Re: [PATCH net-next] nfp: add L4 RSS hashing on UDP traffic

[Simon Horman <simon.horman@corigine.com>]

On Tue, May 23, 2023 at 02:20:05PM -0700, Jakub Kicinski wrote:
> On Tue, 23 May 2023 12:49:06 +0200 Paolo Abeni wrote:
> > > Previously, since the introduction of the driver, RSS hashing
> > > was only performed on the source and destination IP addresses
> > > of UDP packets thereby limiting UDP traffic to a single queue
> > > for multiple connections on the same IP address. The transport
> > > layer is now included in RSS hashing for UDP traffic, which
> > > was not previously the case. The reason behind the previous
> > > limitation is unclear - either a historic limitation of the
> > > NFP device, or an oversight.  
> > 
> > FTR including the transport header in RSS hash for UDP will damage
> > fragmented traffic, but whoever is relaying on fragments nowadays
> > should have already at least a dedicated setup.
> 
> Yup, that's the exact reason it was disabled by default, FWIW.
> 
> The Microsoft spec is not crystal clear on how to handles this:
> https://learn.microsoft.com/en-us/windows-hardware/drivers/network/rss-hashing-types#ndis_hash_ipv4
> There is a note saying:
> 
>   If a NIC receives a packet that has both IP and TCP headers,
>   NDIS_HASH_TCP_IPV4 should not always be used. In the case of a
>   fragmented IP packet, NDIS_HASH_IPV4 must be used. This includes
>   the first fragment which contains both IP and TCP headers.
> 
> While NDIS_HASH_UDP_IPV4 makes no such distinction and talks only about
> "presence" of the header.
> 
> Maybe we should document that device is expected not to use the UDP
> header if MF is set?

Yes, maybe.

Could you suggest where such documentation should go?