From: Jesper Dangaard Brouer <brouer@redhat.com>
To: "Björn Töpel" <bjorn.topel@gmail.com>
Cc: brouer@redhat.com, netdev@vger.kernel.org, ast@kernel.org,
daniel@iogearbox.net, bpf@vger.kernel.org,
magnus.karlsson@gmail.com, magnus.karlsson@intel.com,
jonathan.lemon@gmail.com, ecree@solarflare.com,
thoiland@redhat.com, andrii.nakryiko@gmail.com
Subject: Re: [PATCH bpf-next v3 0/6] Introduce the BPF dispatcher
Date: Mon, 9 Dec 2019 18:00:08 +0100 [thread overview]
Message-ID: <20191209180008.72c98c53@carbon> (raw)
In-Reply-To: <20191209135522.16576-1-bjorn.topel@gmail.com>
On Mon, 9 Dec 2019 14:55:16 +0100
Björn Töpel <bjorn.topel@gmail.com> wrote:
> Performance
> ===========
>
> The tests were performed using the xdp_rxq_info sample program with
> the following command-line:
>
> 1. XDP_DRV:
> # xdp_rxq_info --dev eth0 --action XDP_DROP
> 2. XDP_SKB:
> # xdp_rxq_info --dev eth0 -S --action XDP_DROP
> 3. xdp-perf, from selftests/bpf:
> # test_progs -v -t xdp_perf
>
>
> Run with mitigations=auto
> -------------------------
>
> Baseline:
> 1. 22.0 Mpps
> 2. 3.8 Mpps
> 3. 15 ns
>
> Dispatcher:
> 1. 29.4 Mpps (+34%)
> 2. 4.0 Mpps (+5%)
> 3. 5 ns (+66%)
Thanks for providing these extra measurement points. This is good
work. I just want to remind people that when working at these high
speeds, it is easy to get amazed by a +34% improvement, but we have to
be careful to understand that this is saving approx 10 ns time or
cycles.
In reality cycles or time saved in #2 (3.8 Mpps -> 4.0 Mpps) is larger
(1/3.8-1/4)*1000 = 13.15 ns. Than #1 (22.0 Mpps -> 29.4 Mpps)
(1/22-1/29.4)*1000 = 11.44 ns. Test #3 keeps us honest 15 ns -> 5 ns =
10 ns. The 10 ns improvement is a big deal in XDP context, and also
correspond to my own experience with retpoline (approx 12 ns overhead).
To Bjørn, I would appreciate more digits on your Mpps numbers, so I get
more accuracy on my checks-and-balances I described above. I suspect
the 3.8 Mpps -> 4.0 Mpps will be closer to the other numbers when we
get more accuracy.
> Dispatcher (full; walk all entries, and fallback):
> 1. 20.4 Mpps (-7%)
> 2. 3.8 Mpps
> 3. 18 ns (-20%)
>
> Run with mitigations=off
> ------------------------
>
> Baseline:
> 1. 29.6 Mpps
> 2. 4.1 Mpps
> 3. 5 ns
>
> Dispatcher:
> 1. 30.7 Mpps (+4%)
> 2. 4.1 Mpps
> 3. 5 ns
While +4% sounds good, but could be measurement noise ;-)
(1/29.6-1/30.7)*1000 = 1.21 ns
As both #3 says 5 ns.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2019-12-09 17:00 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-09 13:55 [PATCH bpf-next v3 0/6] Introduce the BPF dispatcher Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 1/6] bpf: move trampoline JIT image allocation to a function Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 2/6] bpf: introduce BPF dispatcher Björn Töpel
2019-12-10 5:50 ` Alexei Starovoitov
2019-12-10 5:54 ` Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 3/6] bpf, xdp: start using the BPF dispatcher for XDP Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 4/6] bpf: start using the BPF dispatcher in BPF_TEST_RUN Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 5/6] selftests: bpf: add xdp_perf test Björn Töpel
2019-12-10 11:05 ` Jesper Dangaard Brouer
2019-12-10 11:56 ` Björn Töpel
2019-12-09 13:55 ` [PATCH bpf-next v3 6/6] bpf, x86: align dispatcher branch targets to 16B Björn Töpel
2019-12-09 15:00 ` [PATCH bpf-next v3 0/6] Introduce the BPF dispatcher Toke Høiland-Jørgensen
2019-12-09 17:42 ` Björn Töpel
2019-12-11 12:38 ` Björn Töpel
2019-12-11 13:17 ` Toke Høiland-Jørgensen
2019-12-09 17:00 ` Jesper Dangaard Brouer [this message]
2019-12-09 17:45 ` Björn Töpel
2019-12-09 19:50 ` Jesper Dangaard Brouer
2019-12-10 19:28 ` Samudrala, Sridhar
2019-12-10 20:04 ` Björn Töpel
2019-12-10 19:59 ` Björn Töpel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191209180008.72c98c53@carbon \
--to=brouer@redhat.com \
--cc=andrii.nakryiko@gmail.com \
--cc=ast@kernel.org \
--cc=bjorn.topel@gmail.com \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=ecree@solarflare.com \
--cc=jonathan.lemon@gmail.com \
--cc=magnus.karlsson@gmail.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=thoiland@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.