Re: [net-next] net: sched: extend flow action with RSS

[Hariprasad Kelam]

> On Fri, Oct 20, 2023 at 2:12 AM Hariprasad Kelam <hkelam@xxxxxxxxxxx>
> wrote:
> >
> > This patch extends current flow action with RSS, such that the user
> > can install flower offloads with action RSS followed by a group id.
> > Since this is done in hardware skip_sw flag is enforced.
> 
> Our typical rule for TC is we need s/w equivalence for offloads. How would
> this work in absence of offload?
> 
[Hari]  
Our typical rule for TC is we need s/w equivalence for offloads. How would this work in absence of offload?

This patch we added as an extension to receive queue selection in hardware.
This patch "act_skbedit: skbedit queue mapping for receive queue" enabled receive queue selection in hardware
and skip_sw is enforced.

Adding stakeholders of this patch, to get their opinion.
sridhar.samudrala@xxxxxxxxx  amritha.nambiar@xxxxxxxxx

incase of RSS, hardware makes decisions about incoming packets before they are even received in the queue.

Thanks,
Hariprasad k




> cheers,
> jamal
> 
> > Example:
> > In a multi rss group supported NIC,
> >
> > rss group #1 flow hash indirection table populated with rx queues 1 to
> > 4 rss group #2 flow hash indirection table populated with rx queues 5
> > to 9
> >
> > $tc filter add dev eth1 ingress protocol ip flower ip_proto tcp
> > dst_port
> > 443 action skbedit rss_group 1 skip_sw
> >
> > Packets destined to tcp port 443 will be distributed among rx queues 1
> > to 4
> >
> > $tc filter add dev eth1 ingress protocol ip flower ip_proto udp
> > dst_port
> > 8080 action skbedit rss_group 2 skip_sw
> >
> > Packets destined to udp port 8080 will be distributed among rx queues
> > 5 to 9