When offload is enabled, vhost needs to access the first mbuf
to get the packet info, e.g. TCP header. So we couldn't delay
the data copy in this case.
Fixes: e5c494a7a22b ("vhost: batch small guest memory copies")
Reported-by: Lei Yao <[email protected]>
Signed-off-by: Tiwei Bie <[email protected]>
---
lib/librte_vhost/virtio_net.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index b75c93c..3c6f1a5 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -987,7 +987,8 @@ copy_desc_to_mbuf(struct virtio_net *dev, struct
vhost_virtqueue *vq,
mbuf_avail = cpy_len;
} else {
if (likely(cpy_len > MAX_BATCH_LEN ||
- copy_nb >= vq->size)) {
+ copy_nb >= vq->size ||
+ (hdr && cur == m))) {
rte_memcpy(rte_pktmbuf_mtod_offset(cur, void *,
mbuf_offset),
(void *)((uintptr_t)(desc_addr +
--
2.7.4
