virtio: VIRTIO_F_NOTIFY_ON_EMPTY to force callback on empty

PATCH In review Rusty Russell 2 participants 6 messages May 30, 2008

Subsystems (5)

Networking Network drivers Documentation virtio core +1 more subsystems

lore

Share & tools

lore

Rusty Russell

author

2008-05-30 05:12 UTC #0 lore raw

PATCH 1/3 virtio: VIRTIO_F_NOTIFY_ON_EMPTY to force callback on empty

virtio allows drivers to suppress callbacks (ie. interrupts) for efficiency (no locking, it's just an optimization).

There's a similar mechanism for the host to suppress notifications coming from the guest: in that case, we ignore the suppression if the ring is completely full.

It turns out that life is simpler if the host similarly ignores callback suppression when the ring is completely empty: the network driver wants to free up old packets in a timely manner, and otherwise has to use a timer to poll.

We have to remove the code which ignores interrupts when the driver has disabled them (again, it had no locking and hence was unreliable anyway).

Signed-off-by Rusty Russell <[email protected]>

---
 drivers/virtio/virtio_ring.c  |    7 -------
 include/linux/virtio_config.h |    4 ++++
 2 files changed, 4 insertions(+), 7 deletions(-)

drivers/virtio/virtio_ring.c +0 -7

diff -r 95a02f0e0e21 drivers/virtio/virtio_ring.c
@@ -253,13 +253,6 @@ irqreturn_t vring_interrupt(int irq, voi
 	if (unlikely(vq->broken))
 		return IRQ_HANDLED;
 
-	/* Other side may have missed us turning off the interrupt,
-	 * but we should preserve disable semantic for virtio users. */
-	if (unlikely(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)) {
-		pr_debug("virtqueue interrupt after disable for %p\n", vq);
-		return IRQ_HANDLED;
-	}
-
 	pr_debug("virtqueue callback for %p (%p)\n", vq, vq->vq.callback);
 	if (vq->vq.callback)
 		vq->vq.callback(&vq->vq);

include/linux/virtio_config.h +4 -0

diff -r 95a02f0e0e21 include/linux/virtio_config.h
@@ -14,6 +14,10 @@
 #define VIRTIO_CONFIG_S_DRIVER_OK	4
 /* We've given up on this device. */
 #define VIRTIO_CONFIG_S_FAILED		0x80
+
+/* Do we get callbacks when the ring is completely used, even if we've
+ * suppressed them? */
+#define VIRTIO_F_NOTIFY_ON_EMPTY	24
 
 #ifdef __KERNEL__
 #include <linux/virtio.h>

Rusty Russell

author

2008-05-30 05:13 UTC #1 lore raw

PATCH 2/3 lguest: implement VIRTIO_F_NOTIFY_ON_EMPTY

This is the lguest implementation of the VIRTIO_F_NOTIFY_ON_EMPTY feature. It is currently only published for network devices, but it is turned on for everyone.

Signed-off-by Rusty Russell <[email protected]>

---
 Documentation/lguest/lguest.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

Documentation/lguest/lguest.c +10 -2

diff -r a2e94bbd1589 Documentation/lguest/lguest.c
@@ -157,6 +157,9 @@ struct virtqueue
 
 	/* The routine to call when the Guest pings us. */
 	void (*handle_output)(int fd, struct virtqueue *me);
+
+	/* Outstanding buffers */
+	unsigned int inflight;
 };
 
 /* Remember the arguments to the program so we can "reboot" */

@@ -702,6 +705,7 @@ static unsigned get_vq_desc(struct virtq
 			errx(1, "Looped descriptor");
 	} while ((i = next_desc(vq, i)) != vq->vring.num);
 
+	vq->inflight++;
 	return head;
 }

@@ -719,6 +723,7 @@ static void add_used(struct virtqueue *v
 	/* Make sure buffer is written before we update index. */
 	wmb();
 	vq->vring.used->idx++;
+	vq->inflight--;
 }
 
 /* This actually sends the interrupt for this virtqueue */

@@ -726,8 +731,9 @@ static void trigger_irq(int fd, struct v
 {
 	unsigned long buf[] = { LHREQ_IRQ, vq->config.irq };
 
-	/* If they don't want an interrupt, don't send one. */
-	if (vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
+	/* If they don't want an interrupt, don't send one, unless empty. */
+	if ((vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
+	    && vq->inflight)
 		return;
 
 	/* Send the Guest an interrupt tell them we used something up. */

@@ -1107,6 +1113,7 @@ static void add_virtqueue(struct device 
 	vq->next = NULL;
 	vq->last_avail_idx = 0;
 	vq->dev = dev;
+	vq->inflight = 0;
 
 	/* Initialize the configuration. */
 	vq->config.num = num_descs;

@@ -1368,6 +1375,7 @@ static void setup_tun_net(const char *ar
 
 	/* Tell Guest what MAC address to use. */
 	add_feature(dev, VIRTIO_NET_F_MAC);
+	add_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY);
 	set_config(dev, sizeof(conf), &conf);
 
 	/* We don't need the socket any more; setup is done. */

Rusty Russell

author

2008-05-30 05:14 UTC #3 lore raw

PATCH 3/3 virtio: use callback on empty in virtio_net

In reply to Rusty Russell

virtio_net uses a timer to free old transmitted packets, rather than leaving callbacks enabled all the time. If the host promises to always notify us when the transmit ring is empty, we can free packets at that point and avoid the timer.

Signed-off-by Rusty Russell <[email protected]>

---
 drivers/net/virtio_net.c |   22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

drivers/net/virtio_net.c +17 -5

diff -r 95a02f0e0e21 drivers/net/virtio_net.c
@@ -44,6 +44,7 @@ struct virtnet_info
 	/* 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. */
 	struct timer_list xmit_free_timer;
 
 	/* Number of input buffers, and max we've ever had. */

@@ -51,6 +52,7 @@ struct virtnet_info
 
 	/* For cleaning up after transmission. */
 	struct tasklet_struct tasklet;
+	bool free_in_tasklet;
 
 	/* Receive & send queues. */
 	struct sk_buff_head recv;

@@ -74,7 +76,7 @@ static void skb_xmit_done(struct virtque
 	/* Suppress further interrupts. */
 	svq->vq_ops->disable_cb(svq);
 
-	/* We were waiting for more output buffers. */
+	/* We were probably waiting for more output buffers. */
 	netif_wake_queue(vi->dev);
 
 	/* Make sure we re-xmit last_xmit_skb: if there are no more packets

@@ -240,6 +242,8 @@ static void free_old_xmit_skbs(struct vi
 	}
 }
 
+/* If the virtio transport doesn't always notify us when all in-flight packets
+ * are consumed, we fall back to using this function on a timer to free them. */
 static void xmit_free(unsigned long data)
 {
 	struct virtnet_info *vi = (void *)data;

@@ -300,7 +304,7 @@ static int xmit_skb(struct virtnet_info 
 	num = skb_to_sgvec(skb, sg+1, 0, skb->len) + 1;
 
 	err = vi->svq->vq_ops->add_buf(vi->svq, sg, num, 0, skb);
-	if (!err)
+	if (!err && !vi->free_in_tasklet)
 		mod_timer(&vi->xmit_free_timer, jiffies + (HZ/10));
 
 	return err;

@@ -315,6 +319,8 @@ static void xmit_tasklet(unsigned long d
 		vi->svq->vq_ops->kick(vi->svq);
 		vi->last_xmit_skb = NULL;
 	}
+	if (vi->free_in_tasklet)
+		free_old_xmit_skbs(vi);
 	netif_tx_unlock_bh(vi->dev);
 }

@@ -455,6 +461,10 @@ static int virtnet_probe(struct virtio_d
 	vi->vdev = vdev;
 	vdev->priv = vi;
 
+	/* If they give us a callback when all buffers are done, we don't need
+	 * the timer. */
+	vi->free_in_tasklet = virtio_has_feature(vdev,VIRTIO_F_NOTIFY_ON_EMPTY);
+
 	/* We expect two virtqueues, receive then send. */
 	vi->rvq = vdev->config->find_vq(vdev, 0, skb_recv_done);
 	if (IS_ERR(vi->rvq)) {

@@ -474,7 +487,8 @@ static int virtnet_probe(struct virtio_d
 
 	tasklet_init(&vi->tasklet, xmit_tasklet, (unsigned long)vi);
 
-	setup_timer(&vi->xmit_free_timer, xmit_free, (unsigned long)vi);
+	if (!vi->free_in_tasklet)
+		setup_timer(&vi->xmit_free_timer, xmit_free, (unsigned long)vi);
 
 	err = register_netdev(dev);
 	if (err) {

@@ -513,7 +527,8 @@ static void virtnet_remove(struct virtio
 	/* Stop all the virtqueues. */
 	vdev->config->reset(vdev);
 
-	del_timer_sync(&vi->xmit_free_timer);
+	if (!vi->free_in_tasklet)
+		del_timer_sync(&vi->xmit_free_timer);
 
 	/* Free our skbs in send and recv queues, if any. */
 	while ((skb = __skb_dequeue(&vi->recv)) != NULL) {

@@ -538,7 +553,7 @@ static unsigned int features[] = {
 static unsigned int features[] = {
 	VIRTIO_NET_F_CSUM, VIRTIO_NET_F_GSO, VIRTIO_NET_F_MAC,
 	VIRTIO_NET_F_HOST_TSO4, VIRTIO_NET_F_HOST_UFO, VIRTIO_NET_F_HOST_TSO6,
-	VIRTIO_NET_F_HOST_ECN,
+	VIRTIO_NET_F_HOST_ECN, VIRTIO_F_NOTIFY_ON_EMPTY,
 };
 
 static struct virtio_driver virtio_net = {

Rusty Russell

author

2008-05-30 05:14 UTC #2 lore raw

PATCH 2/3 lguest: implement VIRTIO_F_NOTIFY_ON_EMPTY

In reply to Rusty Russell

This is the lguest implementation of the VIRTIO_F_NOTIFY_ON_EMPTY feature. It is currently only published for network devices, but it is turned on for everyone.

Signed-off-by Rusty Russell <[email protected]>

---
 Documentation/lguest/lguest.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

Documentation/lguest/lguest.c +10 -2

diff -r a2e94bbd1589 Documentation/lguest/lguest.c
@@ -157,6 +157,9 @@ struct virtqueue
 
 	/* The routine to call when the Guest pings us. */
 	void (*handle_output)(int fd, struct virtqueue *me);
+
+	/* Outstanding buffers */
+	unsigned int inflight;
 };
 
 /* Remember the arguments to the program so we can "reboot" */

@@ -702,6 +705,7 @@ static unsigned get_vq_desc(struct virtq
 			errx(1, "Looped descriptor");
 	} while ((i = next_desc(vq, i)) != vq->vring.num);
 
+	vq->inflight++;
 	return head;
 }

@@ -719,6 +723,7 @@ static void add_used(struct virtqueue *v
 	/* Make sure buffer is written before we update index. */
 	wmb();
 	vq->vring.used->idx++;
+	vq->inflight--;
 }
 
 /* This actually sends the interrupt for this virtqueue */

@@ -726,8 +731,9 @@ static void trigger_irq(int fd, struct v
 {
 	unsigned long buf[] = { LHREQ_IRQ, vq->config.irq };
 
-	/* If they don't want an interrupt, don't send one. */
-	if (vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
+	/* If they don't want an interrupt, don't send one, unless empty. */
+	if ((vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
+	    && vq->inflight)
 		return;
 
 	/* Send the Guest an interrupt tell them we used something up. */

@@ -1107,6 +1113,7 @@ static void add_virtqueue(struct device 
 	vq->next = NULL;
 	vq->last_avail_idx = 0;
 	vq->dev = dev;
+	vq->inflight = 0;
 
 	/* Initialize the configuration. */
 	vq->config.num = num_descs;

@@ -1368,6 +1375,7 @@ static void setup_tun_net(const char *ar
 
 	/* Tell Guest what MAC address to use. */
 	add_feature(dev, VIRTIO_NET_F_MAC);
+	add_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY);
 	set_config(dev, sizeof(conf), &conf);
 
 	/* We don't need the socket any more; setup is done. */

Rusty Russell

author

2008-05-30 05:16 UTC #4 lore raw

PATCH 3/3 virtio: use callback on empty in virtio_net

Signed-off-by Rusty Russell <[email protected]>

---
 drivers/net/virtio_net.c |   22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

drivers/net/virtio_net.c +17 -5

diff -r 95a02f0e0e21 drivers/net/virtio_net.c
@@ -44,6 +44,7 @@ struct virtnet_info
 	/* 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. */
 	struct timer_list xmit_free_timer;
 
 	/* Number of input buffers, and max we've ever had. */

@@ -51,6 +52,7 @@ struct virtnet_info
 
 	/* For cleaning up after transmission. */
 	struct tasklet_struct tasklet;
+	bool free_in_tasklet;
 
 	/* Receive & send queues. */
 	struct sk_buff_head recv;

@@ -74,7 +76,7 @@ static void skb_xmit_done(struct virtque
 	/* Suppress further interrupts. */
 	svq->vq_ops->disable_cb(svq);
 
-	/* We were waiting for more output buffers. */
+	/* We were probably waiting for more output buffers. */
 	netif_wake_queue(vi->dev);
 
 	/* Make sure we re-xmit last_xmit_skb: if there are no more packets

@@ -240,6 +242,8 @@ static void free_old_xmit_skbs(struct vi
 	}
 }
 
+/* If the virtio transport doesn't always notify us when all in-flight packets
+ * are consumed, we fall back to using this function on a timer to free them. */
 static void xmit_free(unsigned long data)
 {
 	struct virtnet_info *vi = (void *)data;

@@ -300,7 +304,7 @@ static int xmit_skb(struct virtnet_info 
 	num = skb_to_sgvec(skb, sg+1, 0, skb->len) + 1;
 
 	err = vi->svq->vq_ops->add_buf(vi->svq, sg, num, 0, skb);
-	if (!err)
+	if (!err && !vi->free_in_tasklet)
 		mod_timer(&vi->xmit_free_timer, jiffies + (HZ/10));
 
 	return err;

@@ -315,6 +319,8 @@ static void xmit_tasklet(unsigned long d
 		vi->svq->vq_ops->kick(vi->svq);
 		vi->last_xmit_skb = NULL;
 	}
+	if (vi->free_in_tasklet)
+		free_old_xmit_skbs(vi);
 	netif_tx_unlock_bh(vi->dev);
 }

@@ -455,6 +461,10 @@ static int virtnet_probe(struct virtio_d
 	vi->vdev = vdev;
 	vdev->priv = vi;
 
+	/* If they give us a callback when all buffers are done, we don't need
+	 * the timer. */
+	vi->free_in_tasklet = virtio_has_feature(vdev,VIRTIO_F_NOTIFY_ON_EMPTY);
+
 	/* We expect two virtqueues, receive then send. */
 	vi->rvq = vdev->config->find_vq(vdev, 0, skb_recv_done);
 	if (IS_ERR(vi->rvq)) {

@@ -474,7 +487,8 @@ static int virtnet_probe(struct virtio_d
 
 	tasklet_init(&vi->tasklet, xmit_tasklet, (unsigned long)vi);
 
-	setup_timer(&vi->xmit_free_timer, xmit_free, (unsigned long)vi);
+	if (!vi->free_in_tasklet)
+		setup_timer(&vi->xmit_free_timer, xmit_free, (unsigned long)vi);
 
 	err = register_netdev(dev);
 	if (err) {

@@ -513,7 +527,8 @@ static void virtnet_remove(struct virtio
 	/* Stop all the virtqueues. */
 	vdev->config->reset(vdev);
 
-	del_timer_sync(&vi->xmit_free_timer);
+	if (!vi->free_in_tasklet)
+		del_timer_sync(&vi->xmit_free_timer);
 
 	/* Free our skbs in send and recv queues, if any. */
 	while ((skb = __skb_dequeue(&vi->recv)) != NULL) {

@@ -538,7 +553,7 @@ static unsigned int features[] = {
 static unsigned int features[] = {
 	VIRTIO_NET_F_CSUM, VIRTIO_NET_F_GSO, VIRTIO_NET_F_MAC,
 	VIRTIO_NET_F_HOST_TSO4, VIRTIO_NET_F_HOST_UFO, VIRTIO_NET_F_HOST_TSO6,
-	VIRTIO_NET_F_HOST_ECN,
+	VIRTIO_NET_F_HOST_ECN, VIRTIO_F_NOTIFY_ON_EMPTY,
 };
 
 static struct virtio_driver virtio_net = {

Christian Borntraeger

2008-05-30 08:59 UTC #5 lore raw

In reply to Rusty Russell

Am Freitag, 30. Mai 2008 schrieb Rusty Russell:

Quoted 22 lines

It turns out that life is simpler if the host similarly ignores callback suppression when the ring is completely empty: the network driver wants to free up old packets in a timely manner, and otherwise has to use a timer to poll.
We have to remove the code which ignores interrupts when the driver has disabled them (again, it had no locking and hence was unreliable anyway).
Signed-off-by Rusty Russell <[email protected]>
---
 drivers/virtio/virtio_ring.c  |    7 -------
 include/linux/virtio_config.h |    4 ++++
 2 files changed, 4 insertions(+), 7 deletions(-)
diff -r 95a02f0e0e21 drivers/virtio/virtio_ring.c
drivers/virtio/virtio_ring.c +0 -0
@@ -253,13 +253,6 @@ irqreturn_t vring_interrupt(int irq, voi
 	if (unlikely(vq->broken))
 		return IRQ_HANDLED;
 

drivers/virtio/virtio_ring.c irqreturn_t vring_interrupt(int irq, voi · line 258

Quoted 10 lines

drivers/virtio/virtio_ring.c +0 -7

-	/* Other side may have missed us turning off the interrupt,
-	 * but we should preserve disable semantic for virtio users. */
-	if (unlikely(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)) {
-		pr_debug("virtqueue interrupt after disable for %p\n", vq);
-		return IRQ_HANDLED;
-	}
-
 	pr_debug("virtqueue callback for %p (%p)\n", vq, vq->vq.callback);
 	if (vq->vq.callback)
 		vq->vq.callback(&vq->vq);

that part makes sense.

include/linux/virtio_config.h line 20

Quoted 11 lines

diff -r 95a02f0e0e21 include/linux/virtio_config.h

include/linux/virtio_config.h +4 -0

@@ -14,6 +14,10 @@
 #define VIRTIO_CONFIG_S_DRIVER_OK	4
 /* We've given up on this device. */
 #define VIRTIO_CONFIG_S_FAILED		0x80
+
+/* Do we get callbacks when the ring is completely used, even if we've
+ * suppressed them? */
+#define VIRTIO_F_NOTIFY_ON_EMPTY	24

24?

Does that mean the first 24 bits are driver owned and 24-31 are for the common virtio code?

Christian