Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 8958f574 authored by Rusty Russell's avatar Rusty Russell
Browse files

virtio_net: return NETDEV_TX_BUSY instead of queueing an extra skb. 



This effectively reverts 99ffc696
"virtio: wean net driver off NETDEV_TX_BUSY".

The complexity of queuing an skb (setting a tasklet to re-xmit) is
questionable, especially once we get rid of the other reason for the
tasklet in the next patch.

If the skb won't fit in the tx queue, just return NETDEV_TX_BUSY.
This is frowned upon, so a followup patch uses a more complex solution.

Signed-off-by: default avatarRusty Russell <rusty@rustcorp.com.au>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
parent 2b5bbe3b

drivers/net/virtio_net.c

+10 −36
Original line number Original line Diff line number Diff line
@@ -48,9 +48,6 @@ struct virtnet_info 
	struct napi_struct napi;

	struct napi_struct napi;

	unsigned int status;

	unsigned int status;





	/* The skb we couldn't send because buffers were full. */

	struct sk_buff *last_xmit_skb;



	/* If we need to free in a timer, this is it. */

	/* If we need to free in a timer, this is it. */

	struct timer_list xmit_free_timer;

	struct timer_list xmit_free_timer;
@@ -120,8 +117,7 @@ static void skb_xmit_done(struct virtqueue *svq) 
	/* We were probably waiting for more output buffers. */

	/* We were probably waiting for more output buffers. */

	netif_wake_queue(vi->dev);

	netif_wake_queue(vi->dev);





	/* Make sure we re-xmit last_xmit_skb: if there are no more packets

	if (vi->free_in_tasklet)

	 * queued, start_xmit won't be called. */

		tasklet_schedule(&vi->tasklet);

		tasklet_schedule(&vi->tasklet);

}

}
@@ -543,11 +539,6 @@ static void xmit_tasklet(unsigned long data) 
	struct virtnet_info *vi = (void *)data;

	struct virtnet_info *vi = (void *)data;





	netif_tx_lock_bh(vi->dev);

	netif_tx_lock_bh(vi->dev);

	if (vi->last_xmit_skb && xmit_skb(vi, vi->last_xmit_skb) >= 0) {

		vi->svq->vq_ops->kick(vi->svq);

		vi->last_xmit_skb = NULL;

	}

	if (vi->free_in_tasklet)

	free_old_xmit_skbs(vi);

	free_old_xmit_skbs(vi);

	netif_tx_unlock_bh(vi->dev);

	netif_tx_unlock_bh(vi->dev);

}

}
@@ -560,28 +551,16 @@ again: 
	/* Free up any pending old buffers before queueing new ones. */

	/* Free up any pending old buffers before queueing new ones. */

	free_old_xmit_skbs(vi);

	free_old_xmit_skbs(vi);





	/* If we has a buffer left over from last time, send it now. */

	if (unlikely(vi->last_xmit_skb) &&

	    xmit_skb(vi, vi->last_xmit_skb) < 0)

		goto stop_queue;



	vi->last_xmit_skb = NULL;



	/* Put new one in send queue and do transmit */

	/* Put new one in send queue and do transmit */

	if (likely(skb)) {

	__skb_queue_head(&vi->send, skb);

	__skb_queue_head(&vi->send, skb);

		if (xmit_skb(vi, skb) < 0) {

	if (likely(xmit_skb(vi, skb) >= 0)) {

			vi->last_xmit_skb = skb;

			skb = NULL;

			goto stop_queue;

		}

	}

done:

		vi->svq->vq_ops->kick(vi->svq);

		vi->svq->vq_ops->kick(vi->svq);

		return NETDEV_TX_OK;

		return NETDEV_TX_OK;

	}





stop_queue:

	/* Ring too full for this packet, remove it from queue again. */

	pr_debug("%s: virtio not prepared to send\n", dev->name);

	pr_debug("%s: virtio not prepared to send\n", dev->name);

	__skb_unlink(skb, &vi->send);

	netif_stop_queue(dev);

	netif_stop_queue(dev);





	/* Activate callback for using skbs: if this returns false it

	/* Activate callback for using skbs: if this returns false it
@@ -591,12 +570,7 @@ stop_queue: 
		netif_start_queue(dev);

		netif_start_queue(dev);

		goto again;

		goto again;

	}

	}

	if (skb) {

	return NETDEV_TX_BUSY;

		/* Drop this skb: we only queue one. */

		vi->dev->stats.tx_dropped++;

		kfree_skb(skb);

	}

	goto done;

}

}





static int virtnet_set_mac_address(struct net_device *dev, void *p)

static int virtnet_set_mac_address(struct net_device *dev, void *p)