Add support for used_event feature, and utilize it to
reduce the number of interrupts and exits for the guest.
Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
---
hw/vhost_net.c | 6 ++++
hw/virtio.c | 92 ++++++++++++++++++++++++++++++++++++++++++++++++++-----
hw/virtio.h | 9 +++++-
3 files changed, 97 insertions(+), 10 deletions(-)
diff --git a/hw/vhost_net.c b/hw/vhost_net.c
index 7e94f61..41841d9 100644
--- a/hw/vhost_net.c
+++ b/hw/vhost_net.c
@@ -50,6 +50,9 @@ uint64_t vhost_net_get_features(struct vhost_net *net, uint64_t features)
if (!(net->dev.features & (1 << VIRTIO_RING_F_INDIRECT_DESC))) {
features &= ~(1 << VIRTIO_RING_F_INDIRECT_DESC);
}
+ if (!(net->dev.features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
+ features &= ~(1 << VIRTIO_RING_F_EVENT_IDX);
+ }
if (!(net->dev.features & (1 << VIRTIO_NET_F_MRG_RXBUF))) {
features &= ~(1 << VIRTIO_NET_F_MRG_RXBUF);
}
@@ -65,6 +68,9 @@ void vhost_net_ack_features(struct vhost_net *net, uint64_t features)
if (features & (1 << VIRTIO_RING_F_INDIRECT_DESC)) {
net->dev.acked_features |= (1 << VIRTIO_RING_F_INDIRECT_DESC);
}
+ if (features & (1 << VIRTIO_RING_F_EVENT_IDX)) {
+ net->dev.acked_features |= (1 << VIRTIO_RING_F_EVENT_IDX);
+ }
if (features & (1 << VIRTIO_NET_F_MRG_RXBUF)) {
net->dev.acked_features |= (1 << VIRTIO_NET_F_MRG_RXBUF);
}
diff --git a/hw/virtio.c b/hw/virtio.c
index b9acbd4..d51ac93 100644
--- a/hw/virtio.c
+++ b/hw/virtio.c
@@ -72,7 +72,17 @@ struct VirtQueue
VRing vring;
target_phys_addr_t pa;
uint16_t last_avail_idx;
+ /* Last used index value we have signalled on */
+ uint16_t signalled_used;
+
+ /* Last used index value we have signalled on */
+ bool signalled_used_valid;
+
+ /* Notification enabled? */
+ bool notification;
+
int inuse;
+
uint16_t vector;
void (*handle_output)(VirtIODevice *vdev, VirtQueue *vq);
VirtIODevice *vdev;
@@ -141,6 +151,11 @@ static inline uint16_t vring_avail_ring(VirtQueue *vq, int i)
return lduw_phys(pa);
}
+static inline uint16_t vring_used_event(VirtQueue *vq)
+{
+ return vring_avail_ring(vq, vq->vring.num);
+}
+
static inline void vring_used_ring_id(VirtQueue *vq, int i, uint32_t val)
{
target_phys_addr_t pa;
@@ -162,11 +177,11 @@ static uint16_t vring_used_idx(VirtQueue *vq)
return lduw_phys(pa);
}
-static inline void vring_used_idx_increment(VirtQueue *vq, uint16_t val)
+static inline void vring_used_idx_set(VirtQueue *vq, uint16_t val)
{
target_phys_addr_t pa;
pa = vq->vring.used + offsetof(VRingUsed, idx);
- stw_phys(pa, vring_used_idx(vq) + val);
+ stw_phys(pa, val);
}
static inline void vring_used_flags_set_bit(VirtQueue *vq, int mask)
@@ -183,12 +198,26 @@ static inline void vring_used_flags_unset_bit(VirtQueue *vq, int mask)
stw_phys(pa, lduw_phys(pa) & ~mask);
}
+static inline void vring_avail_event(VirtQueue *vq, uint16_t val)
+{
+ target_phys_addr_t pa;
+ if (!vq->notification) {
+ return;
+ }
+ pa = vq->vring.used + offsetof(VRingUsed, ring[vq->vring.num]);
+ stw_phys(pa, val);
+}
+
void virtio_queue_set_notification(VirtQueue *vq, int enable)
{
- if (enable)
+ vq->notification = enable;
+ if (vq->vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX)) {
+ vring_avail_event(vq, vring_avail_idx(vq));
+ } else if (enable) {
vring_used_flags_unset_bit(vq, VRING_USED_F_NO_NOTIFY);
- else
+ } else {
vring_used_flags_set_bit(vq, VRING_USED_F_NO_NOTIFY);
+ }
}
int virtio_queue_ready(VirtQueue *vq)
@@ -234,11 +263,16 @@ void virtqueue_fill(VirtQueue *vq, const VirtQueueElement *elem,
void virtqueue_flush(VirtQueue *vq, unsigned int count)
{
+ uint16_t old, new;
/* Make sure buffer is written before we update index. */
wmb();
trace_virtqueue_flush(vq, count);
- vring_used_idx_increment(vq, count);
+ old = vring_used_idx(vq);
+ new = old + count;
+ vring_used_idx_set(vq, new);
vq->inuse -= count;
+ if (unlikely((int16_t)(new - vq->signalled_used) < (uint16_t)(new - old)))
+ vq->signalled_used_valid = false;
}
void virtqueue_push(VirtQueue *vq, const VirtQueueElement *elem,
@@ -395,6 +429,9 @@ int virtqueue_pop(VirtQueue *vq, VirtQueueElement *elem)
max = vq->vring.num;
i = head = virtqueue_get_head(vq, vq->last_avail_idx++);
+ if (vq->vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX)) {
+ vring_avail_event(vq, vring_avail_idx(vq));
+ }
if (vring_desc_flags(desc_pa, i) & VRING_DESC_F_INDIRECT) {
if (vring_desc_len(desc_pa, i) % sizeof(VRingDesc)) {
@@ -478,6 +515,9 @@ void virtio_reset(void *opaque)
vdev->vq[i].last_avail_idx = 0;
vdev->vq[i].pa = 0;
vdev->vq[i].vector = VIRTIO_NO_VECTOR;
+ vdev->vq[i].signalled_used = 0;
+ vdev->vq[i].signalled_used_valid = false;
+ vdev->vq[i].notification = true;
}
}
@@ -629,13 +669,45 @@ void virtio_irq(VirtQueue *vq)
virtio_notify_vector(vq->vdev, vq->vector);
}
-void virtio_notify(VirtIODevice *vdev, VirtQueue *vq)
+/* Assuming a given event_idx value from the other size, if
+ * we have just incremented index from old to new_idx,
+ * should we trigger an event? */
+static inline int vring_need_event(uint16_t event, uint16_t new, uint16_t old)
+{
+ /* Note: Xen has similar logic for notification hold-off
+ * in include/xen/interface/io/ring.h with req_event and req_prod
+ * corresponding to event_idx + 1 and new respectively.
+ * Note also that req_event and req_prod in Xen start at 1,
+ * event indexes in virtio start at 0. */
+ return (uint16_t)(new - event - 1) < (uint16_t)(new - old);
+}
+
+static bool vring_notify(VirtIODevice *vdev, VirtQueue *vq)
{
+ uint16_t old, new;
+ bool v;
/* Always notify when queue is empty (when feature acknowledge) */
- if ((vring_avail_flags(vq) & VRING_AVAIL_F_NO_INTERRUPT) &&
- (!(vdev->guest_features & (1 << VIRTIO_F_NOTIFY_ON_EMPTY)) ||
- (vq->inuse || vring_avail_idx(vq) != vq->last_avail_idx)))
+ if (((vdev->guest_features & (1 << VIRTIO_F_NOTIFY_ON_EMPTY)) &&
+ !vq->inuse && vring_avail_idx(vq) == vq->last_avail_idx)) {
+ return true;
+ }
+
+ if (!(vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
+ return !(vring_avail_flags(vq) & VRING_AVAIL_F_NO_INTERRUPT);
+ }
+
+ v = vq->signalled_used_valid;
+ vq->signalled_used_valid = true;
+ old = vq->signalled_used;
+ new = vq->signalled_used = vring_used_idx(vq);
+ return !v || vring_need_event(vring_used_event(vq), new, old);
+}
+
+void virtio_notify(VirtIODevice *vdev, VirtQueue *vq)
+{
+ if (!vring_notify(vdev, vq)) {
return;
+ }
trace_virtio_notify(vdev, vq);
vdev->isr |= 0x01;
@@ -732,6 +804,8 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f)
vdev->vq[i].vring.num = qemu_get_be32(f);
vdev->vq[i].pa = qemu_get_be64(f);
qemu_get_be16s(f, &vdev->vq[i].last_avail_idx);
+ vdev->vq[i].signalled_used_valid = false;
+ vdev->vq[i].notification = true;
if (vdev->vq[i].pa) {
uint16_t nheads;
diff --git a/hw/virtio.h b/hw/virtio.h
index 9517b97..3438a91 100644
--- a/hw/virtio.h
+++ b/hw/virtio.h
@@ -46,6 +46,11 @@
#define VIRTIO_F_NOTIFY_ON_EMPTY 24
/* We support indirect buffer descriptors */
#define VIRTIO_RING_F_INDIRECT_DESC 28
+/* The Guest publishes the used index for which it expects an interrupt
+ * at the end of the avail ring. Host should ignore the avail->flags field. */
+/* The Host publishes the avail index for which it expects a kick
+ * at the end of the used ring. Guest should ignore the used->flags field. */
+#define VIRTIO_RING_F_EVENT_IDX 29
/* A guest should never accept this. It implies negotiation is broken. */
#define VIRTIO_F_BAD_FEATURE 30
@@ -213,7 +218,9 @@ void virtio_serial_exit(VirtIODevice *vdev);
#define DEFINE_VIRTIO_COMMON_FEATURES(_state, _field) \
DEFINE_PROP_BIT64("indirect_desc", _state, _field, \
- VIRTIO_RING_F_INDIRECT_DESC, true)
+ VIRTIO_RING_F_INDIRECT_DESC, true), \
+ DEFINE_PROP_BIT64("event_idx", _state, _field, \
+ VIRTIO_RING_F_EVENT_IDX, true)
target_phys_addr_t virtio_queue_get_desc_addr(VirtIODevice *vdev, int n);
target_phys_addr_t virtio_queue_get_avail_addr(VirtIODevice *vdev, int n);
--
1.7.5.53.gc233e
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
