verbs.c

		dprintk("RPC:       %s: recovering MR %p\n", __func__, mw);
		mw->mw_xprt->rx_ia.ri_ops->ro_recover_mr(mw);

		spin_lock(&buf->rb_recovery_lock);
	}
	spin_unlock(&buf->rb_recovery_lock);
}

void
rpcrdma_defer_mr_recovery(struct rpcrdma_mw *mw)
{
	struct rpcrdma_xprt *r_xprt = mw->mw_xprt;
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;

	spin_lock(&buf->rb_recovery_lock);
	rpcrdma_push_mw(mw, &buf->rb_stale_mrs);
	spin_unlock(&buf->rb_recovery_lock);

	schedule_delayed_work(&buf->rb_recovery_worker, 0);
}

static void
rpcrdma_create_mrs(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
	unsigned int count;
	LIST_HEAD(free);
	LIST_HEAD(all);

	for (count = 0; count < 32; count++) {
		struct rpcrdma_mw *mw;
		int rc;

		mw = kzalloc(sizeof(*mw), GFP_KERNEL);
		if (!mw)
			break;

		rc = ia->ri_ops->ro_init_mr(ia, mw);
		if (rc) {
			kfree(mw);
			break;
		}

		mw->mw_xprt = r_xprt;

		list_add(&mw->mw_list, &free);
		list_add(&mw->mw_all, &all);
	}

	spin_lock(&buf->rb_mwlock);
	list_splice(&free, &buf->rb_mws);
	list_splice(&all, &buf->rb_all);
	r_xprt->rx_stats.mrs_allocated += count;
	spin_unlock(&buf->rb_mwlock);

	dprintk("RPC:       %s: created %u MRs\n", __func__, count);
}

static void
rpcrdma_mr_refresh_worker(struct work_struct *work)
{
	struct rpcrdma_buffer *buf = container_of(work, struct rpcrdma_buffer,
						  rb_refresh_worker.work);
	struct rpcrdma_xprt *r_xprt = container_of(buf, struct rpcrdma_xprt,
						   rx_buf);

	rpcrdma_create_mrs(r_xprt);
}

struct rpcrdma_req *
rpcrdma_create_req(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_buffer *buffer = &r_xprt->rx_buf;
	struct rpcrdma_req *req;

	req = kzalloc(sizeof(*req), GFP_KERNEL);
	if (req == NULL)
		return ERR_PTR(-ENOMEM);

	spin_lock(&buffer->rb_reqslock);
	list_add(&req->rl_all, &buffer->rb_allreqs);
	spin_unlock(&buffer->rb_reqslock);
	req->rl_buffer = &r_xprt->rx_buf;
	INIT_LIST_HEAD(&req->rl_registered);
	return req;
}

/**
 * rpcrdma_create_rep - Allocate an rpcrdma_rep object
 * @r_xprt: controlling transport
 *
 * Returns 0 on success or a negative errno on failure.
 */
int
rpcrdma_create_rep(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_create_data_internal *cdata = &r_xprt->rx_data;
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
	struct rpcrdma_rep *rep;
	int rc;

	rc = -ENOMEM;
	rep = kzalloc(sizeof(*rep), GFP_KERNEL);
	if (rep == NULL)
		goto out;

	rep->rr_rdmabuf = rpcrdma_alloc_regbuf(cdata->inline_rsize,
					       DMA_FROM_DEVICE, GFP_KERNEL);
	if (IS_ERR(rep->rr_rdmabuf)) {
		rc = PTR_ERR(rep->rr_rdmabuf);
		goto out_free;
	}
	xdr_buf_init(&rep->rr_hdrbuf, rep->rr_rdmabuf->rg_base,
		     rdmab_length(rep->rr_rdmabuf));

	rep->rr_cqe.done = rpcrdma_wc_receive;
	rep->rr_rxprt = r_xprt;
	INIT_WORK(&rep->rr_work, rpcrdma_deferred_completion);
	rep->rr_recv_wr.next = NULL;
	rep->rr_recv_wr.wr_cqe = &rep->rr_cqe;
	rep->rr_recv_wr.sg_list = &rep->rr_rdmabuf->rg_iov;
	rep->rr_recv_wr.num_sge = 1;

	spin_lock(&buf->rb_lock);
	list_add(&rep->rr_list, &buf->rb_recv_bufs);
	spin_unlock(&buf->rb_lock);
	return 0;

out_free:
	kfree(rep);
out:
	dprintk("RPC:       %s: reply buffer %d alloc failed\n",
		__func__, rc);
	return rc;
}

int
rpcrdma_buffer_create(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
	int i, rc;

	buf->rb_max_requests = r_xprt->rx_data.max_requests;
	buf->rb_bc_srv_max_requests = 0;
	spin_lock_init(&buf->rb_mwlock);
	spin_lock_init(&buf->rb_lock);
	spin_lock_init(&buf->rb_recovery_lock);
	INIT_LIST_HEAD(&buf->rb_mws);
	INIT_LIST_HEAD(&buf->rb_all);
	INIT_LIST_HEAD(&buf->rb_stale_mrs);
	INIT_DELAYED_WORK(&buf->rb_refresh_worker,
			  rpcrdma_mr_refresh_worker);
	INIT_DELAYED_WORK(&buf->rb_recovery_worker,
			  rpcrdma_mr_recovery_worker);

	rpcrdma_create_mrs(r_xprt);

	INIT_LIST_HEAD(&buf->rb_send_bufs);
	INIT_LIST_HEAD(&buf->rb_allreqs);
	spin_lock_init(&buf->rb_reqslock);
	for (i = 0; i < buf->rb_max_requests; i++) {
		struct rpcrdma_req *req;

		req = rpcrdma_create_req(r_xprt);
		if (IS_ERR(req)) {
			dprintk("RPC:       %s: request buffer %d alloc"
				" failed\n", __func__, i);
			rc = PTR_ERR(req);
			goto out;
		}
		list_add(&req->rl_list, &buf->rb_send_bufs);
	}

	INIT_LIST_HEAD(&buf->rb_recv_bufs);
	for (i = 0; i <= buf->rb_max_requests; i++) {
		rc = rpcrdma_create_rep(r_xprt);
		if (rc)
			goto out;
	}

	rc = rpcrdma_sendctxs_create(r_xprt);
	if (rc)
		goto out;

	return 0;
out:
	rpcrdma_buffer_destroy(buf);
	return rc;
}

static struct rpcrdma_req *
rpcrdma_buffer_get_req_locked(struct rpcrdma_buffer *buf)
{
	struct rpcrdma_req *req;

	req = list_first_entry(&buf->rb_send_bufs,
			       struct rpcrdma_req, rl_list);
	list_del_init(&req->rl_list);
	return req;
}

static struct rpcrdma_rep *
rpcrdma_buffer_get_rep_locked(struct rpcrdma_buffer *buf)
{
	struct rpcrdma_rep *rep;

	rep = list_first_entry(&buf->rb_recv_bufs,
			       struct rpcrdma_rep, rr_list);
	list_del(&rep->rr_list);
	return rep;
}

static void
rpcrdma_destroy_rep(struct rpcrdma_rep *rep)
{
	rpcrdma_free_regbuf(rep->rr_rdmabuf);
	kfree(rep);
}

void
rpcrdma_destroy_req(struct rpcrdma_req *req)
{
	rpcrdma_free_regbuf(req->rl_recvbuf);
	rpcrdma_free_regbuf(req->rl_sendbuf);
	rpcrdma_free_regbuf(req->rl_rdmabuf);
	kfree(req);
}

static void
rpcrdma_destroy_mrs(struct rpcrdma_buffer *buf)
{
	struct rpcrdma_xprt *r_xprt = container_of(buf, struct rpcrdma_xprt,
						   rx_buf);
	struct rpcrdma_ia *ia = rdmab_to_ia(buf);
	struct rpcrdma_mw *mw;
	unsigned int count;

	count = 0;
	spin_lock(&buf->rb_mwlock);
	while (!list_empty(&buf->rb_all)) {
		mw = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
		list_del(&mw->mw_all);

		spin_unlock(&buf->rb_mwlock);
		ia->ri_ops->ro_release_mr(mw);
		count++;
		spin_lock(&buf->rb_mwlock);
	}
	spin_unlock(&buf->rb_mwlock);
	r_xprt->rx_stats.mrs_allocated = 0;

	dprintk("RPC:       %s: released %u MRs\n", __func__, count);
}

void
rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf)
{
	cancel_delayed_work_sync(&buf->rb_recovery_worker);
	cancel_delayed_work_sync(&buf->rb_refresh_worker);

	rpcrdma_sendctxs_destroy(buf);

	while (!list_empty(&buf->rb_recv_bufs)) {
		struct rpcrdma_rep *rep;

		rep = rpcrdma_buffer_get_rep_locked(buf);
		rpcrdma_destroy_rep(rep);
	}
	buf->rb_send_count = 0;

	spin_lock(&buf->rb_reqslock);
	while (!list_empty(&buf->rb_allreqs)) {
		struct rpcrdma_req *req;

		req = list_first_entry(&buf->rb_allreqs,
				       struct rpcrdma_req, rl_all);
		list_del(&req->rl_all);

		spin_unlock(&buf->rb_reqslock);
		rpcrdma_destroy_req(req);
		spin_lock(&buf->rb_reqslock);
	}
	spin_unlock(&buf->rb_reqslock);
	buf->rb_recv_count = 0;

	rpcrdma_destroy_mrs(buf);
}

struct rpcrdma_mw *
rpcrdma_get_mw(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
	struct rpcrdma_mw *mw = NULL;

	spin_lock(&buf->rb_mwlock);
	if (!list_empty(&buf->rb_mws))
		mw = rpcrdma_pop_mw(&buf->rb_mws);
	spin_unlock(&buf->rb_mwlock);

	if (!mw)
		goto out_nomws;
	return mw;

out_nomws:
	dprintk("RPC:       %s: no MWs available\n", __func__);
	if (r_xprt->rx_ep.rep_connected != -ENODEV)
		schedule_delayed_work(&buf->rb_refresh_worker, 0);

	/* Allow the reply handler and refresh worker to run */
	cond_resched();

	return NULL;
}

void
rpcrdma_put_mw(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mw *mw)
{
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;

	spin_lock(&buf->rb_mwlock);
	rpcrdma_push_mw(mw, &buf->rb_mws);
	spin_unlock(&buf->rb_mwlock);
}

static struct rpcrdma_rep *
rpcrdma_buffer_get_rep(struct rpcrdma_buffer *buffers)
{
	/* If an RPC previously completed without a reply (say, a
	 * credential problem or a soft timeout occurs) then hold off
	 * on supplying more Receive buffers until the number of new
	 * pending RPCs catches up to the number of posted Receives.
	 */
	if (unlikely(buffers->rb_send_count < buffers->rb_recv_count))
		return NULL;

	if (unlikely(list_empty(&buffers->rb_recv_bufs)))
		return NULL;
	buffers->rb_recv_count++;
	return rpcrdma_buffer_get_rep_locked(buffers);
}

/*
 * Get a set of request/reply buffers.
 *
 * Reply buffer (if available) is attached to send buffer upon return.
 */
struct rpcrdma_req *
rpcrdma_buffer_get(struct rpcrdma_buffer *buffers)
{
	struct rpcrdma_req *req;

	spin_lock(&buffers->rb_lock);
	if (list_empty(&buffers->rb_send_bufs))
		goto out_reqbuf;
	buffers->rb_send_count++;
	req = rpcrdma_buffer_get_req_locked(buffers);
	req->rl_reply = rpcrdma_buffer_get_rep(buffers);
	spin_unlock(&buffers->rb_lock);
	return req;

out_reqbuf:
	spin_unlock(&buffers->rb_lock);
	pr_warn("RPC:       %s: out of request buffers\n", __func__);
	return NULL;
}

/*
 * Put request/reply buffers back into pool.
 * Pre-decrement counter/array index.
 */
void
rpcrdma_buffer_put(struct rpcrdma_req *req)
{
	struct rpcrdma_buffer *buffers = req->rl_buffer;
	struct rpcrdma_rep *rep = req->rl_reply;

	req->rl_reply = NULL;

	spin_lock(&buffers->rb_lock);
	buffers->rb_send_count--;
	list_add_tail(&req->rl_list, &buffers->rb_send_bufs);
	if (rep) {
		buffers->rb_recv_count--;
		list_add_tail(&rep->rr_list, &buffers->rb_recv_bufs);
	}
	spin_unlock(&buffers->rb_lock);
}

/*
 * Recover reply buffers from pool.
 * This happens when recovering from disconnect.
 */
void
rpcrdma_recv_buffer_get(struct rpcrdma_req *req)
{
	struct rpcrdma_buffer *buffers = req->rl_buffer;

	spin_lock(&buffers->rb_lock);
	req->rl_reply = rpcrdma_buffer_get_rep(buffers);
	spin_unlock(&buffers->rb_lock);
}

/*
 * Put reply buffers back into pool when not attached to
 * request. This happens in error conditions.
 */
void
rpcrdma_recv_buffer_put(struct rpcrdma_rep *rep)
{
	struct rpcrdma_buffer *buffers = &rep->rr_rxprt->rx_buf;

	spin_lock(&buffers->rb_lock);
	buffers->rb_recv_count--;
	list_add_tail(&rep->rr_list, &buffers->rb_recv_bufs);
	spin_unlock(&buffers->rb_lock);
}

/**
 * rpcrdma_alloc_regbuf - allocate and DMA-map memory for SEND/RECV buffers
 * @size: size of buffer to be allocated, in bytes
 * @direction: direction of data movement
 * @flags: GFP flags
 *
 * Returns an ERR_PTR, or a pointer to a regbuf, a buffer that
 * can be persistently DMA-mapped for I/O.
 *
 * xprtrdma uses a regbuf for posting an outgoing RDMA SEND, or for
 * receiving the payload of RDMA RECV operations. During Long Calls
 * or Replies they may be registered externally via ro_map.
 */
struct rpcrdma_regbuf *
rpcrdma_alloc_regbuf(size_t size, enum dma_data_direction direction,
		     gfp_t flags)
{
	struct rpcrdma_regbuf *rb;

	rb = kmalloc(sizeof(*rb) + size, flags);
	if (rb == NULL)
		return ERR_PTR(-ENOMEM);

	rb->rg_device = NULL;
	rb->rg_direction = direction;
	rb->rg_iov.length = size;

	return rb;
}

/**
 * __rpcrdma_map_regbuf - DMA-map a regbuf
 * @ia: controlling rpcrdma_ia
 * @rb: regbuf to be mapped
 */
bool
__rpcrdma_dma_map_regbuf(struct rpcrdma_ia *ia, struct rpcrdma_regbuf *rb)
{
	struct ib_device *device = ia->ri_device;

	if (rb->rg_direction == DMA_NONE)
		return false;

	rb->rg_iov.addr = ib_dma_map_single(device,
					    (void *)rb->rg_base,
					    rdmab_length(rb),
					    rb->rg_direction);
	if (ib_dma_mapping_error(device, rdmab_addr(rb)))
		return false;

	rb->rg_device = device;
	rb->rg_iov.lkey = ia->ri_pd->local_dma_lkey;
	return true;
}

static void
rpcrdma_dma_unmap_regbuf(struct rpcrdma_regbuf *rb)
{
	if (!rpcrdma_regbuf_is_mapped(rb))
		return;

	ib_dma_unmap_single(rb->rg_device, rdmab_addr(rb),
			    rdmab_length(rb), rb->rg_direction);
	rb->rg_device = NULL;
}

/**
 * rpcrdma_free_regbuf - deregister and free registered buffer
 * @rb: regbuf to be deregistered and freed
 */
void
rpcrdma_free_regbuf(struct rpcrdma_regbuf *rb)
{
	if (!rb)
		return;

	rpcrdma_dma_unmap_regbuf(rb);
	kfree(rb);
}

/*
 * Prepost any receive buffer, then post send.
 *
 * Receive buffer is donated to hardware, reclaimed upon recv completion.
 */
int
rpcrdma_ep_post(struct rpcrdma_ia *ia,
		struct rpcrdma_ep *ep,
		struct rpcrdma_req *req)
{
	struct ib_send_wr *send_wr = &req->rl_sendctx->sc_wr;
	struct ib_send_wr *send_wr_fail;
	int rc;

	if (req->rl_reply) {
		rc = rpcrdma_ep_post_recv(ia, req->rl_reply);
		if (rc)
			return rc;
		req->rl_reply = NULL;
	}

	dprintk("RPC:       %s: posting %d s/g entries\n",
		__func__, send_wr->num_sge);

	if (!ep->rep_send_count ||
	    test_bit(RPCRDMA_REQ_F_TX_RESOURCES, &req->rl_flags)) {
		send_wr->send_flags |= IB_SEND_SIGNALED;
		ep->rep_send_count = ep->rep_send_batch;
	} else {
		send_wr->send_flags &= ~IB_SEND_SIGNALED;
		--ep->rep_send_count;
	}
	rc = ib_post_send(ia->ri_id->qp, send_wr, &send_wr_fail);
	if (rc)
		goto out_postsend_err;
	return 0;

out_postsend_err:
	pr_err("rpcrdma: RDMA Send ib_post_send returned %i\n", rc);
	return -ENOTCONN;
}

int
rpcrdma_ep_post_recv(struct rpcrdma_ia *ia,
		     struct rpcrdma_rep *rep)
{
	struct ib_recv_wr *recv_wr_fail;
	int rc;

	if (!rpcrdma_dma_map_regbuf(ia, rep->rr_rdmabuf))
		goto out_map;
	rc = ib_post_recv(ia->ri_id->qp, &rep->rr_recv_wr, &recv_wr_fail);
	if (rc)
		goto out_postrecv;
	return 0;

out_map:
	pr_err("rpcrdma: failed to DMA map the Receive buffer\n");
	return -EIO;

out_postrecv:
	pr_err("rpcrdma: ib_post_recv returned %i\n", rc);
	return -ENOTCONN;
}

/**
 * rpcrdma_ep_post_extra_recv - Post buffers for incoming backchannel requests
 * @r_xprt: transport associated with these backchannel resources
 * @min_reqs: minimum number of incoming requests expected
 *
 * Returns zero if all requested buffers were posted, or a negative errno.
 */
int
rpcrdma_ep_post_extra_recv(struct rpcrdma_xprt *r_xprt, unsigned int count)
{
	struct rpcrdma_buffer *buffers = &r_xprt->rx_buf;
	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
	struct rpcrdma_rep *rep;
	int rc;

	while (count--) {
		spin_lock(&buffers->rb_lock);
		if (list_empty(&buffers->rb_recv_bufs))
			goto out_reqbuf;
		rep = rpcrdma_buffer_get_rep_locked(buffers);
		spin_unlock(&buffers->rb_lock);

		rc = rpcrdma_ep_post_recv(ia, rep);
		if (rc)
			goto out_rc;
	}

	return 0;

out_reqbuf:
	spin_unlock(&buffers->rb_lock);
	pr_warn("%s: no extra receive buffers\n", __func__);
	return -ENOMEM;

out_rc:
	rpcrdma_recv_buffer_put(rep);
	return rc;
}