 		ret = pass_accept_req(dev, rpl_skb);
 		if (!ret)
 			kfree_skb(rpl_skb);
@@ -2857,10 +2876,10 @@ static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos)
 	struct tcp_options_received tmp_opt;
 
 	/* Store values from cpl_rx_pkt in temporary location. */
-	vlantag = cpl->vlan;
-	len = cpl->len;
-	l2info  = cpl->l2info;
-	hdr_len = cpl->hdr_len;
+	vlantag = (__force u16) cpl->vlan;
+	len = (__force u16) cpl->len;
+	l2info  = (__force u32) cpl->l2info;
+	hdr_len = (__force u16) cpl->hdr_len;
 	intf = cpl->iff;
 
 	__skb_pull(skb, sizeof(*req) + sizeof(struct rss_header));
@@ -2871,19 +2890,24 @@ static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos)
 	 */
 	memset(&tmp_opt, 0, sizeof(tmp_opt));
 	tcp_clear_options(&tmp_opt);
-	tcp_parse_options(skb, &tmp_opt, 0, 0, NULL);
+	tcp_parse_options(skb, &tmp_opt, NULL, 0, NULL);
 
 	req = (struct cpl_pass_accept_req *)__skb_push(skb, sizeof(*req));
 	memset(req, 0, sizeof(*req));
 	req->l2info = cpu_to_be16(V_SYN_INTF(intf) |
-			 V_SYN_MAC_IDX(G_RX_MACIDX(htonl(l2info))) |
+			 V_SYN_MAC_IDX(G_RX_MACIDX(
+			 (__force int) htonl(l2info))) |
 			 F_SYN_XACT_MATCH);
-	req->hdr_len = cpu_to_be32(V_SYN_RX_CHAN(G_RX_CHAN(htonl(l2info))) |
-				V_TCP_HDR_LEN(G_RX_TCPHDR_LEN(htons(hdr_len))) |
-				V_IP_HDR_LEN(G_RX_IPHDR_LEN(htons(hdr_len))) |
-				V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(htonl(l2info))));
-	req->vlan = vlantag;
-	req->len = len;
+	req->hdr_len = cpu_to_be32(V_SYN_RX_CHAN(G_RX_CHAN(
+					(__force int) htonl(l2info))) |
+				   V_TCP_HDR_LEN(G_RX_TCPHDR_LEN(
+					(__force int) htons(hdr_len))) |
+				   V_IP_HDR_LEN(G_RX_IPHDR_LEN(
+					(__force int) htons(hdr_len))) |
+				   V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(
+					(__force int) htonl(l2info))));
+	req->vlan = (__force __be16) vlantag;
+	req->len = (__force __be16) len;
 	req->tos_stid = cpu_to_be32(PASS_OPEN_TID(stid) |
 				    PASS_OPEN_TOS(tos));
 	req->tcpopt.mss = htons(tmp_opt.mss_clamp);
@@ -2912,7 +2936,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	req->op_compl = htonl(V_WR_OP(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL(1));
 	req->len16_pkd = htonl(FW_WR_LEN16(DIV_ROUND_UP(sizeof(*req), 16)));
 	req->le.version_cpl = htonl(F_FW_OFLD_CONNECTION_WR_CPL);
-	req->le.filter = filter;
+	req->le.filter = (__force __be32) filter;
 	req->le.lport = lport;
 	req->le.pport = rport;
 	req->le.u.ipv4.lip = laddr;
@@ -2938,7 +2962,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	 * TP will ignore any value > 0 for MSS index.
 	 */
 	req->tcb.opt0 = cpu_to_be64(V_MSS_IDX(0xF));
-	req->cookie = cpu_to_be64((u64)skb);
+	req->cookie = (unsigned long)skb;
 
 	set_wr_txq(req_skb, CPL_PRIORITY_CONTROL, port_id);
 	cxgb4_ofld_send(dev->rdev.lldi.ports[0], req_skb);
@@ -2988,7 +3012,8 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb)
 	/*
 	 * Calculate the server tid from filter hit index from cpl_rx_pkt.
 	 */
-	stid = cpu_to_be32(rss->hash_val) - dev->rdev.lldi.tids->sftid_base
+	stid = (__force int) cpu_to_be32((__force u32) rss->hash_val)
+					  - dev->rdev.lldi.tids->sftid_base
 					  + dev->rdev.lldi.tids->nstids;
 
 	lep = (struct c4iw_ep *)lookup_stid(dev->rdev.lldi.tids, stid);
@@ -3049,10 +3074,10 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	step = dev->rdev.lldi.nrxq / dev->rdev.lldi.nchan;
 	rss_qid = dev->rdev.lldi.rxq_ids[pi->port_id * step];
-	window = htons(tcph->window);
+	window = (__force u16) htons((__force u16)tcph->window);
 
 	/* Calcuate filter portion for LE region. */
-	filter = cpu_to_be32(select_ntuple(dev, dst, e));
+	filter = (__force unsigned int) cpu_to_be32(select_ntuple(dev, dst, e));
 
 	/*
 	 * Synthesize the cpl_pass_accept_req. We have everything except the
@@ -3175,11 +3200,16 @@ static DECLARE_WORK(skb_work, process_work);
 static void ep_timeout(unsigned long arg)
 {
 	struct c4iw_ep *ep = (struct c4iw_ep *)arg;
+	int kickit = 0;
 
 	spin_lock(&timeout_lock);
-	list_add_tail(&ep->entry, &timeout_list);
+	if (!test_and_set_bit(TIMEOUT, &ep->com.flags)) {
+		list_add_tail(&ep->entry, &timeout_list);
+		kickit = 1;
+	}
 	spin_unlock(&timeout_lock);
-	queue_work(workq, &skb_work);
+	if (kickit)
+		queue_work(workq, &skb_work);
 }
 
 /*
@@ -3268,8 +3298,14 @@ static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	/*
 	 * Wake up any threads in rdma_init() or rdma_fini().
+	 * However, if we are on MPAv2 and want to retry with MPAv1
+	 * then, don't wake up yet.
 	 */
-	c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
+	if (mpa_rev == 2 && !ep->tried_with_mpa_v1) {
+		if (ep->com.state != MPA_REQ_SENT)
+			c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
+	} else
+		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
 	sched(dev, skb);
 	return 0;
 }

drivers/infiniband/hw/cxgb4/device.c

@@ -533,7 +533,7 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 	PDBG("udb len 0x%x udb base %p db_reg %p gts_reg %p qpshift %lu "
 	     "qpmask 0x%x cqshift %lu cqmask 0x%x\n",
 	     (unsigned)pci_resource_len(rdev->lldi.pdev, 2),
-	     (void *)pci_resource_start(rdev->lldi.pdev, 2),
+	     (void *)(unsigned long)pci_resource_start(rdev->lldi.pdev, 2),
 	     rdev->lldi.db_reg,
 	     rdev->lldi.gts_reg,
 	     rdev->qpshift, rdev->qpmask,
@@ -797,7 +797,8 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp,
 		       "RSS %#llx, FL %#llx, len %u\n",
 		       pci_name(ctx->lldi.pdev), gl->va,
 		       (unsigned long long)be64_to_cpu(*rsp),
-		       (unsigned long long)be64_to_cpu(*(u64 *)gl->va),
+		       (unsigned long long)be64_to_cpu(
+		       *(__force __be64 *)gl->va),
 		       gl->tot_len);
 
 		return 0;

drivers/infiniband/hw/cxgb4/ev.c

@@ -46,9 +46,11 @@ static void post_qp_event(struct c4iw_dev *dev, struct c4iw_cq *chp,
 
 	if ((qhp->attr.state == C4IW_QP_STATE_ERROR) ||
 	    (qhp->attr.state == C4IW_QP_STATE_TERMINATE)) {
-		PDBG("%s AE received after RTS - "
-		     "qp state %d qpid 0x%x status 0x%x\n", __func__,
-		     qhp->attr.state, qhp->wq.sq.qid, CQE_STATUS(err_cqe));
+		pr_err("%s AE after RTS - qpid 0x%x opcode %d status 0x%x "\
+		       "type %d wrid.hi 0x%x wrid.lo 0x%x\n",
+		       __func__, CQE_QPID(err_cqe), CQE_OPCODE(err_cqe),
+		       CQE_STATUS(err_cqe), CQE_TYPE(err_cqe),
+		       CQE_WRID_HI(err_cqe), CQE_WRID_LOW(err_cqe));
 		return;
 	}
 

drivers/infiniband/hw/cxgb4/iw_cxgb4.h

@@ -716,6 +716,8 @@ enum c4iw_ep_flags {
 	ABORT_REQ_IN_PROGRESS	= 1,
 	RELEASE_RESOURCES	= 2,
 	CLOSE_SENT		= 3,
+	TIMEOUT                 = 4,
+	QP_REFERENCED           = 5,
 };
 
 enum c4iw_ep_history {
@@ -866,7 +868,7 @@ struct ib_fast_reg_page_list *c4iw_alloc_fastreg_pbl(
 					int page_list_len);
 struct ib_mr *c4iw_alloc_fast_reg_mr(struct ib_pd *pd, int pbl_depth);
 int c4iw_dealloc_mw(struct ib_mw *mw);
-struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd);
+struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd, enum ib_mw_type type);
 struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start,
 					   u64 length, u64 virt, int acc,
 					   struct ib_udata *udata);

drivers/infiniband/hw/cxgb4/mem.c

@@ -650,7 +650,7 @@ struct ib_mr *c4iw_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	return ERR_PTR(err);
 }
 
-struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd)
+struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd, enum ib_mw_type type)
 {
 	struct c4iw_dev *rhp;
 	struct c4iw_pd *php;
@@ -659,6 +659,9 @@ struct ib_mw *c4iw_alloc_mw(struct ib_pd *pd)
 	u32 stag = 0;
 	int ret;
 
+	if (type != IB_MW_TYPE_1)
+		return ERR_PTR(-EINVAL);
+
 	php = to_c4iw_pd(pd);
 	rhp = php->rhp;
 	mhp = kzalloc(sizeof(*mhp), GFP_KERNEL);

drivers/infiniband/hw/cxgb4/qp.c

@@ -1383,6 +1383,7 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp,
 	qhp->ep = NULL;
 	set_state(qhp, C4IW_QP_STATE_ERROR);
 	free = 1;
+	abort = 1;
 	wake_up(&qhp->wait);
 	BUG_ON(!ep);
 	flush_qp(qhp);

drivers/infiniband/hw/ehca/ehca_iverbs.h

@@ -95,7 +95,7 @@ int ehca_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr);
 
 int ehca_dereg_mr(struct ib_mr *mr);
 
-struct ib_mw *ehca_alloc_mw(struct ib_pd *pd);
+struct ib_mw *ehca_alloc_mw(struct ib_pd *pd, enum ib_mw_type type);
 
 int ehca_bind_mw(struct ib_qp *qp, struct ib_mw *mw,
 		 struct ib_mw_bind *mw_bind);

drivers/infiniband/hw/ehca/ehca_mrmw.c

@@ -688,7 +688,7 @@ int ehca_dereg_mr(struct ib_mr *mr)
 
 /*----------------------------------------------------------------------*/
 
-struct ib_mw *ehca_alloc_mw(struct ib_pd *pd)
+struct ib_mw *ehca_alloc_mw(struct ib_pd *pd, enum ib_mw_type type)
 {
 	struct ib_mw *ib_mw;
 	u64 h_ret;
@@ -698,6 +698,9 @@ struct ib_mw *ehca_alloc_mw(struct ib_pd *pd)
 		container_of(pd->device, struct ehca_shca, ib_device);
 	struct ehca_mw_hipzout_parms hipzout;
 
+	if (type != IB_MW_TYPE_1)
+		return ERR_PTR(-EINVAL);
+
 	e_mw = ehca_mw_new();
 	if (!e_mw) {
 		ib_mw = ERR_PTR(-ENOMEM);

drivers/infiniband/hw/mlx4/mad.c

@@ -1999,16 +1999,17 @@ int mlx4_ib_init_sriov(struct mlx4_ib_dev *dev)
 			goto demux_err;
 		err = mlx4_ib_alloc_demux_ctx(dev, &dev->sriov.demux[i], i + 1);
 		if (err)
-			goto demux_err;
+			goto free_pv;
 	}
 	mlx4_ib_master_tunnels(dev, 1);
 	return 0;
 
+free_pv:
+	free_pv_object(dev, mlx4_master_func_num(dev->dev), i + 1);
 demux_err:
-	while (i > 0) {
+	while (--i >= 0) {
 		free_pv_object(dev, mlx4_master_func_num(dev->dev), i + 1);
 		mlx4_ib_free_demux_ctx(&dev->sriov.demux[i]);
-		--i;
 	}
 	mlx4_ib_device_unregister_sysfs(dev);
 

drivers/infiniband/hw/mlx4/main.c

@@ -137,6 +137,14 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 		props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS;
 	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC)
 		props->device_cap_flags |= IB_DEVICE_XRC;
+	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW)
+		props->device_cap_flags |= IB_DEVICE_MEM_WINDOW;
+	if (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN) {
+		if (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_WIN_TYPE_2B)
+			props->device_cap_flags |= IB_DEVICE_MEM_WINDOW_TYPE_2B;
+		else
+			props->device_cap_flags |= IB_DEVICE_MEM_WINDOW_TYPE_2A;
+	}
 
 	props->vendor_id	   = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
 		0xffffff;
@@ -1434,6 +1442,17 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 		ibdev->ib_dev.dealloc_fmr	= mlx4_ib_fmr_dealloc;
 	}
 
+	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
+	    dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN) {
+		ibdev->ib_dev.alloc_mw = mlx4_ib_alloc_mw;
+		ibdev->ib_dev.bind_mw = mlx4_ib_bind_mw;
+		ibdev->ib_dev.dealloc_mw = mlx4_ib_dealloc_mw;
+
+		ibdev->ib_dev.uverbs_cmd_mask |=
+			(1ull << IB_USER_VERBS_CMD_ALLOC_MW) |
+			(1ull << IB_USER_VERBS_CMD_DEALLOC_MW);
+	}
+
 	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC) {
 		ibdev->ib_dev.alloc_xrcd = mlx4_ib_alloc_xrcd;
 		ibdev->ib_dev.dealloc_xrcd = mlx4_ib_dealloc_xrcd;
@@ -1601,8 +1620,7 @@ static void do_slave_init(struct mlx4_ib_dev *ibdev, int slave, int do_init)
 		spin_unlock_irqrestore(&ibdev->sriov.going_down_lock, flags);
 	}
 out:
-	if (dm)
-		kfree(dm);
+	kfree(dm);
 	return;
 }
 

drivers/infiniband/hw/mlx4/mlx4_ib.h

@@ -116,6 +116,11 @@ struct mlx4_ib_mr {
 	struct ib_umem	       *umem;
 };
 
+struct mlx4_ib_mw {
+	struct ib_mw		ibmw;
+	struct mlx4_mw		mmw;
+};
+
 struct mlx4_ib_fast_reg_page_list {
 	struct ib_fast_reg_page_list	ibfrpl;
 	__be64			       *mapped_page_list;
@@ -533,6 +538,11 @@ static inline struct mlx4_ib_mr *to_mmr(struct ib_mr *ibmr)
 	return container_of(ibmr, struct mlx4_ib_mr, ibmr);
 }
 
+static inline struct mlx4_ib_mw *to_mmw(struct ib_mw *ibmw)
+{
+	return container_of(ibmw, struct mlx4_ib_mw, ibmw);
+}
+
 static inline struct mlx4_ib_fast_reg_page_list *to_mfrpl(struct ib_fast_reg_page_list *ibfrpl)
 {
 	return container_of(ibfrpl, struct mlx4_ib_fast_reg_page_list, ibfrpl);
@@ -581,6 +591,10 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 				  u64 virt_addr, int access_flags,
 				  struct ib_udata *udata);
 int mlx4_ib_dereg_mr(struct ib_mr *mr);
+struct ib_mw *mlx4_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type);
+int mlx4_ib_bind_mw(struct ib_qp *qp, struct ib_mw *mw,
+		    struct ib_mw_bind *mw_bind);
+int mlx4_ib_dealloc_mw(struct ib_mw *mw);
 struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
 					int max_page_list_len);
 struct ib_fast_reg_page_list *mlx4_ib_alloc_fast_reg_page_list(struct ib_device *ibdev,
@@ -652,12 +666,12 @@ int __mlx4_ib_query_gid(struct ib_device *ibdev, u8 port, int index,
 int mlx4_ib_resolve_grh(struct mlx4_ib_dev *dev, const struct ib_ah_attr *ah_attr,
 			u8 *mac, int *is_mcast, u8 port);
 
-static inline int mlx4_ib_ah_grh_present(struct mlx4_ib_ah *ah)
+static inline bool mlx4_ib_ah_grh_present(struct mlx4_ib_ah *ah)
 {
 	u8 port = be32_to_cpu(ah->av.ib.port_pd) >> 24 & 3;
 
 	if (rdma_port_get_link_layer(ah->ibah.device, port) == IB_LINK_LAYER_ETHERNET)
-		return 1;
+		return true;
 
 	return !!(ah->av.ib.g_slid & 0x80);
 }

drivers/infiniband/hw/mlx4/mr.c

@@ -41,9 +41,19 @@ static u32 convert_access(int acc)
 	       (acc & IB_ACCESS_REMOTE_WRITE  ? MLX4_PERM_REMOTE_WRITE : 0) |
 	       (acc & IB_ACCESS_REMOTE_READ   ? MLX4_PERM_REMOTE_READ  : 0) |
 	       (acc & IB_ACCESS_LOCAL_WRITE   ? MLX4_PERM_LOCAL_WRITE  : 0) |
+	       (acc & IB_ACCESS_MW_BIND	      ? MLX4_PERM_BIND_MW      : 0) |
 	       MLX4_PERM_LOCAL_READ;
 }
 
+static enum mlx4_mw_type to_mlx4_type(enum ib_mw_type type)
+{
+	switch (type) {
+	case IB_MW_TYPE_1:	return MLX4_MW_TYPE_1;
+	case IB_MW_TYPE_2:	return MLX4_MW_TYPE_2;
+	default:		return -1;
+	}
+}
+
 struct ib_mr *mlx4_ib_get_dma_mr(struct ib_pd *pd, int acc)
 {
 	struct mlx4_ib_mr *mr;
@@ -68,7 +78,7 @@ struct ib_mr *mlx4_ib_get_dma_mr(struct ib_pd *pd, int acc)
 	return &mr->ibmr;
 
 err_mr:
-	mlx4_mr_free(to_mdev(pd->device)->dev, &mr->mmr);
+	(void) mlx4_mr_free(to_mdev(pd->device)->dev, &mr->mmr);
 
 err_free:
 	kfree(mr);
@@ -163,7 +173,7 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 	return &mr->ibmr;
 
 err_mr:
-	mlx4_mr_free(to_mdev(pd->device)->dev, &mr->mmr);
+	(void) mlx4_mr_free(to_mdev(pd->device)->dev, &mr->mmr);
 
 err_umem:
 	ib_umem_release(mr->umem);
@@ -177,8 +187,11 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 int mlx4_ib_dereg_mr(struct ib_mr *ibmr)
 {
 	struct mlx4_ib_mr *mr = to_mmr(ibmr);
+	int ret;
 
-	mlx4_mr_free(to_mdev(ibmr->device)->dev, &mr->mmr);
+	ret = mlx4_mr_free(to_mdev(ibmr->device)->dev, &mr->mmr);
+	if (ret)
+		return ret;
 	if (mr->umem)
 		ib_umem_release(mr->umem);
 	kfree(mr);
@@ -186,6 +199,70 @@ int mlx4_ib_dereg_mr(struct ib_mr *ibmr)
 	return 0;
 }
 
+struct ib_mw *mlx4_ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type)
+{
+	struct mlx4_ib_dev *dev = to_mdev(pd->device);
+	struct mlx4_ib_mw *mw;
+	int err;
+
+	mw = kmalloc(sizeof(*mw), GFP_KERNEL);
+	if (!mw)
+		return ERR_PTR(-ENOMEM);
+
+	err = mlx4_mw_alloc(dev->dev, to_mpd(pd)->pdn,
+			    to_mlx4_type(type), &mw->mmw);
+	if (err)
+		goto err_free;
+
+	err = mlx4_mw_enable(dev->dev, &mw->mmw);
+	if (err)
+		goto err_mw;
+
+	mw->ibmw.rkey = mw->mmw.key;
+
+	return &mw->ibmw;
+
+err_mw:
+	mlx4_mw_free(dev->dev, &mw->mmw);
+
+err_free:
+	kfree(mw);
+
+	return ERR_PTR(err);
+}
+
+int mlx4_ib_bind_mw(struct ib_qp *qp, struct ib_mw *mw,
+		    struct ib_mw_bind *mw_bind)
+{
+	struct ib_send_wr  wr;
+	struct ib_send_wr *bad_wr;
+	int ret;
+
+	memset(&wr, 0, sizeof(wr));
+	wr.opcode               = IB_WR_BIND_MW;
+	wr.wr_id                = mw_bind->wr_id;
+	wr.send_flags           = mw_bind->send_flags;
+	wr.wr.bind_mw.mw        = mw;
+	wr.wr.bind_mw.bind_info = mw_bind->bind_info;
+	wr.wr.bind_mw.rkey      = ib_inc_rkey(mw->rkey);
+
+	ret = mlx4_ib_post_send(qp, &wr, &bad_wr);
+	if (!ret)
+		mw->rkey = wr.wr.bind_mw.rkey;
+
+	return ret;
+}
+
+int mlx4_ib_dealloc_mw(struct ib_mw *ibmw)
+{
+	struct mlx4_ib_mw *mw = to_mmw(ibmw);
+
+	mlx4_mw_free(to_mdev(ibmw->device)->dev, &mw->mmw);
+	kfree(mw);
+
+	return 0;
+}
+
 struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
 					int max_page_list_len)
 {
@@ -212,7 +289,7 @@ struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
 	return &mr->ibmr;
 
 err_mr:
-	mlx4_mr_free(dev->dev, &mr->mmr);
+	(void) mlx4_mr_free(dev->dev, &mr->mmr);
 
 err_free:
 	kfree(mr);
@@ -291,7 +368,7 @@ struct ib_fmr *mlx4_ib_fmr_alloc(struct ib_pd *pd, int acc,
 	return &fmr->ibfmr;
 
 err_mr:
-	mlx4_mr_free(to_mdev(pd->device)->dev, &fmr->mfmr.mr);
+	(void) mlx4_mr_free(to_mdev(pd->device)->dev, &fmr->mfmr.mr);
 
 err_free:
 	kfree(fmr);

drivers/infiniband/hw/mlx4/qp.c

@@ -104,6 +104,7 @@ static const __be32 mlx4_ib_opcode[] = {
 	[IB_WR_FAST_REG_MR]			= cpu_to_be32(MLX4_OPCODE_FMR),
 	[IB_WR_MASKED_ATOMIC_CMP_AND_SWP]	= cpu_to_be32(MLX4_OPCODE_MASKED_ATOMIC_CS),
 	[IB_WR_MASKED_ATOMIC_FETCH_AND_ADD]	= cpu_to_be32(MLX4_OPCODE_MASKED_ATOMIC_FA),
+	[IB_WR_BIND_MW]				= cpu_to_be32(MLX4_OPCODE_BIND_MW),
 };
 
 static struct mlx4_ib_sqp *to_msqp(struct mlx4_ib_qp *mqp)
@@ -1746,11 +1747,11 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
 	int header_size;
 	int spc;
 	int i;
-	int is_eth;
-	int is_vlan = 0;
-	int is_grh;
-	u16 vlan;
 	int err = 0;
+	u16 vlan = 0xffff;
+	bool is_eth;
+	bool is_vlan = false;
+	bool is_grh;
 
 	send_size = 0;
 	for (i = 0; i < wr->num_sge; ++i)
@@ -1953,9 +1954,12 @@ static int mlx4_wq_overflow(struct mlx4_ib_wq *wq, int nreq, struct ib_cq *ib_cq
 
 static __be32 convert_access(int acc)
 {
-	return (acc & IB_ACCESS_REMOTE_ATOMIC ? cpu_to_be32(MLX4_WQE_FMR_PERM_ATOMIC)       : 0) |
-	       (acc & IB_ACCESS_REMOTE_WRITE  ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_WRITE) : 0) |
-	       (acc & IB_ACCESS_REMOTE_READ   ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_READ)  : 0) |
+	return (acc & IB_ACCESS_REMOTE_ATOMIC ?
+		cpu_to_be32(MLX4_WQE_FMR_AND_BIND_PERM_ATOMIC)       : 0) |
+	       (acc & IB_ACCESS_REMOTE_WRITE  ?
+		cpu_to_be32(MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_WRITE) : 0) |
+	       (acc & IB_ACCESS_REMOTE_READ   ?
+		cpu_to_be32(MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_READ)  : 0) |
 	       (acc & IB_ACCESS_LOCAL_WRITE   ? cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_WRITE)  : 0) |
 		cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_READ);
 }
@@ -1981,12 +1985,28 @@ static void set_fmr_seg(struct mlx4_wqe_fmr_seg *fseg, struct ib_send_wr *wr)
 	fseg->reserved[1]	= 0;
 }
 
+static void set_bind_seg(struct mlx4_wqe_bind_seg *bseg, struct ib_send_wr *wr)
+{
+	bseg->flags1 =
+		convert_access(wr->wr.bind_mw.bind_info.mw_access_flags) &
+		cpu_to_be32(MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_READ  |
+			    MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_WRITE |
+			    MLX4_WQE_FMR_AND_BIND_PERM_ATOMIC);
+	bseg->flags2 = 0;
+	if (wr->wr.bind_mw.mw->type == IB_MW_TYPE_2)
+		bseg->flags2 |= cpu_to_be32(MLX4_WQE_BIND_TYPE_2);
+	if (wr->wr.bind_mw.bind_info.mw_access_flags & IB_ZERO_BASED)
+		bseg->flags2 |= cpu_to_be32(MLX4_WQE_BIND_ZERO_BASED);
+	bseg->new_rkey = cpu_to_be32(wr->wr.bind_mw.rkey);
+	bseg->lkey = cpu_to_be32(wr->wr.bind_mw.bind_info.mr->lkey);
+	bseg->addr = cpu_to_be64(wr->wr.bind_mw.bind_info.addr);
+	bseg->length = cpu_to_be64(wr->wr.bind_mw.bind_info.length);
+}
+
 static void set_local_inv_seg(struct mlx4_wqe_local_inval_seg *iseg, u32 rkey)
 {
-	iseg->flags	= 0;
-	iseg->mem_key	= cpu_to_be32(rkey);
-	iseg->guest_id	= 0;
-	iseg->pa	= 0;
+	memset(iseg, 0, sizeof(*iseg));
+	iseg->mem_key = cpu_to_be32(rkey);
 }
 
 static __always_inline void set_raddr_seg(struct mlx4_wqe_raddr_seg *rseg,
@@ -2291,6 +2311,13 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 				size += sizeof (struct mlx4_wqe_fmr_seg) / 16;
 				break;
 
+			case IB_WR_BIND_MW:
+				ctrl->srcrb_flags |=
+					cpu_to_be32(MLX4_WQE_CTRL_STRONG_ORDER);
+				set_bind_seg(wqe, wr);
+				wqe  += sizeof(struct mlx4_wqe_bind_seg);
+				size += sizeof(struct mlx4_wqe_bind_seg) / 16;
+				break;
 			default:
 				/* No extra segments required for sends */
 				break;

drivers/infiniband/hw/mlx4/sysfs.c

@@ -732,7 +732,7 @@ int mlx4_ib_device_register_sysfs(struct mlx4_ib_dev *dev)
 	dev->ports_parent =
 		kobject_create_and_add("ports",
 				       kobject_get(dev->iov_parent));
-	if (!dev->iov_parent) {
+	if (!dev->ports_parent) {
 		ret = -ENOMEM;
 		goto err_ports;
 	}

drivers/infiniband/hw/nes/nes_verbs.c

@@ -55,7 +55,8 @@ static void nes_unregister_ofa_device(struct nes_ib_device *nesibdev);
 /**
  * nes_alloc_mw
  */
-static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) {
+static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd, enum ib_mw_type type)
+{
 	struct nes_pd *nespd = to_nespd(ibpd);
 	struct nes_vnic *nesvnic = to_nesvnic(ibpd->device);
 	struct nes_device *nesdev = nesvnic->nesdev;
@@ -71,6 +72,9 @@ static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) {
 	u32 driver_key = 0;
 	u8 stag_key = 0;
 
+	if (type != IB_MW_TYPE_1)
+		return ERR_PTR(-EINVAL);
+
 	get_random_bytes(&next_stag_index, sizeof(next_stag_index));
 	stag_key = (u8)next_stag_index;
 
@@ -244,20 +248,19 @@ static int nes_bind_mw(struct ib_qp *ibqp, struct ib_mw *ibmw,
 	if (ibmw_bind->send_flags & IB_SEND_SIGNALED)
 		wqe_misc |= NES_IWARP_SQ_WQE_SIGNALED_COMPL;
 
-	if (ibmw_bind->mw_access_flags & IB_ACCESS_REMOTE_WRITE) {
+	if (ibmw_bind->bind_info.mw_access_flags & IB_ACCESS_REMOTE_WRITE)
 		wqe_misc |= NES_CQP_STAG_RIGHTS_REMOTE_WRITE;
-	}
-	if (ibmw_bind->mw_access_flags & IB_ACCESS_REMOTE_READ) {
+	if (ibmw_bind->bind_info.mw_access_flags & IB_ACCESS_REMOTE_READ)
 		wqe_misc |= NES_CQP_STAG_RIGHTS_REMOTE_READ;
-	}
 
 	set_wqe_32bit_value(wqe->wqe_words, NES_IWARP_SQ_WQE_MISC_IDX, wqe_misc);
-	set_wqe_32bit_value(wqe->wqe_words, NES_IWARP_SQ_BIND_WQE_MR_IDX, ibmw_bind->mr->lkey);
+	set_wqe_32bit_value(wqe->wqe_words, NES_IWARP_SQ_BIND_WQE_MR_IDX,
+			    ibmw_bind->bind_info.mr->lkey);
 	set_wqe_32bit_value(wqe->wqe_words, NES_IWARP_SQ_BIND_WQE_MW_IDX, ibmw->rkey);
 	set_wqe_32bit_value(wqe->wqe_words, NES_IWARP_SQ_BIND_WQE_LENGTH_LOW_IDX,
-			ibmw_bind->length);
+			ibmw_bind->bind_info.length);
 	wqe->wqe_words[NES_IWARP_SQ_BIND_WQE_LENGTH_HIGH_IDX] = 0;
-	u64temp = (u64)ibmw_bind->addr;
+	u64temp = (u64)ibmw_bind->bind_info.addr;
 	set_wqe_64bit_value(wqe->wqe_words, NES_IWARP_SQ_BIND_WQE_VA_FBO_LOW_IDX, u64temp);
 
 	head++;

drivers/infiniband/hw/qib/qib_qp.c

@@ -268,8 +268,9 @@ static void remove_qp(struct qib_ibdev *dev, struct qib_qp *qp)
 				qpp = &q->next)
 			if (q == qp) {
 				atomic_dec(&qp->refcount);
-				*qpp = qp->next;
-				rcu_assign_pointer(qp->next, NULL);
+				rcu_assign_pointer(*qpp,
+					rcu_dereference_protected(qp->next,
+					 lockdep_is_held(&dev->qpt_lock)));
 				break;
 			}
 	}

drivers/infiniband/ulp/ipoib/ipoib.h

@@ -117,6 +117,8 @@ enum {
 #define	IPOIB_OP_CM     (0)
 #endif
 
+#define IPOIB_QPN_MASK ((__force u32) cpu_to_be32(0xFFFFFF))
+
 /* structs */
 
 struct ipoib_header {
@@ -760,4 +762,6 @@ extern int ipoib_debug_level;
 
 #define IPOIB_QPN(ha) (be32_to_cpup((__be32 *) ha) & 0xffffff)
 
+extern const char ipoib_driver_version[];
+
 #endif /* _IPOIB_H */

drivers/infiniband/ulp/ipoib/ipoib_ethtool.c

@@ -39,7 +39,24 @@
 static void ipoib_get_drvinfo(struct net_device *netdev,
 			      struct ethtool_drvinfo *drvinfo)
 {
-	strlcpy(drvinfo->driver, "ipoib", sizeof(drvinfo->driver));
+	struct ipoib_dev_priv *priv = netdev_priv(netdev);
+	struct ib_device_attr *attr;
+
+	attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+	if (attr && !ib_query_device(priv->ca, attr))
+		snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+			 "%d.%d.%d", (int)(attr->fw_ver >> 32),
+			 (int)(attr->fw_ver >> 16) & 0xffff,
+			 (int)attr->fw_ver & 0xffff);
+	kfree(attr);
+
+	strlcpy(drvinfo->bus_info, dev_name(priv->ca->dma_device),
+		sizeof(drvinfo->bus_info));
+
+	strlcpy(drvinfo->version, ipoib_driver_version,
+		sizeof(drvinfo->version));
+
+	strlcpy(drvinfo->driver, "ib_ipoib", sizeof(drvinfo->driver));
 }
 
 static int ipoib_get_coalesce(struct net_device *dev,

drivers/infiniband/ulp/ipoib/ipoib_main.c

@@ -49,9 +49,14 @@
 #include <linux/jhash.h>
 #include <net/arp.h>
 
+#define DRV_VERSION "1.0.0"
+
+const char ipoib_driver_version[] = DRV_VERSION;
+
 MODULE_AUTHOR("Roland Dreier");
 MODULE_DESCRIPTION("IP-over-InfiniBand net driver");
 MODULE_LICENSE("Dual BSD/GPL");
+MODULE_VERSION(DRV_VERSION);
 
 int ipoib_sendq_size __read_mostly = IPOIB_TX_RING_SIZE;
 int ipoib_recvq_size __read_mostly = IPOIB_RX_RING_SIZE;
@@ -505,6 +510,9 @@ static void path_rec_completion(int status,
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
+	if (IS_ERR_OR_NULL(ah))
+		ipoib_del_neighs_by_gid(dev, path->pathrec.dgid.raw);
+
 	if (old_ah)
 		ipoib_put_ah(old_ah);
 
@@ -844,10 +852,10 @@ static u32 ipoib_addr_hash(struct ipoib_neigh_hash *htbl, u8 *daddr)
 	 * different subnets.
 	 */
 	 /* qpn octets[1:4) & port GUID octets[12:20) */
-	u32 *daddr_32 = (u32 *) daddr;
+	u32 *d32 = (u32 *) daddr;
 	u32 hv;
 
-	hv = jhash_3words(daddr_32[3], daddr_32[4], 0xFFFFFF & daddr_32[0], 0);
+	hv = jhash_3words(d32[3], d32[4], IPOIB_QPN_MASK & d32[0], 0);
 	return hv & htbl->mask;
 }
 
@@ -1688,6 +1696,8 @@ static void ipoib_remove_one(struct ib_device *device)
 		return;
 
 	dev_list = ib_get_client_data(device, &ipoib_client);
+	if (!dev_list)
+		return;
 
 	list_for_each_entry_safe(priv, tmp, dev_list, list) {
 		ib_unregister_event_handler(&priv->event_handler);

drivers/infiniband/ulp/iser/iscsi_iser.h

@@ -94,7 +94,7 @@
 
 					/* support up to 512KB in one RDMA */
 #define ISCSI_ISER_SG_TABLESIZE         (0x80000 >> SHIFT_4K)
-#define ISER_DEF_CMD_PER_LUN		128
+#define ISER_DEF_CMD_PER_LUN		ISCSI_DEF_XMIT_CMDS_MAX
 
 /* QP settings */
 /* Maximal bounds on received asynchronous PDUs */

drivers/infiniband/ulp/iser/iser_memory.c

@@ -369,10 +369,11 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *iser_task,
 	regd_buf = &iser_task->rdma_regd[cmd_dir];
 
 	aligned_len = iser_data_buf_aligned_len(mem, ibdev);
-	if (aligned_len != mem->dma_nents) {
+	if (aligned_len != mem->dma_nents ||
+	    (!ib_conn->fmr_pool && mem->dma_nents > 1)) {
 		iscsi_conn->fmr_unalign_cnt++;
-		iser_warn("rdma alignment violation %d/%d aligned\n",
-			 aligned_len, mem->size);
+		iser_warn("rdma alignment violation (%d/%d aligned) or FMR not supported\n",
+			  aligned_len, mem->size);
 		iser_data_buf_dump(mem, ibdev);
 
 		/* unmap the command data before accessing it */
@@ -404,7 +405,7 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *iser_task,
 	} else { /* use FMR for multiple dma entries */
 		iser_page_vec_build(mem, ib_conn->page_vec, ibdev);
 		err = iser_reg_page_vec(ib_conn, ib_conn->page_vec, &regd_buf->reg);
-		if (err) {
+		if (err && err != -EAGAIN) {
 			iser_data_buf_dump(mem, ibdev);
 			iser_err("mem->dma_nents = %d (dlength = 0x%x)\n",
 				 mem->dma_nents,

drivers/infiniband/ulp/iser/iser_verbs.c

@@ -242,10 +242,14 @@ static int iser_create_ib_conn_res(struct iser_conn *ib_conn)
 				    IB_ACCESS_REMOTE_READ);
 
 	ib_conn->fmr_pool = ib_create_fmr_pool(device->pd, &params);
-	if (IS_ERR(ib_conn->fmr_pool)) {
-		ret = PTR_ERR(ib_conn->fmr_pool);
+	ret = PTR_ERR(ib_conn->fmr_pool);
+	if (IS_ERR(ib_conn->fmr_pool) && ret != -ENOSYS) {
 		ib_conn->fmr_pool = NULL;
 		goto out_err;
+	} else if (ret == -ENOSYS) {
+		ib_conn->fmr_pool = NULL;
+		iser_warn("FMRs are not supported, using unaligned mode\n");
+		ret = 0;
 	}
 
 	memset(&init_attr, 0, sizeof init_attr);

drivers/infiniband/ulp/srp/ib_srp.c

@@ -700,23 +700,24 @@ static int srp_reconnect_target(struct srp_target_port *target)
 	struct Scsi_Host *shost = target->scsi_host;
 	int i, ret;
 
-	if (target->state != SRP_TARGET_LIVE)
-		return -EAGAIN;
-
 	scsi_target_block(&shost->shost_gendev);
 
 	srp_disconnect_target(target);
 	/*
-	 * Now get a new local CM ID so that we avoid confusing the
-	 * target in case things are really fouled up.
+	 * Now get a new local CM ID so that we avoid confusing the target in
+	 * case things are really fouled up. Doing so also ensures that all CM
+	 * callbacks will have finished before a new QP is allocated.
 	 */
 	ret = srp_new_cm_id(target);
-	if (ret)
-		goto unblock;
-
-	ret = srp_create_target_ib(target);
-	if (ret)
-		goto unblock;
+	/*
+	 * Whether or not creating a new CM ID succeeded, create a new
+	 * QP. This guarantees that all completion callback function
+	 * invocations have finished before request resetting starts.
+	 */
+	if (ret == 0)
+		ret = srp_create_target_ib(target);
+	else
+		srp_create_target_ib(target);
 
 	for (i = 0; i < SRP_CMD_SQ_SIZE; ++i) {
 		struct srp_request *req = &target->req_ring[i];
@@ -728,11 +729,12 @@ static int srp_reconnect_target(struct srp_target_port *target)
 	for (i = 0; i < SRP_SQ_SIZE; ++i)
 		list_add(&target->tx_ring[i]->list, &target->free_tx);
 
-	ret = srp_connect_target(target);
+	if (ret == 0)
+		ret = srp_connect_target(target);
 
-unblock:
 	scsi_target_unblock(&shost->shost_gendev, ret == 0 ? SDEV_RUNNING :
 			    SDEV_TRANSPORT_OFFLINE);
+	target->transport_offline = !!ret;
 
 	if (ret)
 		goto err;
@@ -1352,6 +1354,12 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	unsigned long flags;
 	int len;
 
+	if (unlikely(target->transport_offline)) {
+		scmnd->result = DID_NO_CONNECT << 16;
+		scmnd->scsi_done(scmnd);
+		return 0;
+	}
+
 	spin_lock_irqsave(&target->lock, flags);
 	iu = __srp_get_tx_iu(target, SRP_IU_CMD);
 	if (!iu)
@@ -1695,6 +1703,9 @@ static int srp_send_tsk_mgmt(struct srp_target_port *target,
 	struct srp_iu *iu;
 	struct srp_tsk_mgmt *tsk_mgmt;
 
+	if (!target->connected || target->qp_in_error)
+		return -1;
+
 	init_completion(&target->tsk_mgmt_done);
 
 	spin_lock_irq(&target->lock);
@@ -1736,7 +1747,7 @@ static int srp_abort(struct scsi_cmnd *scmnd)
 
 	shost_printk(KERN_ERR, target->scsi_host, "SRP abort called\n");
 
-	if (!req || target->qp_in_error || !srp_claim_req(target, req, scmnd))
+	if (!req || !srp_claim_req(target, req, scmnd))
 		return FAILED;
 	srp_send_tsk_mgmt(target, req->index, scmnd->device->lun,
 			  SRP_TSK_ABORT_TASK);
@@ -1754,8 +1765,6 @@ static int srp_reset_device(struct scsi_cmnd *scmnd)
 
 	shost_printk(KERN_ERR, target->scsi_host, "SRP reset_device called\n");
 
-	if (target->qp_in_error)
-		return FAILED;
 	if (srp_send_tsk_mgmt(target, SRP_TAG_NO_REQ, scmnd->device->lun,
 			      SRP_TSK_LUN_RESET))
 		return FAILED;
@@ -1972,7 +1981,6 @@ static int srp_add_target(struct srp_host *host, struct srp_target_port *target)
 	spin_unlock(&host->target_lock);
 
 	target->state = SRP_TARGET_LIVE;
-	target->connected = false;
 
 	scsi_scan_target(&target->scsi_host->shost_gendev,
 			 0, target->scsi_id, SCAN_WILD_CARD, 0);

drivers/infiniband/ulp/srp/ib_srp.h

@@ -140,6 +140,7 @@ struct srp_target_port {
 	unsigned int		cmd_sg_cnt;
 	unsigned int		indirect_size;
 	bool			allow_ext_sg;
+	bool			transport_offline;
 
 	/* Everything above this point is used in the hot path of
 	 * command processing. Try to keep them packed into cachelines.

drivers/net/ethernet/mellanox/mlx4/en_main.c

@@ -198,7 +198,7 @@ static void mlx4_en_remove(struct mlx4_dev *dev, void *endev_ptr)
 
 	flush_workqueue(mdev->workqueue);
 	destroy_workqueue(mdev->workqueue);
-	mlx4_mr_free(dev, &mdev->mr);
+	(void) mlx4_mr_free(dev, &mdev->mr);
 	iounmap(mdev->uar_map);
 	mlx4_uar_free(dev, &mdev->priv_uar);
 	mlx4_pd_free(dev, mdev->priv_pdn);
@@ -303,7 +303,7 @@ static void *mlx4_en_add(struct mlx4_dev *dev)
 	return mdev;
 
 err_mr:
-	mlx4_mr_free(dev, &mdev->mr);
+	(void) mlx4_mr_free(dev, &mdev->mr);
 err_map:
 	if (!mdev->uar_map)
 		iounmap(mdev->uar_map);

drivers/net/ethernet/mellanox/mlx4/fw.c

@@ -762,15 +762,19 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	u64	flags;
 	int	err = 0;
 	u8	field;
+	u32	bmme_flags;
 
 	err = mlx4_cmd_box(dev, 0, outbox->dma, 0, 0, MLX4_CMD_QUERY_DEV_CAP,
 			   MLX4_CMD_TIME_CLASS_A, MLX4_CMD_NATIVE);
 	if (err)
 		return err;
 
-	/* add port mng change event capability unconditionally to slaves */
+	/* add port mng change event capability and disable mw type 1
+	 * unconditionally to slaves
+	 */
 	MLX4_GET(flags, outbox->buf, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
 	flags |= MLX4_DEV_CAP_FLAG_PORT_MNG_CHG_EV;
+	flags &= ~MLX4_DEV_CAP_FLAG_MEM_WINDOW;
 	MLX4_PUT(outbox->buf, flags, QUERY_DEV_CAP_EXT_FLAGS_OFFSET);
 
 	/* For guests, report Blueflame disabled */
@@ -778,6 +782,11 @@ int mlx4_QUERY_DEV_CAP_wrapper(struct mlx4_dev *dev, int slave,
 	field &= 0x7f;
 	MLX4_PUT(outbox->buf, field, QUERY_DEV_CAP_BF_OFFSET);
 
+	/* For guests, disable mw type 2 */
+	MLX4_GET(bmme_flags, outbox, QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
+	bmme_flags &= ~MLX4_BMME_FLAG_TYPE_2_WIN;
+	MLX4_PUT(outbox->buf, bmme_flags, QUERY_DEV_CAP_BMME_FLAGS_OFFSET);
+
 	return 0;
 }
 
@@ -1203,6 +1212,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 #define  INIT_HCA_FS_IB_NUM_ADDRS_OFFSET  (INIT_HCA_FS_PARAM_OFFSET + 0x26)
 #define INIT_HCA_TPT_OFFSET		 0x0f0
 #define	 INIT_HCA_DMPT_BASE_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x00)
+#define  INIT_HCA_TPT_MW_OFFSET		 (INIT_HCA_TPT_OFFSET + 0x08)
 #define	 INIT_HCA_LOG_MPT_SZ_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x0b)
 #define	 INIT_HCA_MTT_BASE_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x10)
 #define	 INIT_HCA_CMPT_BASE_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x18)
@@ -1319,6 +1329,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	/* TPT attributes */
 
 	MLX4_PUT(inbox, param->dmpt_base,  INIT_HCA_DMPT_BASE_OFFSET);
+	MLX4_PUT(inbox, param->mw_enabled, INIT_HCA_TPT_MW_OFFSET);
 	MLX4_PUT(inbox, param->log_mpt_sz, INIT_HCA_LOG_MPT_SZ_OFFSET);
 	MLX4_PUT(inbox, param->mtt_base,   INIT_HCA_MTT_BASE_OFFSET);
 	MLX4_PUT(inbox, param->cmpt_base,  INIT_HCA_CMPT_BASE_OFFSET);
@@ -1415,6 +1426,7 @@ int mlx4_QUERY_HCA(struct mlx4_dev *dev,
 	/* TPT attributes */
 
 	MLX4_GET(param->dmpt_base,  outbox, INIT_HCA_DMPT_BASE_OFFSET);
+	MLX4_GET(param->mw_enabled, outbox, INIT_HCA_TPT_MW_OFFSET);
 	MLX4_GET(param->log_mpt_sz, outbox, INIT_HCA_LOG_MPT_SZ_OFFSET);
 	MLX4_GET(param->mtt_base,   outbox, INIT_HCA_MTT_BASE_OFFSET);
 	MLX4_GET(param->cmpt_base,  outbox, INIT_HCA_CMPT_BASE_OFFSET);

drivers/net/ethernet/mellanox/mlx4/fw.h

@@ -170,6 +170,7 @@ struct mlx4_init_hca_param {
 	u8  log_mc_table_sz;
 	u8  log_mpt_sz;
 	u8  log_uar_sz;
+	u8  mw_enabled;  /* Enable memory windows */
 	u8  uar_page_sz; /* log pg sz in 4k chunks */
 	u8  steering_mode; /* for QUERY_HCA */
 	u64 dev_cap_enabled;

drivers/net/ethernet/mellanox/mlx4/main.c

@@ -1431,6 +1431,10 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
 		init_hca.uar_page_sz = PAGE_SHIFT - 12;
+		init_hca.mw_enabled = 0;
+		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
+		    dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN)
+			init_hca.mw_enabled = INIT_HCA_TPT_MW_ENABLE;
 
 		err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size);
 		if (err)

drivers/net/ethernet/mellanox/mlx4/mlx4.h

@@ -60,6 +60,8 @@
 #define MLX4_FS_MGM_LOG_ENTRY_SIZE	7
 #define MLX4_FS_NUM_MCG			(1 << 17)
 
+#define INIT_HCA_TPT_MW_ENABLE          (1 << 7)
+
 #define MLX4_NUM_UP		8
 #define MLX4_NUM_TC		8
 #define MLX4_RATELIMIT_UNITS 3 /* 100 Mbps */
@@ -113,10 +115,10 @@ enum {
 	MLX4_NUM_CMPTS		= MLX4_CMPT_NUM_TYPE << MLX4_CMPT_SHIFT
 };
 
-enum mlx4_mr_state {
-	MLX4_MR_DISABLED = 0,
-	MLX4_MR_EN_HW,
-	MLX4_MR_EN_SW
+enum mlx4_mpt_state {
+	MLX4_MPT_DISABLED = 0,
+	MLX4_MPT_EN_HW,
+	MLX4_MPT_EN_SW
 };
 
 #define MLX4_COMM_TIME		10000
@@ -263,6 +265,22 @@ struct mlx4_icm_table {
 	struct mlx4_icm	      **icm;
 };
 
+#define MLX4_MPT_FLAG_SW_OWNS	    (0xfUL << 28)
+#define MLX4_MPT_FLAG_FREE	    (0x3UL << 28)
+#define MLX4_MPT_FLAG_MIO	    (1 << 17)
+#define MLX4_MPT_FLAG_BIND_ENABLE   (1 << 15)
+#define MLX4_MPT_FLAG_PHYSICAL	    (1 <<  9)
+#define MLX4_MPT_FLAG_REGION	    (1 <<  8)
+
+#define MLX4_MPT_PD_FLAG_FAST_REG   (1 << 27)
+#define MLX4_MPT_PD_FLAG_RAE	    (1 << 28)
+#define MLX4_MPT_PD_FLAG_EN_INV	    (3 << 24)
+
+#define MLX4_MPT_QP_FLAG_BOUND_QP   (1 << 7)
+
+#define MLX4_MPT_STATUS_SW		0xF0
+#define MLX4_MPT_STATUS_HW		0x00
+
 /*
  * Must be packed because mtt_seg is 64 bits but only aligned to 32 bits.
  */
@@ -863,10 +881,10 @@ int __mlx4_cq_alloc_icm(struct mlx4_dev *dev, int *cqn);
 void __mlx4_cq_free_icm(struct mlx4_dev *dev, int cqn);
 int __mlx4_srq_alloc_icm(struct mlx4_dev *dev, int *srqn);
 void __mlx4_srq_free_icm(struct mlx4_dev *dev, int srqn);
-int __mlx4_mr_reserve(struct mlx4_dev *dev);
-void __mlx4_mr_release(struct mlx4_dev *dev, u32 index);
-int __mlx4_mr_alloc_icm(struct mlx4_dev *dev, u32 index);
-void __mlx4_mr_free_icm(struct mlx4_dev *dev, u32 index);
+int __mlx4_mpt_reserve(struct mlx4_dev *dev);
+void __mlx4_mpt_release(struct mlx4_dev *dev, u32 index);
+int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index);
+void __mlx4_mpt_free_icm(struct mlx4_dev *dev, u32 index);
 u32 __mlx4_alloc_mtt_range(struct mlx4_dev *dev, int order);
 void __mlx4_free_mtt_range(struct mlx4_dev *dev, u32 first_seg, int order);
 

drivers/net/ethernet/mellanox/mlx4/mr.c

@@ -44,20 +44,6 @@
 #include "mlx4.h"
 #include "icm.h"
 
-#define MLX4_MPT_FLAG_SW_OWNS	    (0xfUL << 28)
-#define MLX4_MPT_FLAG_FREE	    (0x3UL << 28)
-#define MLX4_MPT_FLAG_MIO	    (1 << 17)
-#define MLX4_MPT_FLAG_BIND_ENABLE   (1 << 15)
-#define MLX4_MPT_FLAG_PHYSICAL	    (1 <<  9)
-#define MLX4_MPT_FLAG_REGION	    (1 <<  8)
-
-#define MLX4_MPT_PD_FLAG_FAST_REG   (1 << 27)
-#define MLX4_MPT_PD_FLAG_RAE	    (1 << 28)
-#define MLX4_MPT_PD_FLAG_EN_INV	    (3 << 24)
-
-#define MLX4_MPT_STATUS_SW		0xF0
-#define MLX4_MPT_STATUS_HW		0x00
-
 static u32 mlx4_buddy_alloc(struct mlx4_buddy *buddy, int order)
 {
 	int o;
@@ -321,7 +307,7 @@ static int mlx4_mr_alloc_reserved(struct mlx4_dev *dev, u32 mridx, u32 pd,
 	mr->size       = size;
 	mr->pd	       = pd;
 	mr->access     = access;
-	mr->enabled    = MLX4_MR_DISABLED;
+	mr->enabled    = MLX4_MPT_DISABLED;
 	mr->key	       = hw_index_to_key(mridx);
 
 	return mlx4_mtt_init(dev, npages, page_shift, &mr->mtt);
@@ -335,14 +321,14 @@ static int mlx4_WRITE_MTT(struct mlx4_dev *dev,
 			MLX4_CMD_TIME_CLASS_A,  MLX4_CMD_WRAPPED);
 }
 
-int __mlx4_mr_reserve(struct mlx4_dev *dev)
+int __mlx4_mpt_reserve(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 
 	return mlx4_bitmap_alloc(&priv->mr_table.mpt_bitmap);
 }
 
-static int mlx4_mr_reserve(struct mlx4_dev *dev)
+static int mlx4_mpt_reserve(struct mlx4_dev *dev)
 {
 	u64 out_param;
 
@@ -353,17 +339,17 @@ static int mlx4_mr_reserve(struct mlx4_dev *dev)
 			return -1;
 		return get_param_l(&out_param);
 	}
-	return  __mlx4_mr_reserve(dev);
+	return  __mlx4_mpt_reserve(dev);
 }
 
-void __mlx4_mr_release(struct mlx4_dev *dev, u32 index)
+void __mlx4_mpt_release(struct mlx4_dev *dev, u32 index)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 
 	mlx4_bitmap_free(&priv->mr_table.mpt_bitmap, index);
 }
 
-static void mlx4_mr_release(struct mlx4_dev *dev, u32 index)
+static void mlx4_mpt_release(struct mlx4_dev *dev, u32 index)
 {
 	u64 in_param;
 
@@ -376,17 +362,17 @@ static void mlx4_mr_release(struct mlx4_dev *dev, u32 index)
 				  index);
 		return;
 	}
-	__mlx4_mr_release(dev, index);
+	__mlx4_mpt_release(dev, index);
 }
 
-int __mlx4_mr_alloc_icm(struct mlx4_dev *dev, u32 index)
+int __mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index)
 {
 	struct mlx4_mr_table *mr_table = &mlx4_priv(dev)->mr_table;
 
 	return mlx4_table_get(dev, &mr_table->dmpt_table, index);
 }
 
-static int mlx4_mr_alloc_icm(struct mlx4_dev *dev, u32 index)
+static int mlx4_mpt_alloc_icm(struct mlx4_dev *dev, u32 index)
 {
 	u64 param;
 
@@ -397,17 +383,17 @@ static int mlx4_mr_alloc_icm(struct mlx4_dev *dev, u32 index)
 							MLX4_CMD_TIME_CLASS_A,
 							MLX4_CMD_WRAPPED);
 	}
-	return __mlx4_mr_alloc_icm(dev, index);
+	return __mlx4_mpt_alloc_icm(dev, index);
 }
 
-void __mlx4_mr_free_icm(struct mlx4_dev *dev, u32 index)
+void __mlx4_mpt_free_icm(struct mlx4_dev *dev, u32 index)
 {
 	struct mlx4_mr_table *mr_table = &mlx4_priv(dev)->mr_table;
 
 	mlx4_table_put(dev, &mr_table->dmpt_table, index);
 }
 
-static void mlx4_mr_free_icm(struct mlx4_dev *dev, u32 index)
+static void mlx4_mpt_free_icm(struct mlx4_dev *dev, u32 index)
 {
 	u64 in_param;
 
@@ -420,7 +406,7 @@ static void mlx4_mr_free_icm(struct mlx4_dev *dev, u32 index)
 				  index);
 		return;
 	}
-	return __mlx4_mr_free_icm(dev, index);
+	return __mlx4_mpt_free_icm(dev, index);
 }
 
 int mlx4_mr_alloc(struct mlx4_dev *dev, u32 pd, u64 iova, u64 size, u32 access,
@@ -429,41 +415,52 @@ int mlx4_mr_alloc(struct mlx4_dev *dev, u32 pd, u64 iova, u64 size, u32 access,
 	u32 index;
 	int err;
 
-	index = mlx4_mr_reserve(dev);
+	index = mlx4_mpt_reserve(dev);
 	if (index == -1)
 		return -ENOMEM;
 
 	err = mlx4_mr_alloc_reserved(dev, index, pd, iova, size,
 				     access, npages, page_shift, mr);
 	if (err)
-		mlx4_mr_release(dev, index);
+		mlx4_mpt_release(dev, index);
 
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx4_mr_alloc);
 
-static void mlx4_mr_free_reserved(struct mlx4_dev *dev, struct mlx4_mr *mr)
+static int mlx4_mr_free_reserved(struct mlx4_dev *dev, struct mlx4_mr *mr)
 {
 	int err;
 
-	if (mr->enabled == MLX4_MR_EN_HW) {
+	if (mr->enabled == MLX4_MPT_EN_HW) {
 		err = mlx4_HW2SW_MPT(dev, NULL,
 				     key_to_hw_index(mr->key) &
 				     (dev->caps.num_mpts - 1));
-		if (err)
-			mlx4_warn(dev, "xxx HW2SW_MPT failed (%d)\n", err);
+		if (err) {
+			mlx4_warn(dev, "HW2SW_MPT failed (%d),", err);
+			mlx4_warn(dev, "MR has MWs bound to it.\n");
+			return err;
+		}
 
-		mr->enabled = MLX4_MR_EN_SW;
+		mr->enabled = MLX4_MPT_EN_SW;
 	}
 	mlx4_mtt_cleanup(dev, &mr->mtt);
+
+	return 0;
 }
 
-void mlx4_mr_free(struct mlx4_dev *dev, struct mlx4_mr *mr)
+int mlx4_mr_free(struct mlx4_dev *dev, struct mlx4_mr *mr)
 {
-	mlx4_mr_free_reserved(dev, mr);
+	int ret;
+
+	ret = mlx4_mr_free_reserved(dev, mr);
+	if (ret)
+		return ret;
 	if (mr->enabled)
-		mlx4_mr_free_icm(dev, key_to_hw_index(mr->key));
-	mlx4_mr_release(dev, key_to_hw_index(mr->key));
+		mlx4_mpt_free_icm(dev, key_to_hw_index(mr->key));
+	mlx4_mpt_release(dev, key_to_hw_index(mr->key));
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx4_mr_free);
 
@@ -473,7 +470,7 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
 	struct mlx4_mpt_entry *mpt_entry;
 	int err;
 
-	err = mlx4_mr_alloc_icm(dev, key_to_hw_index(mr->key));
+	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mr->key));
 	if (err)
 		return err;
 
@@ -520,7 +517,7 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
 		mlx4_warn(dev, "SW2HW_MPT failed (%d)\n", err);
 		goto err_cmd;
 	}
-	mr->enabled = MLX4_MR_EN_HW;
+	mr->enabled = MLX4_MPT_EN_HW;
 
 	mlx4_free_cmd_mailbox(dev, mailbox);
 
@@ -530,7 +527,7 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
 	mlx4_free_cmd_mailbox(dev, mailbox);
 
 err_table:
-	mlx4_mr_free_icm(dev, key_to_hw_index(mr->key));
+	mlx4_mpt_free_icm(dev, key_to_hw_index(mr->key));
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx4_mr_enable);
@@ -657,6 +654,101 @@ int mlx4_buf_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 }
 EXPORT_SYMBOL_GPL(mlx4_buf_write_mtt);
 
+int mlx4_mw_alloc(struct mlx4_dev *dev, u32 pd, enum mlx4_mw_type type,
+		  struct mlx4_mw *mw)
+{
+	u32 index;
+
+	if ((type == MLX4_MW_TYPE_1 &&
+	     !(dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW)) ||
+	     (type == MLX4_MW_TYPE_2 &&
+	     !(dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN)))
+		return -ENOTSUPP;
+
+	index = mlx4_mpt_reserve(dev);
+	if (index == -1)
+		return -ENOMEM;
+
+	mw->key	    = hw_index_to_key(index);
+	mw->pd      = pd;
+	mw->type    = type;
+	mw->enabled = MLX4_MPT_DISABLED;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_mw_alloc);
+
+int mlx4_mw_enable(struct mlx4_dev *dev, struct mlx4_mw *mw)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_mpt_entry *mpt_entry;
+	int err;
+
+	err = mlx4_mpt_alloc_icm(dev, key_to_hw_index(mw->key));
+	if (err)
+		return err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox)) {
+		err = PTR_ERR(mailbox);
+		goto err_table;
+	}
+	mpt_entry = mailbox->buf;
+
+	memset(mpt_entry, 0, sizeof(*mpt_entry));
+
+	/* Note that the MLX4_MPT_FLAG_REGION bit in mpt_entry->flags is turned
+	 * off, thus creating a memory window and not a memory region.
+	 */
+	mpt_entry->key	       = cpu_to_be32(key_to_hw_index(mw->key));
+	mpt_entry->pd_flags    = cpu_to_be32(mw->pd);
+	if (mw->type == MLX4_MW_TYPE_2) {
+		mpt_entry->flags    |= cpu_to_be32(MLX4_MPT_FLAG_FREE);
+		mpt_entry->qpn       = cpu_to_be32(MLX4_MPT_QP_FLAG_BOUND_QP);
+		mpt_entry->pd_flags |= cpu_to_be32(MLX4_MPT_PD_FLAG_EN_INV);
+	}
+
+	err = mlx4_SW2HW_MPT(dev, mailbox,
+			     key_to_hw_index(mw->key) &
+			     (dev->caps.num_mpts - 1));
+	if (err) {
+		mlx4_warn(dev, "SW2HW_MPT failed (%d)\n", err);
+		goto err_cmd;
+	}
+	mw->enabled = MLX4_MPT_EN_HW;
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+
+	return 0;
+
+err_cmd:
+	mlx4_free_cmd_mailbox(dev, mailbox);
+
+err_table:
+	mlx4_mpt_free_icm(dev, key_to_hw_index(mw->key));
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_mw_enable);
+
+void mlx4_mw_free(struct mlx4_dev *dev, struct mlx4_mw *mw)
+{
+	int err;
+
+	if (mw->enabled == MLX4_MPT_EN_HW) {
+		err = mlx4_HW2SW_MPT(dev, NULL,
+				     key_to_hw_index(mw->key) &
+				     (dev->caps.num_mpts - 1));
+		if (err)
+			mlx4_warn(dev, "xxx HW2SW_MPT failed (%d)\n", err);
+
+		mw->enabled = MLX4_MPT_EN_SW;
+	}
+	if (mw->enabled)
+		mlx4_mpt_free_icm(dev, key_to_hw_index(mw->key));
+	mlx4_mpt_release(dev, key_to_hw_index(mw->key));
+}
+EXPORT_SYMBOL_GPL(mlx4_mw_free);
+
 int mlx4_init_mr_table(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
@@ -831,7 +923,7 @@ int mlx4_fmr_alloc(struct mlx4_dev *dev, u32 pd, u32 access, int max_pages,
 	return 0;
 
 err_free:
-	mlx4_mr_free(dev, &fmr->mr);
+	(void) mlx4_mr_free(dev, &fmr->mr);
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx4_fmr_alloc);
@@ -882,17 +974,21 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
 		       err);
 		return;
 	}
-	fmr->mr.enabled = MLX4_MR_EN_SW;
+	fmr->mr.enabled = MLX4_MPT_EN_SW;
 }
 EXPORT_SYMBOL_GPL(mlx4_fmr_unmap);
 
 int mlx4_fmr_free(struct mlx4_dev *dev, struct mlx4_fmr *fmr)
 {
+	int ret;
+
 	if (fmr->maps)
 		return -EBUSY;
 
-	mlx4_mr_free(dev, &fmr->mr);
-	fmr->mr.enabled = MLX4_MR_DISABLED;
+	ret = mlx4_mr_free(dev, &fmr->mr);
+	if (ret)
+		return ret;
+	fmr->mr.enabled = MLX4_MPT_DISABLED;
 
 	return 0;
 }

drivers/net/ethernet/mellanox/mlx4/resource_tracker.c

@@ -1231,14 +1231,14 @@ static int mpt_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 
 	switch (op) {
 	case RES_OP_RESERVE:
-		index = __mlx4_mr_reserve(dev);
+		index = __mlx4_mpt_reserve(dev);
 		if (index == -1)
 			break;
 		id = index & mpt_mask(dev);
 
 		err = add_res_range(dev, slave, id, 1, RES_MPT, index);
 		if (err) {
-			__mlx4_mr_release(dev, index);
+			__mlx4_mpt_release(dev, index);
 			break;
 		}
 		set_param_l(out_param, index);
@@ -1251,7 +1251,7 @@ static int mpt_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 		if (err)
 			return err;
 
-		err = __mlx4_mr_alloc_icm(dev, mpt->key);
+		err = __mlx4_mpt_alloc_icm(dev, mpt->key);
 		if (err) {
 			res_abort_move(dev, slave, RES_MPT, id);
 			return err;
@@ -1586,7 +1586,7 @@ static int mpt_free_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 		err = rem_res_range(dev, slave, id, 1, RES_MPT, 0);
 		if (err)
 			break;
-		__mlx4_mr_release(dev, index);
+		__mlx4_mpt_release(dev, index);
 		break;
 	case RES_OP_MAP_ICM:
 			index = get_param_l(&in_param);
@@ -1596,7 +1596,7 @@ static int mpt_free_res(struct mlx4_dev *dev, int slave, int op, int cmd,
 			if (err)
 				return err;
 
-			__mlx4_mr_free_icm(dev, mpt->key);
+			__mlx4_mpt_free_icm(dev, mpt->key);
 			res_end_move(dev, slave, RES_MPT, id);
 			return err;
 		break;
@@ -1796,6 +1796,26 @@ static int mr_get_mtt_size(struct mlx4_mpt_entry *mpt)
 	return be32_to_cpu(mpt->mtt_sz);
 }
 
+static u32 mr_get_pd(struct mlx4_mpt_entry *mpt)
+{
+	return be32_to_cpu(mpt->pd_flags) & 0x00ffffff;
+}
+
+static int mr_is_fmr(struct mlx4_mpt_entry *mpt)
+{
+	return be32_to_cpu(mpt->pd_flags) & MLX4_MPT_PD_FLAG_FAST_REG;
+}
+
+static int mr_is_bind_enabled(struct mlx4_mpt_entry *mpt)
+{
+	return be32_to_cpu(mpt->flags) & MLX4_MPT_FLAG_BIND_ENABLE;
+}
+
+static int mr_is_region(struct mlx4_mpt_entry *mpt)
+{
+	return be32_to_cpu(mpt->flags) & MLX4_MPT_FLAG_REGION;
+}
+
 static int qp_get_mtt_addr(struct mlx4_qp_context *qpc)
 {
 	return be32_to_cpu(qpc->mtt_base_addr_l) & 0xfffffff8;
@@ -1856,12 +1876,41 @@ int mlx4_SW2HW_MPT_wrapper(struct mlx4_dev *dev, int slave,
 	int mtt_base = mr_get_mtt_addr(inbox->buf) / dev->caps.mtt_entry_sz;
 	int phys;
 	int id;
+	u32 pd;
+	int pd_slave;
 
 	id = index & mpt_mask(dev);
 	err = mr_res_start_move_to(dev, slave, id, RES_MPT_HW, &mpt);
 	if (err)
 		return err;
 
+	/* Disable memory windows for VFs. */
+	if (!mr_is_region(inbox->buf)) {
+		err = -EPERM;
+		goto ex_abort;
+	}
+
+	/* Make sure that the PD bits related to the slave id are zeros. */
+	pd = mr_get_pd(inbox->buf);
+	pd_slave = (pd >> 17) & 0x7f;
+	if (pd_slave != 0 && pd_slave != slave) {
+		err = -EPERM;
+		goto ex_abort;
+	}
+
+	if (mr_is_fmr(inbox->buf)) {
+		/* FMR and Bind Enable are forbidden in slave devices. */
+		if (mr_is_bind_enabled(inbox->buf)) {
+			err = -EPERM;
+			goto ex_abort;
+		}
+		/* FMR and Memory Windows are also forbidden. */
+		if (!mr_is_region(inbox->buf)) {
+			err = -EPERM;
+			goto ex_abort;
+		}
+	}
+
 	phys = mr_phys_mpt(inbox->buf);
 	if (!phys) {
 		err = get_res(dev, slave, mtt_base, RES_MTT, &mtt);
@@ -3480,7 +3529,7 @@ static void rem_slave_mrs(struct mlx4_dev *dev, int slave)
 			while (state != 0) {
 				switch (state) {
 				case RES_MPT_RESERVED:
-					__mlx4_mr_release(dev, mpt->key);
+					__mlx4_mpt_release(dev, mpt->key);
 					spin_lock_irq(mlx4_tlock(dev));
 					rb_erase(&mpt->com.node,
 						 &tracker->res_tree[RES_MPT]);
@@ -3491,7 +3540,7 @@ static void rem_slave_mrs(struct mlx4_dev *dev, int slave)
 					break;
 
 				case RES_MPT_MAPPED:
-					__mlx4_mr_free_icm(dev, mpt->key);
+					__mlx4_mpt_free_icm(dev, mpt->key);
 					state = RES_MPT_RESERVED;
 					break;
 

include/linux/mlx4/device.h

@@ -171,6 +171,7 @@ enum {
 #define MLX4_ATTR_EXTENDED_PORT_INFO	cpu_to_be16(0xff90)
 
 enum {
+	MLX4_BMME_FLAG_WIN_TYPE_2B	= 1 <<  1,
 	MLX4_BMME_FLAG_LOCAL_INV	= 1 <<  6,
 	MLX4_BMME_FLAG_REMOTE_INV	= 1 <<  7,
 	MLX4_BMME_FLAG_TYPE_2_WIN	= 1 <<  9,
@@ -238,7 +239,8 @@ enum {
 	MLX4_PERM_LOCAL_WRITE	= 1 << 11,
 	MLX4_PERM_REMOTE_READ	= 1 << 12,
 	MLX4_PERM_REMOTE_WRITE	= 1 << 13,
-	MLX4_PERM_ATOMIC	= 1 << 14
+	MLX4_PERM_ATOMIC	= 1 << 14,
+	MLX4_PERM_BIND_MW	= 1 << 15,
 };
 
 enum {
@@ -504,6 +506,18 @@ struct mlx4_mr {
 	int			enabled;
 };
 
+enum mlx4_mw_type {
+	MLX4_MW_TYPE_1 = 1,
+	MLX4_MW_TYPE_2 = 2,
+};
+
+struct mlx4_mw {
+	u32			key;
+	u32			pd;
+	enum mlx4_mw_type	type;
+	int			enabled;
+};
+
 struct mlx4_fmr {
 	struct mlx4_mr		mr;
 	struct mlx4_mpt_entry  *mpt;
@@ -802,8 +816,12 @@ u64 mlx4_mtt_addr(struct mlx4_dev *dev, struct mlx4_mtt *mtt);
 
 int mlx4_mr_alloc(struct mlx4_dev *dev, u32 pd, u64 iova, u64 size, u32 access,
 		  int npages, int page_shift, struct mlx4_mr *mr);
-void mlx4_mr_free(struct mlx4_dev *dev, struct mlx4_mr *mr);
+int mlx4_mr_free(struct mlx4_dev *dev, struct mlx4_mr *mr);
 int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr);
+int mlx4_mw_alloc(struct mlx4_dev *dev, u32 pd, enum mlx4_mw_type type,
+		  struct mlx4_mw *mw);
+void mlx4_mw_free(struct mlx4_dev *dev, struct mlx4_mw *mw);
+int mlx4_mw_enable(struct mlx4_dev *dev, struct mlx4_mw *mw);
 int mlx4_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 		   int start_index, int npages, u64 *page_list);
 int mlx4_buf_write_mtt(struct mlx4_dev *dev, struct mlx4_mtt *mtt,

include/linux/mlx4/qp.h

@@ -265,6 +265,11 @@ struct mlx4_wqe_lso_seg {
 	__be32			header[0];
 };
 
+enum mlx4_wqe_bind_seg_flags2 {
+	MLX4_WQE_BIND_ZERO_BASED = (1 << 30),
+	MLX4_WQE_BIND_TYPE_2     = (1 << 31),
+};
+
 struct mlx4_wqe_bind_seg {
 	__be32			flags1;
 	__be32			flags2;
@@ -277,9 +282,9 @@ struct mlx4_wqe_bind_seg {
 enum {
 	MLX4_WQE_FMR_PERM_LOCAL_READ	= 1 << 27,
 	MLX4_WQE_FMR_PERM_LOCAL_WRITE	= 1 << 28,
-	MLX4_WQE_FMR_PERM_REMOTE_READ	= 1 << 29,
-	MLX4_WQE_FMR_PERM_REMOTE_WRITE	= 1 << 30,
-	MLX4_WQE_FMR_PERM_ATOMIC	= 1 << 31
+	MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_READ	= 1 << 29,
+	MLX4_WQE_FMR_AND_BIND_PERM_REMOTE_WRITE	= 1 << 30,
+	MLX4_WQE_FMR_AND_BIND_PERM_ATOMIC	= 1 << 31
 };
 
 struct mlx4_wqe_fmr_seg {
@@ -304,12 +309,10 @@ struct mlx4_wqe_fmr_ext_seg {
 };
 
 struct mlx4_wqe_local_inval_seg {
-	__be32			flags;
-	u32			reserved1;
+	u64			reserved1;
 	__be32			mem_key;
-	u32			reserved2[2];
-	__be32			guest_id;
-	__be64			pa;
+	u32			reserved2;
+	u64			reserved3[2];
 };
 
 struct mlx4_wqe_raddr_seg {

include/rdma/ib_verbs.h

@@ -115,6 +115,8 @@ enum ib_device_cap_flags {
 	IB_DEVICE_XRC			= (1<<20),
 	IB_DEVICE_MEM_MGT_EXTENSIONS	= (1<<21),
 	IB_DEVICE_BLOCK_MULTICAST_LOOPBACK = (1<<22),
+	IB_DEVICE_MEM_WINDOW_TYPE_2A	= (1<<23),
+	IB_DEVICE_MEM_WINDOW_TYPE_2B	= (1<<24)
 };
 
 enum ib_atomic_cap {
@@ -715,6 +717,11 @@ enum ib_mig_state {
 	IB_MIG_ARMED
 };
 
+enum ib_mw_type {
+	IB_MW_TYPE_1 = 1,
+	IB_MW_TYPE_2 = 2
+};
+
 struct ib_qp_attr {
 	enum ib_qp_state	qp_state;
 	enum ib_qp_state	cur_qp_state;
@@ -758,6 +765,7 @@ enum ib_wr_opcode {
 	IB_WR_FAST_REG_MR,
 	IB_WR_MASKED_ATOMIC_CMP_AND_SWP,
 	IB_WR_MASKED_ATOMIC_FETCH_AND_ADD,
+	IB_WR_BIND_MW,
 };
 
 enum ib_send_flags {
@@ -780,6 +788,23 @@ struct ib_fast_reg_page_list {
 	unsigned int		max_page_list_len;
 };
 
+/**
+ * struct ib_mw_bind_info - Parameters for a memory window bind operation.
+ * @mr: A memory region to bind the memory window to.
+ * @addr: The address where the memory window should begin.
+ * @length: The length of the memory window, in bytes.
+ * @mw_access_flags: Access flags from enum ib_access_flags for the window.
+ *
+ * This struct contains the shared parameters for type 1 and type 2
+ * memory window bind operations.
+ */
+struct ib_mw_bind_info {
+	struct ib_mr   *mr;
+	u64		addr;
+	u64		length;
+	int		mw_access_flags;
+};
+
 struct ib_send_wr {
 	struct ib_send_wr      *next;
 	u64			wr_id;
@@ -823,6 +848,12 @@ struct ib_send_wr {
 			int				access_flags;
 			u32				rkey;
 		} fast_reg;
+		struct {
+			struct ib_mw            *mw;
+			/* The new rkey for the memory window. */
+			u32                      rkey;
+			struct ib_mw_bind_info   bind_info;
+		} bind_mw;
 	} wr;
 	u32			xrc_remote_srq_num;	/* XRC TGT QPs only */
 };
@@ -839,7 +870,8 @@ enum ib_access_flags {
 	IB_ACCESS_REMOTE_WRITE	= (1<<1),
 	IB_ACCESS_REMOTE_READ	= (1<<2),
 	IB_ACCESS_REMOTE_ATOMIC	= (1<<3),
-	IB_ACCESS_MW_BIND	= (1<<4)
+	IB_ACCESS_MW_BIND	= (1<<4),
+	IB_ZERO_BASED		= (1<<5)
 };
 
 struct ib_phys_buf {
@@ -862,13 +894,16 @@ enum ib_mr_rereg_flags {
 	IB_MR_REREG_ACCESS	= (1<<2)
 };
 
+/**
+ * struct ib_mw_bind - Parameters for a type 1 memory window bind operation.
+ * @wr_id:      Work request id.
+ * @send_flags: Flags from ib_send_flags enum.
+ * @bind_info:  More parameters of the bind operation.
+ */
 struct ib_mw_bind {
-	struct ib_mr   *mr;
-	u64		wr_id;
-	u64		addr;
-	u32		length;
-	int		send_flags;
-	int		mw_access_flags;
+	u64                    wr_id;
+	int                    send_flags;
+	struct ib_mw_bind_info bind_info;
 };
 
 struct ib_fmr_attr {
@@ -991,6 +1026,7 @@ struct ib_mw {
 	struct ib_pd		*pd;
 	struct ib_uobject	*uobject;
 	u32			rkey;
+	enum ib_mw_type         type;
 };
 
 struct ib_fmr {
@@ -1202,7 +1238,8 @@ struct ib_device {
 						    int num_phys_buf,
 						    int mr_access_flags,
 						    u64 *iova_start);
-	struct ib_mw *             (*alloc_mw)(struct ib_pd *pd);
+	struct ib_mw *             (*alloc_mw)(struct ib_pd *pd,
+					       enum ib_mw_type type);
 	int                        (*bind_mw)(struct ib_qp *qp,
 					      struct ib_mw *mw,
 					      struct ib_mw_bind *mw_bind);
@@ -2019,6 +2056,8 @@ int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr);
  * ib_dereg_mr - Deregisters a memory region and removes it from the
  *   HCA translation table.
  * @mr: The memory region to deregister.
+ *
+ * This function can fail, if the memory region has memory windows bound to it.
  */
 int ib_dereg_mr(struct ib_mr *mr);
 
@@ -2070,11 +2109,23 @@ static inline void ib_update_fast_reg_key(struct ib_mr *mr, u8 newkey)
 	mr->rkey = (mr->rkey & 0xffffff00) | newkey;
 }
 
+/**
+ * ib_inc_rkey - increments the key portion of the given rkey. Can be used
+ * for calculating a new rkey for type 2 memory windows.
+ * @rkey - the rkey to increment.
+ */
+static inline u32 ib_inc_rkey(u32 rkey)
+{
+	const u32 mask = 0x000000ff;
+	return ((rkey + 1) & mask) | (rkey & ~mask);
+}
+
 /**
  * ib_alloc_mw - Allocates a memory window.
  * @pd: The protection domain associated with the memory window.
+ * @type: The type of the memory window (1 or 2).
  */
-struct ib_mw *ib_alloc_mw(struct ib_pd *pd);
+struct ib_mw *ib_alloc_mw(struct ib_pd *pd, enum ib_mw_type type);
 
 /**
  * ib_bind_mw - Posts a work request to the send queue of the specified
@@ -2084,6 +2135,10 @@ struct ib_mw *ib_alloc_mw(struct ib_pd *pd);
  * @mw: The memory window to bind.
  * @mw_bind: Specifies information about the memory window, including
  *   its address range, remote access rights, and associated memory region.
+ *
+ * If there is no immediate error, the function will update the rkey member
+ * of the mw parameter to its new value. The bind operation can still fail
+ * asynchronously.
  */
 static inline int ib_bind_mw(struct ib_qp *qp,
 			     struct ib_mw *mw,

include/uapi/rdma/ib_user_verbs.h

@@ -261,6 +261,22 @@ struct ib_uverbs_dereg_mr {
 	__u32 mr_handle;
 };
 
+struct ib_uverbs_alloc_mw {
+	__u64 response;
+	__u32 pd_handle;
+	__u8  mw_type;
+	__u8  reserved[3];
+};
+
+struct ib_uverbs_alloc_mw_resp {
+	__u32 mw_handle;
+	__u32 rkey;
+};
+
+struct ib_uverbs_dealloc_mw {
+	__u32 mw_handle;
+};
+
 struct ib_uverbs_create_comp_channel {
 	__u64 response;
 };

net/sunrpc/xprtrdma/verbs.c

@@ -1086,7 +1086,7 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep,
 	case RPCRDMA_MEMWINDOWS:
 		/* Allocate one extra request's worth, for full cycling */
 		for (i = (buf->rb_max_requests+1) * RPCRDMA_MAX_SEGS; i; i--) {
-			r->r.mw = ib_alloc_mw(ia->ri_pd);
+			r->r.mw = ib_alloc_mw(ia->ri_pd, IB_MW_TYPE_1);
 			if (IS_ERR(r->r.mw)) {
 				rc = PTR_ERR(r->r.mw);
 				dprintk("RPC:       %s: ib_alloc_mw"
@@ -1673,12 +1673,12 @@ rpcrdma_register_memwin_external(struct rpcrdma_mr_seg *seg,
 
 	*nsegs = 1;
 	rpcrdma_map_one(ia, seg, writing);
-	param.mr = ia->ri_bind_mem;
+	param.bind_info.mr = ia->ri_bind_mem;
 	param.wr_id = 0ULL;	/* no send cookie */
-	param.addr = seg->mr_dma;
-	param.length = seg->mr_len;
+	param.bind_info.addr = seg->mr_dma;
+	param.bind_info.length = seg->mr_len;
 	param.send_flags = 0;
-	param.mw_access_flags = mem_priv;
+	param.bind_info.mw_access_flags = mem_priv;
 
 	DECR_CQCOUNT(&r_xprt->rx_ep);
 	rc = ib_bind_mw(ia->ri_id->qp, seg->mr_chunk.rl_mw->r.mw, &param);
@@ -1690,7 +1690,7 @@ rpcrdma_register_memwin_external(struct rpcrdma_mr_seg *seg,
 		rpcrdma_unmap_one(ia, seg);
 	} else {
 		seg->mr_rkey = seg->mr_chunk.rl_mw->r.mw->rkey;
-		seg->mr_base = param.addr;
+		seg->mr_base = param.bind_info.addr;
 		seg->mr_nsegs = 1;
 	}
 	return rc;
@@ -1706,10 +1706,10 @@ rpcrdma_deregister_memwin_external(struct rpcrdma_mr_seg *seg,
 	int rc;
 
 	BUG_ON(seg->mr_nsegs != 1);
-	param.mr = ia->ri_bind_mem;
-	param.addr = 0ULL;	/* unbind */
-	param.length = 0;
-	param.mw_access_flags = 0;
+	param.bind_info.mr = ia->ri_bind_mem;
+	param.bind_info.addr = 0ULL;	/* unbind */
+	param.bind_info.length = 0;
+	param.bind_info.mw_access_flags = 0;
 	if (*r) {
 		param.wr_id = (u64) (unsigned long) *r;
 		param.send_flags = IB_SEND_SIGNALED;