d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

vhost: fix descriptor count in async packed ring

Browse Source 
When vhost receive packets from the front-end using packed virtqueue, it
might use multiple descriptors for one packet, so we need calculate and
record the descriptor number for each packet to update available
descriptor counter and used descriptor counter, and rollback when DMA
ring is full.

Fixes: fe8477ebbd ("vhost: support async packed ring dequeue")
Cc: stable@dpdk.org

Signed-off-by: Cheng Jiang <cheng1.jiang@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Chenbo Xia <chenbo.xia@intel.com>
This commit is contained in:
Cheng Jiang 2022-10-11 03:08:02 +00:00 committed by Chenbo Xia
parent 830f7e7907
commit 5c3a69879e
1 changed files with 15 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
lib/vhost/virtio_net.c
View File

@ -3548,14 +3548,15 @@ virtio_dev_tx_async_split_compliant(struct virtio_net *dev,
} }
static __rte_always_inline void static __rte_always_inline void
vhost_async_shadow_dequeue_single_packed(struct vhost_virtqueue *vq, uint16_t buf_id) vhost_async_shadow_dequeue_single_packed(struct vhost_virtqueue *vq,
uint16_t buf_id, uint16_t count)
{ {
struct vhost_async *async = vq->async; struct vhost_async *async = vq->async;
uint16_t idx = async->buffer_idx_packed; uint16_t idx = async->buffer_idx_packed;
async->buffers_packed[idx].id = buf_id; async->buffers_packed[idx].id = buf_id;
async->buffers_packed[idx].len = 0; async->buffers_packed[idx].len = 0;
async->buffers_packed[idx].count = 1; async->buffers_packed[idx].count = count;
async->buffer_idx_packed++; async->buffer_idx_packed++;
if (async->buffer_idx_packed >= vq->size) if (async->buffer_idx_packed >= vq->size)
@ -3576,6 +3577,8 @@ virtio_dev_tx_async_single_packed(struct virtio_net *dev,
uint16_t nr_vec = 0; uint16_t nr_vec = 0;
uint32_t buf_len; uint32_t buf_len;
struct buf_vector buf_vec[BUF_VECTOR_MAX]; struct buf_vector buf_vec[BUF_VECTOR_MAX];
struct vhost_async *async = vq->async;
struct async_inflight_info *pkts_info = async->pkts_info;
static bool allocerr_warned; static bool allocerr_warned;
if (unlikely(fill_vec_buf_packed(dev, vq, vq->last_avail_idx, &desc_count, if (unlikely(fill_vec_buf_packed(dev, vq, vq->last_avail_idx, &desc_count,
@ -3604,8 +3607,12 @@ virtio_dev_tx_async_single_packed(struct virtio_net *dev,
return -1; return -1;
} }
pkts_info[slot_idx].descs = desc_count;
/* update async shadow packed ring */ /* update async shadow packed ring */
vhost_async_shadow_dequeue_single_packed(vq, buf_id); vhost_async_shadow_dequeue_single_packed(vq, buf_id, desc_count);
vq_inc_last_avail_packed(vq, desc_count);
return err; return err;
} }
@ -3644,9 +3651,6 @@ virtio_dev_tx_async_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
} }
pkts_info[slot_idx].mbuf = pkt; pkts_info[slot_idx].mbuf = pkt;
vq_inc_last_avail_packed(vq, 1);
} }
n_xfer = vhost_async_dma_transfer(dev, vq, dma_id, vchan_id, async->pkts_idx, n_xfer = vhost_async_dma_transfer(dev, vq, dma_id, vchan_id, async->pkts_idx,
@ -3657,6 +3661,8 @@ virtio_dev_tx_async_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
pkt_err = pkt_idx - n_xfer; pkt_err = pkt_idx - n_xfer;
if (unlikely(pkt_err)) { if (unlikely(pkt_err)) {
uint16_t descs_err = 0;
pkt_idx -= pkt_err; pkt_idx -= pkt_err;
/** /**
@ -3673,10 +3679,10 @@ virtio_dev_tx_async_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
} }
/* recover available ring */ /* recover available ring */
if (vq->last_avail_idx >= pkt_err) { if (vq->last_avail_idx >= descs_err) {
vq->last_avail_idx -= pkt_err; vq->last_avail_idx -= descs_err;
} else { } else {
vq->last_avail_idx += vq->size - pkt_err; vq->last_avail_idx += vq->size - descs_err;
vq->avail_wrap_counter ^= 1; vq->avail_wrap_counter ^= 1;
} }
} }