Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 21 May 2013 12:29:27 +0200
From:      Marko Zec <zec@fer.hr>
To:        <freebsd-net@freebsd.org>
Cc:        liujie <liujie@263.net>
Subject:   Re: netmap bridge can tranmit big packet in line rate ?
Message-ID:  <201305211229.28182.zec@fer.hr>
In-Reply-To: <1369128316187-5813643.post@n5.nabble.com>
References:  <1369036008085-5813346.post@n5.nabble.com> <CA%2BhQ2%2BhURSWVxTcxCW7Jmd6n1y54ZVC3uUJO1uzcmHCeDuJVFA@mail.gmail.com> <1369128316187-5813643.post@n5.nabble.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 21 May 2013 11:25:16 liujie wrote:
> Hi, Prof.Luigi RIZZO
>
>  Firstly i should thank you for netmap. I tried to send a e-mail to you
> yestoday, but it was rejected.
>
>  I used two machines to test netmap bridge. all with i7-2600 cpu and
> intel 82599 dual-interfaces card.
>
>  One worked as sender and receiver with pkt-gen, the other worked as
> bridge with bridge.c.
>
>  as you said,I feeled comfous too when i saw the big packet performance
> dropped, i tried to change the memory parameters of netmap(netmap_mem1.c
> netmap_mem2.c),but it seemed that  can not resove the problem.
>   60-byte packet send 14882289 pps  recv  13994753 pps
>   124-byte         send   8445770 pps  recv    7628942 pps
>   252-byte         send   4529819 pps  recv     3757843 pps
>   508-byte         send    2350815 pps  recv    1645647 pps
>   1514-byte       send    814288 pps     recv  489133 pps
>
>  sender command: pkt-gen -i ix0 -t   500000000 -l 60
>  receiver command: pkt-gen -i ix1 -r  500000000
> bridge(other machine) command:bridge -i ix0 -i ix1
>
>  can sender and receiver on a same machine ?

Most likely the PCIe path between the dual-ported card and the CPU is the 
bottleneck.  Depending on the chipset and motherboard design, it may be 
that your card is only using 4 instead of 8 PCIe lanes, because some of the 
lanes may be "shared" with another PCIe slot, such as the one in which a 
graphic card is plugged in.

You can also try experimenting with slightly overclocking the PCIe bus if 
the BIOS permits that - I had no problems overclocking the ixgbe and an AMD 
PCIe chipset by 25%.

Marko



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201305211229.28182.zec>