Merging upstream version 5.10.216.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

12 files changed, 82 insertions, 38 deletions

diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c
index db1a25fbe..26c66685a 100644
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -33,6 +33,7 @@ MODULE_AUTHOR("Sean Hefty");
 MODULE_DESCRIPTION("InfiniBand CM");
 MODULE_LICENSE("Dual BSD/GPL");
 
+#define CM_DESTROY_ID_WAIT_TIMEOUT 10000 /* msecs */
 static const char * const ibcm_rej_reason_strs[] = {
 	[IB_CM_REJ_NO_QP]			= "no QP",
 	[IB_CM_REJ_NO_EEC]			= "no EEC",
@@ -1056,13 +1057,26 @@ static void cm_reset_to_idle(struct cm_id_private *cm_id_priv)
 	}
 }
 
+static noinline void cm_destroy_id_wait_timeout(struct ib_cm_id *cm_id,
+						enum ib_cm_state old_state)
+{
+	struct cm_id_private *cm_id_priv;
+
+	cm_id_priv = container_of(cm_id, struct cm_id_private, id);
+	pr_err("%s: cm_id=%p timed out. state %d -> %d, refcnt=%d\n", __func__,
+	       cm_id, old_state, cm_id->state, refcount_read(&cm_id_priv->refcount));
+}
+
 static void cm_destroy_id(struct ib_cm_id *cm_id, int err)
 {
 	struct cm_id_private *cm_id_priv;
+	enum ib_cm_state old_state;
 	struct cm_work *work;
+	int ret;
 
 	cm_id_priv = container_of(cm_id, struct cm_id_private, id);
 	spin_lock_irq(&cm_id_priv->lock);
+	old_state = cm_id->state;
 retest:
 	switch (cm_id->state) {
 	case IB_CM_LISTEN:
@@ -1171,7 +1185,14 @@ retest:
 
 	xa_erase(&cm.local_id_table, cm_local_id(cm_id->local_id));
 	cm_deref_id(cm_id_priv);
-	wait_for_completion(&cm_id_priv->comp);
+	do {
+		ret = wait_for_completion_timeout(&cm_id_priv->comp,
+						  msecs_to_jiffies(
+						  CM_DESTROY_ID_WAIT_TIMEOUT));
+		if (!ret) /* timeout happened */
+			cm_destroy_id_wait_timeout(cm_id, old_state);
+	} while (!ret);
+
 	while ((work = cm_dequeue_work(cm_id_priv)) != NULL)
 		cm_free_work(work);
 
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 3c29fd04b..94c3bad72 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -1686,7 +1686,7 @@ static int assign_client_id(struct ib_client *client)
 {
 	int ret;
 
-	down_write(&clients_rwsem);
+	lockdep_assert_held(&clients_rwsem);
 	/*
 	 * The add/remove callbacks must be called in FIFO/LIFO order. To
 	 * achieve this we assign client_ids so they are sorted in
@@ -1695,14 +1695,11 @@ static int assign_client_id(struct ib_client *client)
 	client->client_id = highest_client_id;
 	ret = xa_insert(&clients, client->client_id, client, GFP_KERNEL);
 	if (ret)
-		goto out;
+		return ret;
 
 	highest_client_id++;
 	xa_set_mark(&clients, client->client_id, CLIENT_REGISTERED);
-
-out:
-	up_write(&clients_rwsem);
-	return ret;
+	return 0;
 }
 
 static void remove_client_id(struct ib_client *client)
@@ -1732,25 +1729,35 @@ int ib_register_client(struct ib_client *client)
 {
 	struct ib_device *device;
 	unsigned long index;
+	bool need_unreg = false;
 	int ret;
 
 	refcount_set(&client->uses, 1);
 	init_completion(&client->uses_zero);
+
+	/*
+	 * The devices_rwsem is held in write mode to ensure that a racing
+	 * ib_register_device() sees a consisent view of clients and devices.
+	 */
+	down_write(&devices_rwsem);
+	down_write(&clients_rwsem);
 	ret = assign_client_id(client);
 	if (ret)
-		return ret;
+		goto out;
 
-	down_read(&devices_rwsem);
+	need_unreg = true;
 	xa_for_each_marked (&devices, index, device, DEVICE_REGISTERED) {
 		ret = add_client_context(device, client);
-		if (ret) {
-			up_read(&devices_rwsem);
-			ib_unregister_client(client);
-			return ret;
-		}
+		if (ret)
+			goto out;
 	}
-	up_read(&devices_rwsem);
-	return 0;
+	ret = 0;
+out:
+	up_write(&clients_rwsem);
+	up_write(&devices_rwsem);
+	if (need_unreg && ret)
+		ib_unregister_client(client);
+	return ret;
 }
 EXPORT_SYMBOL(ib_register_client);
 
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 2a973a139..a0d7777ac 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -1711,7 +1711,7 @@ int bnxt_re_modify_srq(struct ib_srq *ib_srq, struct ib_srq_attr *srq_attr,
 	switch (srq_attr_mask) {
 	case IB_SRQ_MAX_WR:
 		/* SRQ resize is not supported */
-		break;
+		return -EINVAL;
 	case IB_SRQ_LIMIT:
 		/* Change the SRQ threshold */
 		if (srq_attr->srq_limit > srq->qplib_srq.max_wqe)
@@ -1726,13 +1726,12 @@ int bnxt_re_modify_srq(struct ib_srq *ib_srq, struct ib_srq_attr *srq_attr,
 		/* On success, update the shadow */
 		srq->srq_limit = srq_attr->srq_limit;
 		/* No need to Build and send response back to udata */
-		break;
+		return 0;
 	default:
 		ibdev_err(&rdev->ibdev,
 			  "Unsupported srq_attr_mask 0x%x", srq_attr_mask);
 		return -EINVAL;
 	}
-	return 0;
 }
 
 int bnxt_re_query_srq(struct ib_srq *ib_srq, struct ib_srq_attr *srq_attr)
diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c
index 60eb3a645..969004258 100644
--- a/drivers/infiniband/hw/hfi1/pio.c
+++ b/drivers/infiniband/hw/hfi1/pio.c
@@ -2131,7 +2131,7 @@ int init_credit_return(struct hfi1_devdata *dd)
 				   "Unable to allocate credit return DMA range for NUMA %d\n",
 				   i);
 			ret = -ENOMEM;
-			goto done;
+			goto free_cr_base;
 		}
 	}
 	set_dev_node(&dd->pcidev->dev, dd->node);
@@ -2139,6 +2139,10 @@ int init_credit_return(struct hfi1_devdata *dd)
 	ret = 0;
 done:
 	return ret;
+
+free_cr_base:
+	free_credit_return(dd);
+	goto done;
 }
 
 void free_credit_return(struct hfi1_devdata *dd)
diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c
index 2dc97de43..68a8557e9 100644
--- a/drivers/infiniband/hw/hfi1/sdma.c
+++ b/drivers/infiniband/hw/hfi1/sdma.c
@@ -3200,7 +3200,7 @@ int _pad_sdma_tx_descs(struct hfi1_devdata *dd, struct sdma_txreq *tx)
 {
 	int rval = 0;
 
-	if ((unlikely(tx->num_desc + 1 == tx->desc_limit))) {
+	if ((unlikely(tx->num_desc == tx->desc_limit))) {
 		rval = _extend_sdma_tx_descs(dd, tx);
 		if (rval) {
 			__sdma_txclean(dd, tx);
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index a56ebdc15..f67ebd9f3 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -2780,7 +2780,7 @@ DECLARE_UVERBS_NAMED_METHOD(
 	MLX5_IB_METHOD_DEVX_OBJ_MODIFY,
 	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_OBJ_MODIFY_HANDLE,
 			UVERBS_IDR_ANY_OBJECT,
-			UVERBS_ACCESS_WRITE,
+			UVERBS_ACCESS_READ,
 			UA_MANDATORY),
 	UVERBS_ATTR_PTR_IN(
 		MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_IN,
diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c
index cca7a4a6b..7f12a9b05 100644
--- a/drivers/infiniband/hw/mlx5/mad.c
+++ b/drivers/infiniband/hw/mlx5/mad.c
@@ -166,7 +166,8 @@ static int process_pma_cmd(struct mlx5_ib_dev *dev, u8 port_num,
 		mdev = dev->mdev;
 		mdev_port_num = 1;
 	}
-	if (MLX5_CAP_GEN(dev->mdev, num_ports) == 1) {
+	if (MLX5_CAP_GEN(dev->mdev, num_ports) == 1 &&
+	    !mlx5_core_mp_enabled(mdev)) {
 		/* set local port to one for Function-Per-Port HCA. */
 		mdev = dev->mdev;
 		mdev_port_num = 1;
diff --git a/drivers/infiniband/hw/mlx5/wr.c b/drivers/infiniband/hw/mlx5/wr.c
index d6038fb6c..19fd440a6 100644
--- a/drivers/infiniband/hw/mlx5/wr.c
+++ b/drivers/infiniband/hw/mlx5/wr.c
@@ -128,7 +128,7 @@ static void set_eth_seg(const struct ib_send_wr *wr, struct mlx5_ib_qp *qp,
 		 */
 		copysz = min_t(u64, *cur_edge - (void *)eseg->inline_hdr.start,
 			       left);
-		memcpy(eseg->inline_hdr.start, pdata, copysz);
+		memcpy(eseg->inline_hdr.data, pdata, copysz);
 		stride = ALIGN(sizeof(struct mlx5_wqe_eth_seg) -
 			       sizeof(eseg->inline_hdr.start) + copysz, 16);
 		*size += stride / 16;
diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c
index 3543b9af1..d382ac211 100644
--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -1865,8 +1865,17 @@ static int qedr_create_user_qp(struct qedr_dev *dev,
 		/* RQ - read access only (0) */
 		rc = qedr_init_user_queue(udata, dev, &qp->urq, ureq.rq_addr,
 					  ureq.rq_len, true, 0, alloc_and_init);
-		if (rc)
+		if (rc) {
+			ib_umem_release(qp->usq.umem);
+			qp->usq.umem = NULL;
+			if (rdma_protocol_roce(&dev->ibdev, 1)) {
+				qedr_free_pbl(dev, &qp->usq.pbl_info,
+					      qp->usq.pbl_tbl);
+			} else {
+				kfree(qp->usq.pbl_tbl);
+			}
 			return rc;
+		}
 	}
 
 	memset(&in_params, 0, sizeof(in_params));
diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c
index 95f0de0c8..0505c81aa 100644
--- a/drivers/infiniband/sw/rxe/rxe.c
+++ b/drivers/infiniband/sw/rxe/rxe.c
@@ -35,6 +35,8 @@ void rxe_dealloc(struct ib_device *ib_dev)
 
 	if (rxe->tfm)
 		crypto_free_shash(rxe->tfm);
+
+	mutex_destroy(&rxe->usdev_lock);
 }
 
 /* initialize rxe device parameters */
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 86e4ed64e..e009123c7 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -542,21 +542,18 @@ static int ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast)
 			/* SM supports sendonly-fullmember, otherwise fallback to full-member */
 			rec.join_state = SENDONLY_FULLMEMBER_JOIN;
 	}
-	spin_unlock_irq(&priv->lock);
 
 	multicast = ib_sa_join_multicast(&ipoib_sa_client, priv->ca, priv->port,
-					 &rec, comp_mask, GFP_KERNEL,
+					 &rec, comp_mask, GFP_ATOMIC,
 					 ipoib_mcast_join_complete, mcast);
-	spin_lock_irq(&priv->lock);
 	if (IS_ERR(multicast)) {
 		ret = PTR_ERR(multicast);
 		ipoib_warn(priv, "ib_sa_join_multicast failed, status %d\n", ret);
 		/* Requeue this join task with a backoff delay */
 		__ipoib_mcast_schedule_join_thread(priv, mcast, 1);
 		clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
-		spin_unlock_irq(&priv->lock);
 		complete(&mcast->done);
-		spin_lock_irq(&priv->lock);
+		return ret;
 	}
 	return 0;
 }
diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c
index 983f59c87..960f870a9 100644
--- a/drivers/infiniband/ulp/srpt/ib_srpt.c
+++ b/drivers/infiniband/ulp/srpt/ib_srpt.c
@@ -79,12 +79,16 @@ module_param(srpt_srq_size, int, 0444);
 MODULE_PARM_DESC(srpt_srq_size,
 		 "Shared receive queue (SRQ) size.");
 
+static int srpt_set_u64_x(const char *buffer, const struct kernel_param *kp)
+{
+	return kstrtou64(buffer, 16, (u64 *)kp->arg);
+}
 static int srpt_get_u64_x(char *buffer, const struct kernel_param *kp)
 {
 	return sprintf(buffer, "0x%016llx\n", *(u64 *)kp->arg);
 }
-module_param_call(srpt_service_guid, NULL, srpt_get_u64_x, &srpt_service_guid,
-		  0444);
+module_param_call(srpt_service_guid, srpt_set_u64_x, srpt_get_u64_x,
+		  &srpt_service_guid, 0444);
 MODULE_PARM_DESC(srpt_service_guid,
 		 "Using this value for ioc_guid, id_ext, and cm_listen_id instead of using the node_guid of the first HCA.");
 
@@ -210,10 +214,12 @@ static const char *get_ch_state_name(enum rdma_ch_state s)
 /**
  * srpt_qp_event - QP event callback function
  * @event: Description of the event that occurred.
- * @ch: SRPT RDMA channel.
+ * @ptr: SRPT RDMA channel.
  */
-static void srpt_qp_event(struct ib_event *event, struct srpt_rdma_ch *ch)
+static void srpt_qp_event(struct ib_event *event, void *ptr)
 {
+	struct srpt_rdma_ch *ch = ptr;
+
 	pr_debug("QP event %d on ch=%p sess_name=%s-%d state=%s\n",
 		 event->event, ch, ch->sess_name, ch->qp->qp_num,
 		 get_ch_state_name(ch->state));
@@ -1803,8 +1809,7 @@ retry:
 	ch->cq_size = ch->rq_size + sq_size;
 
 	qp_init->qp_context = (void *)ch;
-	qp_init->event_handler
-		= (void(*)(struct ib_event *, void*))srpt_qp_event;
+	qp_init->event_handler = srpt_qp_event;
 	qp_init->send_cq = ch->cq;
 	qp_init->recv_cq = ch->cq;
 	qp_init->sq_sig_type = IB_SIGNAL_REQ_WR;
@@ -3200,7 +3205,6 @@ static int srpt_add_one(struct ib_device *device)
 
 	INIT_IB_EVENT_HANDLER(&sdev->event_handler, sdev->device,
 			      srpt_event_handler);
-	ib_register_event_handler(&sdev->event_handler);
 
 	for (i = 1; i <= sdev->device->phys_port_cnt; i++) {
 		sport = &sdev->port[i - 1];
@@ -3223,6 +3227,7 @@ static int srpt_add_one(struct ib_device *device)
 		}
 	}
 
+	ib_register_event_handler(&sdev->event_handler);
 	spin_lock(&srpt_dev_lock);
 	list_add_tail(&sdev->list, &srpt_dev_list);
 	spin_unlock(&srpt_dev_lock);
@@ -3233,7 +3238,6 @@ static int srpt_add_one(struct ib_device *device)
 
 err_port:
 	srpt_unregister_mad_agent(sdev, i);
-	ib_unregister_event_handler(&sdev->event_handler);
 err_cm:
 	if (sdev->cm_id)
 		ib_destroy_cm_id(sdev->cm_id);