From: "Toke Høiland-Jørgensen" <toke@redhat.com>
To: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: David Miller <davem@davemloft.net>,
netdev@vger.kernel.org, Daniel Borkmann <daniel@iogearbox.net>,
Alexei Starovoitov <ast@kernel.org>,
brouer@redhat.com
Subject: Re: [PATCH net-next 2/2] xdp: Add devmap_idx map type for looking up devices by ifindex
Date: Thu, 21 Feb 2019 16:50:13 +0100 [thread overview]
Message-ID: <87o975i1je.fsf@toke.dk> (raw)
In-Reply-To: <20190221162348.51d739a0@carbon>
Jesper Dangaard Brouer <brouer@redhat.com> writes:
> On Thu, 21 Feb 2019 12:56:54 +0100
> Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
>> The default maps used by xdp_redirect() are changed to use the new map
>> type, which means that xdp_redirect() is no longer limited to ifindex < 64,
>> but instead to 64 total simultaneous interfaces per network namespace. This
>> also provides an easy way to compare the performance of devmap and
>> devmap_idx:
>>
>> xdp_redirect_map (devmap): 8394560 pkt/s
>> xdp_redirect (devmap_idx): 8179480 pkt/s
>>
>> Difference: 215080 pkt/s or 3.1 nanoseconds per packet.
>
> (1/8394560-1/8179480)*10^9 = -3.13239 ns
>
> But was the xdp_redirect_map code-path affected from patch 1/1?
> (1/8412754-1/8394560)*10^9 = -0.2576 ns
>
> It doesn't look like any performance regression to xdp_redirect_map
> from these code changes :-)
Nope, the difference between the two patches is just random noise; the
numbers vary more between runs (or even between samples in the same
run), which is why I didn't mention that difference.
-Toke
next prev parent reply other threads:[~2019-02-21 15:50 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-21 11:56 [PATCH net-next 1/2] xdp: Always use a devmap for XDP_REDIRECT to a device Toke Høiland-Jørgensen
2019-02-21 11:56 ` [PATCH net-next 2/2] xdp: Add devmap_idx map type for looking up devices by ifindex Toke Høiland-Jørgensen
2019-02-21 15:23 ` Jesper Dangaard Brouer
2019-02-21 15:50 ` Toke Høiland-Jørgensen [this message]
2019-02-21 21:49 ` Jakub Kicinski
2019-02-21 23:02 ` Toke Høiland-Jørgensen
2019-02-22 0:32 ` Jakub Kicinski
2019-02-22 9:47 ` Toke Høiland-Jørgensen
2019-02-22 21:30 ` Jakub Kicinski
2019-02-23 11:52 ` Toke Høiland-Jørgensen
2019-02-23 23:19 ` kbuild test robot
2019-02-23 23:28 ` kbuild test robot
2019-02-21 15:19 ` [PATCH net-next 1/2] xdp: Always use a devmap for XDP_REDIRECT to a device Jesper Dangaard Brouer
2019-02-21 15:52 ` Toke Høiland-Jørgensen
2019-02-22 0:36 ` Jakub Kicinski
2019-02-22 10:13 ` Toke Høiland-Jørgensen
2019-02-22 21:37 ` Jakub Kicinski
2019-02-23 10:43 ` Jesper Dangaard Brouer
2019-02-23 12:11 ` Toke Høiland-Jørgensen
2019-02-25 18:47 ` Jakub Kicinski
2019-02-26 11:00 ` Toke Høiland-Jørgensen
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=87o975i1je.fsf@toke.dk \
--to=toke@redhat.com \
--cc=ast@kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=netdev@vger.kernel.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).