RE: [PATCH] net/ice: support vxlan gpe tunnel offload

["Zhang, Qi Z" <qi.z.zhang@intel.com>]

> -----Original Message-----
> From: Ye, MingjinX <mingjinx.ye@intel.com>
> Sent: Tuesday, October 18, 2022 5:29 PM
> To: dev@dpdk.org
> Cc: stable@dpdk.org; Zhou, YidingX <yidingx.zhou@intel.com>; Ye, MingjinX
> <mingjinx.ye@intel.com>; Yang, Qiming <qiming.yang@intel.com>; Zhang, Qi
> Z <qi.z.zhang@intel.com>; Andrew Rybchenko
> <andrew.rybchenko@oktetlabs.ru>; Matz, Olivier
> <olivier.matz@6wind.com>; Somnath Kotur
> <somnath.kotur@broadcom.com>; Ajit Khaparde
> <ajit.khaparde@broadcom.com>
> Subject: [PATCH] net/ice: support vxlan gpe tunnel offload
> 
> PMD does not support VXLAN_GPE tunnel offloading, therefore, it will cause
> the tx queues overflow and stop working when sending such packets.
> 
> This patch adds support for the vxlan gpe.
> 

Is that an issue on the Tx descriptor if ICE_TXD_CTX_UDP_TUNNELING is not configured with a VxLAN GPE packet?
Better to describe the reproduce scenario a little bit here

> Fixes: daa02b5cddbb ("mbuf: add namespace to offload flags")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Mingjin Ye <mingjinx.ye@intel.com>
> ---
>  drivers/net/ice/ice_rxtx.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/net/ice/ice_rxtx.c b/drivers/net/ice/ice_rxtx.c index
> 697251c603..0a2b0376ac 100644
> --- a/drivers/net/ice/ice_rxtx.c
> +++ b/drivers/net/ice/ice_rxtx.c
> @@ -2690,6 +2690,7 @@ ice_parse_tunneling_params(uint64_t ol_flags,
>  		/* for non UDP / GRE tunneling, set to 00b */
>  		break;
>  	case RTE_MBUF_F_TX_TUNNEL_VXLAN:
> +	case RTE_MBUF_F_TX_TUNNEL_VXLAN_GPE:
>  	case RTE_MBUF_F_TX_TUNNEL_GTP:
>  	case RTE_MBUF_F_TX_TUNNEL_GENEVE:
>  		*cd_tunneling |= ICE_TXD_CTX_UDP_TUNNELING;
> --
> 2.34.1