Merge branches 'cxgb4' and 'mlx5' into k.o/for-4.8

drivers/infiniband/core/iwcm.c

···
 
 /*
  * Release a reference on cm_id. If the last reference is being
- * released, enable the waiting thread (in iw_destroy_cm_id) to
- * get woken up, and return 1 if a thread is already waiting.
+ * released, free the cm_id and return 1.
  */
 static int iwcm_deref_id(struct iwcm_id_private *cm_id_priv)
 {
 	BUG_ON(atomic_read(&cm_id_priv->refcount)==0);
 	if (atomic_dec_and_test(&cm_id_priv->refcount)) {
 		BUG_ON(!list_empty(&cm_id_priv->work_list));
-		complete(&cm_id_priv->destroy_comp);
+		free_cm_id(cm_id_priv);
 		return 1;
 	}
 
···
 static void rem_ref(struct iw_cm_id *cm_id)
 {
 	struct iwcm_id_private *cm_id_priv;
-	int cb_destroy;
 
 	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
 
-	/*
-	 * Test bit before deref in case the cm_id gets freed on another
-	 * thread.
-	 */
-	cb_destroy = test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
-	if (iwcm_deref_id(cm_id_priv) && cb_destroy) {
-		BUG_ON(!list_empty(&cm_id_priv->work_list));
-		free_cm_id(cm_id_priv);
-	}
+	(void)iwcm_deref_id(cm_id_priv);
 }
 
 static int cm_event_handler(struct iw_cm_id *cm_id, struct iw_cm_event *event);
···
 	wait_event(cm_id_priv->connect_wait,
 		   !test_bit(IWCM_F_CONNECT_WAIT, &cm_id_priv->flags));
 
+	/*
+	 * Since we're deleting the cm_id, drop any events that
+	 * might arrive before the last dereference.
+	 */
+	set_bit(IWCM_F_DROP_EVENTS, &cm_id_priv->flags);
+
 	spin_lock_irqsave(&cm_id_priv->lock, flags);
 	switch (cm_id_priv->state) {
 	case IW_CM_STATE_LISTEN:
···
 	struct iwcm_id_private *cm_id_priv;
 
 	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
-	BUG_ON(test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags));
-
 	destroy_cm_id(cm_id);
-
-	wait_for_completion(&cm_id_priv->destroy_comp);
-
-	free_cm_id(cm_id_priv);
 }
 EXPORT_SYMBOL(iw_destroy_cm_id);
 
···
 	ret = cm_id->cm_handler(cm_id, iw_event);
 	if (ret) {
 		iw_cm_reject(cm_id, NULL, 0);
-		set_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
-		destroy_cm_id(cm_id);
-		if (atomic_read(&cm_id_priv->refcount)==0)
-			free_cm_id(cm_id_priv);
+		iw_destroy_cm_id(cm_id);
 	}
 
 out:
···
 	unsigned long flags;
 	int empty;
 	int ret = 0;
-	int destroy_id;
 
 	spin_lock_irqsave(&cm_id_priv->lock, flags);
 	empty = list_empty(&cm_id_priv->work_list);
···
 		put_work(work);
 		spin_unlock_irqrestore(&cm_id_priv->lock, flags);
 
-		ret = process_event(cm_id_priv, &levent);
-		if (ret) {
-			set_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
-			destroy_cm_id(&cm_id_priv->id);
-		}
-		BUG_ON(atomic_read(&cm_id_priv->refcount)==0);
-		destroy_id = test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
-		if (iwcm_deref_id(cm_id_priv)) {
-			if (destroy_id) {
-				BUG_ON(!list_empty(&cm_id_priv->work_list));
-				free_cm_id(cm_id_priv);
-			}
+		if (!test_bit(IWCM_F_DROP_EVENTS, &cm_id_priv->flags)) {
+			ret = process_event(cm_id_priv, &levent);
+			if (ret)
+				destroy_cm_id(&cm_id_priv->id);
+		} else
+			pr_debug("dropping event %d\n", levent.event);
+		if (iwcm_deref_id(cm_id_priv))
 			return;
-		}
 		if (empty)
 			return;
 		spin_lock_irqsave(&cm_id_priv->lock, flags);

drivers/infiniband/core/iwcm.h

···
 	struct list_head work_free_list;
 };
 
-#define IWCM_F_CALLBACK_DESTROY   1
+#define IWCM_F_DROP_EVENTS	  1
 #define IWCM_F_CONNECT_WAIT       2
 
 #endif /* IWCM_H */

drivers/infiniband/core/netlink.c

···
 int ibnl_unicast(struct sk_buff *skb, struct nlmsghdr *nlh,
 			__u32 pid)
 {
-	return nlmsg_unicast(nls, skb, pid);
+	int err;
+
+	err = netlink_unicast(nls, skb, pid, 0);
+	return (err < 0) ? err : 0;
 }
 EXPORT_SYMBOL(ibnl_unicast);
 
···
 		return -ENOMEM;
 	}
 
+	nls->sk_sndtimeo = 10 * HZ;
 	return 0;
 }
 

drivers/infiniband/hw/cxgb3/iwch_cm.c

···
 	state_set(&child_ep->com, CONNECTING);
 	child_ep->com.tdev = tdev;
 	child_ep->com.cm_id = NULL;
-	child_ep->com.local_addr.sin_family = PF_INET;
+	child_ep->com.local_addr.sin_family = AF_INET;
 	child_ep->com.local_addr.sin_port = req->local_port;
 	child_ep->com.local_addr.sin_addr.s_addr = req->local_ip;
-	child_ep->com.remote_addr.sin_family = PF_INET;
+	child_ep->com.remote_addr.sin_family = AF_INET;
 	child_ep->com.remote_addr.sin_port = req->peer_port;
 	child_ep->com.remote_addr.sin_addr.s_addr = req->peer_ip;
 	get_ep(&parent_ep->com);

drivers/infiniband/hw/cxgb4/cm.c

···
 		PDBG("%s last streaming msg ack ep %p tid %u state %u "
 		     "initiator %u freeing skb\n", __func__, ep, ep->hwtid,
 		     state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0);
+		mutex_lock(&ep->com.mutex);
 		kfree_skb(ep->mpa_skb);
 		ep->mpa_skb = NULL;
-		mutex_lock(&ep->com.mutex);
 		if (test_bit(STOP_MPA_TIMER, &ep->com.flags))
 			stop_ep_timer(ep);
 		mutex_unlock(&ep->com.mutex);
···
 			ep->com.state = ABORTING;
 		else {
 			ep->com.state = CLOSING;
+
+			/*
+			 * if we close before we see the fw4_ack() then we fix
+			 * up the timer state since we're reusing it.
+			 */
+			if (ep->mpa_skb &&
+			    test_bit(STOP_MPA_TIMER, &ep->com.flags)) {
+				clear_bit(STOP_MPA_TIMER, &ep->com.flags);
+				stop_ep_timer(ep);
+			}
 			start_ep_timer(ep);
 		}
 		set_bit(CLOSE_SENT, &ep->com.flags);

drivers/infiniband/hw/cxgb4/iw_cxgb4.h

···
 	struct t4_wq wq;
 	spinlock_t lock;
 	struct mutex mutex;
-	atomic_t refcnt;
+	struct kref kref;
 	wait_queue_head_t wait;
 	struct timer_list timer;
 	int sq_sig_all;

drivers/infiniband/hw/cxgb4/mem.c

···
 
 	mhp->dereg_skb = alloc_skb(SGE_MAX_WR_LEN, GFP_KERNEL);
 	if (!mhp->dereg_skb) {
-		kfree(mhp);
-		return ERR_PTR(-ENOMEM);
+		ret = -ENOMEM;
+		goto free_mhp;
 	}
 
 	ret = allocate_window(&rhp->rdev, &stag, php->pdid);
-	if (ret) {
-		kfree(mhp->dereg_skb);
-		kfree(mhp);
-		return ERR_PTR(ret);
-	}
+	if (ret)
+		goto free_skb;
 	mhp->rhp = rhp;
 	mhp->attr.pdid = php->pdid;
 	mhp->attr.type = FW_RI_STAG_MW;
···
 	mmid = (stag) >> 8;
 	mhp->ibmw.rkey = stag;
 	if (insert_handle(rhp, &rhp->mmidr, mhp, mmid)) {
-		deallocate_window(&rhp->rdev, mhp->attr.stag, mhp->dereg_skb);
-		kfree(mhp->dereg_skb);
-		kfree(mhp);
-		return ERR_PTR(-ENOMEM);
+		ret = -ENOMEM;
+		goto dealloc_win;
 	}
 	PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag);
 	return &(mhp->ibmw);
+
+dealloc_win:
+	deallocate_window(&rhp->rdev, mhp->attr.stag, mhp->dereg_skb);
+free_skb:
+	kfree_skb(mhp->dereg_skb);
+free_mhp:
+	kfree(mhp);
+	return ERR_PTR(ret);
 }
 
 int c4iw_dealloc_mw(struct ib_mw *mw)
···
 	mmid = (mw->rkey) >> 8;
 	remove_handle(rhp, &rhp->mmidr, mmid);
 	deallocate_window(&rhp->rdev, mhp->attr.stag, mhp->dereg_skb);
+	kfree_skb(mhp->dereg_skb);
 	kfree(mhp);
 	PDBG("%s ib_mw %p mmid 0x%x ptr %p\n", __func__, mw, mmid, mhp);
 	return 0;

drivers/infiniband/hw/cxgb4/qp.c

···
 	return 0;
 }
 
+void _free_qp(struct kref *kref)
+{
+	struct c4iw_qp *qhp;
+
+	qhp = container_of(kref, struct c4iw_qp, kref);
+	PDBG("%s qhp %p\n", __func__, qhp);
+	kfree(qhp);
+}
+
 void c4iw_qp_add_ref(struct ib_qp *qp)
 {
 	PDBG("%s ib_qp %p\n", __func__, qp);
-	atomic_inc(&(to_c4iw_qp(qp)->refcnt));
+	kref_get(&to_c4iw_qp(qp)->kref);
 }
 
 void c4iw_qp_rem_ref(struct ib_qp *qp)
 {
 	PDBG("%s ib_qp %p\n", __func__, qp);
-	if (atomic_dec_and_test(&(to_c4iw_qp(qp)->refcnt)))
-		wake_up(&(to_c4iw_qp(qp)->wait));
+	kref_put(&to_c4iw_qp(qp)->kref, _free_qp);
 }
 
 static void add_to_fc_list(struct list_head *head, struct list_head *entry)
···
 	wait_event(qhp->wait, !qhp->ep);
 
 	remove_handle(rhp, &rhp->qpidr, qhp->wq.sq.qid);
-	atomic_dec(&qhp->refcnt);
-	wait_event(qhp->wait, !atomic_read(&qhp->refcnt));
 
 	spin_lock_irq(&rhp->lock);
 	if (!list_empty(&qhp->db_fc_entry))
···
 	destroy_qp(&rhp->rdev, &qhp->wq,
 		   ucontext ? &ucontext->uctx : &rhp->rdev.uctx);
 
+	c4iw_qp_rem_ref(ib_qp);
+
 	PDBG("%s ib_qp %p qpid 0x%0x\n", __func__, ib_qp, qhp->wq.sq.qid);
-	kfree(qhp);
 	return 0;
 }
 
···
 	init_completion(&qhp->rq_drained);
 	mutex_init(&qhp->mutex);
 	init_waitqueue_head(&qhp->wait);
-	atomic_set(&qhp->refcnt, 1);
+	kref_init(&qhp->kref);
 
 	ret = insert_handle(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid);
 	if (ret)
···
 	return 0;
 }
 
+static void move_qp_to_err(struct c4iw_qp *qp)
+{
+	struct c4iw_qp_attributes attrs = { .next_state = C4IW_QP_STATE_ERROR };
+
+	(void)c4iw_modify_qp(qp->rhp, qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
+}
+
 void c4iw_drain_sq(struct ib_qp *ibqp)
 {
 	struct c4iw_qp *qp = to_c4iw_qp(ibqp);
 	unsigned long flag;
 	bool need_to_wait;
 
+	move_qp_to_err(qp);
 	spin_lock_irqsave(&qp->lock, flag);
 	need_to_wait = !t4_sq_empty(&qp->wq);
 	spin_unlock_irqrestore(&qp->lock, flag);
···
 	unsigned long flag;
 	bool need_to_wait;
 
+	move_qp_to_err(qp);
 	spin_lock_irqsave(&qp->lock, flag);
 	need_to_wait = !t4_rq_empty(&qp->wq);
 	spin_unlock_irqrestore(&qp->lock, flag);

drivers/infiniband/hw/mlx5/gsi.c

···
 	return MLX5_CAP_GEN(dev->mdev, set_deth_sqpn);
 }
 
-static u32 next_outstanding(struct mlx5_ib_gsi_qp *gsi, u32 index)
-{
-	return ++index % gsi->cap.max_send_wr;
-}
-
-#define for_each_outstanding_wr(gsi, index) \
-	for (index = gsi->outstanding_ci; index != gsi->outstanding_pi; \
-	     index = next_outstanding(gsi, index))
-
 /* Call with gsi->lock locked */
 static void generate_completions(struct mlx5_ib_gsi_qp *gsi)
 {
···
 	struct mlx5_ib_gsi_wr *wr;
 	u32 index;
 
-	for_each_outstanding_wr(gsi, index) {
-		wr = &gsi->outstanding_wrs[index];
+	for (index = gsi->outstanding_ci; index != gsi->outstanding_pi;
+	     index++) {
+		wr = &gsi->outstanding_wrs[index % gsi->cap.max_send_wr];
 
 		if (!wr->completed)
 			break;
···
 		return -ENOMEM;
 	}
 
-	gsi_wr = &gsi->outstanding_wrs[gsi->outstanding_pi];
-	gsi->outstanding_pi = next_outstanding(gsi, gsi->outstanding_pi);
+	gsi_wr = &gsi->outstanding_wrs[gsi->outstanding_pi %
+				       gsi->cap.max_send_wr];
+	gsi->outstanding_pi++;
 
 	if (!wc) {
 		memset(&gsi_wr->wc, 0, sizeof(gsi_wr->wc));

drivers/infiniband/hw/mlx5/main.c

···
 	return ret;
 }
 
-static const char const *names[] = {
+static const char * const names[] = {
 	"rx_write_requests",
 	"rx_read_requests",
 	"rx_atomic_requests",

drivers/infiniband/hw/mlx5/qp.c

···
 		struct mlx5_ib_port *mibport = &dev->port[port_num];
 
 		context->qp_counter_set_usr_page |=
-			cpu_to_be32(mibport->q_cnt_id << 16);
+			cpu_to_be32((u32)(mibport->q_cnt_id) << 24);
 	}
 
 	if (!ibqp->uobject && cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT)