|SUCCESS| [V0,1/1] Add VXLAN TSO function

dpdklab at iol.unh.edu dpdklab at iol.unh.edu
Wed May 20 14:55:51 CEST 2026


Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/164184

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-05-20

Commit: (from series.patch)
Author: Feifei Wang <wff_light at vip.163.com>

Subject: [V0 1/1] Add VXLAN TSO function

This patch adds the RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO flag to the
hinic3 driver's supported TX offload capabilities.

> diff --git a/drivers/net/hinic3/hinic3_ethdev.c b/drivers/net/hinic3/hinic3_ethdev.c
> index f4eb788..fbadb1e 100644
> --- a/drivers/net/hinic3/hinic3_ethdev.c
> +++ b/drivers/net/hinic3/hinic3_ethdev.c
> @@ -696,7 +696,7 @@ hinic3_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
>  		RTE_ETH_TX_OFFLOAD_VLAN_INSERT | RTE_ETH_TX_OFFLOAD_IPV4_CKSUM |
>  		RTE_ETH_TX_OFFLOAD_UDP_CKSUM | RTE_ETH_TX_OFFLOAD_TCP_CKSUM |
>  		RTE_ETH_TX_OFFLOAD_SCTP_CKSUM |
> -		RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM |
> +		RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM | RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO |
>  		RTE_ETH_TX_OFFLOAD_TCP_TSO | RTE_ETH_TX_OFFLOAD_MULTI_SEGS;

The line exceeds the 80-character limit. Can this line be wrapped to
comply with the coding style guide?

Additionally, does the hinic3 hardware actually support VXLAN TSO? The
patch adds the capability flag but provides no evidence of underlying
hardware support or corresponding code changes in the TX path to handle
VXLAN TSO. Is there validation that the hardware can perform tunnel
segmentation for VXLAN-encapsulated packets?

The function hinic3_dev_tnl_tso_support() is called immediately after
this change. Does that function already handle VXLAN TSO configuration,
or does it need modification to enable this offload capability properly?

>  	if (nic_dev->feature_cap & NIC_F_HTN_CMDQ)
>  		hinic3_dev_tnl_tso_support(info, nic_dev);


More information about the test-report mailing list