On Tue, 15 Sep 2009, Eric Dumazet wrote:

2.6.31 is actually faster than 2.6.22 on the bench you provided.

Well at high packet rates which were not the topic.

Must be specific to the hardware I guess ?

Huh? Even your loopback numbers did show the regression up to 10k.

As text size presumably is bigger in 2.6.31, fetching code
in cpu caches to handle 10 packets per second is what we call
a cold path anyway.

Ok so its an accepted regression? This is a significant reason not to use
newer versions of kernels for latency critical applications that may have
to send a packet once in a while for notification. The latency is doubled
(1G) / tripled / quadrupled (IB) vs 2.6.22.

If you want to make it a fast path, you want to make sure code its
always hot in cpu caches, and find a way to inject packets into
the kernel to make sure cpu keep the path hot.

Oh, gosh.