Re: [dpdk-dev] Question about 'rxm->hash.rss' and 'mb->hash.fdir'

[Ferruh Yigit <ferruh.yigit@intel.com>]

On 6/30/2021 3:45 AM, Min Hu (Connor) wrote:
> Hi, all
>     one question about 'rxm->hash.rss' and 'mb->hash.fdir'.
> 
>     In Rx recv packets function,
>     'rxm->hash.rss' will report rss hash result from Rx desc.
>     'rxm->hash.fdir' will report filter identifier from Rx desc.
> 
>     But function implementation differs from some PMDs. for example:
>     i40e, MLX5 report the two at the same time if pkt_flags is set,like:
> ******************************************
>         if (pkt_flags & PKT_RX_RSS_HASH) {
>             rxm->hash.rss =            
> rte_le_to_cpu_32(rxd.wb.qword0.hi_dword.rss);
>         }
>         if (pkt_flags & PKT_RX_FDIR) {
>             mb->hash.fdir.hi =
>         rte_le_to_cpu_32(rxdp->wb.qword3.hi_dword.fd_id);
>         }
> ********************************************
> 
>     While, ixgbe only report one of the two. like:
> ******************************************
>         if (likely(pkt_flags & PKT_RX_RSS_HASH))
>             mb->hash.rss = rte_le_to_cpu_32(
>                 rxdp[j].wb.lower.hi_dword.rss);
>         else if (pkt_flags & PKT_RX_FDIR) {
>             mb->hash.fdir.hash = rte_le_to_cpu_16(
>                 rxdp[j].wb.lower.hi_dword.csum_ip.csum) &
>                 IXGBE_ATR_HASH_MASK;
>             mb->hash.fdir.id = rte_le_to_cpu_16(
>                 rxdp[j].wb.lower.hi_dword.csum_ip.ip_id);
>         }
> ********************************************
>     So, what is application scenario for 'rxm->hash.rss' and 'mb->hash.fdir',
> that is, why the two should be reported? How about reporting the two at the same
> time?
>     Thanks for  your reply.


Hi Connor,

mbuf->hash is union, so it is not possible to set both 'hash.rss' & 'hash.fdir'.

I assume for i40e & mlx5 case 'pkt_flags' indicate which one is valid and only
one is set in practice. Cc'ed driver mainteriners for more comment.