Hello,
Please don't top post.
On Mon, Sep 18, 2023 at 11:47 AM Dexia Li <dexia...@jaguarmicro.com> wrote:
>
> Hi，david
>
> In dpdk 22.11 rte_mbuf_core.h file, there are notes as follows:
>
> /**
> * TCP segmentation offload. To enable this offload feature for a
> * packet to be transmitted on hardware supporting TSO:
> * - set the RTE_MBUF_F_TX_TCP_SEG flag in mbuf->ol_flags (this flag implies
> * RTE_MBUF_F_TX_TCP_CKSUM)
> * - set the flag RTE_MBUF_F_TX_IPV4 or RTE_MBUF_F_TX_IPV6
> * - if it's IPv4, set the RTE_MBUF_F_TX_IP_CKSUM flag
> * - fill the mbuf offload information: l2_len, l3_len, l4_len, tso_segsz
> */
>
> And also, in testpmd example csum engine file csumonly.c, we can see
> func process_inner_cksums partial code as follows
>
> if (tso_segsz)
> ol_flags |= RTE_MBUF_F_TX_TCP_SEG;
> else if (tx_offloads & RTE_ETH_TX_OFFLOAD_TCP_CKSUM) {
> ol_flags |= RTE_MBUF_F_TX_TCP_CKSUM;
> } else {
> if (info->is_tunnel)
> l4_off = info->outer_l2_len +
> info->outer_l3_len +
> info->l2_len + info->l3_len;
> else
> l4_off = info->l2_len + info->l3_len;
> tcp_hdr->cksum = 0;
> tcp_hdr->cksum =
> get_udptcp_checksum(m, l3_hdr, l4_off,
> info->ethertype);
> }
>
> We can see when RTE_MBUF_F_TX_TCP_SEG is set, and
> RTE_ETH_TX_OFFLOAD_TCP_CKSUM is not set.
I don't see the point of copying this code.
Let me repeat what I said in my previous mail: asking for TSO means
implicitly that the hardware will do TCP checksum and the DPDK API
leaves some leeway to applications about this fact.
And specifically for the iavf driver, I see no bug in this area of the code.
To be frank, I found some bug in the net/ice and net/iavf drivers, but
it is not related to setting both ol_flags, and nothing is to be
changed on OVS side to fix this different issue I identified.
Please provide the details of which driver and what the issue is, in
your case, when both flags are set.
--
David Marchand
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
