1 files changed, 127 insertions, 60 deletions

diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 9040340fad19..34bf6f4eef4a 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -202,8 +202,13 @@ static void free_long_term_buff(struct ibmvnic_adapter *adapter,
 	if (!ltb->buff)
 		return;
 
+	/* VIOS automatically unmaps the long term buffer at remote
+	 * end for the following resets:
+	 * FAILOVER, MOBILITY, TIMEOUT.
+	 */
 	if (adapter->reset_reason != VNIC_RESET_FAILOVER &&
-	    adapter->reset_reason != VNIC_RESET_MOBILITY)
+	    adapter->reset_reason != VNIC_RESET_MOBILITY &&
+	    adapter->reset_reason != VNIC_RESET_TIMEOUT)
 		send_request_unmap(adapter, ltb->map_id);
 	dma_free_coherent(dev, ltb->size, ltb->buff, ltb->addr);
 }
@@ -689,8 +694,11 @@ static int init_tx_pools(struct net_device *netdev)
 
 	adapter->tso_pool = kcalloc(tx_subcrqs,
 				    sizeof(struct ibmvnic_tx_pool), GFP_KERNEL);
-	if (!adapter->tso_pool)
+	if (!adapter->tso_pool) {
+		kfree(adapter->tx_pool);
+		adapter->tx_pool = NULL;
 		return -1;
+	}
 
 	adapter->num_active_tx_pools = tx_subcrqs;
 
@@ -1076,8 +1084,7 @@ static int __ibmvnic_open(struct net_device *netdev)
 
 	rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_UP);
 	if (rc) {
-		for (i = 0; i < adapter->req_rx_queues; i++)
-			napi_disable(&adapter->napi[i]);
+		ibmvnic_napi_disable(adapter);
 		release_resources(adapter);
 		return rc;
 	}
@@ -1254,10 +1261,8 @@ static int __ibmvnic_close(struct net_device *netdev)
 
 	adapter->state = VNIC_CLOSING;
 	rc = set_link_state(adapter, IBMVNIC_LOGICAL_LNK_DN);
-	if (rc)
-		return rc;
 	adapter->state = VNIC_CLOSED;
-	return 0;
+	return rc;
 }
 
 static int ibmvnic_close(struct net_device *netdev)
@@ -1457,8 +1462,6 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 	netdev_tx_t ret = NETDEV_TX_OK;
 
 	if (test_bit(0, &adapter->resetting)) {
-		if (!netif_subqueue_stopped(netdev, skb))
-			netif_stop_subqueue(netdev, queue_num);
 		dev_kfree_skb_any(skb);
 
 		tx_send_failed++;
@@ -1520,6 +1523,9 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 		skb_copy_from_linear_data(skb, dst, skb->len);
 	}
 
+	/* post changes to long_term_buff *dst before VIOS accessing it */
+	dma_wmb();
+
 	tx_pool->consumer_index =
 	    (tx_pool->consumer_index + 1) % tx_pool->num_buffers;
 
@@ -1747,10 +1753,9 @@ static int ibmvnic_set_mac(struct net_device *netdev, void *p)
 	if (!is_valid_ether_addr(addr->sa_data))
 		return -EADDRNOTAVAIL;
 
-	if (adapter->state != VNIC_PROBED) {
-		ether_addr_copy(adapter->mac_addr, addr->sa_data);
+	ether_addr_copy(adapter->mac_addr, addr->sa_data);
+	if (adapter->state != VNIC_PROBED)
 		rc = __ibmvnic_set_mac(netdev, addr->sa_data);
-	}
 
 	return rc;
 }
@@ -1845,7 +1850,7 @@ static int do_reset(struct ibmvnic_adapter *adapter,
 	u64 old_num_rx_queues, old_num_tx_queues;
 	u64 old_num_rx_slots, old_num_tx_slots;
 	struct net_device *netdev = adapter->netdev;
-	int i, rc;
+	int rc;
 
 	netdev_dbg(adapter->netdev, "Re-setting driver (%d)\n",
 		   rwi->reset_reason);
@@ -1990,10 +1995,6 @@ static int do_reset(struct ibmvnic_adapter *adapter,
 	/* refresh device's multicast list */
 	ibmvnic_set_multi(netdev);
 
-	/* kick napi */
-	for (i = 0; i < adapter->req_rx_queues; i++)
-		napi_schedule(&adapter->napi[i]);
-
 	if (adapter->reset_reason == VNIC_RESET_FAILOVER ||
 	    adapter->reset_reason == VNIC_RESET_MOBILITY) {
 		call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, netdev);
@@ -2247,8 +2248,10 @@ static int ibmvnic_reset(struct ibmvnic_adapter *adapter,
 	 * flush reset queue and process this reset
 	 */
 	if (adapter->force_reset_recovery && !list_empty(&adapter->rwi_list)) {
-		list_for_each_safe(entry, tmp_entry, &adapter->rwi_list)
+		list_for_each_safe(entry, tmp_entry, &adapter->rwi_list) {
 			list_del(entry);
+			kfree(list_entry(entry, struct ibmvnic_rwi, list));
+		}
 	}
 	rwi->reset_reason = reason;
 	list_add_tail(&rwi->list, &adapter->rwi_list);
@@ -2342,6 +2345,8 @@ restart_poll:
 		offset = be16_to_cpu(next->rx_comp.off_frame_data);
 		flags = next->rx_comp.flags;
 		skb = rx_buff->skb;
+		/* load long_term_buff before copying to skb */
+		dma_rmb();
 		skb_copy_to_linear_data(skb, rx_buff->data + offset,
 					length);
 
@@ -3255,11 +3260,25 @@ static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry)
 	struct device *dev = &adapter->vdev->dev;
 	union ibmvnic_crq crq;
 	int max_entries;
+	int cap_reqs;
+
+	/* We send out 6 or 7 REQUEST_CAPABILITY CRQs below (depending on
+	 * the PROMISC flag). Initialize this count upfront. When the tasklet
+	 * receives a response to all of these, it will send the next protocol
+	 * message (QUERY_IP_OFFLOAD).
+	 */
+	if (!(adapter->netdev->flags & IFF_PROMISC) ||
+	    adapter->promisc_supported)
+		cap_reqs = 7;
+	else
+		cap_reqs = 6;
 
 	if (!retry) {
 		/* Sub-CRQ entries are 32 byte long */
 		int entries_page = 4 * PAGE_SIZE / (sizeof(u64) * 4);
 
+		atomic_set(&adapter->running_cap_crqs, cap_reqs);
+
 		if (adapter->min_tx_entries_per_subcrq > entries_page ||
 		    adapter->min_rx_add_entries_per_subcrq > entries_page) {
 			dev_err(dev, "Fatal, invalid entries per sub-crq\n");
@@ -3320,44 +3339,45 @@ static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry)
 					adapter->opt_rx_comp_queues;
 
 		adapter->req_rx_add_queues = adapter->max_rx_add_queues;
+	} else {
+		atomic_add(cap_reqs, &adapter->running_cap_crqs);
 	}
-
 	memset(&crq, 0, sizeof(crq));
 	crq.request_capability.first = IBMVNIC_CRQ_CMD;
 	crq.request_capability.cmd = REQUEST_CAPABILITY;
 
 	crq.request_capability.capability = cpu_to_be16(REQ_TX_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_tx_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_rx_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_ADD_QUEUES);
 	crq.request_capability.number = cpu_to_be64(adapter->req_rx_add_queues);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_TX_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
 	    cpu_to_be64(adapter->req_tx_entries_per_subcrq);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_RX_ADD_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
 	    cpu_to_be64(adapter->req_rx_add_entries_per_subcrq);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_MTU);
 	crq.request_capability.number = cpu_to_be64(adapter->req_mtu);
-	atomic_inc(&adapter->running_cap_crqs);
+	cap_reqs--;
 	ibmvnic_send_crq(adapter, &crq);
 
 	if (adapter->netdev->flags & IFF_PROMISC) {
@@ -3365,16 +3385,21 @@ static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry)
 			crq.request_capability.capability =
 			    cpu_to_be16(PROMISC_REQUESTED);
 			crq.request_capability.number = cpu_to_be64(1);
-			atomic_inc(&adapter->running_cap_crqs);
+			cap_reqs--;
 			ibmvnic_send_crq(adapter, &crq);
 		}
 	} else {
 		crq.request_capability.capability =
 		    cpu_to_be16(PROMISC_REQUESTED);
 		crq.request_capability.number = cpu_to_be64(0);
-		atomic_inc(&adapter->running_cap_crqs);
+		cap_reqs--;
 		ibmvnic_send_crq(adapter, &crq);
 	}
+
+	/* Keep at end to catch any discrepancy between expected and actual
+	 * CRQs sent.
+	 */
+	WARN_ON(cap_reqs != 0);
 }
 
 static int pending_scrq(struct ibmvnic_adapter *adapter,
@@ -3779,118 +3804,132 @@ static void send_map_query(struct ibmvnic_adapter *adapter)
 static void send_cap_queries(struct ibmvnic_adapter *adapter)
 {
 	union ibmvnic_crq crq;
+	int cap_reqs;
+
+	/* We send out 25 QUERY_CAPABILITY CRQs below.  Initialize this count
+	 * upfront. When the tasklet receives a response to all of these, it
+	 * can send out the next protocol messaage (REQUEST_CAPABILITY).
+	 */
+	cap_reqs = 25;
+
+	atomic_set(&adapter->running_cap_crqs, cap_reqs);
 
-	atomic_set(&adapter->running_cap_crqs, 0);
 	memset(&crq, 0, sizeof(crq));
 	crq.query_capability.first = IBMVNIC_CRQ_CMD;
 	crq.query_capability.cmd = QUERY_CAPABILITY;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_TX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_RX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_RX_ADD_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_TX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_RX_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_RX_ADD_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MIN_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MIN_RX_ADD_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MAX_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 	    cpu_to_be16(MAX_RX_ADD_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(TCP_IP_OFFLOAD);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(PROMISC_SUPPORTED);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MIN_MTU);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_MTU);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_MULTICAST_FILTERS);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(VLAN_HEADER_INSERTION);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(RX_VLAN_HEADER_INSERTION);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(MAX_TX_SG_ENTRIES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(RX_SG_SUPPORTED);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(OPT_TX_COMP_SUB_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(OPT_RX_COMP_QUEUES);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_RX_BUFADD_Q_PER_RX_COMP_Q);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_TX_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability =
 			cpu_to_be16(OPT_RXBA_ENTRIES_PER_SUBCRQ);
-	atomic_inc(&adapter->running_cap_crqs);
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
 
 	crq.query_capability.capability = cpu_to_be16(TX_RX_DESC_REQ);
-	atomic_inc(&adapter->running_cap_crqs);
+
 	ibmvnic_send_crq(adapter, &crq);
+	cap_reqs--;
+
+	/* Keep at end to catch any discrepancy between expected and actual
+	 * CRQs sent.
+	 */
+	WARN_ON(cap_reqs != 0);
 }
 
 static void handle_vpd_size_rsp(union ibmvnic_crq *crq,
@@ -4157,6 +4196,8 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq,
 	char *name;
 
 	atomic_dec(&adapter->running_cap_crqs);
+	netdev_dbg(adapter->netdev, "Outstanding request-caps: %d\n",
+		   atomic_read(&adapter->running_cap_crqs));
 	switch (be16_to_cpu(crq->request_capability_rsp.capability)) {
 	case REQ_TX_QUEUES:
 		req_value = &adapter->req_tx_queues;
@@ -4272,6 +4313,14 @@ static int handle_login_rsp(union ibmvnic_crq *login_rsp_crq,
 		return 0;
 	}
 
+	if (adapter->failover_pending) {
+		adapter->init_done_rc = -EAGAIN;
+		netdev_dbg(netdev, "Failover pending, ignoring login response\n");
+		complete(&adapter->init_done);
+		/* login response buffer will be released on reset */
+		return 0;
+	}
+
 	netdev->mtu = adapter->req_mtu - ETH_HLEN;
 
 	netdev_dbg(adapter->netdev, "Login Response Buffer:\n");
@@ -4595,7 +4644,22 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq,
 				complete(&adapter->init_done);
 				adapter->init_done_rc = -EIO;
 			}
-			ibmvnic_reset(adapter, VNIC_RESET_FAILOVER);
+			rc = ibmvnic_reset(adapter, VNIC_RESET_FAILOVER);
+			if (rc && rc != -EBUSY) {
+				/* We were unable to schedule the failover
+				 * reset either because the adapter was still
+				 * probing (eg: during kexec) or we could not
+				 * allocate memory. Clear the failover_pending
+				 * flag since no one else will. We ignore
+				 * EBUSY because it means either FAILOVER reset
+				 * is already scheduled or the adapter is
+				 * being removed.
+				 */
+				netdev_err(netdev,
+					   "Error %ld scheduling failover reset\n",
+					   rc);
+				adapter->failover_pending = false;
+			}
 			break;
 		case IBMVNIC_CRQ_INIT_COMPLETE:
 			dev_info(dev, "Partner initialization complete\n");
@@ -4752,15 +4816,15 @@ static void ibmvnic_tasklet(void *data)
 	while (!done) {
 		/* Pull all the valid messages off the CRQ */
 		while ((crq = ibmvnic_next_crq(adapter)) != NULL) {
+			/* This barrier makes sure ibmvnic_next_crq()'s
+			 * crq->generic.first & IBMVNIC_CRQ_CMD_RSP is loaded
+			 * before ibmvnic_handle_crq()'s
+			 * switch(gen_crq->first) and switch(gen_crq->cmd).
+			 */
+			dma_rmb();
 			ibmvnic_handle_crq(crq, adapter);
 			crq->generic.first = 0;
 		}
-
-		/* remain in tasklet until all
-		 * capabilities responses are received
-		 */
-		if (!adapter->wait_capability)
-			done = true;
 	}
 	/* if capabilities CRQ's were sent in this tasklet, the following
 	 * tasklet must wait until all responses are received
@@ -4902,6 +4966,9 @@ static int init_crq_queue(struct ibmvnic_adapter *adapter)
 	crq->cur = 0;
 	spin_lock_init(&crq->lock);
 
+	/* process any CRQs that were queued before we enabled interrupts */
+	tasklet_schedule(&adapter->tasklet);
+
 	return retrc;
 
 req_irq_failed: