From: Eric Dumazet <eric.dumazet@gmail.com>
To: Rui <wirelesser@gmail.com>
Cc: netdev@vger.kernel.org
Subject: Re: multi bpf filter will impact performance?
Date: Wed, 01 Dec 2010 05:03:35 +0100 [thread overview]
Message-ID: <1291176215.2856.383.camel@edumazet-laptop> (raw)
In-Reply-To: <AANLkTi=-iuGJDzELJpu26KtKr77uDx0xNCc8ibEidJ4h@mail.gmail.com>
Le mercredi 01 décembre 2010 à 11:48 +0800, Rui a écrit :
> if RPS can spread the load into 4 separate cpus, how about the
> "packet_rcv(or tpacket_rcv)" ? will they run in parallel?
>
> as I use 'tcpdump(PACKET_MMAP)' to copy the packet to user space, I
> expect there are simultaneous packet_rcv running in each CPU to put
> the packet into ringbuffer.
Yes, the filter code wan run in parallel with no particular slowdown,
since code and bpf data is shared by all cpus (no writes)
But its rather important for performance that each cpu store packets
into its own packet socket or ring buffer, to avoid false sharing
slowdowns.
With such a setup (split packets to four cpus, then make sure one cpu
deliver packets to one particular PACKET socket/ring buffer), it should
really be fast enough.
next prev parent reply other threads:[~2010-12-01 4:03 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-30 9:22 multi bpf filter will impact performance? Rui
2010-11-30 9:34 ` Eric Dumazet
[not found] ` <AANLkTi=VpmnrXTBNV7McQm6mq9ULT7KTKbM8_hLPoL=2@mail.gmail.com>
[not found] ` <1291127670.2904.96.camel@edumazet-laptop>
2010-12-01 3:48 ` Rui
2010-12-01 4:03 ` Eric Dumazet [this message]
2010-12-01 7:45 ` [PATCH net-next-2.6] filter: add SKF_AD_RXHASH and SKF_AD_CPU Eric Dumazet
2010-12-01 8:03 ` Changli Gao
2010-12-06 21:02 ` David Miller
2010-12-03 9:40 ` multi bpf filter will impact performance? Junchang Wang
2010-12-01 7:36 ` Changli Gao
2010-12-01 7:47 ` Eric Dumazet
2010-12-01 7:59 ` Changli Gao
2010-12-01 8:09 ` Eric Dumazet
2010-12-01 8:15 ` Changli Gao
2010-12-01 8:42 ` Eric Dumazet
2010-12-01 17:22 ` Hagen Paul Pfeifer
2010-12-01 18:18 ` David Miller
2010-12-01 18:24 ` David Miller
2010-12-01 18:24 ` Eric Dumazet
2010-12-01 18:44 ` David Miller
2010-12-01 19:48 ` Eric Dumazet
2010-12-01 20:23 ` David Miller
2010-12-01 20:45 ` [PATCH net-next-2.6] filter: add a security check at install time Eric Dumazet
2010-12-02 2:30 ` Changli Gao
2010-12-02 6:46 ` Eric Dumazet
2010-12-02 8:11 ` Changli Gao
2010-12-02 8:53 ` Eric Dumazet
2010-12-02 9:00 ` Eric Dumazet
2010-12-02 9:10 ` Changli Gao
2010-12-02 9:54 ` Eric Dumazet
2010-12-02 10:10 ` Changli Gao
2010-12-02 11:15 ` Eric Dumazet
2010-12-02 11:29 ` Changli Gao
2010-12-02 13:14 ` Eric Dumazet
2010-12-02 10:59 ` Changli Gao
2010-12-06 21:07 ` David Miller
2010-12-03 6:32 ` multi bpf filter will impact performance? Eric Dumazet
2010-12-05 20:53 ` PATCH] filter: fix sk_filter rcu handling Eric Dumazet
2010-12-05 21:08 ` Andi Kleen
2010-12-05 21:28 ` Eric Dumazet
2010-12-06 17:29 ` David Miller
2010-11-30 10:01 ` multi bpf filter will impact performance? Eric Dumazet
2010-11-30 11:17 ` Eric Dumazet
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=1291176215.2856.383.camel@edumazet-laptop \
--to=eric.dumazet@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=wirelesser@gmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox