On 2019/10/28 4:17, David Miller wrote:

From: Toke Høiland-Jørgensen <redacted>
Date: Sun, 27 Oct 2019 16:24:24 +0100

quoted

The results in the paper also shows somewhat disappointing performance
for the eBPF implementation, but that is not too surprising given that
it's implemented as a TC eBPF hook, not an XDP program. I seem to recall
that this was also one of the things puzzling to me back when this was
presented...

Also, no attempt was made to dyanamically optimize the data structures
and code generated in response to features actually used.

That's the big error.

The full OVS key is huge, OVS is really quite a monster.

But people don't use the entire key, nor do they use the totality of
the data paths.

So just doing a 1-to-1 translation of the OVS datapath into BPF makes
absolutely no sense whatsoever and it is guaranteed to have worse
performance.

Agree that 1-to-1 translation would result in worse performance.
What I'm doing now is just supporting subset of keys, only very basic ones.
This does not accelerate all usages so dynamic program generation certainly
has value. What is difficult is that basically flow insertion is triggered
by datapath packet reception and it causes latency spike. Going through
bpf verifier on each new flow packet reception on datapath does not look
feasible so we need to come up with something to avoid this.

Toshiaki Makita