Re: [PATCH net-next-2.6] net: Xmit Packet Steering (XPS)

Tom Herbert a écrit :

It probably is a special case, but addresses a real problem.  Like rps,
this addresses the case where a single queue NIC is the bottleneck on a
system.  Anything that takes work off the interrupting core and spreads
it to other cores may alleviate the bottleneck. 

Yes, and this brings back an idea I mentioned earlier in April, for a multicast
problem we had with AthenaCR.

XPS could be 'started' only if caller runs from ksoftirqd, eg we are in a stress
situation, and additional cost of XPS is nothing compared to huge gains we have
to spread part of work to more CPUS.

(Costs of XPS includes : ~500 bytes of ICACHE, and IPI)