Thread (26 messages) flat view 26 messages, 4 authors, 2022-08-26

Re: [RFC v2 1/7] vhost: expose used buffers

From: Guo Zhi <hidden>
Date: 2022-08-26 03:11:04
Also in: kvm, lkml


----- Original Message -----
From: "jasowang" <redacted>
To: "Guo Zhi" <redacted>, "eperezma" <eperezma@redhat.com>, "sgarzare" <sgarzare@redhat.com>, "Michael
Tsirkin" [off-list ref]
Cc: "netdev" <redacted>, "linux-kernel" <redacted>, "kvm list" <redacted>,
"virtualization" [off-list ref]
Sent: Thursday, August 25, 2022 3:01:31 PM
Subject: Re: [RFC v2 1/7] vhost: expose used buffers
在 2022/8/17 21:57, Guo Zhi 写道:
quoted
Follow VIRTIO 1.1 spec, only writing out a single used ring for a batch
of descriptors.

Signed-off-by: Guo Zhi <redacted>
---
  drivers/vhost/vhost.c | 14 ++++++++++++--
  drivers/vhost/vhost.h |  1 +
  2 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 40097826cff0..7b20fa5a46c3 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -2376,10 +2376,20 @@ static int __vhost_add_used_n(struct vhost_virtqueue
*vq,
  	vring_used_elem_t __user *used;
  	u16 old, new;
  	int start;
+	int copy_n = count;
  
+	/**
+	 * If in order feature negotiated, devices can notify the use of a batch of
buffers to
+	 * the driver by only writing out a single used ring entry with the id
corresponding
+	 * to the head entry of the descriptor chain describing the last buffer in the
batch.
+	 */
+	if (vhost_has_feature(vq, VIRTIO_F_IN_ORDER)) {
+		copy_n = 1;
+		heads = &heads[count - 1];

Do we need to check whether or not the buffer is fully used before doing
this?
This is the caller / user of __vhost_add_used_n's duty to make sure all buffer is fully used.
The device will only batch if all heads are fully used.
quoted
+	}
  	start = vq->last_used_idx & (vq->num - 1);
  	used = vq->used->ring + start;
-	if (vhost_put_used(vq, heads, start, count)) {
+	if (vhost_put_used(vq, heads, start, copy_n)) {
  		vq_err(vq, "Failed to write used");
  		return -EFAULT;
  	}
@@ -2410,7 +2420,7 @@ int vhost_add_used_n(struct vhost_virtqueue *vq, struct
vring_used_elem *heads,
  
  	start = vq->last_used_idx & (vq->num - 1);
  	n = vq->num - start;
-	if (n < count) {
+	if (n < count && !vhost_has_feature(vq, VIRTIO_F_IN_ORDER)) {
  		r = __vhost_add_used_n(vq, heads, n);
  		if (r < 0)
  			return r;
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index d9109107af08..0d5c49a30421 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -236,6 +236,7 @@ enum {
  	VHOST_FEATURES = (1ULL << VIRTIO_F_NOTIFY_ON_EMPTY) |
  			 (1ULL << VIRTIO_RING_F_INDIRECT_DESC) |
  			 (1ULL << VIRTIO_RING_F_EVENT_IDX) |
+			 (1ULL << VIRTIO_F_IN_ORDER) |
  			 (1ULL << VHOST_F_LOG_ALL) |

Are we sure all vhost devices can support in-order (especially the SCSI)?

It looks better to start from a device specific one.

Thanks
Sorry for my mistake, I will change it.
quoted
  			 (1ULL << VIRTIO_F_ANY_LAYOUT) |
  			 (1ULL << VIRTIO_F_VERSION_1)
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help