dpdk LRO功能总结

dpdk版本：dpdk-stable-16.11.1

程序路径：example/vhost

启用dpdk的LRO

static struct rte_eth_conf default_port_conf = {  
    .rxmode = {  
        .mq_mode = ETH_MQ_RX_RSS,  
        .max_rx_pkt_len = ETHER_MAX_LEN,  
        .split_hdr_size = 0, /**< hdr buf size */  
        .header_split   = 0, /**< Header Split disabled */  
        .hw_ip_checksum = 0, /**< IP checksum offload enabled */  
        .hw_vlan_filter = 0, /**< VLAN filtering disabled */  
        .hw_vlan_strip  = 0, /**< VLAN strip disabled. */  
        .hw_vlan_extend = 0, /**< Extended VLAN disabled. */  
        .jumbo_frame    = 0, /**< Jumbo Frame Support disabled */  
        .hw_strip_crc   = 1, /**< CRC stripped by hardware 这个也得打开*/   
        .enable_lro     = 1, /*打开LRO功能*/
    },  
    .rx_adv_conf = {  
        .rss_conf = {  
            .rss_key = default_rsskey_40bytes,  
            .rss_key_len = 40,  
            .rss_hf = ETH_RSS_PROTO_MASK,  
        },  
    },  
    .txmode = {  
        .mq_mode = ETH_MQ_TX_NONE,  
    },  
};  

rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf);

通过上面代码，dpdk就使能了LRO功能

mbuf的默认长度是2048，当网卡收到大于2048的数据包时，会把数据包存入多个mbuf中，通过mbuf->next关联

但现在的dpdk代码虽然把数据包合并了，却没有计算合并后的数据包校验和，导致在vm中的数据直接丢掉。

vhost接口：

在virtio_enqueue_offload函数中设置VIRTIO_NET_HDR_F_DATA_VALID，可以使上层不检查数据包的校验和。

static void
virtio_enqueue_offload(struct rte_mbuf *m_buf, struct virtio_net_hdr *net_hdr)
{
        uint64_t csum_l4 = m_buf->ol_flags & PKT_TX_L4_MASK;
ffload

        if (m_buf->ol_flags & PKT_TX_TCP_SEG)
                csum_l4 |= PKT_TX_TCP_CKSUM;

        net_hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID;
#if 0

        if (csum_l4) {
                net_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
                net_hdr->csum_start = m_buf->l2_len + m_buf->l3_len;

                switch (csum_l4) {
                case PKT_TX_TCP_CKSUM:
                        net_hdr->csum_offset = (offsetof(struct tcp_hdr,
                                                cksum));
                        break;
                case PKT_TX_UDP_CKSUM:
                        net_hdr->csum_offset = (offsetof(struct udp_hdr,
                                                dgram_cksum));
                        break;
                case PKT_TX_SCTP_CKSUM:
                        net_hdr->csum_offset = (offsetof(struct sctp_hdr,
                                                cksum));
。。。。
}

OVS中：

在netdev_dpdk_filter_packet_len函数里会检查数据包的长度，超过max_packet_len的数据都会被丢弃

static int
netdev_dpdk_filter_packet_len(struct netdev_dpdk *dev, struct rte_mbuf **pkts,
                              int pkt_cnt)
{

    int i = 0;
    int cnt = 0;
    struct rte_mbuf *pkt;

    for (i = 0; i < pkt_cnt; i++) {
        pkt = pkts[i];
        if (OVS_UNLIKELY(pkt->pkt_len > dev->max_packet_len)) {
            VLOG_WARN_RL(&rl, "%s: Too big size %" PRIu32 " max_packet_len %d",
                         dev->up.name, pkt->pkt_len, dev->max_packet_len);
             rte_pktmbuf_free(pkt);
             continue;
             pkt->data_len = 2044;
        }

        if (OVS_UNLIKELY(i != cnt)) {
            pkts[cnt] = pkt;
        }
        cnt++;
    }

    return cnt;
}