Re: [RFC] Introduce to batch variants of accept() and epoll_ctl() syscall

Eric Dumazet [off-list ref] writes:

I believe accept() is the problem here, because it contends with the
softirq processing the tcp session handshake.

The MOSBENCH people some time ago did a per CPU accept queue. This is
probably overkill, but there are clearly some scaling problems here
with enough cores.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only