ipoib_ib.c

linux-2.6.15.6

	addr = dma_map_single(priv->ca->dma_device, skb->data, skb->len,
			      DMA_TO_DEVICE);
	pci_unmap_addr_set(tx_req, mapping, addr);

	if (unlikely(post_send(priv, priv->tx_head & (IPOIB_TX_RING_SIZE - 1),
			       address->ah, qpn, addr, skb->len))) {
		ipoib_warn(priv, "post_send failed\n");
		++priv->stats.tx_errors;
		dma_unmap_single(priv->ca->dma_device, addr, skb->len,
				 DMA_TO_DEVICE);
		dev_kfree_skb_any(skb);
	} else {
		dev->trans_start = jiffies;

		address->last_send = priv->tx_head;
		++priv->tx_head;

		if (priv->tx_head - priv->tx_tail == IPOIB_TX_RING_SIZE) {
			ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n");
			netif_stop_queue(dev);
		}
	}
}

static void __ipoib_reap_ah(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);
	struct ipoib_ah *ah, *tah;
	LIST_HEAD(remove_list);

	spin_lock_irq(&priv->lock);
	list_for_each_entry_safe(ah, tah, &priv->dead_ahs, list)
		if ((int) priv->tx_tail - (int) ah->last_send >= 0) {
			list_del(&ah->list);
			list_add_tail(&ah->list, &remove_list);
		}
	spin_unlock_irq(&priv->lock);

	list_for_each_entry_safe(ah, tah, &remove_list, list) {
		ipoib_dbg(priv, "Reaping ah %p\n", ah->ah);
		ib_destroy_ah(ah->ah);
		kfree(ah);
	}
}

void ipoib_reap_ah(void *dev_ptr)
{
	struct net_device *dev = dev_ptr;
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	__ipoib_reap_ah(dev);

	if (!test_bit(IPOIB_STOP_REAPER, &priv->flags))
		queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task, HZ);
}

int ipoib_ib_dev_open(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);
	int ret;

	ret = ipoib_init_qp(dev);
	if (ret) {
		ipoib_warn(priv, "ipoib_init_qp returned %d\n", ret);
		return -1;
	}

	ret = ipoib_ib_post_receives(dev);
	if (ret) {
		ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret);
		return -1;
	}

	clear_bit(IPOIB_STOP_REAPER, &priv->flags);
	queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task, HZ);

	return 0;
}

int ipoib_ib_dev_up(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	set_bit(IPOIB_FLAG_OPER_UP, &priv->flags);

	return ipoib_mcast_start_thread(dev);
}

int ipoib_ib_dev_down(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	ipoib_dbg(priv, "downing ib_dev\n");

	clear_bit(IPOIB_FLAG_OPER_UP, &priv->flags);
	netif_carrier_off(dev);

	/* Shutdown the P_Key thread if still active */
	if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
		down(&pkey_sem);
		set_bit(IPOIB_PKEY_STOP, &priv->flags);
		cancel_delayed_work(&priv->pkey_task);
		up(&pkey_sem);
		flush_workqueue(ipoib_workqueue);
	}

	ipoib_mcast_stop_thread(dev, 1);

	/*
	 * Flush the multicast groups first so we stop any multicast joins. The
	 * completion thread may have already died and we may deadlock waiting
	 * for the completion thread to finish some multicast joins.
	 */
	ipoib_mcast_dev_flush(dev);

	/* Delete broadcast and local addresses since they will be recreated */
	ipoib_mcast_dev_down(dev);

	ipoib_flush_paths(dev);

	return 0;
}

static int recvs_pending(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);
	int pending = 0;
	int i;

	for (i = 0; i < IPOIB_RX_RING_SIZE; ++i)
		if (priv->rx_ring[i].skb)
			++pending;

	return pending;
}

int ipoib_ib_dev_stop(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);
	struct ib_qp_attr qp_attr;
	unsigned long begin;
	struct ipoib_tx_buf *tx_req;
	int i;

	/*
	 * Move our QP to the error state and then reinitialize in
	 * when all work requests have completed or have been flushed.
	 */
	qp_attr.qp_state = IB_QPS_ERR;
	if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE))
		ipoib_warn(priv, "Failed to modify QP to ERROR state\n");

	/* Wait for all sends and receives to complete */
	begin = jiffies;

	while (priv->tx_head != priv->tx_tail || recvs_pending(dev)) {
		if (time_after(jiffies, begin + 5 * HZ)) {
			ipoib_warn(priv, "timing out; %d sends %d receives not completed\n",
				   priv->tx_head - priv->tx_tail, recvs_pending(dev));

			/*
			 * assume the HW is wedged and just free up
			 * all our pending work requests.
			 */
			while ((int) priv->tx_tail - (int) priv->tx_head < 0) {
				tx_req = &priv->tx_ring[priv->tx_tail &
							(IPOIB_TX_RING_SIZE - 1)];
				dma_unmap_single(priv->ca->dma_device,
						 pci_unmap_addr(tx_req, mapping),
						 tx_req->skb->len,
						 DMA_TO_DEVICE);
				dev_kfree_skb_any(tx_req->skb);
				++priv->tx_tail;
			}

			for (i = 0; i < IPOIB_RX_RING_SIZE; ++i)
				if (priv->rx_ring[i].skb) {
					dma_unmap_single(priv->ca->dma_device,
							 pci_unmap_addr(&priv->rx_ring[i],
									mapping),
							 IPOIB_BUF_SIZE,
							 DMA_FROM_DEVICE);
					dev_kfree_skb_any(priv->rx_ring[i].skb);
					priv->rx_ring[i].skb = NULL;
				}

			goto timeout;
		}

		msleep(1);
	}

	ipoib_dbg(priv, "All sends and receives done.\n");

timeout:
	qp_attr.qp_state = IB_QPS_RESET;
	if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE))
		ipoib_warn(priv, "Failed to modify QP to RESET state\n");

	/* Wait for all AHs to be reaped */
	set_bit(IPOIB_STOP_REAPER, &priv->flags);
	cancel_delayed_work(&priv->ah_reap_task);
	flush_workqueue(ipoib_workqueue);

	begin = jiffies;

	while (!list_empty(&priv->dead_ahs)) {
		__ipoib_reap_ah(dev);

		if (time_after(jiffies, begin + HZ)) {
			ipoib_warn(priv, "timing out; will leak address handles\n");
			break;
		}

		msleep(1);
	}

	return 0;
}

int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	priv->ca = ca;
	priv->port = port;
	priv->qp = NULL;

	if (ipoib_transport_dev_init(dev, ca)) {
		printk(KERN_WARNING "%s: ipoib_transport_dev_init failed\n", ca->name);
		return -ENODEV;
	}

	if (dev->flags & IFF_UP) {
		if (ipoib_ib_dev_open(dev)) {
			ipoib_transport_dev_cleanup(dev);
			return -ENODEV;
		}
	}

	return 0;
}

void ipoib_ib_dev_flush(void *_dev)
{
	struct net_device *dev = (struct net_device *)_dev;
	struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv;

	if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
		return;

	ipoib_dbg(priv, "flushing\n");

	ipoib_ib_dev_down(dev);

	/*
	 * The device could have been brought down between the start and when
	 * we get here, don't bring it back up if it's not configured up
	 */
	if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
		ipoib_ib_dev_up(dev);

	down(&priv->vlan_mutex);

	/* Flush any child interfaces too */
	list_for_each_entry(cpriv, &priv->child_intfs, list)
		ipoib_ib_dev_flush(&cpriv->dev);

	up(&priv->vlan_mutex);
}

void ipoib_ib_dev_cleanup(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	ipoib_dbg(priv, "cleaning up ib_dev\n");

	ipoib_mcast_stop_thread(dev, 1);

	/* Delete the broadcast address and the local address */
	ipoib_mcast_dev_down(dev);

	ipoib_transport_dev_cleanup(dev);
}

/*
 * Delayed P_Key Assigment Interim Support
 *
 * The following is initial implementation of delayed P_Key assigment
 * mechanism. It is using the same approach implemented for the multicast
 * group join. The single goal of this implementation is to quickly address
 * Bug #2507. This implementation will probably be removed when the P_Key
 * change async notification is available.
 */

static void ipoib_pkey_dev_check_presence(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);
	u16 pkey_index = 0;

	if (ib_find_cached_pkey(priv->ca, priv->port, priv->pkey, &pkey_index))
		clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
	else
		set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
}

void ipoib_pkey_poll(void *dev_ptr)
{
	struct net_device *dev = dev_ptr;
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	ipoib_pkey_dev_check_presence(dev);

	if (test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags))
		ipoib_open(dev);
	else {
		down(&pkey_sem);
		if (!test_bit(IPOIB_PKEY_STOP, &priv->flags))
			queue_delayed_work(ipoib_workqueue,
					   &priv->pkey_task,
					   HZ);
		up(&pkey_sem);
	}
}

int ipoib_pkey_dev_delay_open(struct net_device *dev)
{
	struct ipoib_dev_priv *priv = netdev_priv(dev);

	/* Look for the interface pkey value in the IB Port P_Key table and */
	/* set the interface pkey assigment flag                            */
	ipoib_pkey_dev_check_presence(dev);

	/* P_Key value not assigned yet - start polling */
	if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
		down(&pkey_sem);
		clear_bit(IPOIB_PKEY_STOP, &priv->flags);
		queue_delayed_work(ipoib_workqueue,
				   &priv->pkey_task,
				   HZ);
		up(&pkey_sem);
		return 1;
	}

	return 0;
}