2018-02-14 02:44:41

by Jason Wang

[permalink] [raw]
Subject: [PATCH RFC 0/2] Packed ring for vhost

Hi all:

This RFC implement a subset of packed ring which was described at
https://github.com/oasis-tcs/virtio-docs/blob/master/virtio-v1.1-packed-wd07.pdf
. The code were tested with pmd implement by Jens at
http://dpdk.org/ml/archives/dev/2018-January/089417.html. Minor
change was needed for pmd codes to kick virtqueue since it assumes a
busy polling backend.

Test were done between localhost and guest. Testpmd (rxonly) in guest
reports 2.4Mpps. Testpmd (txonly) repots about 2.1Mpps.

It's not a complete implemention, here's what were missed:

- Device Area
- Driver Area
- Descriptor indirection
- Zerocopy may not be functional
- Migration path is not tested
- Vhost devices except for net
- vIOMMU can not work (mainly because the metadata prefetch is not
implemented).
- See FIXME/TODO in the codes for more details
- No batching or other optimizations were implemented

For a quick prototype, this series open code the tracking of warp
counter and descriptor index at net device. This will be addressed in
the future by:

- Move get_rx_bufs() from net.c to vhost.c
- Let vhost_get_vq_desc() returns vring_used_elem instad of head id

With the above, we can hide the internal (at least part of) vring
layout from specific device.

Please review.

Thanks

Jason Wang (2):
virtio: introduce packed ring defines
vhost: packed ring support

drivers/vhost/net.c | 14 +-
drivers/vhost/vhost.c | 351 ++++++++++++++++++++++++++++++++++---
drivers/vhost/vhost.h | 6 +-
include/uapi/linux/virtio_config.h | 9 +
include/uapi/linux/virtio_ring.h | 17 ++
5 files changed, 369 insertions(+), 28 deletions(-)

--
2.7.4



2018-02-14 02:44:41

by Jason Wang

[permalink] [raw]
Subject: Re: [PATCH RFC 0/2] Packed ring for vhost



On 2018年02月14日 10:37, Jason Wang wrote:
> Hi all:
>
> This RFC implement a subset of packed ring which was described at
> https://github.com/oasis-tcs/virtio-docs/blob/master/virtio-v1.1-packed-wd07.pdf
> . The code were tested with pmd implement by Jens at
> http://dpdk.org/ml/archives/dev/2018-January/089417.html. Minor
> change was needed for pmd codes to kick virtqueue since it assumes a
> busy polling backend.
>
> Test were done between localhost and guest. Testpmd (rxonly) in guest
> reports 2.4Mpps. Testpmd (txonly) repots about 2.1Mpps.
>
> It's not a complete implemention, here's what were missed:
>
> - Device Area
> - Driver Area
> - Descriptor indirection
> - Zerocopy may not be functional
> - Migration path is not tested
> - Vhost devices except for net
> - vIOMMU can not work (mainly because the metadata prefetch is not
> implemented).
> - See FIXME/TODO in the codes for more details
> - No batching or other optimizations were implemented
>
> For a quick prototype, this series open code the tracking of warp
> counter and descriptor index at net device. This will be addressed in
> the future by:
>
> - Move get_rx_bufs() from net.c to vhost.c
> - Let vhost_get_vq_desc() returns vring_used_elem instad of head id
>
> With the above, we can hide the internal (at least part of) vring
> layout from specific device.
>
> Please review.
>
> Thanks

It's near spring festival in China, will probably reply after the holiday.

Thanks

2018-02-14 02:48:22

by Jason Wang

[permalink] [raw]
Subject: [PATCH RFC 2/2] vhost: packed ring support

Signed-off-by: Jason Wang <[email protected]>
---
drivers/vhost/net.c | 14 +-
drivers/vhost/vhost.c | 351 ++++++++++++++++++++++++++++++++++++++++++++++----
drivers/vhost/vhost.h | 6 +-
3 files changed, 343 insertions(+), 28 deletions(-)

diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index c613d2e..65b27c9 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -67,7 +67,8 @@ enum {
VHOST_NET_FEATURES = VHOST_FEATURES |
(1ULL << VHOST_NET_F_VIRTIO_NET_HDR) |
(1ULL << VIRTIO_NET_F_MRG_RXBUF) |
- (1ULL << VIRTIO_F_IOMMU_PLATFORM)
+ (1ULL << VIRTIO_F_IOMMU_PLATFORM) |
+ (1ULL << VIRTIO_F_RING_PACKED)
};

enum {
@@ -473,6 +474,7 @@ static void handle_tx(struct vhost_net *net)
struct socket *sock;
struct vhost_net_ubuf_ref *uninitialized_var(ubufs);
bool zcopy, zcopy_used;
+ struct vring_used_elem used;

mutex_lock(&vq->mutex);
sock = vq->private_data;
@@ -494,6 +496,8 @@ static void handle_tx(struct vhost_net *net)
vhost_zerocopy_signal_used(net, vq);


+ used.idx = vq->last_avail_idx & (vq->num - 1);
+ used.wrap_counter = vq->used_wrap_counter;
head = vhost_net_tx_get_vq_desc(net, vq, vq->iov,
ARRAY_SIZE(vq->iov),
&out, &in);
@@ -515,6 +519,8 @@ static void handle_tx(struct vhost_net *net)
}
/* Skip header. TODO: support TSO. */
len = iov_length(vq->iov, out);
+ used.id = head;
+ used.len = 0;
iov_iter_init(&msg.msg_iter, WRITE, vq->iov, out, len);
iov_iter_advance(&msg.msg_iter, hdr_size);
/* Sanity check */
@@ -576,7 +582,7 @@ static void handle_tx(struct vhost_net *net)
pr_debug("Truncated TX packet: "
" len %d != %zd\n", err, len);
if (!zcopy_used)
- vhost_add_used_and_signal(&net->dev, vq, head, 0);
+ vhost_add_used_and_signal_n(&net->dev, vq, &used, 1);
else
vhost_zerocopy_signal_used(net, vq);
vhost_net_tx_packet(net);
@@ -691,6 +697,8 @@ static int get_rx_bufs(struct vhost_virtqueue *vq,
r = -ENOBUFS;
goto err;
}
+ heads[headcount].idx = vq->last_avail_idx & (vq->num - 1);
+ heads[headcount].wrap_counter = vq->used_wrap_counter;
r = vhost_get_vq_desc(vq, vq->iov + seg,
ARRAY_SIZE(vq->iov) - seg, &out,
&in, log, log_num);
@@ -780,6 +788,8 @@ static void handle_rx(struct vhost_net *net)
vq_log = unlikely(vhost_has_feature(vq, VHOST_F_LOG_ALL)) ?
vq->log : NULL;
mergeable = vhost_has_feature(vq, VIRTIO_NET_F_MRG_RXBUF);
+ /* FIXME: workaround for current dpdk prototype */
+ mergeable = false;

while ((sock_len = vhost_net_rx_peek_head_len(net, sock->sk))) {
sock_len += sock_hlen;
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 2db5af8..5667d03 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -324,6 +324,7 @@ static void vhost_vq_reset(struct vhost_dev *dev,
vhost_reset_is_le(vq);
vhost_disable_cross_endian(vq);
vq->busyloop_timeout = 0;
+ vq->used_wrap_counter = true;
vq->umem = NULL;
vq->iotlb = NULL;
__vhost_vq_meta_reset(vq);
@@ -1136,10 +1137,22 @@ static int vhost_iotlb_miss(struct vhost_virtqueue *vq, u64 iova, int access)
return 0;
}

-static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
- struct vring_desc __user *desc,
- struct vring_avail __user *avail,
- struct vring_used __user *used)
+static int vq_access_ok_packed(struct vhost_virtqueue *vq, unsigned int num,
+ struct vring_desc __user *desc,
+ struct vring_avail __user *avail,
+ struct vring_used __user *used)
+{
+ struct vring_desc_packed *packed = (struct vring_desc_packed *)desc;
+
+ /* FIXME: check device area and driver area */
+ return access_ok(VERIFY_READ, packed, num * sizeof(*packed)) &&
+ access_ok(VERIFY_WRITE, packed, num * sizeof(*packed));
+}
+
+static int vq_access_ok_split(struct vhost_virtqueue *vq, unsigned int num,
+ struct vring_desc __user *desc,
+ struct vring_avail __user *avail,
+ struct vring_used __user *used)

{
size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
@@ -1151,6 +1164,17 @@ static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
sizeof *used + num * sizeof *used->ring + s);
}

+static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
+ struct vring_desc __user *desc,
+ struct vring_avail __user *avail,
+ struct vring_used __user *used)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vq_access_ok_packed(vq, num, desc, avail, used);
+ else
+ return vq_access_ok_split(vq, num, desc, avail, used);
+}
+
static void vhost_vq_meta_update(struct vhost_virtqueue *vq,
const struct vhost_umem_node *node,
int type)
@@ -1763,6 +1787,9 @@ int vhost_vq_init_access(struct vhost_virtqueue *vq)

vhost_init_is_le(vq);

+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return 0;
+
r = vhost_update_used_flags(vq);
if (r)
goto err;
@@ -1947,18 +1974,136 @@ static int get_indirect(struct vhost_virtqueue *vq,
return 0;
}

-/* This looks in the virtqueue and for the first available buffer, and converts
- * it to an iovec for convenient access. Since descriptors consist of some
- * number of output then some number of input descriptors, it's actually two
- * iovecs, but we pack them into one and note how many of each there were.
- *
- * This function returns the descriptor number found, or vq->num (which is
- * never a valid descriptor number) if none was found. A negative code is
- * returned on error. */
-int vhost_get_vq_desc(struct vhost_virtqueue *vq,
- struct iovec iov[], unsigned int iov_size,
- unsigned int *out_num, unsigned int *in_num,
- struct vhost_log *log, unsigned int *log_num)
+#define DESC_AVAIL (1 << VRING_DESC_F_AVAIL)
+#define DESC_USED (1 << VRING_DESC_F_USED)
+static bool desc_is_avail(struct vhost_virtqueue *vq,
+ struct vring_desc_packed *desc)
+{
+ if (vq->used_wrap_counter)
+ if ((desc->flags & DESC_AVAIL) && !(desc->flags & DESC_USED))
+ return true;
+ if (vq->used_wrap_counter == false)
+ if (!(desc->flags & DESC_AVAIL) && (desc->flags & DESC_USED))
+ return true;
+
+ return false;
+}
+
+static void set_desc_used(struct vhost_virtqueue *vq,
+ struct vring_desc_packed *desc, bool wrap_counter)
+{
+ __virtio16 flags = desc->flags;
+
+ if (wrap_counter) {
+ desc->flags |= cpu_to_vhost16(vq, DESC_AVAIL);
+ desc->flags |= cpu_to_vhost16(vq, DESC_USED);
+ } else {
+ desc->flags &= ~cpu_to_vhost16(vq, DESC_AVAIL);
+ desc->flags &= ~cpu_to_vhost16(vq, DESC_USED);
+ }
+
+ desc->flags = flags;
+}
+
+static int vhost_get_vq_desc_packed(struct vhost_virtqueue *vq,
+ struct iovec iov[], unsigned int iov_size,
+ unsigned int *out_num, unsigned int *in_num,
+ struct vhost_log *log,
+ unsigned int *log_num)
+{
+ struct vring_desc_packed desc;
+ int ret, access, i;
+ u16 avail_idx = vq->last_avail_idx;
+
+ /* When we start there are none of either input nor output. */
+ *out_num = *in_num = 0;
+ if (unlikely(log))
+ *log_num = 0;
+
+ do {
+ unsigned int iov_count = *in_num + *out_num;
+
+ i = vq->last_avail_idx & (vq->num - 1);
+ ret = vhost_copy_from_user(vq, &desc, vq->desc_packed + i,
+ sizeof(desc));
+ if (unlikely(ret)) {
+ vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
+ i, vq->desc_packed + i);
+ return -EFAULT;
+ }
+
+ if (!desc_is_avail(vq, &desc)) {
+ /* If there's nothing new since last we looked, return
+ * invalid.
+ */
+ if (likely(avail_idx == vq->last_avail_idx))
+ return vq->num;
+ }
+
+ /* Only start to read descriptor after we're sure it was
+ * available.
+ */
+ smp_rmb();
+
+ /* FIXME: support indirect */
+ if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_INDIRECT)) {
+ vq_err(vq, "Indirect descriptor is not supported\n");
+ return -EFAULT;
+ }
+
+ if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE))
+ access = VHOST_ACCESS_WO;
+ else
+ access = VHOST_ACCESS_RO;
+ ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr),
+ vhost32_to_cpu(vq, desc.len),
+ iov + iov_count, iov_size - iov_count,
+ access);
+ if (unlikely(ret < 0)) {
+ if (ret != -EAGAIN)
+ vq_err(vq, "Translation failure %d idx %d\n",
+ ret, i);
+ return ret;
+ }
+
+ if (access == VHOST_ACCESS_WO) {
+ /* If this is an input descriptor,
+ * increment that count.
+ */
+ *in_num += ret;
+ if (unlikely(log)) {
+ log[*log_num].addr =
+ vhost64_to_cpu(vq, desc.addr);
+ log[*log_num].len =
+ vhost32_to_cpu(vq, desc.len);
+ ++*log_num;
+ }
+ } else {
+ /* If it's an output descriptor, they're all supposed
+ * to come before any input descriptors.
+ */
+ if (unlikely(*in_num)) {
+ vq_err(vq, "Desc out after in: idx %d\n",
+ i);
+ return -EINVAL;
+ }
+ *out_num += ret;
+ }
+
+ /* On success, increment avail index. */
+ if ((++vq->last_avail_idx & (vq->num - 1)) == 0)
+ vq->used_wrap_counter ^= 1;
+
+ /* If this descriptor says it doesn't chain, we're done. */
+ } while (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_NEXT));
+
+ return desc.id;
+}
+
+static int vhost_get_vq_desc_split(struct vhost_virtqueue *vq,
+ struct iovec iov[], unsigned int iov_size,
+ unsigned int *out_num, unsigned int *in_num,
+ struct vhost_log *log, unsigned int *log_num)
{
struct vring_desc desc;
unsigned int i, head, found = 0;
@@ -2096,6 +2241,30 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq,
BUG_ON(!(vq->used_flags & VRING_USED_F_NO_NOTIFY));
return head;
}
+
+/* This looks in the virtqueue and for the first available buffer, and converts
+ * it to an iovec for convenient access. Since descriptors consist of some
+ * number of output then some number of input descriptors, it's actually two
+ * iovecs, but we pack them into one and note how many of each there were.
+ *
+ * This function returns the descriptor number found, or vq->num (which is
+ * never a valid descriptor number) if none was found. A negative code is
+ * returned on error.
+ */
+int vhost_get_vq_desc(struct vhost_virtqueue *vq,
+ struct iovec iov[], unsigned int iov_size,
+ unsigned int *out_num, unsigned int *in_num,
+ struct vhost_log *log, unsigned int *log_num)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vhost_get_vq_desc_packed(vq, iov, iov_size,
+ out_num, in_num,
+ log, log_num);
+ else
+ return vhost_get_vq_desc_split(vq, iov, iov_size,
+ out_num, in_num,
+ log, log_num);
+}
EXPORT_SYMBOL_GPL(vhost_get_vq_desc);

/* Reverse the effect of vhost_get_vq_desc. Useful for error handling. */
@@ -2161,10 +2330,48 @@ static int __vhost_add_used_n(struct vhost_virtqueue *vq,
return 0;
}

-/* After we've used one of their buffers, we tell them about it. We'll then
- * want to notify the guest, using eventfd. */
-int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
- unsigned count)
+static int vhost_add_used_n_packed(struct vhost_virtqueue *vq,
+ struct vring_used_elem *heads,
+ unsigned int count)
+{
+ struct vring_desc_packed desc = {
+ .addr = 0,
+ .flags = 0,
+ };
+ int i, ret;
+
+ for (i = 0; i < count; i++) {
+ desc.id = heads[i].id;
+ desc.len = heads[i].len;
+ set_desc_used(vq, &desc, heads[i].wrap_counter);
+
+ /* Update flags etc before desc is written */
+ smp_mb();
+
+ ret = vhost_copy_to_user(vq, vq->desc_packed + heads[i].idx,
+ &desc, sizeof(desc));
+ if (unlikely(ret)) {
+ vq_err(vq, "Failed to set descriptor: idx %d addr %p\n",
+ heads[i].idx, vq->desc_packed + heads[i].idx);
+ return -EFAULT;
+ }
+ if (unlikely(vq->log_used)) {
+ /* Make sure desc is written before update log. */
+ smp_wmb();
+ log_write(vq->log_base,
+ vq->log_addr + heads[i].idx * sizeof(desc),
+ sizeof(desc));
+ if (vq->log_ctx)
+ eventfd_signal(vq->log_ctx, 1);
+ }
+ }
+
+ return 0;
+}
+
+static int vhost_add_used_n_split(struct vhost_virtqueue *vq,
+ struct vring_used_elem *heads,
+ unsigned int count)
{
int start, n, r;

@@ -2196,6 +2403,19 @@ int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
}
return r;
}
+
+/* After we've used one of their buffers, we tell them about it. We'll then
+ * want to notify the guest, using eventfd.
+ */
+int vhost_add_used_n(struct vhost_virtqueue *vq,
+ struct vring_used_elem *heads,
+ unsigned int count)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vhost_add_used_n_packed(vq, heads, count);
+ else
+ return vhost_add_used_n_split(vq, heads, count);
+}
EXPORT_SYMBOL_GPL(vhost_add_used_n);

static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
@@ -2203,6 +2423,11 @@ static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
__u16 old, new;
__virtio16 event;
bool v;
+
+ /* FIXME: check driver area */
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return false;
+
/* Flush out used index updates. This is paired
* with the barrier that the Guest executes when enabling
* interrupts. */
@@ -2265,7 +2490,8 @@ void vhost_add_used_and_signal_n(struct vhost_dev *dev,
EXPORT_SYMBOL_GPL(vhost_add_used_and_signal_n);

/* return true if we're sure that avaiable ring is empty */
-bool vhost_vq_avail_empty(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+static bool vhost_vq_avail_empty_split(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
{
__virtio16 avail_idx;
int r;
@@ -2280,10 +2506,61 @@ bool vhost_vq_avail_empty(struct vhost_dev *dev, struct vhost_virtqueue *vq)

return vq->avail_idx == vq->last_avail_idx;
}
+
+/* FIXME: unify codes with vhost_enable_notify_packed() */
+static bool vhost_vq_avail_empty_packed(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
+{
+ struct vring_desc_packed desc;
+ int ret, i = vq->last_avail_idx & (vq->num - 1);
+
+ ret = vhost_copy_from_user(vq, &desc, vq->desc_packed + i,
+ sizeof(desc));
+ if (unlikely(ret)) {
+ vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
+ i, vq->desc_packed + i);
+ return -EFAULT;
+ }
+
+ /* Read flag after desc is read */
+ smp_mb();
+
+ return desc_is_avail(vq, &desc);
+}
+
+bool vhost_vq_avail_empty(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vhost_vq_avail_empty_packed(dev, vq);
+ else
+ return vhost_vq_avail_empty_split(dev, vq);
+}
EXPORT_SYMBOL_GPL(vhost_vq_avail_empty);

-/* OK, now we need to know about added descriptors. */
-bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+static bool vhost_enable_notify_packed(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
+{
+ struct vring_desc_packed desc;
+ int ret, i = vq->last_avail_idx & (vq->num - 1);
+
+ /* FIXME: disable notification through device area */
+
+ ret = vhost_copy_from_user(vq, &desc, vq->desc_packed + i,
+ sizeof(desc));
+ if (unlikely(ret)) {
+ vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
+ i, vq->desc_packed + i);
+ return -EFAULT;
+ }
+
+ /* Read flag after desc is read */
+ smp_mb();
+
+ return desc_is_avail(vq, &desc);
+}
+
+static bool vhost_enable_notify_split(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
{
__virtio16 avail_idx;
int r;
@@ -2318,10 +2595,25 @@ bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)

return vhost16_to_cpu(vq, avail_idx) != vq->avail_idx;
}
+
+/* OK, now we need to know about added descriptors. */
+bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vhost_enable_notify_packed(dev, vq);
+ else
+ return vhost_enable_notify_split(dev, vq);
+}
EXPORT_SYMBOL_GPL(vhost_enable_notify);

-/* We don't need to be notified again. */
-void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+static void vhost_disable_notify_packed(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
+{
+ /* FIXME: disable notification through device area */
+}
+
+static void vhost_disable_notify_split(struct vhost_dev *dev,
+ struct vhost_virtqueue *vq)
{
int r;

@@ -2335,6 +2627,15 @@ void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
&vq->used->flags, r);
}
}
+
+/* We don't need to be notified again. */
+void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
+{
+ if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED))
+ return vhost_disable_notify_packed(dev, vq);
+ else
+ return vhost_disable_notify_split(dev, vq);
+}
EXPORT_SYMBOL_GPL(vhost_disable_notify);

/* Create a new message. */
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index ac4b605..cf6533a 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -87,7 +87,10 @@ struct vhost_virtqueue {
/* The actual ring of buffers. */
struct mutex mutex;
unsigned int num;
- struct vring_desc __user *desc;
+ union {
+ struct vring_desc __user *desc;
+ struct vring_desc_packed __user *desc_packed;
+ };
struct vring_avail __user *avail;
struct vring_used __user *used;
const struct vhost_umem_node *meta_iotlb[VHOST_NUM_ADDRS];
@@ -144,6 +147,7 @@ struct vhost_virtqueue {
bool user_be;
#endif
u32 busyloop_timeout;
+ bool used_wrap_counter;
};

struct vhost_msg_node {
--
2.7.4


2018-02-14 02:48:34

by Michael S. Tsirkin

[permalink] [raw]
Subject: Re: [PATCH RFC 0/2] Packed ring for vhost

On Wed, Feb 14, 2018 at 10:37:07AM +0800, Jason Wang wrote:
> Hi all:
>
> This RFC implement a subset of packed ring which was described at
> https://github.com/oasis-tcs/virtio-docs/blob/master/virtio-v1.1-packed-wd07.pdf
> . The code were tested with pmd implement by Jens at
> http://dpdk.org/ml/archives/dev/2018-January/089417.html. Minor
> change was needed for pmd codes to kick virtqueue since it assumes a
> busy polling backend.
>
> Test were done between localhost and guest. Testpmd (rxonly) in guest
> reports 2.4Mpps. Testpmd (txonly) repots about 2.1Mpps.

How does this compare with the split ring design?

> It's not a complete implemention, here's what were missed:
>
> - Device Area
> - Driver Area
> - Descriptor indirection
> - Zerocopy may not be functional
> - Migration path is not tested
> - Vhost devices except for net
> - vIOMMU can not work (mainly because the metadata prefetch is not
> implemented).
> - See FIXME/TODO in the codes for more details
> - No batching or other optimizations were implemented

ioeventfd for PIO/mmio/s390.

> For a quick prototype, this series open code the tracking of warp
> counter and descriptor index at net device. This will be addressed in
> the future by:
>
> - Move get_rx_bufs() from net.c to vhost.c
> - Let vhost_get_vq_desc() returns vring_used_elem instad of head id
>
> With the above, we can hide the internal (at least part of) vring
> layout from specific device.
>
> Please review.
>
> Thanks
>
> Jason Wang (2):
> virtio: introduce packed ring defines
> vhost: packed ring support
>
> drivers/vhost/net.c | 14 +-
> drivers/vhost/vhost.c | 351 ++++++++++++++++++++++++++++++++++---
> drivers/vhost/vhost.h | 6 +-
> include/uapi/linux/virtio_config.h | 9 +
> include/uapi/linux/virtio_ring.h | 17 ++
> 5 files changed, 369 insertions(+), 28 deletions(-)
>
> --
> 2.7.4

2018-02-14 02:49:38

by Jason Wang

[permalink] [raw]
Subject: [PATCH RFC 1/2] virtio: introduce packed ring defines

Signed-off-by: Jason Wang <[email protected]>
---
include/uapi/linux/virtio_config.h | 9 +++++++++
include/uapi/linux/virtio_ring.h | 17 +++++++++++++++++
2 files changed, 26 insertions(+)

diff --git a/include/uapi/linux/virtio_config.h b/include/uapi/linux/virtio_config.h
index 308e209..5903d51 100644
--- a/include/uapi/linux/virtio_config.h
+++ b/include/uapi/linux/virtio_config.h
@@ -71,4 +71,13 @@
* this is for compatibility with legacy systems.
*/
#define VIRTIO_F_IOMMU_PLATFORM 33
+
+#define VIRTIO_F_RING_PACKED 34
+
+/*
+ * This feature indicates that all buffers are used by the device in
+ * the same order in which they have been made available.
+ */
+#define VIRTIO_F_IN_ORDER 35
+
#endif /* _UAPI_LINUX_VIRTIO_CONFIG_H */
diff --git a/include/uapi/linux/virtio_ring.h b/include/uapi/linux/virtio_ring.h
index 6d5d5fa..a169b53 100644
--- a/include/uapi/linux/virtio_ring.h
+++ b/include/uapi/linux/virtio_ring.h
@@ -43,6 +43,8 @@
#define VRING_DESC_F_WRITE 2
/* This means the buffer contains a list of buffer descriptors. */
#define VRING_DESC_F_INDIRECT 4
+#define VRING_DESC_F_AVAIL 7
+#define VRING_DESC_F_USED 15

/* The Host uses this in used->flags to advise the Guest: don't kick me when
* you add a buffer. It's unreliable, so it's simply an optimization. Guest
@@ -62,6 +64,17 @@
* at the end of the used ring. Guest should ignore the used->flags field. */
#define VIRTIO_RING_F_EVENT_IDX 29

+struct vring_desc_packed {
+ /* Buffer Address. */
+ __virtio64 addr;
+ /* Buffer Length. */
+ __virtio32 len;
+ /* Buffer ID. */
+ __virtio16 id;
+ /* The flags depending on descriptor type. */
+ __virtio16 flags;
+};
+
/* Virtio ring descriptors: 16 bytes. These can chain together via "next". */
struct vring_desc {
/* Address (guest-physical). */
@@ -86,6 +99,10 @@ struct vring_used_elem {
__virtio32 id;
/* Total length of the descriptor chain which was used (written to) */
__virtio32 len;
+ /* Index of the descriptor that needs to write, used by packed ring. */
+ u16 idx;
+ /* Wrap counter for this used desc, used by packed ring. */
+ bool wrap_counter;
};

struct vring_used {
--
2.7.4


2018-02-14 03:49:23

by Jason Wang

[permalink] [raw]
Subject: Re: [PATCH RFC 0/2] Packed ring for vhost



On 2018年02月14日 10:47, Michael S. Tsirkin wrote:
> On Wed, Feb 14, 2018 at 10:37:07AM +0800, Jason Wang wrote:
>> Hi all:
>>
>> This RFC implement a subset of packed ring which was described at
>> https://github.com/oasis-tcs/virtio-docs/blob/master/virtio-v1.1-packed-wd07.pdf
>> . The code were tested with pmd implement by Jens at
>> http://dpdk.org/ml/archives/dev/2018-January/089417.html. Minor
>> change was needed for pmd codes to kick virtqueue since it assumes a
>> busy polling backend.
>>
>> Test were done between localhost and guest. Testpmd (rxonly) in guest
>> reports 2.4Mpps. Testpmd (txonly) repots about 2.1Mpps.
> How does this compare with the split ring design?

No obvious difference (+-5%). I believe we reach the bottleneck of vhost.

>
>> It's not a complete implemention, here's what were missed:
>>
>> - Device Area
>> - Driver Area
>> - Descriptor indirection
>> - Zerocopy may not be functional
>> - Migration path is not tested
>> - Vhost devices except for net
>> - vIOMMU can not work (mainly because the metadata prefetch is not
>> implemented).
>> - See FIXME/TODO in the codes for more details
>> - No batching or other optimizations were implemented
> ioeventfd for PIO/mmio/s390.
>

Probably, but this is not the stuffs of packed ring I think.

Thanks

2018-02-27 09:08:33

by Tiwei Bie

[permalink] [raw]
Subject: Re: [PATCH RFC 2/2] vhost: packed ring support

On Wed, Feb 14, 2018 at 10:37:09AM +0800, Jason Wang wrote:
[...]
> +static void set_desc_used(struct vhost_virtqueue *vq,
> + struct vring_desc_packed *desc, bool wrap_counter)
> +{
> + __virtio16 flags = desc->flags;
> +
> + if (wrap_counter) {
> + desc->flags |= cpu_to_vhost16(vq, DESC_AVAIL);
> + desc->flags |= cpu_to_vhost16(vq, DESC_USED);
> + } else {
> + desc->flags &= ~cpu_to_vhost16(vq, DESC_AVAIL);
> + desc->flags &= ~cpu_to_vhost16(vq, DESC_USED);
> + }
> +
> + desc->flags = flags;

The `desc->flags` is restored after the change.

> +}
> +
> +static int vhost_get_vq_desc_packed(struct vhost_virtqueue *vq,
> + struct iovec iov[], unsigned int iov_size,
> + unsigned int *out_num, unsigned int *in_num,
> + struct vhost_log *log,
> + unsigned int *log_num)
> +{
> + struct vring_desc_packed desc;
> + int ret, access, i;
> + u16 avail_idx = vq->last_avail_idx;
> +
> + /* When we start there are none of either input nor output. */
> + *out_num = *in_num = 0;
> + if (unlikely(log))
> + *log_num = 0;
> +
> + do {
> + unsigned int iov_count = *in_num + *out_num;
> +
> + i = vq->last_avail_idx & (vq->num - 1);

The queue size may not be a power of 2 in packed ring.

Best regards,
Tiwei Bie

2018-02-28 02:39:42

by Jason Wang

[permalink] [raw]
Subject: Re: [PATCH RFC 2/2] vhost: packed ring support



On 2018年02月27日 17:03, Tiwei Bie wrote:
> On Wed, Feb 14, 2018 at 10:37:09AM +0800, Jason Wang wrote:
> [...]
>> +static void set_desc_used(struct vhost_virtqueue *vq,
>> + struct vring_desc_packed *desc, bool wrap_counter)
>> +{
>> + __virtio16 flags = desc->flags;
>> +
>> + if (wrap_counter) {
>> + desc->flags |= cpu_to_vhost16(vq, DESC_AVAIL);
>> + desc->flags |= cpu_to_vhost16(vq, DESC_USED);
>> + } else {
>> + desc->flags &= ~cpu_to_vhost16(vq, DESC_AVAIL);
>> + desc->flags &= ~cpu_to_vhost16(vq, DESC_USED);
>> + }
>> +
>> + desc->flags = flags;
> The `desc->flags` is restored after the change.

Right, will fix.

>> +}
>> +
>> +static int vhost_get_vq_desc_packed(struct vhost_virtqueue *vq,
>> + struct iovec iov[], unsigned int iov_size,
>> + unsigned int *out_num, unsigned int *in_num,
>> + struct vhost_log *log,
>> + unsigned int *log_num)
>> +{
>> + struct vring_desc_packed desc;
>> + int ret, access, i;
>> + u16 avail_idx = vq->last_avail_idx;
>> +
>> + /* When we start there are none of either input nor output. */
>> + *out_num = *in_num = 0;
>> + if (unlikely(log))
>> + *log_num = 0;
>> +
>> + do {
>> + unsigned int iov_count = *in_num + *out_num;
>> +
>> + i = vq->last_avail_idx & (vq->num - 1);
> The queue size may not be a power of 2 in packed ring.

Will fix this too.

Thanks

> Best regards,
> Tiwei Bie