11 files changed, 34 insertions, 654 deletions

diff --git a/drivers/staging/rdma/hfi1/Makefile b/drivers/staging/rdma/hfi1/Makefile
index 55077f396cf9..7797f2c0f01f 100644
--- a/drivers/staging/rdma/hfi1/Makefile
+++ b/drivers/staging/rdma/hfi1/Makefile
@@ -7,7 +7,7 @@
 #
 obj-$(CONFIG_INFINIBAND_HFI1) += hfi1.o
 
-hfi1-y := chip.o cq.o device.o diag.o driver.o efivar.o eprom.o file_ops.o firmware.o \
+hfi1-y := chip.o device.o diag.o driver.o efivar.o eprom.o file_ops.o firmware.o \
 	init.o intr.o mad.o pcie.o pio.o pio_copy.o \
 	qp.o qsfp.o rc.o ruc.o sdma.o srq.o sysfs.o trace.o twsi.o \
 	uc.o ud.o user_exp_rcv.o user_pages.o user_sdma.o verbs_mcast.o verbs.o
diff --git a/drivers/staging/rdma/hfi1/cq.c b/drivers/staging/rdma/hfi1/cq.c
deleted file mode 100644
index 25d1a2a25fee..000000000000
--- a/drivers/staging/rdma/hfi1/cq.c
+++ /dev/null
@@ -1,558 +0,0 @@
-/*
- *
- * This file is provided under a dual BSD/GPLv2 license.  When using or
- * redistributing this file, you may do so under either license.
- *
- * GPL LICENSE SUMMARY
- *
- * Copyright(c) 2015 Intel Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * General Public License for more details.
- *
- * BSD LICENSE
- *
- * Copyright(c) 2015 Intel Corporation.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *
- *  - Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- *  - Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in
- *    the documentation and/or other materials provided with the
- *    distribution.
- *  - Neither the name of Intel Corporation nor the names of its
- *    contributors may be used to endorse or promote products derived
- *    from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
- * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
- * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
- * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
- * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
- * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
- * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
- * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
- * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- */
-
-#include <linux/err.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <linux/kthread.h>
-
-#include "verbs.h"
-#include "hfi.h"
-
-/**
- * hfi1_cq_enter - add a new entry to the completion queue
- * @cq: completion queue
- * @entry: work completion entry to add
- * @sig: true if @entry is a solicited entry
- *
- * This may be called with qp->s_lock held.
- */
-void hfi1_cq_enter(struct hfi1_cq *cq, struct ib_wc *entry, int solicited)
-{
-	struct hfi1_cq_wc *wc;
-	unsigned long flags;
-	u32 head;
-	u32 next;
-
-	spin_lock_irqsave(&cq->lock, flags);
-
-	/*
-	 * Note that the head pointer might be writable by user processes.
-	 * Take care to verify it is a sane value.
-	 */
-	wc = cq->queue;
-	head = wc->head;
-	if (head >= (unsigned) cq->ibcq.cqe) {
-		head = cq->ibcq.cqe;
-		next = 0;
-	} else
-		next = head + 1;
-	if (unlikely(next == wc->tail)) {
-		spin_unlock_irqrestore(&cq->lock, flags);
-		if (cq->ibcq.event_handler) {
-			struct ib_event ev;
-
-			ev.device = cq->ibcq.device;
-			ev.element.cq = &cq->ibcq;
-			ev.event = IB_EVENT_CQ_ERR;
-			cq->ibcq.event_handler(&ev, cq->ibcq.cq_context);
-		}
-		return;
-	}
-	if (cq->ip) {
-		wc->uqueue[head].wr_id = entry->wr_id;
-		wc->uqueue[head].status = entry->status;
-		wc->uqueue[head].opcode = entry->opcode;
-		wc->uqueue[head].vendor_err = entry->vendor_err;
-		wc->uqueue[head].byte_len = entry->byte_len;
-		wc->uqueue[head].ex.imm_data =
-			(__u32 __force)entry->ex.imm_data;
-		wc->uqueue[head].qp_num = entry->qp->qp_num;
-		wc->uqueue[head].src_qp = entry->src_qp;
-		wc->uqueue[head].wc_flags = entry->wc_flags;
-		wc->uqueue[head].pkey_index = entry->pkey_index;
-		wc->uqueue[head].slid = entry->slid;
-		wc->uqueue[head].sl = entry->sl;
-		wc->uqueue[head].dlid_path_bits = entry->dlid_path_bits;
-		wc->uqueue[head].port_num = entry->port_num;
-		/* Make sure entry is written before the head index. */
-		smp_wmb();
-	} else
-		wc->kqueue[head] = *entry;
-	wc->head = next;
-
-	if (cq->notify == IB_CQ_NEXT_COMP ||
-	    (cq->notify == IB_CQ_SOLICITED &&
-	     (solicited || entry->status != IB_WC_SUCCESS))) {
-		struct kthread_worker *worker;
-		/*
-		 * This will cause send_complete() to be called in
-		 * another thread.
-		 */
-		smp_read_barrier_depends(); /* see hfi1_cq_exit */
-		worker = cq->dd->worker;
-		if (likely(worker)) {
-			cq->notify = IB_CQ_NONE;
-			cq->triggered++;
-			queue_kthread_work(worker, &cq->comptask);
-		}
-	}
-
-	spin_unlock_irqrestore(&cq->lock, flags);
-}
-
-/**
- * hfi1_poll_cq - poll for work completion entries
- * @ibcq: the completion queue to poll
- * @num_entries: the maximum number of entries to return
- * @entry: pointer to array where work completions are placed
- *
- * Returns the number of completion entries polled.
- *
- * This may be called from interrupt context.  Also called by ib_poll_cq()
- * in the generic verbs code.
- */
-int hfi1_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry)
-{
-	struct hfi1_cq *cq = to_icq(ibcq);
-	struct hfi1_cq_wc *wc;
-	unsigned long flags;
-	int npolled;
-	u32 tail;
-
-	/* The kernel can only poll a kernel completion queue */
-	if (cq->ip) {
-		npolled = -EINVAL;
-		goto bail;
-	}
-
-	spin_lock_irqsave(&cq->lock, flags);
-
-	wc = cq->queue;
-	tail = wc->tail;
-	if (tail > (u32) cq->ibcq.cqe)
-		tail = (u32) cq->ibcq.cqe;
-	for (npolled = 0; npolled < num_entries; ++npolled, ++entry) {
-		if (tail == wc->head)
-			break;
-		/* The kernel doesn't need a RMB since it has the lock. */
-		*entry = wc->kqueue[tail];
-		if (tail >= cq->ibcq.cqe)
-			tail = 0;
-		else
-			tail++;
-	}
-	wc->tail = tail;
-
-	spin_unlock_irqrestore(&cq->lock, flags);
-
-bail:
-	return npolled;
-}
-
-static void send_complete(struct kthread_work *work)
-{
-	struct hfi1_cq *cq = container_of(work, struct hfi1_cq, comptask);
-
-	/*
-	 * The completion handler will most likely rearm the notification
-	 * and poll for all pending entries.  If a new completion entry
-	 * is added while we are in this routine, queue_work()
-	 * won't call us again until we return so we check triggered to
-	 * see if we need to call the handler again.
-	 */
-	for (;;) {
-		u8 triggered = cq->triggered;
-
-		/*
-		 * IPoIB connected mode assumes the callback is from a
-		 * soft IRQ. We simulate this by blocking "bottom halves".
-		 * See the implementation for ipoib_cm_handle_tx_wc(),
-		 * netif_tx_lock_bh() and netif_tx_lock().
-		 */
-		local_bh_disable();
-		cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context);
-		local_bh_enable();
-
-		if (cq->triggered == triggered)
-			return;
-	}
-}
-
-/**
- * hfi1_create_cq - create a completion queue
- * @ibdev: the device this completion queue is attached to
- * @attr: creation attributes
- * @context: unused by the driver
- * @udata: user data for libibverbs.so
- *
- * Returns a pointer to the completion queue or negative errno values
- * for failure.
- *
- * Called by ib_create_cq() in the generic verbs code.
- */
-struct ib_cq *hfi1_create_cq(
-	struct ib_device *ibdev,
-	const struct ib_cq_init_attr *attr,
-	struct ib_ucontext *context,
-	struct ib_udata *udata)
-{
-	struct hfi1_ibdev *dev = to_idev(ibdev);
-	struct hfi1_cq *cq;
-	struct hfi1_cq_wc *wc;
-	struct ib_cq *ret;
-	u32 sz;
-	unsigned int entries = attr->cqe;
-
-	if (attr->flags)
-		return ERR_PTR(-EINVAL);
-
-	if (entries < 1 || entries > hfi1_max_cqes)
-		return ERR_PTR(-EINVAL);
-
-	/* Allocate the completion queue structure. */
-	cq = kmalloc(sizeof(*cq), GFP_KERNEL);
-	if (!cq)
-		return ERR_PTR(-ENOMEM);
-
-	/*
-	 * Allocate the completion queue entries and head/tail pointers.
-	 * This is allocated separately so that it can be resized and
-	 * also mapped into user space.
-	 * We need to use vmalloc() in order to support mmap and large
-	 * numbers of entries.
-	 */
-	sz = sizeof(*wc);
-	if (udata && udata->outlen >= sizeof(__u64))
-		sz += sizeof(struct ib_uverbs_wc) * (entries + 1);
-	else
-		sz += sizeof(struct ib_wc) * (entries + 1);
-	wc = vmalloc_user(sz);
-	if (!wc) {
-		ret = ERR_PTR(-ENOMEM);
-		goto bail_cq;
-	}
-
-	/*
-	 * Return the address of the WC as the offset to mmap.
-	 * See hfi1_mmap() for details.
-	 */
-	if (udata && udata->outlen >= sizeof(__u64)) {
-		int err;
-
-		cq->ip = rvt_create_mmap_info(&dev->rdi, sz, context, wc);
-		if (!cq->ip) {
-			ret = ERR_PTR(-ENOMEM);
-			goto bail_wc;
-		}
-
-		err = ib_copy_to_udata(udata, &cq->ip->offset,
-				       sizeof(cq->ip->offset));
-		if (err) {
-			ret = ERR_PTR(err);
-			goto bail_ip;
-		}
-	} else
-		cq->ip = NULL;
-
-	spin_lock(&dev->n_cqs_lock);
-	if (dev->n_cqs_allocated == hfi1_max_cqs) {
-		spin_unlock(&dev->n_cqs_lock);
-		ret = ERR_PTR(-ENOMEM);
-		goto bail_ip;
-	}
-
-	dev->n_cqs_allocated++;
-	spin_unlock(&dev->n_cqs_lock);
-
-	if (cq->ip) {
-		spin_lock_irq(&dev->rdi.pending_lock);
-		list_add(&cq->ip->pending_mmaps, &dev->rdi.pending_mmaps);
-		spin_unlock_irq(&dev->rdi.pending_lock);
-	}
-
-	/*
-	 * ib_create_cq() will initialize cq->ibcq except for cq->ibcq.cqe.
-	 * The number of entries should be >= the number requested or return
-	 * an error.
-	 */
-	cq->dd = dd_from_dev(dev);
-	cq->ibcq.cqe = entries;
-	cq->notify = IB_CQ_NONE;
-	cq->triggered = 0;
-	spin_lock_init(&cq->lock);
-	init_kthread_work(&cq->comptask, send_complete);
-	wc->head = 0;
-	wc->tail = 0;
-	cq->queue = wc;
-
-	ret = &cq->ibcq;
-
-	goto done;
-
-bail_ip:
-	kfree(cq->ip);
-bail_wc:
-	vfree(wc);
-bail_cq:
-	kfree(cq);
-done:
-	return ret;
-}
-
-/**
- * hfi1_destroy_cq - destroy a completion queue
- * @ibcq: the completion queue to destroy.
- *
- * Returns 0 for success.
- *
- * Called by ib_destroy_cq() in the generic verbs code.
- */
-int hfi1_destroy_cq(struct ib_cq *ibcq)
-{
-	struct hfi1_ibdev *dev = to_idev(ibcq->device);
-	struct hfi1_cq *cq = to_icq(ibcq);
-
-	flush_kthread_work(&cq->comptask);
-	spin_lock(&dev->n_cqs_lock);
-	dev->n_cqs_allocated--;
-	spin_unlock(&dev->n_cqs_lock);
-	if (cq->ip)
-		kref_put(&cq->ip->ref, rvt_release_mmap_info);
-	else
-		vfree(cq->queue);
-	kfree(cq);
-
-	return 0;
-}
-
-/**
- * hfi1_req_notify_cq - change the notification type for a completion queue
- * @ibcq: the completion queue
- * @notify_flags: the type of notification to request
- *
- * Returns 0 for success.
- *
- * This may be called from interrupt context.  Also called by
- * ib_req_notify_cq() in the generic verbs code.
- */
-int hfi1_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags notify_flags)
-{
-	struct hfi1_cq *cq = to_icq(ibcq);
-	unsigned long flags;
-	int ret = 0;
-
-	spin_lock_irqsave(&cq->lock, flags);
-	/*
-	 * Don't change IB_CQ_NEXT_COMP to IB_CQ_SOLICITED but allow
-	 * any other transitions (see C11-31 and C11-32 in ch. 11.4.2.2).
-	 */
-	if (cq->notify != IB_CQ_NEXT_COMP)
-		cq->notify = notify_flags & IB_CQ_SOLICITED_MASK;
-
-	if ((notify_flags & IB_CQ_REPORT_MISSED_EVENTS) &&
-	    cq->queue->head != cq->queue->tail)
-		ret = 1;
-
-	spin_unlock_irqrestore(&cq->lock, flags);
-
-	return ret;
-}
-
-/**
- * hfi1_resize_cq - change the size of the CQ
- * @ibcq: the completion queue
- *
- * Returns 0 for success.
- */
-int hfi1_resize_cq(struct ib_cq *ibcq, int cqe, struct ib_udata *udata)
-{
-	struct hfi1_cq *cq = to_icq(ibcq);
-	struct hfi1_cq_wc *old_wc;
-	struct hfi1_cq_wc *wc;
-	u32 head, tail, n;
-	int ret;
-	u32 sz;
-
-	if (cqe < 1 || cqe > hfi1_max_cqes) {
-		ret = -EINVAL;
-		goto bail;
-	}
-
-	/*
-	 * Need to use vmalloc() if we want to support large #s of entries.
-	 */
-	sz = sizeof(*wc);
-	if (udata && udata->outlen >= sizeof(__u64))
-		sz += sizeof(struct ib_uverbs_wc) * (cqe + 1);
-	else
-		sz += sizeof(struct ib_wc) * (cqe + 1);
-	wc = vmalloc_user(sz);
-	if (!wc) {
-		ret = -ENOMEM;
-		goto bail;
-	}
-
-	/* Check that we can write the offset to mmap. */
-	if (udata && udata->outlen >= sizeof(__u64)) {
-		__u64 offset = 0;
-
-		ret = ib_copy_to_udata(udata, &offset, sizeof(offset));
-		if (ret)
-			goto bail_free;
-	}
-
-	spin_lock_irq(&cq->lock);
-	/*
-	 * Make sure head and tail are sane since they
-	 * might be user writable.
-	 */
-	old_wc = cq->queue;
-	head = old_wc->head;
-	if (head > (u32) cq->ibcq.cqe)
-		head = (u32) cq->ibcq.cqe;
-	tail = old_wc->tail;
-	if (tail > (u32) cq->ibcq.cqe)
-		tail = (u32) cq->ibcq.cqe;
-	if (head < tail)
-		n = cq->ibcq.cqe + 1 + head - tail;
-	else
-		n = head - tail;
-	if (unlikely((u32)cqe < n)) {
-		ret = -EINVAL;
-		goto bail_unlock;
-	}
-	for (n = 0; tail != head; n++) {
-		if (cq->ip)
-			wc->uqueue[n] = old_wc->uqueue[tail];
-		else
-			wc->kqueue[n] = old_wc->kqueue[tail];
-		if (tail == (u32) cq->ibcq.cqe)
-			tail = 0;
-		else
-			tail++;
-	}
-	cq->ibcq.cqe = cqe;
-	wc->head = n;
-	wc->tail = 0;
-	cq->queue = wc;
-	spin_unlock_irq(&cq->lock);
-
-	vfree(old_wc);
-
-	if (cq->ip) {
-		struct hfi1_ibdev *dev = to_idev(ibcq->device);
-		struct rvt_mmap_info *ip = cq->ip;
-
-		rvt_update_mmap_info(&dev->rdi, ip, sz, wc);
-
-		/*
-		 * Return the offset to mmap.
-		 * See hfi1_mmap() for details.
-		 */
-		if (udata && udata->outlen >= sizeof(__u64)) {
-			ret = ib_copy_to_udata(udata, &ip->offset,
-					       sizeof(ip->offset));
-			if (ret)
-				goto bail;
-		}
-
-		spin_lock_irq(&dev->rdi.pending_lock);
-		if (list_empty(&ip->pending_mmaps))
-			list_add(&ip->pending_mmaps, &dev->rdi.pending_mmaps);
-		spin_unlock_irq(&dev->rdi.pending_lock);
-	}
-
-	ret = 0;
-	goto bail;
-
-bail_unlock:
-	spin_unlock_irq(&cq->lock);
-bail_free:
-	vfree(wc);
-bail:
-	return ret;
-}
-
-int hfi1_cq_init(struct hfi1_devdata *dd)
-{
-	int ret = 0;
-	int cpu;
-	struct task_struct *task;
-
-	if (dd->worker)
-		return 0;
-	dd->worker = kzalloc(sizeof(*dd->worker), GFP_KERNEL);
-	if (!dd->worker)
-		return -ENOMEM;
-	init_kthread_worker(dd->worker);
-	task = kthread_create_on_node(
-		kthread_worker_fn,
-		dd->worker,
-		dd->assigned_node_id,
-		"hfi1_cq%d", dd->unit);
-	if (IS_ERR(task))
-		goto task_fail;
-	cpu = cpumask_first(cpumask_of_node(dd->assigned_node_id));
-	kthread_bind(task, cpu);
-	wake_up_process(task);
-out:
-	return ret;
-task_fail:
-	ret = PTR_ERR(task);
-	kfree(dd->worker);
-	dd->worker = NULL;
-	goto out;
-}
-
-void hfi1_cq_exit(struct hfi1_devdata *dd)
-{
-	struct kthread_worker *worker;
-
-	worker = dd->worker;
-	if (!worker)
-		return;
-	/* blocks future queuing from send_complete() */
-	dd->worker = NULL;
-	smp_wmb(); /* See hfi1_cq_enter */
-	flush_kthread_worker(worker);
-	kthread_stop(worker->task);
-	kfree(worker);
-}
diff --git a/drivers/staging/rdma/hfi1/hfi.h b/drivers/staging/rdma/hfi1/hfi.h
index e5f3451fecc2..e6a5fede0c02 100644
--- a/drivers/staging/rdma/hfi1/hfi.h
+++ b/drivers/staging/rdma/hfi1/hfi.h
@@ -1009,8 +1009,6 @@ struct hfi1_devdata {
 	u16 psxmitwait_check_rate;
 	/* high volume overflow errors deferred to tasklet */
 	struct tasklet_struct error_tasklet;
-	/* per device cq worker */
-	struct kthread_worker *worker;
 
 	/* MSI-X information */
 	struct hfi1_msix_entry *msix_entries;
diff --git a/drivers/staging/rdma/hfi1/init.c b/drivers/staging/rdma/hfi1/init.c
index b4076b22af14..d1cb2c854f9e 100644
--- a/drivers/staging/rdma/hfi1/init.c
+++ b/drivers/staging/rdma/hfi1/init.c
@@ -765,7 +765,6 @@ int hfi1_init(struct hfi1_devdata *dd, int reinit)
 	/* enable chip even if we have an error, so we can debug cause */
 	enable_chip(dd);
 
-	ret = hfi1_cq_init(dd);
 done:
 	/*
 	 * Set status even if port serdes is not initialized
@@ -1312,7 +1311,6 @@ static void cleanup_device_data(struct hfi1_devdata *dd)
 	kfree(dd->boardname);
 	vfree(dd->events);
 	vfree(dd->status);
-	hfi1_cq_exit(dd);
 }
 
 /*
diff --git a/drivers/staging/rdma/hfi1/qp.c b/drivers/staging/rdma/hfi1/qp.c
index a336d2a40d58..5e50dea12de9 100644
--- a/drivers/staging/rdma/hfi1/qp.c
+++ b/drivers/staging/rdma/hfi1/qp.c
@@ -304,7 +304,7 @@ int hfi1_error_qp(struct rvt_qp *qp, enum ib_wc_status err)
 	if (test_and_clear_bit(RVT_R_WRID_VALID, &qp->r_aflags)) {
 		wc.wr_id = qp->r_wr_id;
 		wc.status = err;
-		hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc, 1);
+		rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
 	}
 	wc.status = IB_WC_WR_FLUSH_ERR;
 
@@ -327,7 +327,7 @@ int hfi1_error_qp(struct rvt_qp *qp, enum ib_wc_status err)
 			wc.wr_id = get_rwqe_ptr(&qp->r_rq, tail)->wr_id;
 			if (++tail >= qp->r_rq.size)
 				tail = 0;
-			hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc, 1);
+			rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
 		}
 		wq->tail = tail;
 
diff --git a/drivers/staging/rdma/hfi1/rc.c b/drivers/staging/rdma/hfi1/rc.c
index bd504decc46d..e80a09261dcc 100644
--- a/drivers/staging/rdma/hfi1/rc.c
+++ b/drivers/staging/rdma/hfi1/rc.c
@@ -1040,7 +1040,7 @@ void hfi1_rc_send_complete(struct rvt_qp *qp, struct hfi1_ib_header *hdr)
 			wc.opcode = ib_hfi1_wc_opcode[wqe->wr.opcode];
 			wc.byte_len = wqe->length;
 			wc.qp = &qp->ibqp;
-			hfi1_cq_enter(to_icq(qp->ibqp.send_cq), &wc, 0);
+			rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0);
 		}
 		if (++qp->s_last >= qp->s_size)
 			qp->s_last = 0;
@@ -1097,7 +1097,7 @@ static struct rvt_swqe *do_rc_completion(struct rvt_qp *qp,
 			wc.opcode = ib_hfi1_wc_opcode[wqe->wr.opcode];
 			wc.byte_len = wqe->length;
 			wc.qp = &qp->ibqp;
-			hfi1_cq_enter(to_icq(qp->ibqp.send_cq), &wc, 0);
+			rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc, 0);
 		}
 		if (++qp->s_last >= qp->s_size)
 			qp->s_last = 0;
@@ -2157,8 +2157,8 @@ send_last:
 		wc.dlid_path_bits = 0;
 		wc.port_num = 0;
 		/* Signal completion event if the solicited bit is set. */
-		hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc,
-			      (bth0 & IB_BTH_SOLICITED) != 0);
+		rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
+			     (bth0 & IB_BTH_SOLICITED) != 0);
 		break;
 
 	case OP(RDMA_WRITE_FIRST):
diff --git a/drivers/staging/rdma/hfi1/ruc.c b/drivers/staging/rdma/hfi1/ruc.c
index 0b324b17bf09..c659cf806992 100644
--- a/drivers/staging/rdma/hfi1/ruc.c
+++ b/drivers/staging/rdma/hfi1/ruc.c
@@ -138,7 +138,7 @@ bad_lkey:
 	wc.opcode = IB_WC_RECV;
 	wc.qp = &qp->ibqp;
 	/* Signal solicited completion event. */
-	hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc, 1);
+	rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
 	ret = 0;
 bail:
 	return ret;
@@ -566,8 +566,8 @@ again:
 	wc.sl = qp->remote_ah_attr.sl;
 	wc.port_num = 1;
 	/* Signal completion event if the solicited bit is set. */
-	hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc,
-		      wqe->wr.send_flags & IB_SEND_SOLICITED);
+	rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
+		     wqe->wr.send_flags & IB_SEND_SOLICITED);
 
 send_comp:
 	spin_lock_irqsave(&sqp->s_lock, flags);
@@ -909,8 +909,8 @@ void hfi1_send_complete(struct rvt_qp *qp, struct rvt_swqe *wqe,
 		wc.qp = &qp->ibqp;
 		if (status == IB_WC_SUCCESS)
 			wc.byte_len = wqe->length;
-		hfi1_cq_enter(to_icq(qp->ibqp.send_cq), &wc,
-			      status != IB_WC_SUCCESS);
+		rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.send_cq), &wc,
+			     status != IB_WC_SUCCESS);
 	}
 
 	last = qp->s_last;
diff --git a/drivers/staging/rdma/hfi1/uc.c b/drivers/staging/rdma/hfi1/uc.c
index 0935182d4ac9..75cded36ec17 100644
--- a/drivers/staging/rdma/hfi1/uc.c
+++ b/drivers/staging/rdma/hfi1/uc.c
@@ -469,9 +469,9 @@ last_imm:
 		wc.dlid_path_bits = 0;
 		wc.port_num = 0;
 		/* Signal completion event if the solicited bit is set. */
-		hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc,
-			      (ohdr->bth[0] &
-				cpu_to_be32(IB_BTH_SOLICITED)) != 0);
+		rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
+			     (ohdr->bth[0] &
+			      cpu_to_be32(IB_BTH_SOLICITED)) != 0);
 		break;
 
 	case OP(RDMA_WRITE_FIRST):
diff --git a/drivers/staging/rdma/hfi1/ud.c b/drivers/staging/rdma/hfi1/ud.c
index a0e62229d7a1..a4746e83999d 100644
--- a/drivers/staging/rdma/hfi1/ud.c
+++ b/drivers/staging/rdma/hfi1/ud.c
@@ -247,8 +247,8 @@ static void ud_loopback(struct rvt_qp *sqp, struct rvt_swqe *swqe)
 	wc.dlid_path_bits = ah_attr->dlid & ((1 << ppd->lmc) - 1);
 	wc.port_num = qp->port_num;
 	/* Signal completion event if the solicited bit is set. */
-	hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc,
-		      swqe->wr.send_flags & IB_SEND_SOLICITED);
+	rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
+		     swqe->wr.send_flags & IB_SEND_SOLICITED);
 	ibp->rvp.n_loop_pkts++;
 bail_unlock:
 	spin_unlock_irqrestore(&qp->r_lock, flags);
@@ -878,9 +878,9 @@ void hfi1_ud_rcv(struct hfi1_packet *packet)
 		dlid & ((1 << ppd_from_ibp(ibp)->lmc) - 1);
 	wc.port_num = qp->port_num;
 	/* Signal completion event if the solicited bit is set. */
-	hfi1_cq_enter(to_icq(qp->ibqp.recv_cq), &wc,
-		      (ohdr->bth[0] &
-			cpu_to_be32(IB_BTH_SOLICITED)) != 0);
+	rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc,
+		     (ohdr->bth[0] &
+		      cpu_to_be32(IB_BTH_SOLICITED)) != 0);
 	return;
 
 drop:
diff --git a/drivers/staging/rdma/hfi1/verbs.c b/drivers/staging/rdma/hfi1/verbs.c
index 70af487c990f..ee969d0dcd77 100644
--- a/drivers/staging/rdma/hfi1/verbs.c
+++ b/drivers/staging/rdma/hfi1/verbs.c
@@ -1719,7 +1719,6 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd)
 
 	/* Only need to initialize non-zero fields. */
 
-	spin_lock_init(&dev->n_cqs_lock);
 	spin_lock_init(&dev->n_qps_lock);
 	spin_lock_init(&dev->n_srqs_lock);
 	spin_lock_init(&dev->n_mcast_grps_lock);
@@ -1816,11 +1815,11 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd)
 	ibdev->post_send = post_send;
 	ibdev->post_recv = post_receive;
 	ibdev->post_srq_recv = hfi1_post_srq_receive;
-	ibdev->create_cq = hfi1_create_cq;
-	ibdev->destroy_cq = hfi1_destroy_cq;
-	ibdev->resize_cq = hfi1_resize_cq;
-	ibdev->poll_cq = hfi1_poll_cq;
-	ibdev->req_notify_cq = hfi1_req_notify_cq;
+	ibdev->create_cq = NULL;
+	ibdev->destroy_cq = NULL;
+	ibdev->resize_cq = NULL;
+	ibdev->poll_cq = NULL;
+	ibdev->req_notify_cq = NULL;
 	ibdev->get_dma_mr = NULL;
 	ibdev->reg_user_mr = NULL;
 	ibdev->dereg_mr = NULL;
@@ -1860,14 +1859,20 @@ int hfi1_register_ib_device(struct hfi1_devdata *dd)
 	dd->verbs_dev.rdi.dparms.qos_shift = dd->qos_shift;
 	dd->verbs_dev.rdi.dparms.qpn_res_start = kdeth_qp << 16;
 	dd->verbs_dev.rdi.dparms.qpn_res_end =
-		dd->verbs_dev.rdi.dparms.qpn_res_start + 65535;
+	dd->verbs_dev.rdi.dparms.qpn_res_start + 65535;
 	dd->verbs_dev.rdi.driver_f.qp_priv_alloc = qp_priv_alloc;
 	dd->verbs_dev.rdi.driver_f.qp_priv_free = qp_priv_free;
 	dd->verbs_dev.rdi.driver_f.free_all_qps = free_all_qps;
 	dd->verbs_dev.rdi.driver_f.notify_qp_reset = notify_qp_reset;
 
+	/* completeion queue */
+	snprintf(dd->verbs_dev.rdi.dparms.cq_name,
+		 sizeof(dd->verbs_dev.rdi.dparms.cq_name),
+		 "hfi1_cq%d", dd->unit);
+	dd->verbs_dev.rdi.dparms.node = dd->assigned_node_id;
+
 	/* misc settings */
-	dd->verbs_dev.rdi.flags = RVT_FLAG_CQ_INIT_DRIVER;
+	dd->verbs_dev.rdi.flags = 0; /* Let rdmavt handle it all */
 	dd->verbs_dev.rdi.dparms.lkey_table_size = hfi1_lkey_table_size;
 	dd->verbs_dev.rdi.dparms.nports = dd->num_pports;
 	dd->verbs_dev.rdi.dparms.npkeys = hfi1_get_npkeys(dd);
diff --git a/drivers/staging/rdma/hfi1/verbs.h b/drivers/staging/rdma/hfi1/verbs.h
index c22f0d13ad7f..ef8fb13d2af2 100644
--- a/drivers/staging/rdma/hfi1/verbs.h
+++ b/drivers/staging/rdma/hfi1/verbs.h
@@ -64,6 +64,7 @@
 #include <rdma/ib_mad.h>
 #include <rdma/rdma_vt.h>
 #include <rdma/rdmavt_qp.h>
+#include <rdma/rdmavt_cq.h>
 
 struct hfi1_ctxtdata;
 struct hfi1_pportdata;
@@ -81,12 +82,6 @@ struct hfi1_packet;
  */
 #define HFI1_UVERBS_ABI_VERSION       2
 
-/*
- * Define an ib_cq_notify value that is not valid so we know when CQ
- * notifications are armed.
- */
-#define IB_CQ_NONE      (IB_CQ_NEXT_COMP + 1)
-
 #define IB_SEQ_NAK	(3 << 29)
 
 /* AETH NAK opcode values */
@@ -236,35 +231,6 @@ struct hfi1_mcast {
 };
 
 /*
- * This structure is used to contain the head pointer, tail pointer,
- * and completion queue entries as a single memory allocation so
- * it can be mmap'ed into user space.
- */
-struct hfi1_cq_wc {
-	u32 head;               /* index of next entry to fill */
-	u32 tail;               /* index of next ib_poll_cq() entry */
-	union {
-		/* these are actually size ibcq.cqe + 1 */
-		struct ib_uverbs_wc uqueue[0];
-		struct ib_wc kqueue[0];
-	};
-};
-
-/*
- * The completion queue structure.
- */
-struct hfi1_cq {
-	struct ib_cq ibcq;
-	struct kthread_work comptask;
-	struct hfi1_devdata *dd;
-	spinlock_t lock; /* protect changes in this struct */
-	u8 notify;
-	u8 triggered;
-	struct hfi1_cq_wc *queue;
-	struct rvt_mmap_info *ip;
-};
-
-/*
  * hfi1 specific data structures that will be hidden from rvt after the queue
  * pair is made common
  */
@@ -363,8 +329,6 @@ struct hfi1_ibdev {
 	u64 n_kmem_wait;
 	u64 n_send_schedule;
 
-	u32 n_cqs_allocated;    /* number of CQs allocated for device */
-	spinlock_t n_cqs_lock;
 	u32 n_qps_allocated;    /* number of QPs allocated for device */
 	spinlock_t n_qps_lock;
 	u32 n_srqs_allocated;   /* number of SRQs allocated for device */
@@ -395,11 +359,6 @@ struct hfi1_verbs_counters {
 	u32 vl15_dropped;
 };
 
-static inline struct hfi1_cq *to_icq(struct ib_cq *ibcq)
-{
-	return container_of(ibcq, struct hfi1_cq, ibcq);
-}
-
 static inline struct rvt_qp *to_iqp(struct ib_qp *ibqp)
 {
 	return container_of(ibqp, struct rvt_qp, ibqp);
@@ -563,28 +522,6 @@ int hfi1_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr);
 
 int hfi1_destroy_srq(struct ib_srq *ibsrq);
 
-int hfi1_cq_init(struct hfi1_devdata *dd);
-
-void hfi1_cq_exit(struct hfi1_devdata *dd);
-
-void hfi1_cq_enter(struct hfi1_cq *cq, struct ib_wc *entry, int sig);
-
-int hfi1_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *entry);
-
-struct ib_cq *hfi1_create_cq(
-	struct ib_device *ibdev,
-	const struct ib_cq_init_attr *attr,
-	struct ib_ucontext *context,
-	struct ib_udata *udata);
-
-int hfi1_destroy_cq(struct ib_cq *ibcq);
-
-int hfi1_req_notify_cq(
-	struct ib_cq *ibcq,
-	enum ib_cq_notify_flags notify_flags);
-
-int hfi1_resize_cq(struct ib_cq *ibcq, int cqe, struct ib_udata *udata);
-
 static inline void hfi1_put_ss(struct rvt_sge_state *ss)
 {
 	while (ss->num_sge) {