Displaying 9 results from an estimated 9 matches for "xmit_tasklet".
2009 May 29
1
[PATCH 3/4] virtio_net: don't free buffers in xmit ring
...the
transmitting userspace could block with a full socket until the timer
fired, the skb destructor was called, and they were re-woken.
So we added the VIRTIO_F_NOTIFY_ON_EMPTY feature: supporting devices
send an interrupt (even if normally suppressed) on an empty xmit ring
which makes us schedule xmit_tasklet(). This was a benchmark win.
Unfortunately, VIRTIO_F_NOTIFY_ON_EMPTY makes quite a lot of work: a
host which is faster than the guest will fire the interrupt every xmit
packet (slowing the guest down further). Attempting mitigation in the
host adds overhead of userspace timers (possibly with the...
2009 May 29
1
[PATCH 3/4] virtio_net: don't free buffers in xmit ring
...the
transmitting userspace could block with a full socket until the timer
fired, the skb destructor was called, and they were re-woken.
So we added the VIRTIO_F_NOTIFY_ON_EMPTY feature: supporting devices
send an interrupt (even if normally suppressed) on an empty xmit ring
which makes us schedule xmit_tasklet(). This was a benchmark win.
Unfortunately, VIRTIO_F_NOTIFY_ON_EMPTY makes quite a lot of work: a
host which is faster than the guest will fire the interrupt every xmit
packet (slowing the guest down further). Attempting mitigation in the
host adds overhead of userspace timers (possibly with the...
2008 May 26
7
[PATCH 1/3] virtio: fix virtio_net xmit of freed skb bug
If we fail to transmit a packet, we assume the queue is full and put
the skb into last_xmit_skb. However, if more space frees up before we
xmit it, we loop, and the result can be transmitting the same skb twice.
Fix is simple: set skb to NULL if we've used it in some way, and check
before sending.
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
---
drivers/net/virtio_net.c |
2008 May 26
7
[PATCH 1/3] virtio: fix virtio_net xmit of freed skb bug
If we fail to transmit a packet, we assume the queue is full and put
the skb into last_xmit_skb. However, if more space frees up before we
xmit it, we loop, and the result can be transmitting the same skb twice.
Fix is simple: set skb to NULL if we've used it in some way, and check
before sending.
Signed-off-by: Rusty Russell <rusty at rustcorp.com.au>
---
drivers/net/virtio_net.c |
2008 Jun 08
2
[PATCH 1/4] virtio_net: Fix skb->csum_start computation
From: Mark McLoughlin <markmc at redhat.com>
hdr->csum_start is the offset from the start of the ethernet
header to the transport layer checksum field. skb->csum_start
is the offset from skb->head.
skb_partial_csum_set() assumes that skb->data points to the
ethernet header - i.e. it computes skb->csum_start by adding
the headroom to hdr->csum_start.
Since
2009 May 29
2
[PATCH 2/4] virtio_net: return NETDEV_TX_BUSY instead of queueing an extra skb.
...: if there are no more packets
- * queued, start_xmit won't be called. */
- tasklet_schedule(&vi->tasklet);
+ if (vi->free_in_tasklet)
+ tasklet_schedule(&vi->tasklet);
}
static void receive_skb(struct net_device *dev, struct sk_buff *skb,
@@ -509,12 +505,7 @@ static void xmit_tasklet(unsigned long d
struct virtnet_info *vi = (void *)data;
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_x...
2009 May 29
2
[PATCH 2/4] virtio_net: return NETDEV_TX_BUSY instead of queueing an extra skb.
...: if there are no more packets
- * queued, start_xmit won't be called. */
- tasklet_schedule(&vi->tasklet);
+ if (vi->free_in_tasklet)
+ tasklet_schedule(&vi->tasklet);
}
static void receive_skb(struct net_device *dev, struct sk_buff *skb,
@@ -509,12 +505,7 @@ static void xmit_tasklet(unsigned long d
struct virtnet_info *vi = (void *)data;
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_x...
2009 Sep 21
0
[PATCH 2/6] virtio: make add_buf return capacity remaining
...->len) + 1;
err = vi->svq->vq_ops->add_buf(vi->svq, sg, num, 0, skb);
- if (!err && !vi->free_in_tasklet)
+ if (err >= 0 && !vi->free_in_tasklet)
mod_timer(&vi->xmit_free_timer, jiffies + (HZ/10));
return err;
@@ -538,7 +538,7 @@ static void xmit_tasklet(unsigned long d
struct virtnet_info *vi = (void *)data;
netif_tx_lock_bh(vi->dev);
- if (vi->last_xmit_skb && xmit_skb(vi, vi->last_xmit_skb) == 0) {
+ if (vi->last_xmit_skb && xmit_skb(vi, vi->last_xmit_skb) >= 0) {
vi->svq->vq_ops->kick(vi->s...
2009 Sep 21
0
[PATCH 2/6] virtio: make add_buf return capacity remaining
...->len) + 1;
err = vi->svq->vq_ops->add_buf(vi->svq, sg, num, 0, skb);
- if (!err && !vi->free_in_tasklet)
+ if (err >= 0 && !vi->free_in_tasklet)
mod_timer(&vi->xmit_free_timer, jiffies + (HZ/10));
return err;
@@ -538,7 +538,7 @@ static void xmit_tasklet(unsigned long d
struct virtnet_info *vi = (void *)data;
netif_tx_lock_bh(vi->dev);
- if (vi->last_xmit_skb && xmit_skb(vi, vi->last_xmit_skb) == 0) {
+ if (vi->last_xmit_skb && xmit_skb(vi, vi->last_xmit_skb) >= 0) {
vi->svq->vq_ops->kick(vi->s...