[dpdk-dev] [PATCH v9 00/15] Vhost: add support to packed ring layout

[dpdk-dev] [PATCH v9 00/15] Vhost: add support to packed ring layout

[Maxime Coquelin]

This series is a handover from Jen's "[PATCH v4 00/20]
implement packed virtqueues", which only implements the
vhost side. Virtio PMD implementation will follow in a 
next series.

The series applies on top of previous reworks I posted
during this cycle that merges mergeable and non-mergeable
receive paths, and refactors transmit path to re-use
vector buffers.

I haven't run performance tests for now as the Virtio PMD
side isn't ready.

The series has been tested with Tiwei's series implementing
packed ring support to Kernel's virtio-net driver, and
with Wei series implementing the Qemu side.

To test it, I have used testpmd on host side with a vhost
vdev and a tap vdev forwarding to each other. Transferts
of big random files have been done in both ways with
integrity verified.

Tests have been run with Rx mrg ON/OFF and events suppression
ON/OFF.

Tests have also been run with legacy split ring layout to
ensure no regression have been introduced.

Changes since v8:
=================
- Fix indents (Tiwei)
- Rename struct vring_desc_packed to vring_packed_desc (Tiwei)

Changes since v7:
=================
- Align structs and defines naming with Kernel header (Tiwei)
- Fix event based notifications (Tiwei)
- Fix Clang build issues caused by unused symbols (Tiwei)

Changes since v6:
=================
- Various style cleanups (Tiwei, Jason)
- Simplify event based notification (Jason)
- Build support with future kernels (Tiwei)
- Prevent buffer vectors overflow in map_one_desc (Tiwei)

Changes since v5:
=================
- Remove duplicated VHOST_USER_F_PROTOCOL_FEATURES definition (Tiwei)
- Fix vq_is_ready (Maxime)
- Fix local used index overflow in flush_shadow_used_ring_packed (Tiwei)
- Make desc_is_avail() a bool (Tiwei)
- Improve desc_is_avail() logic (Tiwei)
- Remove split rings addr NULL assignment in the right patch (Tiwei)
- Make buffer id a uint16_t (Tiwei)

Jens Freimann (2):
  vhost: add virtio packed virtqueue defines
  vhost: add helpers for packed virtqueues

Maxime Coquelin (12):
  vhost: clear shadow used table index at flush time
  vhost: make indirect desc table copy desc type agnostic
  vhost: clear batch copy index at copy time
  vhost: extract split ring handling from Rx and Tx functions
  vhost: append shadow used ring function names with split
  vhost: add shadow used ring support for packed rings
  vhost: create descriptor mapping function
  vhost: add vector filling support for packed ring
  vhost: add Rx support for packed ring
  vhost: add Tx support for packed ring
  vhost: add notification for packed ring
  vhost: advertize packed ring layout support

Yuanhan Liu (1):
  vhost: vring address setup for packed queues

 lib/librte_vhost/vhost.c      | 115 ++++++-
 lib/librte_vhost/vhost.h      | 130 ++++++-
 lib/librte_vhost/vhost_user.c | 127 +++++--
 lib/librte_vhost/virtio_net.c | 776 +++++++++++++++++++++++++++++++++---------
 4 files changed, 939 insertions(+), 209 deletions(-)

-- 
2.14.4

[dpdk-dev] [PATCH v9 01/15] vhost: add virtio packed virtqueue defines

[Maxime Coquelin]

From: Jens Freimann <jfreimann@redhat.com>

Signed-off-by: Jens Freimann <jfreimann@redhat.com>
---
 lib/librte_vhost/vhost.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 79e3117d2..83c028a1b 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -192,6 +192,26 @@ struct vhost_msg {
  #define VIRTIO_F_VERSION_1 32
 #endif
 
+/* Declare packed ring related bits for older kernels */
+#ifndef VIRTIO_F_RING_PACKED
+
+#define VIRTIO_F_RING_PACKED 34
+
+#define VRING_DESC_F_NEXT	1
+#define VRING_DESC_F_WRITE	2
+#define VRING_DESC_F_INDIRECT	4
+
+#define VRING_DESC_F_AVAIL	(1ULL << 7)
+#define VRING_DESC_F_USED	(1ULL << 15)
+
+struct vring_packed_desc {
+	uint64_t addr;
+	uint32_t len;
+	uint16_t id;
+	uint16_t flags;
+};
+#endif
+
 /*
  * Available and used descs are in same order
  */
-- 
2.14.4

[dpdk-dev] [PATCH v9 02/15] vhost: add helpers for packed virtqueues

[Maxime Coquelin]

From: Jens Freimann <jfreimann@redhat.com>

Add some helper functions to check descriptor flags
and check if a vring is of type packed.

Signed-off-by: Jens Freimann <jfreimann@redhat.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 83c028a1b..d362823a9 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -344,6 +344,19 @@ struct virtio_net {
 	struct vhost_user_extern_ops extern_ops;
 } __rte_cache_aligned;
 
+static __rte_always_inline bool
+vq_is_packed(struct virtio_net *dev)
+{
+	return dev->features & (1ull << VIRTIO_F_RING_PACKED);
+}
+
+static inline bool
+desc_is_avail(struct vring_packed_desc *desc, bool wrap_counter)
+{
+	return wrap_counter == !!(desc->flags & VRING_DESC_F_AVAIL) &&
+		wrap_counter != !!(desc->flags & VRING_DESC_F_USED);
+}
+
 #define VHOST_LOG_PAGE	4096
 
 /*
-- 
2.14.4

[dpdk-dev] [PATCH v9 03/15] vhost: vring address setup for packed queues

[Maxime Coquelin]

From: Yuanhan Liu <yuanhan.liu@linux.intel.com>

Add code to set up packed queues when enabled.

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Signed-off-by: Jens Freimann <jfreiman@redhat.com>
Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.c      | 43 ++++++++++++++++++++++++++++++++++++++-----
 lib/librte_vhost/vhost.h      |  7 ++++++-
 lib/librte_vhost/vhost_user.c | 39 +++++++++++++++++++++++++++++++++++----
 3 files changed, 79 insertions(+), 10 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 13ce864db..0bf2cc14a 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -115,14 +115,11 @@ free_device(struct virtio_net *dev)
 	rte_free(dev);
 }
 
-int
-vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq)
+static int
+vring_translate_split(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
 	uint64_t req_size, size;
 
-	if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
-		goto out;
-
 	req_size = sizeof(struct vring_desc) * vq->size;
 	size = req_size;
 	vq->desc = (struct vring_desc *)(uintptr_t)vhost_iova_to_vva(dev, vq,
@@ -153,6 +150,40 @@ vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq)
 	if (!vq->used || size != req_size)
 		return -1;
 
+	return 0;
+}
+
+static int
+vring_translate_packed(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+	uint64_t req_size, size;
+
+	req_size = sizeof(struct vring_packed_desc) * vq->size;
+	size = req_size;
+	vq->desc_packed =
+		(struct vring_packed_desc *)(uintptr_t)vhost_iova_to_vva(dev,
+					vq, vq->ring_addrs.desc_user_addr,
+					&size, VHOST_ACCESS_RW);
+	if (!vq->desc_packed || size != req_size)
+		return -1;
+
+	return 0;
+}
+
+int
+vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+
+	if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
+		goto out;
+
+	if (vq_is_packed(dev)) {
+		if (vring_translate_packed(dev, vq) < 0)
+			return -1;
+	} else {
+		if (vring_translate_split(dev, vq) < 0)
+			return -1;
+	}
 out:
 	vq->access_ok = 1;
 
@@ -234,6 +265,8 @@ alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx)
 	dev->virtqueue[vring_idx] = vq;
 	init_vring_queue(dev, vring_idx);
 	rte_spinlock_init(&vq->access_lock);
+	vq->avail_wrap_counter = 1;
+	vq->used_wrap_counter = 1;
 
 	dev->nr_vring += 1;
 
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index d362823a9..b486682c5 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -84,7 +84,10 @@ struct log_cache_entry {
  * Structure contains variables relevant to RX/TX virtqueues.
  */
 struct vhost_virtqueue {
-	struct vring_desc	*desc;
+	union {
+		struct vring_desc	*desc;
+		struct vring_packed_desc   *desc_packed;
+	};
 	struct vring_avail	*avail;
 	struct vring_used	*used;
 	uint32_t		size;
@@ -122,6 +125,8 @@ struct vhost_virtqueue {
 
 	struct batch_copy_elem	*batch_copy_elems;
 	uint16_t		batch_copy_nb_elems;
+	bool			used_wrap_counter;
+	bool			avail_wrap_counter;
 
 	struct log_cache_entry log_cache[VHOST_LOG_CACHE_NR];
 	uint16_t log_cache_nb_elem;
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index bea6a0428..dca43ff00 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -467,6 +467,27 @@ translate_ring_addresses(struct virtio_net *dev, int vq_index)
 	struct vhost_vring_addr *addr = &vq->ring_addrs;
 	uint64_t len;
 
+	if (vq_is_packed(dev)) {
+		len = sizeof(struct vring_packed_desc) * vq->size;
+		vq->desc_packed = (struct vring_packed_desc *) ring_addr_to_vva
+			(dev, vq, addr->desc_user_addr, &len);
+		vq->log_guest_addr = 0;
+		if (vq->desc_packed == NULL ||
+				len != sizeof(struct vring_packed_desc) *
+				vq->size) {
+			RTE_LOG(DEBUG, VHOST_CONFIG,
+				"(%d) failed to map desc_packed ring.\n",
+				dev->vid);
+			return dev;
+		}
+
+		dev = numa_realloc(dev, vq_index);
+		vq = dev->virtqueue[vq_index];
+		addr = &vq->ring_addrs;
+
+		return dev;
+	}
+
 	/* The addresses are converted from QEMU virtual to Vhost virtual. */
 	if (vq->desc && vq->avail && vq->used)
 		return dev;
@@ -875,10 +896,20 @@ vhost_user_set_mem_table(struct virtio_net **pdev, struct VhostUserMsg *pmsg)
 	return -1;
 }
 
-static int
-vq_is_ready(struct vhost_virtqueue *vq)
+static bool
+vq_is_ready(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
-	return vq && vq->desc && vq->avail && vq->used &&
+	bool rings_ok;
+
+	if (!vq)
+		return false;
+
+	if (vq_is_packed(dev))
+		rings_ok = !!vq->desc_packed;
+	else
+		rings_ok = vq->desc && vq->avail && vq->used;
+
+	return rings_ok &&
 	       vq->kickfd != VIRTIO_UNINITIALIZED_EVENTFD &&
 	       vq->callfd != VIRTIO_UNINITIALIZED_EVENTFD;
 }
@@ -895,7 +926,7 @@ virtio_is_ready(struct virtio_net *dev)
 	for (i = 0; i < dev->nr_vring; i++) {
 		vq = dev->virtqueue[i];
 
-		if (!vq_is_ready(vq))
+		if (!vq_is_ready(dev, vq))
 			return 0;
 	}
 
-- 
2.14.4

[dpdk-dev] [PATCH v9 04/15] vhost: clear shadow used table index at flush time

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 82d5d9e17..d6b30899f 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -115,6 +115,7 @@ flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq)
 	vhost_log_cache_sync(dev, vq);
 
 	*(volatile uint16_t *)&vq->used->idx += vq->shadow_used_idx;
+	vq->shadow_used_idx = 0;
 	vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
 		sizeof(vq->used->idx));
 }
@@ -567,7 +568,6 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
 
-	vq->shadow_used_idx = 0;
 	avail_head = *((volatile uint16_t *)&vq->avail->idx);
 	for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
 		uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
@@ -1055,7 +1055,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 		goto out_access_unlock;
 
 	vq->batch_copy_nb_elems = 0;
-	vq->shadow_used_idx = 0;
 
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
 		vhost_user_iotlb_rd_lock(vq);
@@ -1086,7 +1085,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 
 		flush_shadow_used_ring(dev, vq);
 		vhost_vring_call(dev, vq);
-		vq->shadow_used_idx = 0;
 	}
 
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
-- 
2.14.4

[dpdk-dev] [PATCH v9 05/15] vhost: make indirect desc table copy desc type agnostic

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index d6b30899f..3bed77eec 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -37,16 +37,15 @@ is_valid_virt_queue_idx(uint32_t idx, int is_tx, uint32_t nr_vring)
 	return (is_tx ^ (idx & 1)) == 0 && idx < nr_vring;
 }
 
-static __rte_always_inline struct vring_desc *
+static __rte_always_inline void *
 alloc_copy_ind_table(struct virtio_net *dev, struct vhost_virtqueue *vq,
-					 struct vring_desc *desc)
+					 uint64_t desc_addr, uint64_t desc_len)
 {
-	struct vring_desc *idesc;
+	void *idesc;
 	uint64_t src, dst;
-	uint64_t len, remain = desc->len;
-	uint64_t desc_addr = desc->addr;
+	uint64_t len, remain = desc_len;
 
-	idesc = rte_malloc(__func__, desc->len, 0);
+	idesc = rte_malloc(__func__, desc_len, 0);
 	if (unlikely(!idesc))
 		return 0;
 
@@ -72,7 +71,7 @@ alloc_copy_ind_table(struct virtio_net *dev, struct vhost_virtqueue *vq,
 }
 
 static __rte_always_inline void
-free_ind_table(struct vring_desc *idesc)
+free_ind_table(void *idesc)
 {
 	rte_free(idesc);
 }
@@ -251,7 +250,8 @@ fill_vec_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			 * The indirect desc table is not contiguous
 			 * in process VA space, we have to copy it.
 			 */
-			idesc = alloc_copy_ind_table(dev, vq, &vq->desc[idx]);
+			idesc = alloc_copy_ind_table(dev, vq,
+					vq->desc[idx].addr, vq->desc[idx].len);
 			if (unlikely(!idesc))
 				return -1;
 
-- 
2.14.4

[dpdk-dev] [PATCH v9 06/15] vhost: clear batch copy index at copy time

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 3bed77eec..5cc3138d0 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -141,6 +141,8 @@ do_data_copy_enqueue(struct virtio_net *dev, struct vhost_virtqueue *vq)
 		vhost_log_cache_write(dev, vq, elem[i].log_addr, elem[i].len);
 		PRINT_PACKET(dev, (uintptr_t)elem[i].dst, elem[i].len, 0);
 	}
+
+	vq->batch_copy_nb_elems = 0;
 }
 
 static inline void
@@ -152,6 +154,8 @@ do_data_copy_dequeue(struct vhost_virtqueue *vq)
 
 	for (i = 0; i < count; i++)
 		rte_memcpy(elem[i].dst, elem[i].src, elem[i].len);
+
+	vq->batch_copy_nb_elems = 0;
 }
 
 /* avoid write operation when necessary, to lessen cache issues */
@@ -564,8 +568,6 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 	if (count == 0)
 		goto out;
 
-	vq->batch_copy_nb_elems = 0;
-
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
 
 	avail_head = *((volatile uint16_t *)&vq->avail->idx);
@@ -1054,8 +1056,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 	if (unlikely(vq->enabled == 0))
 		goto out_access_unlock;
 
-	vq->batch_copy_nb_elems = 0;
-
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
 		vhost_user_iotlb_rd_lock(vq);
 
-- 
2.14.4

[dpdk-dev] [PATCH v9 07/15] vhost: extract split ring handling from Rx and Tx functions

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 238 +++++++++++++++++++++++-------------------
 1 file changed, 129 insertions(+), 109 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 5cc3138d0..bdfd6ebef 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -226,13 +226,13 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf, struct virtio_net_hdr *net_hdr)
 }
 
 static __rte_always_inline int
-fill_vec_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
-			 uint32_t avail_idx, uint32_t *vec_idx,
+fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
+			 uint32_t avail_idx, uint16_t *vec_idx,
 			 struct buf_vector *buf_vec, uint16_t *desc_chain_head,
 			 uint16_t *desc_chain_len, uint8_t perm)
 {
 	uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
-	uint32_t vec_id = *vec_idx;
+	uint16_t vec_id = *vec_idx;
 	uint32_t len    = 0;
 	uint64_t dlen, desc_avail, desc_iova;
 	struct vring_desc *descs = vq->desc;
@@ -323,13 +323,13 @@ fill_vec_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
  * Returns -1 on fail, 0 on success
  */
 static inline int
-reserve_avail_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
+reserve_avail_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 				uint32_t size, struct buf_vector *buf_vec,
 				uint16_t *num_buffers, uint16_t avail_head,
 				uint16_t *nr_vec)
 {
 	uint16_t cur_idx;
-	uint32_t vec_idx = 0;
+	uint16_t vec_idx = 0;
 	uint16_t max_tries, tries = 0;
 
 	uint16_t head_idx = 0;
@@ -347,7 +347,8 @@ reserve_avail_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		if (unlikely(cur_idx == avail_head))
 			return -1;
 
-		if (unlikely(fill_vec_buf(dev, vq, cur_idx, &vec_idx, buf_vec,
+		if (unlikely(fill_vec_buf_split(dev, vq, cur_idx,
+						&vec_idx, buf_vec,
 						&head_idx, &len,
 						VHOST_ACCESS_RW) < 0))
 			return -1;
@@ -534,48 +535,22 @@ copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
 }
 
 static __rte_always_inline uint32_t
-virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
+virtio_dev_rx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	struct rte_mbuf **pkts, uint32_t count)
 {
-	struct vhost_virtqueue *vq;
 	uint32_t pkt_idx = 0;
 	uint16_t num_buffers;
 	struct buf_vector buf_vec[BUF_VECTOR_MAX];
 	uint16_t avail_head;
 
-	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
-	if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->nr_vring))) {
-		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
-			dev->vid, __func__, queue_id);
-		return 0;
-	}
-
-	vq = dev->virtqueue[queue_id];
-
-	rte_spinlock_lock(&vq->access_lock);
-
-	if (unlikely(vq->enabled == 0))
-		goto out_access_unlock;
-
-	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
-		vhost_user_iotlb_rd_lock(vq);
-
-	if (unlikely(vq->access_ok == 0))
-		if (unlikely(vring_translate(dev, vq) < 0))
-			goto out;
-
-	count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
-	if (count == 0)
-		goto out;
-
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
-
 	avail_head = *((volatile uint16_t *)&vq->avail->idx);
+
 	for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
 		uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
 		uint16_t nr_vec = 0;
 
-		if (unlikely(reserve_avail_buf(dev, vq,
+		if (unlikely(reserve_avail_buf_split(dev, vq,
 						pkt_len, buf_vec, &num_buffers,
 						avail_head, &nr_vec) < 0)) {
 			VHOST_LOG_DEBUG(VHOST_DATA,
@@ -608,6 +583,42 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 		vhost_vring_call(dev, vq);
 	}
 
+	return pkt_idx;
+}
+
+static __rte_always_inline uint32_t
+virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
+	struct rte_mbuf **pkts, uint32_t count)
+{
+	struct vhost_virtqueue *vq;
+
+	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+	if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->nr_vring))) {
+		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
+			dev->vid, __func__, queue_id);
+		return 0;
+	}
+
+	vq = dev->virtqueue[queue_id];
+
+	rte_spinlock_lock(&vq->access_lock);
+
+	if (unlikely(vq->enabled == 0))
+		goto out_access_unlock;
+
+	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
+		vhost_user_iotlb_rd_lock(vq);
+
+	if (unlikely(vq->access_ok == 0))
+		if (unlikely(vring_translate(dev, vq) < 0))
+			goto out;
+
+	count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
+	if (count == 0)
+		goto out;
+
+	count = virtio_dev_rx_split(dev, vq, pkts, count);
+
 out:
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
 		vhost_user_iotlb_rd_unlock(vq);
@@ -615,7 +626,7 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 out_access_unlock:
 	rte_spinlock_unlock(&vq->access_lock);
 
-	return pkt_idx;
+	return count;
 }
 
 uint16_t
@@ -1021,48 +1032,13 @@ restore_mbuf(struct rte_mbuf *m)
 	}
 }
 
-uint16_t
-rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
+static __rte_always_inline uint16_t
+virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
 {
-	struct virtio_net *dev;
-	struct rte_mbuf *rarp_mbuf = NULL;
-	struct vhost_virtqueue *vq;
-	uint32_t i = 0;
+	uint16_t i;
 	uint16_t free_entries;
 
-	dev = get_device(vid);
-	if (!dev)
-		return 0;
-
-	if (unlikely(!(dev->flags & VIRTIO_DEV_BUILTIN_VIRTIO_NET))) {
-		RTE_LOG(ERR, VHOST_DATA,
-			"(%d) %s: built-in vhost net backend is disabled.\n",
-			dev->vid, __func__);
-		return 0;
-	}
-
-	if (unlikely(!is_valid_virt_queue_idx(queue_id, 1, dev->nr_vring))) {
-		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
-			dev->vid, __func__, queue_id);
-		return 0;
-	}
-
-	vq = dev->virtqueue[queue_id];
-
-	if (unlikely(rte_spinlock_trylock(&vq->access_lock) == 0))
-		return 0;
-
-	if (unlikely(vq->enabled == 0))
-		goto out_access_unlock;
-
-	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
-		vhost_user_iotlb_rd_lock(vq);
-
-	if (unlikely(vq->access_ok == 0))
-		if (unlikely(vring_translate(dev, vq) < 0))
-			goto out;
-
 	if (unlikely(dev->dequeue_zero_copy)) {
 		struct zcopy_mbuf *zmbuf, *next;
 		int nr_updated = 0;
@@ -1089,39 +1065,10 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
 
-	/*
-	 * Construct a RARP broadcast packet, and inject it to the "pkts"
-	 * array, to looks like that guest actually send such packet.
-	 *
-	 * Check user_send_rarp() for more information.
-	 *
-	 * broadcast_rarp shares a cacheline in the virtio_net structure
-	 * with some fields that are accessed during enqueue and
-	 * rte_atomic16_cmpset() causes a write if using cmpxchg. This could
-	 * result in false sharing between enqueue and dequeue.
-	 *
-	 * Prevent unnecessary false sharing by reading broadcast_rarp first
-	 * and only performing cmpset if the read indicates it is likely to
-	 * be set.
-	 */
-
-	if (unlikely(rte_atomic16_read(&dev->broadcast_rarp) &&
-			rte_atomic16_cmpset((volatile uint16_t *)
-				&dev->broadcast_rarp.cnt, 1, 0))) {
-
-		rarp_mbuf = rte_net_make_rarp_packet(mbuf_pool, &dev->mac);
-		if (rarp_mbuf == NULL) {
-			RTE_LOG(ERR, VHOST_DATA,
-				"Failed to make RARP packet.\n");
-			return 0;
-		}
-		count -= 1;
-	}
-
 	free_entries = *((volatile uint16_t *)&vq->avail->idx) -
 			vq->last_avail_idx;
 	if (free_entries == 0)
-		goto out;
+		return 0;
 
 	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
 
@@ -1133,10 +1080,10 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 	for (i = 0; i < count; i++) {
 		struct buf_vector buf_vec[BUF_VECTOR_MAX];
 		uint16_t head_idx, dummy_len;
-		uint32_t nr_vec = 0;
+		uint16_t nr_vec = 0;
 		int err;
 
-		if (unlikely(fill_vec_buf(dev, vq,
+		if (unlikely(fill_vec_buf_split(dev, vq,
 						vq->last_avail_idx + i,
 						&nr_vec, buf_vec,
 						&head_idx, &dummy_len,
@@ -1195,6 +1142,79 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 		vhost_vring_call(dev, vq);
 	}
 
+	return i;
+}
+
+uint16_t
+rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
+	struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
+{
+	struct virtio_net *dev;
+	struct rte_mbuf *rarp_mbuf = NULL;
+	struct vhost_virtqueue *vq;
+
+	dev = get_device(vid);
+	if (!dev)
+		return 0;
+
+	if (unlikely(!(dev->flags & VIRTIO_DEV_BUILTIN_VIRTIO_NET))) {
+		RTE_LOG(ERR, VHOST_DATA,
+			"(%d) %s: built-in vhost net backend is disabled.\n",
+			dev->vid, __func__);
+		return 0;
+	}
+
+	if (unlikely(!is_valid_virt_queue_idx(queue_id, 1, dev->nr_vring))) {
+		RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
+			dev->vid, __func__, queue_id);
+		return 0;
+	}
+
+	vq = dev->virtqueue[queue_id];
+
+	if (unlikely(rte_spinlock_trylock(&vq->access_lock) == 0))
+		return 0;
+
+	if (unlikely(vq->enabled == 0))
+		goto out_access_unlock;
+
+	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
+		vhost_user_iotlb_rd_lock(vq);
+
+	if (unlikely(vq->access_ok == 0))
+		if (unlikely(vring_translate(dev, vq) < 0))
+			goto out;
+
+	/*
+	 * Construct a RARP broadcast packet, and inject it to the "pkts"
+	 * array, to looks like that guest actually send such packet.
+	 *
+	 * Check user_send_rarp() for more information.
+	 *
+	 * broadcast_rarp shares a cacheline in the virtio_net structure
+	 * with some fields that are accessed during enqueue and
+	 * rte_atomic16_cmpset() causes a write if using cmpxchg. This could
+	 * result in false sharing between enqueue and dequeue.
+	 *
+	 * Prevent unnecessary false sharing by reading broadcast_rarp first
+	 * and only performing cmpset if the read indicates it is likely to
+	 * be set.
+	 */
+	if (unlikely(rte_atomic16_read(&dev->broadcast_rarp) &&
+			rte_atomic16_cmpset((volatile uint16_t *)
+				&dev->broadcast_rarp.cnt, 1, 0))) {
+
+		rarp_mbuf = rte_net_make_rarp_packet(mbuf_pool, &dev->mac);
+		if (rarp_mbuf == NULL) {
+			RTE_LOG(ERR, VHOST_DATA,
+				"Failed to make RARP packet.\n");
+			return 0;
+		}
+		count -= 1;
+	}
+
+	count = virtio_dev_tx_split(dev, vq, mbuf_pool, pkts, count);
+
 out:
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
 		vhost_user_iotlb_rd_unlock(vq);
@@ -1207,10 +1227,10 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 		 * Inject it to the head of "pkts" array, so that switch's mac
 		 * learning table will get updated first.
 		 */
-		memmove(&pkts[1], pkts, i * sizeof(struct rte_mbuf *));
+		memmove(&pkts[1], pkts, count * sizeof(struct rte_mbuf *));
 		pkts[0] = rarp_mbuf;
-		i += 1;
+		count += 1;
 	}
 
-	return i;
+	return count;
 }
-- 
2.14.4

[dpdk-dev] [PATCH v9 08/15] vhost: append shadow used ring function names with split

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index bdfd6ebef..ae256e062 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -77,8 +77,9 @@ free_ind_table(void *idesc)
 }
 
 static __rte_always_inline void
-do_flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq,
-			  uint16_t to, uint16_t from, uint16_t size)
+do_flush_shadow_used_ring_split(struct virtio_net *dev,
+			struct vhost_virtqueue *vq,
+			uint16_t to, uint16_t from, uint16_t size)
 {
 	rte_memcpy(&vq->used->ring[to],
 			&vq->shadow_used_ring[from],
@@ -89,22 +90,22 @@ do_flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq,
 }
 
 static __rte_always_inline void
-flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq)
+flush_shadow_used_ring_split(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
 	uint16_t used_idx = vq->last_used_idx & (vq->size - 1);
 
 	if (used_idx + vq->shadow_used_idx <= vq->size) {
-		do_flush_shadow_used_ring(dev, vq, used_idx, 0,
+		do_flush_shadow_used_ring_split(dev, vq, used_idx, 0,
 					  vq->shadow_used_idx);
 	} else {
 		uint16_t size;
 
 		/* update used ring interval [used_idx, vq->size] */
 		size = vq->size - used_idx;
-		do_flush_shadow_used_ring(dev, vq, used_idx, 0, size);
+		do_flush_shadow_used_ring_split(dev, vq, used_idx, 0, size);
 
 		/* update the left half used ring interval [0, left_size] */
-		do_flush_shadow_used_ring(dev, vq, 0, size,
+		do_flush_shadow_used_ring_split(dev, vq, 0, size,
 					  vq->shadow_used_idx - size);
 	}
 	vq->last_used_idx += vq->shadow_used_idx;
@@ -120,7 +121,7 @@ flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq)
 }
 
 static __rte_always_inline void
-update_shadow_used_ring(struct vhost_virtqueue *vq,
+update_shadow_used_ring_split(struct vhost_virtqueue *vq,
 			 uint16_t desc_idx, uint16_t len)
 {
 	uint16_t i = vq->shadow_used_idx++;
@@ -353,7 +354,7 @@ reserve_avail_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 						VHOST_ACCESS_RW) < 0))
 			return -1;
 		len = RTE_MIN(len, size);
-		update_shadow_used_ring(vq, head_idx, len);
+		update_shadow_used_ring_split(vq, head_idx, len);
 		size -= len;
 
 		cur_idx++;
@@ -579,7 +580,7 @@ virtio_dev_rx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	do_data_copy_enqueue(dev, vq);
 
 	if (likely(vq->shadow_used_idx)) {
-		flush_shadow_used_ring(dev, vq);
+		flush_shadow_used_ring_split(dev, vq);
 		vhost_vring_call(dev, vq);
 	}
 
@@ -1048,7 +1049,8 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			next = TAILQ_NEXT(zmbuf, next);
 
 			if (mbuf_is_consumed(zmbuf->mbuf)) {
-				update_shadow_used_ring(vq, zmbuf->desc_idx, 0);
+				update_shadow_used_ring_split(vq,
+						zmbuf->desc_idx, 0);
 				nr_updated += 1;
 
 				TAILQ_REMOVE(&vq->zmbuf_list, zmbuf, next);
@@ -1059,7 +1061,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			}
 		}
 
-		flush_shadow_used_ring(dev, vq);
+		flush_shadow_used_ring_split(dev, vq);
 		vhost_vring_call(dev, vq);
 	}
 
@@ -1091,7 +1093,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			break;
 
 		if (likely(dev->dequeue_zero_copy == 0))
-			update_shadow_used_ring(vq, head_idx, 0);
+			update_shadow_used_ring_split(vq, head_idx, 0);
 
 		rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
 
@@ -1138,7 +1140,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		do_data_copy_dequeue(vq);
 		if (unlikely(i < count))
 			vq->shadow_used_idx = i;
-		flush_shadow_used_ring(dev, vq);
+		flush_shadow_used_ring_split(dev, vq);
 		vhost_vring_call(dev, vq);
 	}
 
-- 
2.14.4

[dpdk-dev] [PATCH v9 09/15] vhost: add shadow used ring support for packed rings

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.c      |  9 ++++--
 lib/librte_vhost/vhost.h      | 13 ++++++--
 lib/librte_vhost/vhost_user.c | 64 ++++++++++++++++++++++++++++----------
 lib/librte_vhost/virtio_net.c | 71 +++++++++++++++++++++++++++++++++++++++++--
 4 files changed, 133 insertions(+), 24 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 0bf2cc14a..534f30cf6 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -93,9 +93,12 @@ cleanup_device(struct virtio_net *dev, int destroy)
 }
 
 void
-free_vq(struct vhost_virtqueue *vq)
+free_vq(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
-	rte_free(vq->shadow_used_ring);
+	if (vq_is_packed(dev))
+		rte_free(vq->shadow_used_packed);
+	else
+		rte_free(vq->shadow_used_split);
 	rte_free(vq->batch_copy_elems);
 	rte_mempool_free(vq->iotlb_pool);
 	rte_free(vq);
@@ -110,7 +113,7 @@ free_device(struct virtio_net *dev)
 	uint32_t i;
 
 	for (i = 0; i < dev->nr_vring; i++)
-		free_vq(dev->virtqueue[i]);
+		free_vq(dev, dev->virtqueue[i]);
 
 	rte_free(dev);
 }
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index b486682c5..70f0eebcf 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -80,6 +80,12 @@ struct log_cache_entry {
 	unsigned long val;
 };
 
+struct vring_used_elem_packed {
+	uint16_t id;
+	uint32_t len;
+	uint32_t count;
+};
+
 /**
  * Structure contains variables relevant to RX/TX virtqueues.
  */
@@ -119,7 +125,10 @@ struct vhost_virtqueue {
 	struct zcopy_mbuf	*zmbufs;
 	struct zcopy_mbuf_list	zmbuf_list;
 
-	struct vring_used_elem  *shadow_used_ring;
+	union {
+		struct vring_used_elem  *shadow_used_split;
+		struct vring_used_elem_packed *shadow_used_packed;
+	};
 	uint16_t                shadow_used_idx;
 	struct vhost_vring_addr ring_addrs;
 
@@ -587,7 +596,7 @@ void vhost_destroy_device(int);
 void vhost_destroy_device_notify(struct virtio_net *dev);
 
 void cleanup_vq(struct vhost_virtqueue *vq, int destroy);
-void free_vq(struct vhost_virtqueue *vq);
+void free_vq(struct virtio_net *dev, struct vhost_virtqueue *vq);
 
 int alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx);
 
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index dca43ff00..71d1fe0ac 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -233,7 +233,7 @@ vhost_user_set_features(struct virtio_net *dev, uint64_t features)
 
 			dev->virtqueue[dev->nr_vring] = NULL;
 			cleanup_vq(vq, 1);
-			free_vq(vq);
+			free_vq(dev, vq);
 		}
 	}
 
@@ -282,13 +282,26 @@ vhost_user_set_vring_num(struct virtio_net *dev,
 		TAILQ_INIT(&vq->zmbuf_list);
 	}
 
-	vq->shadow_used_ring = rte_malloc(NULL,
+	if (vq_is_packed(dev)) {
+		vq->shadow_used_packed = rte_malloc(NULL,
+				vq->size *
+				sizeof(struct vring_used_elem_packed),
+				RTE_CACHE_LINE_SIZE);
+		if (!vq->shadow_used_packed) {
+			RTE_LOG(ERR, VHOST_CONFIG,
+					"failed to allocate memory for shadow used ring.\n");
+			return -1;
+		}
+
+	} else {
+		vq->shadow_used_split = rte_malloc(NULL,
 				vq->size * sizeof(struct vring_used_elem),
 				RTE_CACHE_LINE_SIZE);
-	if (!vq->shadow_used_ring) {
-		RTE_LOG(ERR, VHOST_CONFIG,
-			"failed to allocate memory for shadow used ring.\n");
-		return -1;
+		if (!vq->shadow_used_split) {
+			RTE_LOG(ERR, VHOST_CONFIG,
+					"failed to allocate memory for shadow used ring.\n");
+			return -1;
+		}
 	}
 
 	vq->batch_copy_elems = rte_malloc(NULL,
@@ -315,7 +328,8 @@ numa_realloc(struct virtio_net *dev, int index)
 	struct virtio_net *old_dev;
 	struct vhost_virtqueue *old_vq, *vq;
 	struct zcopy_mbuf *new_zmbuf;
-	struct vring_used_elem *new_shadow_used_ring;
+	struct vring_used_elem *new_shadow_used_split;
+	struct vring_used_elem_packed *new_shadow_used_packed;
 	struct batch_copy_elem *new_batch_copy_elems;
 	int ret;
 
@@ -350,13 +364,26 @@ numa_realloc(struct virtio_net *dev, int index)
 			vq->zmbufs = new_zmbuf;
 		}
 
-		new_shadow_used_ring = rte_malloc_socket(NULL,
-			vq->size * sizeof(struct vring_used_elem),
-			RTE_CACHE_LINE_SIZE,
-			newnode);
-		if (new_shadow_used_ring) {
-			rte_free(vq->shadow_used_ring);
-			vq->shadow_used_ring = new_shadow_used_ring;
+		if (vq_is_packed(dev)) {
+			new_shadow_used_packed = rte_malloc_socket(NULL,
+					vq->size *
+					sizeof(struct vring_used_elem_packed),
+					RTE_CACHE_LINE_SIZE,
+					newnode);
+			if (new_shadow_used_packed) {
+				rte_free(vq->shadow_used_packed);
+				vq->shadow_used_packed = new_shadow_used_packed;
+			}
+		} else {
+			new_shadow_used_split = rte_malloc_socket(NULL,
+					vq->size *
+					sizeof(struct vring_used_elem),
+					RTE_CACHE_LINE_SIZE,
+					newnode);
+			if (new_shadow_used_split) {
+				rte_free(vq->shadow_used_split);
+				vq->shadow_used_split = new_shadow_used_split;
+			}
 		}
 
 		new_batch_copy_elems = rte_malloc_socket(NULL,
@@ -1047,8 +1074,13 @@ vhost_user_get_vring_base(struct virtio_net *dev,
 
 	if (dev->dequeue_zero_copy)
 		free_zmbufs(vq);
-	rte_free(vq->shadow_used_ring);
-	vq->shadow_used_ring = NULL;
+	if (vq_is_packed(dev)) {
+		rte_free(vq->shadow_used_packed);
+		vq->shadow_used_packed = NULL;
+	} else {
+		rte_free(vq->shadow_used_split);
+		vq->shadow_used_split = NULL;
+	}
 
 	rte_free(vq->batch_copy_elems);
 	vq->batch_copy_elems = NULL;
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index ae256e062..44b9daf4a 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -82,7 +82,7 @@ do_flush_shadow_used_ring_split(struct virtio_net *dev,
 			uint16_t to, uint16_t from, uint16_t size)
 {
 	rte_memcpy(&vq->used->ring[to],
-			&vq->shadow_used_ring[from],
+			&vq->shadow_used_split[from],
 			size * sizeof(struct vring_used_elem));
 	vhost_log_cache_used_vring(dev, vq,
 			offsetof(struct vring_used, ring[to]),
@@ -126,8 +126,73 @@ update_shadow_used_ring_split(struct vhost_virtqueue *vq,
 {
 	uint16_t i = vq->shadow_used_idx++;
 
-	vq->shadow_used_ring[i].id  = desc_idx;
-	vq->shadow_used_ring[i].len = len;
+	vq->shadow_used_split[i].id  = desc_idx;
+	vq->shadow_used_split[i].len = len;
+}
+
+static __rte_unused __rte_always_inline void
+flush_shadow_used_ring_packed(struct virtio_net *dev,
+			struct vhost_virtqueue *vq)
+{
+	int i;
+	uint16_t used_idx = vq->last_used_idx;
+
+	/* Split loop in two to save memory barriers */
+	for (i = 0; i < vq->shadow_used_idx; i++) {
+		vq->desc_packed[used_idx].id = vq->shadow_used_packed[i].id;
+		vq->desc_packed[used_idx].len = vq->shadow_used_packed[i].len;
+
+		used_idx += vq->shadow_used_packed[i].count;
+		if (used_idx >= vq->size)
+			used_idx -= vq->size;
+	}
+
+	rte_smp_wmb();
+
+	for (i = 0; i < vq->shadow_used_idx; i++) {
+		uint16_t flags;
+
+		if (vq->shadow_used_packed[i].len)
+			flags = VRING_DESC_F_WRITE;
+		else
+			flags = 0;
+
+		if (vq->used_wrap_counter) {
+			flags |= VRING_DESC_F_USED;
+			flags |= VRING_DESC_F_AVAIL;
+		} else {
+			flags &= ~VRING_DESC_F_USED;
+			flags &= ~VRING_DESC_F_AVAIL;
+		}
+
+		vq->desc_packed[vq->last_used_idx].flags = flags;
+
+		vhost_log_cache_used_vring(dev, vq,
+					vq->last_used_idx *
+					sizeof(struct vring_packed_desc),
+					sizeof(struct vring_packed_desc));
+
+		vq->last_used_idx += vq->shadow_used_packed[i].count;
+		if (vq->last_used_idx >= vq->size) {
+			vq->used_wrap_counter ^= 1;
+			vq->last_used_idx -= vq->size;
+		}
+	}
+
+	rte_smp_wmb();
+	vq->shadow_used_idx = 0;
+	vhost_log_cache_sync(dev, vq);
+}
+
+static __rte_unused __rte_always_inline void
+update_shadow_used_ring_packed(struct vhost_virtqueue *vq,
+			 uint16_t desc_idx, uint16_t len, uint16_t count)
+{
+	uint16_t i = vq->shadow_used_idx++;
+
+	vq->shadow_used_packed[i].id  = desc_idx;
+	vq->shadow_used_packed[i].len = len;
+	vq->shadow_used_packed[i].count = count;
 }
 
 static inline void
-- 
2.14.4

[dpdk-dev] [PATCH v9 10/15] vhost: create descriptor mapping function

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 70 ++++++++++++++++++++++++-------------------
 1 file changed, 40 insertions(+), 30 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 44b9daf4a..058786871 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -291,6 +291,40 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf, struct virtio_net_hdr *net_hdr)
 	}
 }
 
+static __rte_always_inline int
+map_one_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
+		struct buf_vector *buf_vec, uint16_t *vec_idx,
+		uint64_t desc_iova, uint64_t desc_len, uint8_t perm)
+{
+	uint16_t vec_id = *vec_idx;
+
+	while (desc_len) {
+		uint64_t desc_addr;
+		uint64_t desc_chunck_len = desc_len;
+
+		if (unlikely(vec_id >= BUF_VECTOR_MAX))
+			return -1;
+
+		desc_addr = vhost_iova_to_vva(dev, vq,
+				desc_iova,
+				&desc_chunck_len,
+				perm);
+		if (unlikely(!desc_addr))
+			return -1;
+
+		buf_vec[vec_id].buf_iova = desc_iova;
+		buf_vec[vec_id].buf_addr = desc_addr;
+		buf_vec[vec_id].buf_len  = desc_chunck_len;
+
+		desc_len -= desc_chunck_len;
+		desc_iova += desc_chunck_len;
+		vec_id++;
+	}
+	*vec_idx = vec_id;
+
+	return 0;
+}
+
 static __rte_always_inline int
 fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			 uint32_t avail_idx, uint16_t *vec_idx,
@@ -300,7 +334,7 @@ fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
 	uint16_t vec_id = *vec_idx;
 	uint32_t len    = 0;
-	uint64_t dlen, desc_avail, desc_iova;
+	uint64_t dlen;
 	struct vring_desc *descs = vq->desc;
 	struct vring_desc *idesc = NULL;
 
@@ -337,37 +371,13 @@ fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			return -1;
 		}
 
-
 		len += descs[idx].len;
-		desc_avail = descs[idx].len;
-		desc_iova = descs[idx].addr;
-
-		while (desc_avail) {
-			uint64_t desc_addr;
-			uint64_t desc_chunck_len = desc_avail;
-
-			if (unlikely(vec_id >= BUF_VECTOR_MAX)) {
-				free_ind_table(idesc);
-				return -1;
-			}
 
-			desc_addr = vhost_iova_to_vva(dev, vq,
-					desc_iova,
-					&desc_chunck_len,
-					perm);
-			if (unlikely(!desc_addr)) {
-				free_ind_table(idesc);
-				return -1;
-			}
-
-			buf_vec[vec_id].buf_iova = desc_iova;
-			buf_vec[vec_id].buf_addr = desc_addr;
-			buf_vec[vec_id].buf_len  = desc_chunck_len;
-			buf_vec[vec_id].desc_idx = idx;
-
-			desc_avail -= desc_chunck_len;
-			desc_iova += desc_chunck_len;
-			vec_id++;
+		if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+						descs[idx].addr, descs[idx].len,
+						perm))) {
+			free_ind_table(idesc);
+			return -1;
 		}
 
 		if ((descs[idx].flags & VRING_DESC_F_NEXT) == 0)
-- 
2.14.4

[dpdk-dev] [PATCH v9 11/15] vhost: add vector filling support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 111 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 111 insertions(+)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 058786871..9171ee733 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -450,6 +450,117 @@ reserve_avail_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	return 0;
 }
 
+static __rte_always_inline int
+fill_vec_buf_packed_indirect(struct virtio_net *dev,
+			struct vhost_virtqueue *vq,
+			struct vring_packed_desc *desc, uint16_t *vec_idx,
+			struct buf_vector *buf_vec, uint16_t *len, uint8_t perm)
+{
+	uint16_t i;
+	uint32_t nr_descs;
+	uint16_t vec_id = *vec_idx;
+	uint64_t dlen;
+	struct vring_packed_desc *descs, *idescs = NULL;
+
+	dlen = desc->len;
+	descs = (struct vring_packed_desc *)(uintptr_t)
+		vhost_iova_to_vva(dev, vq, desc->addr, &dlen, VHOST_ACCESS_RO);
+	if (unlikely(!descs))
+		return -1;
+
+	if (unlikely(dlen < desc->len)) {
+		/*
+		 * The indirect desc table is not contiguous
+		 * in process VA space, we have to copy it.
+		 */
+		idescs = alloc_copy_ind_table(dev, vq, desc->addr, desc->len);
+		if (unlikely(!idescs))
+			return -1;
+
+		descs = idescs;
+	}
+
+	nr_descs =  desc->len / sizeof(struct vring_packed_desc);
+	if (unlikely(nr_descs >= vq->size)) {
+		free_ind_table(idescs);
+		return -1;
+	}
+
+	for (i = 0; i < nr_descs; i++) {
+		if (unlikely(vec_id >= BUF_VECTOR_MAX)) {
+			free_ind_table(idescs);
+			return -1;
+		}
+
+		*len += descs[i].len;
+		if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+						descs[i].addr, descs[i].len,
+						perm)))
+			return -1;
+	}
+	*vec_idx = vec_id;
+
+	if (unlikely(!!idescs))
+		free_ind_table(idescs);
+
+	return 0;
+}
+
+static __rte_unused __rte_always_inline int
+fill_vec_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+				uint16_t avail_idx, uint16_t *desc_count,
+				struct buf_vector *buf_vec, uint16_t *vec_idx,
+				uint16_t *buf_id, uint16_t *len, uint8_t perm)
+{
+	bool wrap_counter = vq->avail_wrap_counter;
+	struct vring_packed_desc *descs = vq->desc_packed;
+	uint16_t vec_id = *vec_idx;
+
+	if (avail_idx < vq->last_avail_idx)
+		wrap_counter ^= 1;
+
+	if (unlikely(!desc_is_avail(&descs[avail_idx], wrap_counter)))
+		return -1;
+
+	*desc_count = 0;
+
+	while (1) {
+		if (unlikely(vec_id >= BUF_VECTOR_MAX))
+			return -1;
+
+		*desc_count += 1;
+		*buf_id = descs[avail_idx].id;
+
+		if (descs[avail_idx].flags & VRING_DESC_F_INDIRECT) {
+			if (unlikely(fill_vec_buf_packed_indirect(dev, vq,
+							&descs[avail_idx],
+							&vec_id, buf_vec,
+							len, perm) < 0))
+				return -1;
+		} else {
+			*len += descs[avail_idx].len;
+
+			if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+							descs[avail_idx].addr,
+							descs[avail_idx].len,
+							perm)))
+				return -1;
+		}
+
+		if ((descs[avail_idx].flags & VRING_DESC_F_NEXT) == 0)
+			break;
+
+		if (++avail_idx >= vq->size) {
+			avail_idx -= vq->size;
+			wrap_counter ^= 1;
+		}
+	}
+
+	*vec_idx = vec_id;
+
+	return 0;
+}
+
 static __rte_always_inline int
 copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			    struct rte_mbuf *m, struct buf_vector *buf_vec,
-- 
2.14.4

[dpdk-dev] [PATCH v9 12/15] vhost: add Rx support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/virtio_net.c | 123 ++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 119 insertions(+), 4 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 9171ee733..ef86e5b40 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -130,7 +130,7 @@ update_shadow_used_ring_split(struct vhost_virtqueue *vq,
 	vq->shadow_used_split[i].len = len;
 }
 
-static __rte_unused __rte_always_inline void
+static __rte_always_inline void
 flush_shadow_used_ring_packed(struct virtio_net *dev,
 			struct vhost_virtqueue *vq)
 {
@@ -184,7 +184,7 @@ flush_shadow_used_ring_packed(struct virtio_net *dev,
 	vhost_log_cache_sync(dev, vq);
 }
 
-static __rte_unused __rte_always_inline void
+static __rte_always_inline void
 update_shadow_used_ring_packed(struct vhost_virtqueue *vq,
 			 uint16_t desc_idx, uint16_t len, uint16_t count)
 {
@@ -506,7 +506,7 @@ fill_vec_buf_packed_indirect(struct virtio_net *dev,
 	return 0;
 }
 
-static __rte_unused __rte_always_inline int
+static __rte_always_inline int
 fill_vec_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
 				uint16_t avail_idx, uint16_t *desc_count,
 				struct buf_vector *buf_vec, uint16_t *vec_idx,
@@ -561,6 +561,65 @@ fill_vec_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	return 0;
 }
 
+/*
+ * Returns -1 on fail, 0 on success
+ */
+static inline int
+reserve_avail_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+				uint32_t size, struct buf_vector *buf_vec,
+				uint16_t *nr_vec, uint16_t *num_buffers,
+				uint16_t *nr_descs)
+{
+	uint16_t avail_idx;
+	uint16_t vec_idx = 0;
+	uint16_t max_tries, tries = 0;
+
+	uint16_t buf_id = 0;
+	uint16_t len = 0;
+	uint16_t desc_count;
+
+	*num_buffers = 0;
+	avail_idx = vq->last_avail_idx;
+
+	if (rxvq_is_mergeable(dev))
+		max_tries = vq->size;
+	else
+		max_tries = 1;
+
+	while (size > 0) {
+		if (unlikely(fill_vec_buf_packed(dev, vq,
+						avail_idx, &desc_count,
+						buf_vec, &vec_idx,
+						&buf_id, &len,
+						VHOST_ACCESS_RO) < 0))
+			return -1;
+
+		len = RTE_MIN(len, size);
+		update_shadow_used_ring_packed(vq, buf_id, len, desc_count);
+		size -= len;
+
+		avail_idx += desc_count;
+		if (avail_idx >= vq->size)
+			avail_idx -= vq->size;
+
+		*nr_descs += desc_count;
+		tries++;
+		*num_buffers += 1;
+
+		/*
+		 * if we tried all available ring items, and still
+		 * can't get enough buf, it means something abnormal
+		 * happened.
+		 */
+		if (unlikely(tries > max_tries))
+			return -1;
+	}
+
+	*nr_vec = vec_idx;
+
+	return 0;
+}
+
 static __rte_always_inline int
 copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
 			    struct rte_mbuf *m, struct buf_vector *buf_vec,
@@ -773,6 +832,59 @@ virtio_dev_rx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	return pkt_idx;
 }
 
+static __rte_always_inline uint32_t
+virtio_dev_rx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+	struct rte_mbuf **pkts, uint32_t count)
+{
+	uint32_t pkt_idx = 0;
+	uint16_t num_buffers;
+	struct buf_vector buf_vec[BUF_VECTOR_MAX];
+
+	for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
+		uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
+		uint16_t nr_vec = 0;
+		uint16_t nr_descs = 0;
+
+		if (unlikely(reserve_avail_buf_packed(dev, vq,
+						pkt_len, buf_vec, &nr_vec,
+						&num_buffers, &nr_descs) < 0)) {
+			VHOST_LOG_DEBUG(VHOST_DATA,
+				"(%d) failed to get enough desc from vring\n",
+				dev->vid);
+			vq->shadow_used_idx -= num_buffers;
+			break;
+		}
+
+		rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
+
+		VHOST_LOG_DEBUG(VHOST_DATA, "(%d) current index %d | end index %d\n",
+			dev->vid, vq->last_avail_idx,
+			vq->last_avail_idx + num_buffers);
+
+		if (copy_mbuf_to_desc(dev, vq, pkts[pkt_idx],
+						buf_vec, nr_vec,
+						num_buffers) < 0) {
+			vq->shadow_used_idx -= num_buffers;
+			break;
+		}
+
+		vq->last_avail_idx += nr_descs;
+		if (vq->last_avail_idx >= vq->size) {
+			vq->last_avail_idx -= vq->size;
+			vq->avail_wrap_counter ^= 1;
+		}
+	}
+
+	do_data_copy_enqueue(dev, vq);
+
+	if (likely(vq->shadow_used_idx)) {
+		flush_shadow_used_ring_packed(dev, vq);
+		vhost_vring_call(dev, vq);
+	}
+
+	return pkt_idx;
+}
+
 static __rte_always_inline uint32_t
 virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 	struct rte_mbuf **pkts, uint32_t count)
@@ -804,7 +916,10 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 	if (count == 0)
 		goto out;
 
-	count = virtio_dev_rx_split(dev, vq, pkts, count);
+	if (vq_is_packed(dev))
+		count = virtio_dev_rx_packed(dev, vq, pkts, count);
+	else
+		count = virtio_dev_rx_split(dev, vq, pkts, count);
 
 out:
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
-- 
2.14.4

[dpdk-dev] [PATCH v9 13/15] vhost: add Tx support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.h      |   1 +
 lib/librte_vhost/virtio_net.c | 119 +++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 119 insertions(+), 1 deletion(-)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 70f0eebcf..a7e602bec 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -56,6 +56,7 @@ struct buf_vector {
 struct zcopy_mbuf {
 	struct rte_mbuf *mbuf;
 	uint32_t desc_idx;
+	uint16_t desc_count;
 	uint16_t in_use;
 
 	TAILQ_ENTRY(zcopy_mbuf) next;
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index ef86e5b40..9f33821f4 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -1448,6 +1448,120 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 	return i;
 }
 
+static __rte_always_inline uint16_t
+virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+	struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
+{
+	uint16_t i;
+
+	rte_prefetch0(&vq->desc_packed[vq->last_avail_idx]);
+
+	if (unlikely(dev->dequeue_zero_copy)) {
+		struct zcopy_mbuf *zmbuf, *next;
+
+		for (zmbuf = TAILQ_FIRST(&vq->zmbuf_list);
+		     zmbuf != NULL; zmbuf = next) {
+			next = TAILQ_NEXT(zmbuf, next);
+
+			if (mbuf_is_consumed(zmbuf->mbuf)) {
+				update_shadow_used_ring_packed(vq,
+						zmbuf->desc_idx,
+						0,
+						zmbuf->desc_count);
+
+				TAILQ_REMOVE(&vq->zmbuf_list, zmbuf, next);
+				restore_mbuf(zmbuf->mbuf);
+				rte_pktmbuf_free(zmbuf->mbuf);
+				put_zmbuf(zmbuf);
+				vq->nr_zmbuf -= 1;
+			}
+		}
+
+		flush_shadow_used_ring_packed(dev, vq);
+		vhost_vring_call(dev, vq);
+	}
+
+	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+
+	count = RTE_MIN(count, MAX_PKT_BURST);
+	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) about to dequeue %u buffers\n",
+			dev->vid, count);
+
+	for (i = 0; i < count; i++) {
+		struct buf_vector buf_vec[BUF_VECTOR_MAX];
+		uint16_t buf_id, dummy_len;
+		uint16_t desc_count, nr_vec = 0;
+		int err;
+
+		if (unlikely(fill_vec_buf_packed(dev, vq,
+						vq->last_avail_idx, &desc_count,
+						buf_vec, &nr_vec,
+						&buf_id, &dummy_len,
+						VHOST_ACCESS_RW) < 0))
+			break;
+
+		if (likely(dev->dequeue_zero_copy == 0))
+			update_shadow_used_ring_packed(vq, buf_id, 0,
+					desc_count);
+
+		rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
+
+		pkts[i] = rte_pktmbuf_alloc(mbuf_pool);
+		if (unlikely(pkts[i] == NULL)) {
+			RTE_LOG(ERR, VHOST_DATA,
+				"Failed to allocate memory for mbuf.\n");
+			break;
+		}
+
+		err = copy_desc_to_mbuf(dev, vq, buf_vec, nr_vec, pkts[i],
+				mbuf_pool);
+		if (unlikely(err)) {
+			rte_pktmbuf_free(pkts[i]);
+			break;
+		}
+
+		if (unlikely(dev->dequeue_zero_copy)) {
+			struct zcopy_mbuf *zmbuf;
+
+			zmbuf = get_zmbuf(vq);
+			if (!zmbuf) {
+				rte_pktmbuf_free(pkts[i]);
+				break;
+			}
+			zmbuf->mbuf = pkts[i];
+			zmbuf->desc_idx = buf_id;
+			zmbuf->desc_count = desc_count;
+
+			/*
+			 * Pin lock the mbuf; we will check later to see
+			 * whether the mbuf is freed (when we are the last
+			 * user) or not. If that's the case, we then could
+			 * update the used ring safely.
+			 */
+			rte_mbuf_refcnt_update(pkts[i], 1);
+
+			vq->nr_zmbuf += 1;
+			TAILQ_INSERT_TAIL(&vq->zmbuf_list, zmbuf, next);
+		}
+
+		vq->last_avail_idx += desc_count;
+		if (vq->last_avail_idx >= vq->size) {
+			vq->last_avail_idx -= vq->size;
+			vq->avail_wrap_counter ^= 1;
+		}
+	}
+
+	if (likely(dev->dequeue_zero_copy == 0)) {
+		do_data_copy_dequeue(vq);
+		if (unlikely(i < count))
+			vq->shadow_used_idx = i;
+		flush_shadow_used_ring_packed(dev, vq);
+		vhost_vring_call(dev, vq);
+	}
+
+	return i;
+}
+
 uint16_t
 rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 	struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
@@ -1516,7 +1630,10 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 		count -= 1;
 	}
 
-	count = virtio_dev_tx_split(dev, vq, mbuf_pool, pkts, count);
+	if (vq_is_packed(dev))
+		count = virtio_dev_tx_packed(dev, vq, mbuf_pool, pkts, count);
+	else
+		count = virtio_dev_tx_split(dev, vq, mbuf_pool, pkts, count);
 
 out:
 	if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
-- 
2.14.4

[dpdk-dev] [PATCH v9 14/15] vhost: add notification for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.c      | 71 +++++++++++++++++++++++++++++++++++------
 lib/librte_vhost/vhost.h      | 73 +++++++++++++++++++++++++++++++++++++++++--
 lib/librte_vhost/vhost_user.c | 24 ++++++++++++++
 lib/librte_vhost/virtio_net.c | 12 +++----
 4 files changed, 162 insertions(+), 18 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 534f30cf6..3c9be10a0 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -163,13 +163,28 @@ vring_translate_packed(struct virtio_net *dev, struct vhost_virtqueue *vq)
 
 	req_size = sizeof(struct vring_packed_desc) * vq->size;
 	size = req_size;
-	vq->desc_packed =
-		(struct vring_packed_desc *)(uintptr_t)vhost_iova_to_vva(dev,
-					vq, vq->ring_addrs.desc_user_addr,
-					&size, VHOST_ACCESS_RW);
+	vq->desc_packed = (struct vring_packed_desc *)(uintptr_t)
+		vhost_iova_to_vva(dev, vq, vq->ring_addrs.desc_user_addr,
+				&size, VHOST_ACCESS_RW);
 	if (!vq->desc_packed || size != req_size)
 		return -1;
 
+	req_size = sizeof(struct vring_packed_desc_event);
+	size = req_size;
+	vq->driver_event = (struct vring_packed_desc_event *)(uintptr_t)
+		vhost_iova_to_vva(dev, vq, vq->ring_addrs.avail_user_addr,
+				&size, VHOST_ACCESS_RW);
+	if (!vq->driver_event || size != req_size)
+		return -1;
+
+	req_size = sizeof(struct vring_packed_desc_event);
+	size = req_size;
+	vq->device_event = (struct vring_packed_desc_event *)(uintptr_t)
+		vhost_iova_to_vva(dev, vq, vq->ring_addrs.used_user_addr,
+				&size, VHOST_ACCESS_RW);
+	if (!vq->device_event || size != req_size)
+		return -1;
+
 	return 0;
 }
 
@@ -270,6 +285,7 @@ alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx)
 	rte_spinlock_init(&vq->access_lock);
 	vq->avail_wrap_counter = 1;
 	vq->used_wrap_counter = 1;
+	vq->signalled_used_valid = false;
 
 	dev->nr_vring += 1;
 
@@ -604,7 +620,11 @@ rte_vhost_vring_call(int vid, uint16_t vring_idx)
 	if (!vq)
 		return -1;
 
-	vhost_vring_call(dev, vq);
+	if (vq_is_packed(dev))
+		vhost_vring_call_packed(dev, vq);
+	else
+		vhost_vring_call_split(dev, vq);
+
 	return 0;
 }
 
@@ -625,19 +645,52 @@ rte_vhost_avail_entries(int vid, uint16_t queue_id)
 	return *(volatile uint16_t *)&vq->avail->idx - vq->last_used_idx;
 }
 
+static inline void
+vhost_enable_notify_split(struct vhost_virtqueue *vq, int enable)
+{
+	if (enable)
+		vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
+	else
+		vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+}
+
+static inline void
+vhost_enable_notify_packed(struct virtio_net *dev,
+		struct vhost_virtqueue *vq, int enable)
+{
+	uint16_t flags;
+
+	if (!enable)
+		vq->device_event->flags = VRING_EVENT_F_DISABLE;
+
+	flags = VRING_EVENT_F_ENABLE;
+	if (dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX)) {
+		flags = VRING_EVENT_F_DESC;
+		vq->device_event->off_wrap = vq->last_avail_idx |
+			vq->avail_wrap_counter << 15;
+	}
+
+	rte_smp_wmb();
+
+	vq->device_event->flags = flags;
+}
+
 int
 rte_vhost_enable_guest_notification(int vid, uint16_t queue_id, int enable)
 {
 	struct virtio_net *dev = get_device(vid);
+	struct vhost_virtqueue *vq;
 
 	if (!dev)
 		return -1;
 
-	if (enable)
-		dev->virtqueue[queue_id]->used->flags &=
-			~VRING_USED_F_NO_NOTIFY;
+	vq = dev->virtqueue[queue_id];
+
+	if (vq_is_packed(dev))
+		vhost_enable_notify_packed(dev, vq, enable);
 	else
-		dev->virtqueue[queue_id]->used->flags |= VRING_USED_F_NO_NOTIFY;
+		vhost_enable_notify_split(vq, enable);
+
 	return 0;
 }
 
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index a7e602bec..760a09c0d 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -95,14 +95,21 @@ struct vhost_virtqueue {
 		struct vring_desc	*desc;
 		struct vring_packed_desc   *desc_packed;
 	};
-	struct vring_avail	*avail;
-	struct vring_used	*used;
+	union {
+		struct vring_avail	*avail;
+		struct vring_packed_desc_event *driver_event;
+	};
+	union {
+		struct vring_used	*used;
+		struct vring_packed_desc_event *device_event;
+	};
 	uint32_t		size;
 
 	uint16_t		last_avail_idx;
 	uint16_t		last_used_idx;
 	/* Last used index we notify to front end. */
 	uint16_t		signalled_used;
+	bool			signalled_used_valid;
 #define VIRTIO_INVALID_EVENTFD		(-1)
 #define VIRTIO_UNINITIALIZED_EVENTFD	(-2)
 
@@ -225,6 +232,15 @@ struct vring_packed_desc {
 	uint16_t id;
 	uint16_t flags;
 };
+
+#define VRING_EVENT_F_ENABLE 0x0
+#define VRING_EVENT_F_DISABLE 0x1
+#define VRING_EVENT_F_DESC 0x2
+
+struct vring_packed_desc_event {
+	uint16_t off_wrap;
+	uint16_t flags;
+};
 #endif
 
 /*
@@ -648,7 +664,7 @@ vhost_need_event(uint16_t event_idx, uint16_t new_idx, uint16_t old)
 }
 
 static __rte_always_inline void
-vhost_vring_call(struct virtio_net *dev, struct vhost_virtqueue *vq)
+vhost_vring_call_split(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
 	/* Flush used->idx update before we read avail->flags. */
 	rte_smp_mb();
@@ -675,4 +691,55 @@ vhost_vring_call(struct virtio_net *dev, struct vhost_virtqueue *vq)
 	}
 }
 
+static __rte_always_inline void
+vhost_vring_call_packed(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+	uint16_t old, new, off, off_wrap;
+	bool signalled_used_valid, kick = false;
+
+	/* Flush used desc update. */
+	rte_smp_mb();
+
+	if (!(dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX))) {
+		if (vq->driver_event->flags !=
+				VRING_EVENT_F_DISABLE)
+			kick = true;
+		goto kick;
+	}
+
+	old = vq->signalled_used;
+	new = vq->last_used_idx;
+	vq->signalled_used = new;
+	signalled_used_valid = vq->signalled_used_valid;
+	vq->signalled_used_valid = true;
+
+	if (vq->driver_event->flags != VRING_EVENT_F_DESC) {
+		if (vq->driver_event->flags != VRING_EVENT_F_DISABLE)
+			kick = true;
+		goto kick;
+	}
+
+	if (unlikely(!signalled_used_valid)) {
+		kick = true;
+		goto kick;
+	}
+
+	rte_smp_rmb();
+
+	off_wrap = vq->driver_event->off_wrap;
+	off = off_wrap & ~(1 << 15);
+
+	if (new <= old)
+		old -= vq->size;
+
+	if (vq->used_wrap_counter != off_wrap >> 15)
+		off -= vq->size;
+
+	if (vhost_need_event(off, new, old))
+		kick = true;
+kick:
+	if (kick)
+		eventfd_write(vq->callfd, (eventfd_t)1);
+}
+
 #endif /* _VHOST_NET_CDEV_H_ */
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 71d1fe0ac..90ab265d1 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -512,6 +512,30 @@ translate_ring_addresses(struct virtio_net *dev, int vq_index)
 		vq = dev->virtqueue[vq_index];
 		addr = &vq->ring_addrs;
 
+		len = sizeof(struct vring_packed_desc_event);
+		vq->driver_event = (struct vring_packed_desc_event *)
+					(uintptr_t)ring_addr_to_vva(dev,
+					vq, addr->avail_user_addr, &len);
+		if (vq->driver_event == NULL ||
+				len != sizeof(struct vring_packed_desc_event)) {
+			RTE_LOG(DEBUG, VHOST_CONFIG,
+				"(%d) failed to find driver area address.\n",
+				dev->vid);
+			return dev;
+		}
+
+		len = sizeof(struct vring_packed_desc_event);
+		vq->device_event = (struct vring_packed_desc_event *)
+					(uintptr_t)ring_addr_to_vva(dev,
+					vq, addr->used_user_addr, &len);
+		if (vq->device_event == NULL ||
+				len != sizeof(struct vring_packed_desc_event)) {
+			RTE_LOG(DEBUG, VHOST_CONFIG,
+				"(%d) failed to find device area address.\n",
+				dev->vid);
+			return dev;
+		}
+
 		return dev;
 	}
 
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 9f33821f4..fb9784d77 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -826,7 +826,7 @@ virtio_dev_rx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 
 	if (likely(vq->shadow_used_idx)) {
 		flush_shadow_used_ring_split(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_split(dev, vq);
 	}
 
 	return pkt_idx;
@@ -879,7 +879,7 @@ virtio_dev_rx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
 
 	if (likely(vq->shadow_used_idx)) {
 		flush_shadow_used_ring_packed(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_packed(dev, vq);
 	}
 
 	return pkt_idx;
@@ -1363,7 +1363,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		}
 
 		flush_shadow_used_ring_split(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_split(dev, vq);
 	}
 
 	rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
@@ -1442,7 +1442,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		if (unlikely(i < count))
 			vq->shadow_used_idx = i;
 		flush_shadow_used_ring_split(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_split(dev, vq);
 	}
 
 	return i;
@@ -1478,7 +1478,7 @@ virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		}
 
 		flush_shadow_used_ring_packed(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_packed(dev, vq);
 	}
 
 	VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
@@ -1556,7 +1556,7 @@ virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
 		if (unlikely(i < count))
 			vq->shadow_used_idx = i;
 		flush_shadow_used_ring_packed(dev, vq);
-		vhost_vring_call(dev, vq);
+		vhost_vring_call_packed(dev, vq);
 	}
 
 	return i;
-- 
2.14.4

[dpdk-dev] [PATCH v9 15/15] vhost: advertize packed ring layout support

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 lib/librte_vhost/vhost.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 760a09c0d..9b0ebb754 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -275,7 +275,8 @@ struct vring_packed_desc_event {
 				(1ULL << VIRTIO_RING_F_EVENT_IDX) | \
 				(1ULL << VIRTIO_NET_F_MTU)  | \
 				(1ULL << VIRTIO_F_IN_ORDER) | \
-				(1ULL << VIRTIO_F_IOMMU_PLATFORM))
+				(1ULL << VIRTIO_F_IOMMU_PLATFORM) | \
+				(1ULL << VIRTIO_F_RING_PACKED))
 
 
 struct guest_page {
-- 
2.14.4

Re: [dpdk-dev] [PATCH v9 00/15] Vhost: add support to packed ring layout

[Tiwei Bie]

On Fri, Jul 06, 2018 at 09:07:07AM +0200, Maxime Coquelin wrote:
> This series is a handover from Jen's "[PATCH v4 00/20]
> implement packed virtqueues", which only implements the
> vhost side. Virtio PMD implementation will follow in a 
> next series.
> 
> The series applies on top of previous reworks I posted
> during this cycle that merges mergeable and non-mergeable
> receive paths, and refactors transmit path to re-use
> vector buffers.
> 
> I haven't run performance tests for now as the Virtio PMD
> side isn't ready.
> 
> The series has been tested with Tiwei's series implementing
> packed ring support to Kernel's virtio-net driver, and
> with Wei series implementing the Qemu side.
> 
> To test it, I have used testpmd on host side with a vhost
> vdev and a tap vdev forwarding to each other. Transferts
> of big random files have been done in both ways with
> integrity verified.
> 
> Tests have been run with Rx mrg ON/OFF and events suppression
> ON/OFF.
> 
> Tests have also been run with legacy split ring layout to
> ensure no regression have been introduced.
> 
> Changes since v8:
> =================
> - Fix indents (Tiwei)
> - Rename struct vring_desc_packed to vring_packed_desc (Tiwei)
> 
> Changes since v7:
> =================
> - Align structs and defines naming with Kernel header (Tiwei)
> - Fix event based notifications (Tiwei)
> - Fix Clang build issues caused by unused symbols (Tiwei)
> 
> Changes since v6:
> =================
> - Various style cleanups (Tiwei, Jason)
> - Simplify event based notification (Jason)
> - Build support with future kernels (Tiwei)
> - Prevent buffer vectors overflow in map_one_desc (Tiwei)
> 
> Changes since v5:
> =================
> - Remove duplicated VHOST_USER_F_PROTOCOL_FEATURES definition (Tiwei)
> - Fix vq_is_ready (Maxime)
> - Fix local used index overflow in flush_shadow_used_ring_packed (Tiwei)
> - Make desc_is_avail() a bool (Tiwei)
> - Improve desc_is_avail() logic (Tiwei)
> - Remove split rings addr NULL assignment in the right patch (Tiwei)
> - Make buffer id a uint16_t (Tiwei)
> 
> Jens Freimann (2):
>   vhost: add virtio packed virtqueue defines
>   vhost: add helpers for packed virtqueues
> 
> Maxime Coquelin (12):
>   vhost: clear shadow used table index at flush time
>   vhost: make indirect desc table copy desc type agnostic
>   vhost: clear batch copy index at copy time
>   vhost: extract split ring handling from Rx and Tx functions
>   vhost: append shadow used ring function names with split
>   vhost: add shadow used ring support for packed rings
>   vhost: create descriptor mapping function
>   vhost: add vector filling support for packed ring
>   vhost: add Rx support for packed ring
>   vhost: add Tx support for packed ring
>   vhost: add notification for packed ring
>   vhost: advertize packed ring layout support
> 
> Yuanhan Liu (1):
>   vhost: vring address setup for packed queues
> 
>  lib/librte_vhost/vhost.c      | 115 ++++++-
>  lib/librte_vhost/vhost.h      | 130 ++++++-
>  lib/librte_vhost/vhost_user.c | 127 +++++--
>  lib/librte_vhost/virtio_net.c | 776 +++++++++++++++++++++++++++++++++---------
>  4 files changed, 939 insertions(+), 209 deletions(-)
> 
> -- 
> 2.14.4
> 

For the series:
Acked-by: Tiwei Bie <tiwei.bie@intel.com>

Re: [dpdk-dev] [PATCH v9 03/15] vhost: vring address setup for packed queues

[Tiwei Bie]

On Fri, Jul 06, 2018 at 09:07:10AM +0200, Maxime Coquelin wrote:
[...]
> diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
> index bea6a0428..dca43ff00 100644
> --- a/lib/librte_vhost/vhost_user.c
> +++ b/lib/librte_vhost/vhost_user.c
> @@ -467,6 +467,27 @@ translate_ring_addresses(struct virtio_net *dev, int vq_index)
>  	struct vhost_vring_addr *addr = &vq->ring_addrs;
>  	uint64_t len;
>  
> +	if (vq_is_packed(dev)) {
> +		len = sizeof(struct vring_packed_desc) * vq->size;
> +		vq->desc_packed = (struct vring_packed_desc *) ring_addr_to_vva

FYI, above code will cause below build error on 32bit platform:

/tmp/dpdk-build/lib/librte_vhost/vhost_user.c: In function ‘translate_ring_addresses’:
/tmp/dpdk-build/lib/librte_vhost/vhost_user.c:472:21: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
   vq->desc_packed = (struct vring_packed_desc *) ring_addr_to_vva
                     ^

I'll fix it while applying.

Re: [dpdk-dev] [PATCH v9 00/15] Vhost: add support to packed ring layout

[Tiwei Bie]

On Fri, Jul 06, 2018 at 09:07:07AM +0200, Maxime Coquelin wrote:
> This series is a handover from Jen's "[PATCH v4 00/20]
> implement packed virtqueues", which only implements the
> vhost side. Virtio PMD implementation will follow in a 
> next series.
> 
> The series applies on top of previous reworks I posted
> during this cycle that merges mergeable and non-mergeable
> receive paths, and refactors transmit path to re-use
> vector buffers.
> 
> I haven't run performance tests for now as the Virtio PMD
> side isn't ready.
> 
> The series has been tested with Tiwei's series implementing
> packed ring support to Kernel's virtio-net driver, and
> with Wei series implementing the Qemu side.
> 
> To test it, I have used testpmd on host side with a vhost
> vdev and a tap vdev forwarding to each other. Transferts
> of big random files have been done in both ways with
> integrity verified.
> 
> Tests have been run with Rx mrg ON/OFF and events suppression
> ON/OFF.
> 
> Tests have also been run with legacy split ring layout to
> ensure no regression have been introduced.
> 
> Changes since v8:
> =================
> - Fix indents (Tiwei)
> - Rename struct vring_desc_packed to vring_packed_desc (Tiwei)
> 
> Changes since v7:
> =================
> - Align structs and defines naming with Kernel header (Tiwei)
> - Fix event based notifications (Tiwei)
> - Fix Clang build issues caused by unused symbols (Tiwei)
> 
> Changes since v6:
> =================
> - Various style cleanups (Tiwei, Jason)
> - Simplify event based notification (Jason)
> - Build support with future kernels (Tiwei)
> - Prevent buffer vectors overflow in map_one_desc (Tiwei)
> 
> Changes since v5:
> =================
> - Remove duplicated VHOST_USER_F_PROTOCOL_FEATURES definition (Tiwei)
> - Fix vq_is_ready (Maxime)
> - Fix local used index overflow in flush_shadow_used_ring_packed (Tiwei)
> - Make desc_is_avail() a bool (Tiwei)
> - Improve desc_is_avail() logic (Tiwei)
> - Remove split rings addr NULL assignment in the right patch (Tiwei)
> - Make buffer id a uint16_t (Tiwei)
> 
> Jens Freimann (2):
>   vhost: add virtio packed virtqueue defines
>   vhost: add helpers for packed virtqueues
> 
> Maxime Coquelin (12):
>   vhost: clear shadow used table index at flush time
>   vhost: make indirect desc table copy desc type agnostic
>   vhost: clear batch copy index at copy time
>   vhost: extract split ring handling from Rx and Tx functions
>   vhost: append shadow used ring function names with split
>   vhost: add shadow used ring support for packed rings
>   vhost: create descriptor mapping function
>   vhost: add vector filling support for packed ring
>   vhost: add Rx support for packed ring
>   vhost: add Tx support for packed ring
>   vhost: add notification for packed ring
>   vhost: advertize packed ring layout support
> 
> Yuanhan Liu (1):
>   vhost: vring address setup for packed queues
> 
>  lib/librte_vhost/vhost.c      | 115 ++++++-
>  lib/librte_vhost/vhost.h      | 130 ++++++-
>  lib/librte_vhost/vhost_user.c | 127 +++++--
>  lib/librte_vhost/virtio_net.c | 776 +++++++++++++++++++++++++++++++++---------
>  4 files changed, 939 insertions(+), 209 deletions(-)
> 
> -- 
> 2.14.4
> 

Applied to dpdk-next-virtio/master, thanks.

Re: [dpdk-dev] [PATCH v9 15/15] vhost: advertize packed ring layout support

[Tiwei Bie]

On Fri, Jul 06, 2018 at 09:07:22AM +0200, Maxime Coquelin wrote:
> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
> ---
>  lib/librte_vhost/vhost.h | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
> index 760a09c0d..9b0ebb754 100644
> --- a/lib/librte_vhost/vhost.h
> +++ b/lib/librte_vhost/vhost.h
> @@ -275,7 +275,8 @@ struct vring_packed_desc_event {
>  				(1ULL << VIRTIO_RING_F_EVENT_IDX) | \
>  				(1ULL << VIRTIO_NET_F_MTU)  | \
>  				(1ULL << VIRTIO_F_IN_ORDER) | \
> -				(1ULL << VIRTIO_F_IOMMU_PLATFORM))
> +				(1ULL << VIRTIO_F_IOMMU_PLATFORM) | \
> +				(1ULL << VIRTIO_F_RING_PACKED))
>  
>  
>  struct guest_page {
> -- 
> 2.14.4
> 

I didn't apply this patch, because the packed ring
support isn't complete, e.g. when doing live migration,
the wrap counter isn't synced. We can advertise this
feature bit when it's fully supported.