#include <rte_udp.h>
#include <rte_sctp.h>
#include <rte_arp.h>
+#include <rte_spinlock.h>
+#include <rte_malloc.h>
#include "iotlb.h"
#include "vhost.h"
return (is_tx ^ (idx & 1)) == 0 && idx < nr_vring;
}
+static __rte_always_inline struct vring_desc *
+alloc_copy_ind_table(struct virtio_net *dev, struct vhost_virtqueue *vq,
+ struct vring_desc *desc)
+{
+ struct vring_desc *idesc;
+ uint64_t src, dst;
+ uint64_t len, remain = desc->len;
+ uint64_t desc_addr = desc->addr;
+
+ idesc = rte_malloc(__func__, desc->len, 0);
+ if (unlikely(!idesc))
+ return 0;
+
+ dst = (uint64_t)(uintptr_t)idesc;
+
+ while (remain) {
+ len = remain;
+ src = vhost_iova_to_vva(dev, vq, desc_addr, &len,
+ VHOST_ACCESS_RO);
+ if (unlikely(!src || !len)) {
+ rte_free(idesc);
+ return 0;
+ }
+
+ rte_memcpy((void *)(uintptr_t)dst, (void *)(uintptr_t)src, len);
+
+ remain -= len;
+ dst += len;
+ desc_addr += len;
+ }
+
+ return idesc;
+}
+
+static __rte_always_inline void
+free_ind_table(struct vring_desc *idesc)
+{
+ rte_free(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)
net_hdr->gso_size = m_buf->tso_segsz;
net_hdr->hdr_len = m_buf->l2_len + m_buf->l3_len
+ m_buf->l4_len;
+ } else if (m_buf->ol_flags & PKT_TX_UDP_SEG) {
+ net_hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP;
+ net_hdr->gso_size = m_buf->tso_segsz;
+ net_hdr->hdr_len = m_buf->l2_len + m_buf->l3_len +
+ m_buf->l4_len;
} else {
ASSIGN_UNLESS_EQUAL(net_hdr->gso_type, 0);
ASSIGN_UNLESS_EQUAL(net_hdr->gso_size, 0);
uint32_t desc_avail, desc_offset;
uint32_t mbuf_avail, mbuf_offset;
uint32_t cpy_len;
+ uint64_t desc_chunck_len;
struct vring_desc *desc;
- uint64_t desc_addr;
+ uint64_t desc_addr, desc_gaddr;
/* A counter to avoid desc dead loop chain */
uint16_t nr_desc = 1;
struct batch_copy_elem *batch_copy = vq->batch_copy_elems;
int error = 0;
desc = &descs[desc_idx];
- desc_addr = vhost_iova_to_vva(dev, vq, desc->addr,
- desc->len, VHOST_ACCESS_RW);
+ desc_chunck_len = desc->len;
+ desc_gaddr = desc->addr;
+ desc_addr = vhost_iova_to_vva(dev, vq, desc_gaddr,
+ &desc_chunck_len, VHOST_ACCESS_RW);
/*
* Checking of 'desc_addr' placed outside of 'unlikely' macro to avoid
* performance issue with some versions of gcc (4.8.4 and 5.3.0) which
rte_prefetch0((void *)(uintptr_t)desc_addr);
- virtio_enqueue_offload(m, (struct virtio_net_hdr *)(uintptr_t)desc_addr);
- vhost_log_write(dev, desc->addr, dev->vhost_hlen);
- PRINT_PACKET(dev, (uintptr_t)desc_addr, dev->vhost_hlen, 0);
+ if (likely(desc_chunck_len >= dev->vhost_hlen)) {
+ virtio_enqueue_offload(m,
+ (struct virtio_net_hdr *)(uintptr_t)desc_addr);
+ PRINT_PACKET(dev, (uintptr_t)desc_addr, dev->vhost_hlen, 0);
+ vhost_log_write(dev, desc_gaddr, dev->vhost_hlen);
+ } else {
+ struct virtio_net_hdr vnet_hdr;
+ uint64_t remain = dev->vhost_hlen;
+ uint64_t len;
+ uint64_t src = (uint64_t)(uintptr_t)&vnet_hdr, dst;
+ uint64_t guest_addr = desc_gaddr;
+
+ virtio_enqueue_offload(m, &vnet_hdr);
+
+ while (remain) {
+ len = remain;
+ dst = vhost_iova_to_vva(dev, vq, guest_addr,
+ &len, VHOST_ACCESS_RW);
+ if (unlikely(!dst || !len)) {
+ error = -1;
+ goto out;
+ }
+
+ rte_memcpy((void *)(uintptr_t)dst,
+ (void *)(uintptr_t)src, len);
+
+ PRINT_PACKET(dev, (uintptr_t)dst, (uint32_t)len, 0);
+ vhost_log_write(dev, guest_addr, len);
+ remain -= len;
+ guest_addr += len;
+ dst += len;
+ }
+ }
- desc_offset = dev->vhost_hlen;
desc_avail = desc->len - dev->vhost_hlen;
+ if (unlikely(desc_chunck_len < dev->vhost_hlen)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr = desc->addr + dev->vhost_hlen;
+ desc_addr = vhost_iova_to_vva(dev,
+ vq, desc_gaddr,
+ &desc_chunck_len,
+ VHOST_ACCESS_RW);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+
+ desc_offset = 0;
+ } else {
+ desc_offset = dev->vhost_hlen;
+ desc_chunck_len -= dev->vhost_hlen;
+ }
mbuf_avail = rte_pktmbuf_data_len(m);
mbuf_offset = 0;
}
desc = &descs[desc->next];
- desc_addr = vhost_iova_to_vva(dev, vq, desc->addr,
- desc->len,
+ desc_chunck_len = desc->len;
+ desc_gaddr = desc->addr;
+ desc_addr = vhost_iova_to_vva(dev, vq, desc_gaddr,
+ &desc_chunck_len,
VHOST_ACCESS_RW);
if (unlikely(!desc_addr)) {
error = -1;
desc_offset = 0;
desc_avail = desc->len;
+ } else if (unlikely(desc_chunck_len == 0)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr += desc_offset;
+ desc_addr = vhost_iova_to_vva(dev,
+ vq, desc_gaddr,
+ &desc_chunck_len, VHOST_ACCESS_RW);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+ desc_offset = 0;
}
- cpy_len = RTE_MIN(desc_avail, mbuf_avail);
+ cpy_len = RTE_MIN(desc_chunck_len, mbuf_avail);
if (likely(cpy_len > MAX_BATCH_LEN || copy_nb >= vq->size)) {
rte_memcpy((void *)((uintptr_t)(desc_addr +
desc_offset)),
rte_pktmbuf_mtod_offset(m, void *, mbuf_offset),
cpy_len);
- vhost_log_write(dev, desc->addr + desc_offset, cpy_len);
+ vhost_log_write(dev, desc_gaddr + desc_offset, cpy_len);
PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset),
cpy_len, 0);
} else {
(void *)((uintptr_t)(desc_addr + desc_offset));
batch_copy[copy_nb].src =
rte_pktmbuf_mtod_offset(m, void *, mbuf_offset);
- batch_copy[copy_nb].log_addr = desc->addr + desc_offset;
+ batch_copy[copy_nb].log_addr = desc_gaddr + desc_offset;
batch_copy[copy_nb].len = cpy_len;
copy_nb++;
}
mbuf_offset += cpy_len;
desc_avail -= cpy_len;
desc_offset += cpy_len;
+ desc_chunck_len -= cpy_len;
}
out:
uint16_t used_idx;
uint32_t i, sz;
- LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+ 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);
}
vq = dev->virtqueue[queue_id];
+
+ rte_spinlock_lock(&vq->access_lock);
+
if (unlikely(vq->enabled == 0))
- return 0;
+ goto out_access_unlock;
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_lock(vq);
if (count == 0)
goto out;
- LOG_DEBUG(VHOST_DATA, "(%d) start_idx %d | end_idx %d\n",
+ VHOST_LOG_DEBUG(VHOST_DATA, "(%d) start_idx %d | end_idx %d\n",
dev->vid, start_idx, start_idx + count);
vq->batch_copy_nb_elems = 0;
rte_prefetch0(&vq->desc[desc_indexes[0]]);
for (i = 0; i < count; i++) {
+ struct vring_desc *idesc = NULL;
uint16_t desc_idx = desc_indexes[i];
int err;
if (vq->desc[desc_idx].flags & VRING_DESC_F_INDIRECT) {
+ uint64_t dlen = vq->desc[desc_idx].len;
descs = (struct vring_desc *)(uintptr_t)
vhost_iova_to_vva(dev,
vq, vq->desc[desc_idx].addr,
- vq->desc[desc_idx].len,
- VHOST_ACCESS_RO);
+ &dlen, VHOST_ACCESS_RO);
if (unlikely(!descs)) {
count = i;
break;
}
+ if (unlikely(dlen < vq->desc[desc_idx].len)) {
+ /*
+ * 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[desc_idx]);
+ if (unlikely(!idesc))
+ break;
+
+ descs = idesc;
+ }
+
desc_idx = 0;
sz = vq->desc[desc_idx].len / sizeof(*descs);
} else {
err = copy_mbuf_to_desc(dev, vq, descs, pkts[i], desc_idx, sz);
if (unlikely(err)) {
count = i;
+ free_ind_table(idesc);
break;
}
if (i + 1 < count)
rte_prefetch0(&vq->desc[desc_indexes[i+1]]);
+
+ if (unlikely(!!idesc))
+ free_ind_table(idesc);
}
do_data_copy_enqueue(dev, vq);
offsetof(struct vring_used, idx),
sizeof(vq->used->idx));
- vhost_vring_call(vq);
+ vhost_vring_call(dev, vq);
out:
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_unlock(vq);
+out_access_unlock:
+ rte_spinlock_unlock(&vq->access_lock);
+
return count;
}
uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
uint32_t vec_id = *vec_idx;
uint32_t len = 0;
+ uint64_t dlen;
struct vring_desc *descs = vq->desc;
+ struct vring_desc *idesc = NULL;
*desc_chain_head = idx;
if (vq->desc[idx].flags & VRING_DESC_F_INDIRECT) {
+ dlen = vq->desc[idx].len;
descs = (struct vring_desc *)(uintptr_t)
vhost_iova_to_vva(dev, vq, vq->desc[idx].addr,
- vq->desc[idx].len,
+ &dlen,
VHOST_ACCESS_RO);
if (unlikely(!descs))
return -1;
+ if (unlikely(dlen < vq->desc[idx].len)) {
+ /*
+ * 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]);
+ if (unlikely(!idesc))
+ return -1;
+
+ descs = idesc;
+ }
+
idx = 0;
}
while (1) {
- if (unlikely(vec_id >= BUF_VECTOR_MAX || idx >= vq->size))
+ if (unlikely(vec_id >= BUF_VECTOR_MAX || idx >= vq->size)) {
+ free_ind_table(idesc);
return -1;
+ }
len += descs[idx].len;
buf_vec[vec_id].buf_addr = descs[idx].addr;
*desc_chain_len = len;
*vec_idx = vec_id;
+ if (unlikely(!!idesc))
+ free_ind_table(idesc);
+
return 0;
}
uint16_t num_buffers)
{
uint32_t vec_idx = 0;
- uint64_t desc_addr;
+ uint64_t desc_addr, desc_gaddr;
uint32_t mbuf_offset, mbuf_avail;
uint32_t desc_offset, desc_avail;
uint32_t cpy_len;
+ uint64_t desc_chunck_len;
uint64_t hdr_addr, hdr_phys_addr;
struct rte_mbuf *hdr_mbuf;
struct batch_copy_elem *batch_copy = vq->batch_copy_elems;
+ struct virtio_net_hdr_mrg_rxbuf tmp_hdr, *hdr = NULL;
uint16_t copy_nb = vq->batch_copy_nb_elems;
int error = 0;
goto out;
}
- desc_addr = vhost_iova_to_vva(dev, vq, buf_vec[vec_idx].buf_addr,
- buf_vec[vec_idx].buf_len,
- VHOST_ACCESS_RW);
+ desc_chunck_len = buf_vec[vec_idx].buf_len;
+ desc_gaddr = buf_vec[vec_idx].buf_addr;
+ desc_addr = vhost_iova_to_vva(dev, vq,
+ desc_gaddr,
+ &desc_chunck_len,
+ VHOST_ACCESS_RW);
if (buf_vec[vec_idx].buf_len < dev->vhost_hlen || !desc_addr) {
error = -1;
goto out;
hdr_mbuf = m;
hdr_addr = desc_addr;
- hdr_phys_addr = buf_vec[vec_idx].buf_addr;
+ if (unlikely(desc_chunck_len < dev->vhost_hlen))
+ hdr = &tmp_hdr;
+ else
+ hdr = (struct virtio_net_hdr_mrg_rxbuf *)(uintptr_t)hdr_addr;
+ hdr_phys_addr = desc_gaddr;
rte_prefetch0((void *)(uintptr_t)hdr_addr);
- LOG_DEBUG(VHOST_DATA, "(%d) RX: num merge buffers %d\n",
+ VHOST_LOG_DEBUG(VHOST_DATA, "(%d) RX: num merge buffers %d\n",
dev->vid, num_buffers);
desc_avail = buf_vec[vec_idx].buf_len - dev->vhost_hlen;
- desc_offset = dev->vhost_hlen;
+ if (unlikely(desc_chunck_len < dev->vhost_hlen)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr += dev->vhost_hlen;
+ desc_addr = vhost_iova_to_vva(dev, vq,
+ desc_gaddr,
+ &desc_chunck_len,
+ VHOST_ACCESS_RW);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+
+ desc_offset = 0;
+ } else {
+ desc_offset = dev->vhost_hlen;
+ desc_chunck_len -= dev->vhost_hlen;
+ }
+
mbuf_avail = rte_pktmbuf_data_len(m);
mbuf_offset = 0;
/* done with current desc buf, get the next one */
if (desc_avail == 0) {
vec_idx++;
+ desc_chunck_len = buf_vec[vec_idx].buf_len;
+ desc_gaddr = buf_vec[vec_idx].buf_addr;
desc_addr =
vhost_iova_to_vva(dev, vq,
- buf_vec[vec_idx].buf_addr,
- buf_vec[vec_idx].buf_len,
+ desc_gaddr,
+ &desc_chunck_len,
VHOST_ACCESS_RW);
if (unlikely(!desc_addr)) {
error = -1;
rte_prefetch0((void *)(uintptr_t)desc_addr);
desc_offset = 0;
desc_avail = buf_vec[vec_idx].buf_len;
+ } else if (unlikely(desc_chunck_len == 0)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr += desc_offset;
+ desc_addr = vhost_iova_to_vva(dev, vq,
+ desc_gaddr,
+ &desc_chunck_len, VHOST_ACCESS_RW);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+ desc_offset = 0;
}
/* done with current mbuf, get the next one */
}
if (hdr_addr) {
- struct virtio_net_hdr_mrg_rxbuf *hdr;
-
- hdr = (struct virtio_net_hdr_mrg_rxbuf *)(uintptr_t)
- hdr_addr;
virtio_enqueue_offload(hdr_mbuf, &hdr->hdr);
ASSIGN_UNLESS_EQUAL(hdr->num_buffers, num_buffers);
- vhost_log_write(dev, hdr_phys_addr, dev->vhost_hlen);
- PRINT_PACKET(dev, (uintptr_t)hdr_addr,
- dev->vhost_hlen, 0);
+ if (unlikely(hdr == &tmp_hdr)) {
+ uint64_t len;
+ uint64_t remain = dev->vhost_hlen;
+ uint64_t src = (uint64_t)(uintptr_t)hdr, dst;
+ uint64_t guest_addr = hdr_phys_addr;
+
+ while (remain) {
+ len = remain;
+ dst = vhost_iova_to_vva(dev, vq,
+ guest_addr, &len,
+ VHOST_ACCESS_RW);
+ if (unlikely(!dst || !len)) {
+ error = -1;
+ goto out;
+ }
+
+ rte_memcpy((void *)(uintptr_t)dst,
+ (void *)(uintptr_t)src,
+ len);
+
+ PRINT_PACKET(dev, (uintptr_t)dst,
+ (uint32_t)len, 0);
+ vhost_log_write(dev, guest_addr, len);
+
+ remain -= len;
+ guest_addr += len;
+ dst += len;
+ }
+ } else {
+ PRINT_PACKET(dev, (uintptr_t)hdr_addr,
+ dev->vhost_hlen, 0);
+ vhost_log_write(dev, hdr_phys_addr,
+ dev->vhost_hlen);
+ }
hdr_addr = 0;
}
- cpy_len = RTE_MIN(desc_avail, mbuf_avail);
+ cpy_len = RTE_MIN(desc_chunck_len, mbuf_avail);
if (likely(cpy_len > MAX_BATCH_LEN || copy_nb >= vq->size)) {
rte_memcpy((void *)((uintptr_t)(desc_addr +
desc_offset)),
rte_pktmbuf_mtod_offset(m, void *, mbuf_offset),
cpy_len);
- vhost_log_write(dev,
- buf_vec[vec_idx].buf_addr + desc_offset,
- cpy_len);
+ vhost_log_write(dev, desc_gaddr + desc_offset, cpy_len);
PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset),
cpy_len, 0);
} else {
(void *)((uintptr_t)(desc_addr + desc_offset));
batch_copy[copy_nb].src =
rte_pktmbuf_mtod_offset(m, void *, mbuf_offset);
- batch_copy[copy_nb].log_addr =
- buf_vec[vec_idx].buf_addr + desc_offset;
+ batch_copy[copy_nb].log_addr = desc_gaddr + desc_offset;
batch_copy[copy_nb].len = cpy_len;
copy_nb++;
}
mbuf_offset += cpy_len;
desc_avail -= cpy_len;
desc_offset += cpy_len;
+ desc_chunck_len -= cpy_len;
}
out:
struct buf_vector buf_vec[BUF_VECTOR_MAX];
uint16_t avail_head;
- LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+ 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);
}
vq = dev->virtqueue[queue_id];
+
+ rte_spinlock_lock(&vq->access_lock);
+
if (unlikely(vq->enabled == 0))
- return 0;
+ goto out_access_unlock;
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_lock(vq);
if (unlikely(reserve_avail_buf_mergeable(dev, vq,
pkt_len, buf_vec, &num_buffers,
avail_head) < 0)) {
- LOG_DEBUG(VHOST_DATA,
+ VHOST_LOG_DEBUG(VHOST_DATA,
"(%d) failed to get enough desc from vring\n",
dev->vid);
vq->shadow_used_idx -= num_buffers;
break;
}
- LOG_DEBUG(VHOST_DATA, "(%d) current index %d | end index %d\n",
+ 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 (likely(vq->shadow_used_idx)) {
flush_shadow_used_ring(dev, vq);
- vhost_vring_call(vq);
+ vhost_vring_call(dev, vq);
}
out:
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_unlock(vq);
+out_access_unlock:
+ rte_spinlock_unlock(&vq->access_lock);
+
return pkt_idx;
}
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 (dev->features & (1 << VIRTIO_NET_F_MRG_RXBUF))
return virtio_dev_merge_rx(dev, queue_id, pkts, count);
else
m->tso_segsz = hdr->gso_size;
m->l4_len = (tcp_hdr->data_off & 0xf0) >> 2;
break;
+ case VIRTIO_NET_HDR_GSO_UDP:
+ m->ol_flags |= PKT_TX_UDP_SEG;
+ m->tso_segsz = hdr->gso_size;
+ m->l4_len = sizeof(struct udp_hdr);
+ break;
default:
RTE_LOG(WARNING, VHOST_DATA,
"unsupported gso type %u.\n", hdr->gso_type);
}
}
-#define RARP_PKT_SIZE 64
-
-static int
-make_rarp_packet(struct rte_mbuf *rarp_mbuf, const struct ether_addr *mac)
-{
- struct ether_hdr *eth_hdr;
- struct arp_hdr *rarp;
-
- if (rarp_mbuf->buf_len < 64) {
- RTE_LOG(WARNING, VHOST_DATA,
- "failed to make RARP; mbuf size too small %u (< %d)\n",
- rarp_mbuf->buf_len, RARP_PKT_SIZE);
- return -1;
- }
-
- /* Ethernet header. */
- eth_hdr = rte_pktmbuf_mtod_offset(rarp_mbuf, struct ether_hdr *, 0);
- memset(eth_hdr->d_addr.addr_bytes, 0xff, ETHER_ADDR_LEN);
- ether_addr_copy(mac, ð_hdr->s_addr);
- eth_hdr->ether_type = htons(ETHER_TYPE_RARP);
-
- /* RARP header. */
- rarp = (struct arp_hdr *)(eth_hdr + 1);
- rarp->arp_hrd = htons(ARP_HRD_ETHER);
- rarp->arp_pro = htons(ETHER_TYPE_IPv4);
- rarp->arp_hln = ETHER_ADDR_LEN;
- rarp->arp_pln = 4;
- rarp->arp_op = htons(ARP_OP_REVREQUEST);
-
- ether_addr_copy(mac, &rarp->arp_data.arp_sha);
- ether_addr_copy(mac, &rarp->arp_data.arp_tha);
- memset(&rarp->arp_data.arp_sip, 0x00, 4);
- memset(&rarp->arp_data.arp_tip, 0x00, 4);
-
- rarp_mbuf->pkt_len = rarp_mbuf->data_len = RARP_PKT_SIZE;
-
- return 0;
-}
-
static __rte_always_inline void
put_zmbuf(struct zcopy_mbuf *zmbuf)
{
struct rte_mempool *mbuf_pool)
{
struct vring_desc *desc;
- uint64_t desc_addr;
+ uint64_t desc_addr, desc_gaddr;
uint32_t desc_avail, desc_offset;
uint32_t mbuf_avail, mbuf_offset;
uint32_t cpy_len;
+ uint64_t desc_chunck_len;
struct rte_mbuf *cur = m, *prev = m;
+ struct virtio_net_hdr tmp_hdr;
struct virtio_net_hdr *hdr = NULL;
/* A counter to avoid desc dead loop chain */
uint32_t nr_desc = 1;
goto out;
}
+ desc_chunck_len = desc->len;
+ desc_gaddr = desc->addr;
desc_addr = vhost_iova_to_vva(dev,
- vq, desc->addr,
- desc->len,
+ vq, desc_gaddr,
+ &desc_chunck_len,
VHOST_ACCESS_RO);
if (unlikely(!desc_addr)) {
error = -1;
}
if (virtio_net_with_host_offload(dev)) {
- hdr = (struct virtio_net_hdr *)((uintptr_t)desc_addr);
- rte_prefetch0(hdr);
+ if (unlikely(desc_chunck_len < sizeof(struct virtio_net_hdr))) {
+ uint64_t len = desc_chunck_len;
+ uint64_t remain = sizeof(struct virtio_net_hdr);
+ uint64_t src = desc_addr;
+ uint64_t dst = (uint64_t)(uintptr_t)&tmp_hdr;
+ uint64_t guest_addr = desc_gaddr;
+
+ /*
+ * No luck, the virtio-net header doesn't fit
+ * in a contiguous virtual area.
+ */
+ while (remain) {
+ len = remain;
+ src = vhost_iova_to_vva(dev, vq,
+ guest_addr, &len,
+ VHOST_ACCESS_RO);
+ if (unlikely(!src || !len)) {
+ error = -1;
+ goto out;
+ }
+
+ rte_memcpy((void *)(uintptr_t)dst,
+ (void *)(uintptr_t)src, len);
+
+ guest_addr += len;
+ remain -= len;
+ dst += len;
+ }
+
+ hdr = &tmp_hdr;
+ } else {
+ hdr = (struct virtio_net_hdr *)((uintptr_t)desc_addr);
+ rte_prefetch0(hdr);
+ }
}
/*
goto out;
}
+ desc_chunck_len = desc->len;
+ desc_gaddr = desc->addr;
desc_addr = vhost_iova_to_vva(dev,
- vq, desc->addr,
- desc->len,
+ vq, desc_gaddr,
+ &desc_chunck_len,
VHOST_ACCESS_RO);
if (unlikely(!desc_addr)) {
error = -1;
nr_desc += 1;
} else {
desc_avail = desc->len - dev->vhost_hlen;
- desc_offset = dev->vhost_hlen;
+
+ if (unlikely(desc_chunck_len < dev->vhost_hlen)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr += dev->vhost_hlen;
+ desc_addr = vhost_iova_to_vva(dev,
+ vq, desc_gaddr,
+ &desc_chunck_len,
+ VHOST_ACCESS_RO);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+
+ desc_offset = 0;
+ } else {
+ desc_offset = dev->vhost_hlen;
+ desc_chunck_len -= dev->vhost_hlen;
+ }
}
rte_prefetch0((void *)(uintptr_t)(desc_addr + desc_offset));
- PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset), desc_avail, 0);
+ PRINT_PACKET(dev, (uintptr_t)(desc_addr + desc_offset),
+ (uint32_t)desc_chunck_len, 0);
mbuf_offset = 0;
mbuf_avail = m->buf_len - RTE_PKTMBUF_HEADROOM;
while (1) {
uint64_t hpa;
- cpy_len = RTE_MIN(desc_avail, mbuf_avail);
+ cpy_len = RTE_MIN(desc_chunck_len, mbuf_avail);
/*
* A desc buf might across two host physical pages that are
* will be copied even though zero copy is enabled.
*/
if (unlikely(dev->dequeue_zero_copy && (hpa = gpa_to_hpa(dev,
- desc->addr + desc_offset, cpy_len)))) {
+ desc_gaddr + desc_offset, cpy_len)))) {
cur->data_len = cpy_len;
cur->data_off = 0;
cur->buf_addr = (void *)(uintptr_t)(desc_addr
} else {
if (likely(cpy_len > MAX_BATCH_LEN ||
copy_nb >= vq->size ||
- (hdr && cur == m))) {
+ (hdr && cur == m) ||
+ desc->len != desc_chunck_len)) {
rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
mbuf_offset),
(void *)((uintptr_t)(desc_addr +
mbuf_avail -= cpy_len;
mbuf_offset += cpy_len;
desc_avail -= cpy_len;
+ desc_chunck_len -= cpy_len;
desc_offset += cpy_len;
/* This desc reaches to its end, get the next one */
goto out;
}
+ desc_chunck_len = desc->len;
+ desc_gaddr = desc->addr;
desc_addr = vhost_iova_to_vva(dev,
- vq, desc->addr,
- desc->len,
+ vq, desc_gaddr,
+ &desc_chunck_len,
VHOST_ACCESS_RO);
if (unlikely(!desc_addr)) {
error = -1;
desc_offset = 0;
desc_avail = desc->len;
- PRINT_PACKET(dev, (uintptr_t)desc_addr, desc->len, 0);
+ PRINT_PACKET(dev, (uintptr_t)desc_addr,
+ (uint32_t)desc_chunck_len, 0);
+ } else if (unlikely(desc_chunck_len == 0)) {
+ desc_chunck_len = desc_avail;
+ desc_gaddr += desc_offset;
+ desc_addr = vhost_iova_to_vva(dev, vq,
+ desc_gaddr,
+ &desc_chunck_len,
+ VHOST_ACCESS_RO);
+ if (unlikely(!desc_addr)) {
+ error = -1;
+ goto out;
+ }
+ desc_offset = 0;
+
+ PRINT_PACKET(dev, (uintptr_t)desc_addr,
+ (uint32_t)desc_chunck_len, 0);
}
/*
vq->used->idx += count;
vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
sizeof(vq->used->idx));
- vhost_vring_call(vq);
+ vhost_vring_call(dev, vq);
}
static __rte_always_inline struct zcopy_mbuf *
return true;
}
+static __rte_always_inline void
+restore_mbuf(struct rte_mbuf *m)
+{
+ uint32_t mbuf_size, priv_size;
+
+ while (m) {
+ priv_size = rte_pktmbuf_priv_size(m->pool);
+ mbuf_size = sizeof(struct rte_mbuf) + priv_size;
+ /* start of buffer is after mbuf structure and priv data */
+
+ m->buf_addr = (char *)m + mbuf_size;
+ m->buf_iova = rte_mempool_virt2iova(m) + mbuf_size;
+ m = m->next;
+ }
+}
+
uint16_t
rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
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);
}
vq = dev->virtqueue[queue_id];
- if (unlikely(vq->enabled == 0))
+
+ if (unlikely(rte_spinlock_trylock(&vq->access_lock) == 0))
return 0;
+ if (unlikely(vq->enabled == 0))
+ goto out_access_unlock;
+
vq->batch_copy_nb_elems = 0;
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
nr_updated += 1;
TAILQ_REMOVE(&vq->zmbuf_list, zmbuf, next);
+ restore_mbuf(zmbuf->mbuf);
rte_pktmbuf_free(zmbuf->mbuf);
put_zmbuf(zmbuf);
vq->nr_zmbuf -= 1;
rte_atomic16_cmpset((volatile uint16_t *)
&dev->broadcast_rarp.cnt, 1, 0))) {
- rarp_mbuf = rte_pktmbuf_alloc(mbuf_pool);
+ rarp_mbuf = rte_net_make_rarp_packet(mbuf_pool, &dev->mac);
if (rarp_mbuf == NULL) {
RTE_LOG(ERR, VHOST_DATA,
- "Failed to allocate memory for mbuf.\n");
+ "Failed to make RARP packet.\n");
return 0;
}
-
- if (make_rarp_packet(rarp_mbuf, &dev->mac)) {
- rte_pktmbuf_free(rarp_mbuf);
- rarp_mbuf = NULL;
- } else {
- count -= 1;
- }
+ count -= 1;
}
free_entries = *((volatile uint16_t *)&vq->avail->idx) -
if (free_entries == 0)
goto out;
- LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+ VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
/* Prefetch available and used ring */
avail_idx = vq->last_avail_idx & (vq->size - 1);
count = RTE_MIN(count, MAX_PKT_BURST);
count = RTE_MIN(count, free_entries);
- LOG_DEBUG(VHOST_DATA, "(%d) about to dequeue %u buffers\n",
+ VHOST_LOG_DEBUG(VHOST_DATA, "(%d) about to dequeue %u buffers\n",
dev->vid, count);
/* Retrieve all of the head indexes first to avoid caching issues. */
/* Prefetch descriptor index. */
rte_prefetch0(&vq->desc[desc_indexes[0]]);
for (i = 0; i < count; i++) {
- struct vring_desc *desc;
+ struct vring_desc *desc, *idesc = NULL;
uint16_t sz, idx;
+ uint64_t dlen;
int err;
if (likely(i + 1 < count))
rte_prefetch0(&vq->desc[desc_indexes[i + 1]]);
if (vq->desc[desc_indexes[i]].flags & VRING_DESC_F_INDIRECT) {
+ dlen = vq->desc[desc_indexes[i]].len;
desc = (struct vring_desc *)(uintptr_t)
vhost_iova_to_vva(dev, vq,
vq->desc[desc_indexes[i]].addr,
- sizeof(*desc),
+ &dlen,
VHOST_ACCESS_RO);
if (unlikely(!desc))
break;
+ if (unlikely(dlen < vq->desc[desc_indexes[i]].len)) {
+ /*
+ * 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[desc_indexes[i]]);
+ if (unlikely(!idesc))
+ break;
+
+ desc = idesc;
+ }
+
rte_prefetch0(desc);
sz = vq->desc[desc_indexes[i]].len / sizeof(*desc);
idx = 0;
if (unlikely(pkts[i] == NULL)) {
RTE_LOG(ERR, VHOST_DATA,
"Failed to allocate memory for mbuf.\n");
+ free_ind_table(idesc);
break;
}
mbuf_pool);
if (unlikely(err)) {
rte_pktmbuf_free(pkts[i]);
+ free_ind_table(idesc);
break;
}
zmbuf = get_zmbuf(vq);
if (!zmbuf) {
rte_pktmbuf_free(pkts[i]);
+ free_ind_table(idesc);
break;
}
zmbuf->mbuf = pkts[i];
vq->nr_zmbuf += 1;
TAILQ_INSERT_TAIL(&vq->zmbuf_list, zmbuf, next);
}
+
+ if (unlikely(!!idesc))
+ free_ind_table(idesc);
}
vq->last_avail_idx += i;
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_unlock(vq);
+out_access_unlock:
+ rte_spinlock_unlock(&vq->access_lock);
+
if (unlikely(rarp_mbuf != NULL)) {
/*
* Inject it to the head of "pkts" array, so that switch's mac
git.droids-corp.org / dpdk.git / blobdiff