From: Paolo Abeni <pab...@redhat.com> Date: Thu, 14 Dec 2017 09:29:46 +0100
> I was unable to measure any performance regression for the hash based > demultiplexing, and I think that the number of function calls is > unchanged in such scenario (with vanilla kernel we have ehash() and > reuseport_select_sock(), with the patched one __reuseport_get_info() > and ehash()). > > I agree you are right about the additional stack usage introduced by > this patch. > > Overall I see we need something better than this. Thanks for checking whether it's slower or not. I wonder if x86-64 is putting the argument parts on the stack anyways...
