[P4-dev] Bmv2 with physical interfaces

yunchen chang y2924uki at gmail.com
Mon Jul 3 00:49:25 EDT 2017

Hello Antonin:

Thanks for the reply.

After I do "make clean", the log will not work.
But the web browser still failed after a while.

My topology is :

mininet> ports
s11  lo:0   s11-eth1:1  eth0:2
(s11-eth1 is virtual interface on h1)
(eth0 is physical interface)

Forwarding rules :  forward the packet "destination MAC address: 1" to port
1, others to port 2.

The steps trigger it is :
1. dhclient the virtual interface on h1.
2. add "nameserver" to /etc/resolv.conf on h1.
3. open web browser on h1

(I have check that "ping" is work on h1.)

I want to know:

In addition to transmit the packet between physical interface and virtual
interface, are there something else should I do?
(ex: change some fields on the packet?)

Thank you.

Best regards,

On Fri, Jun 30, 2017 at 1:07 AM, Antonin Bas <antonin at barefootnetworks.com>

> Hi,
> Can you confirm that you re-build all the bmv2 code after running
> configure again? You need to do "make clean" and then "make" again,
> otherwise the new pre-processor flags won't be taken into account.
> Thanks,
> Antonin
> On Thu, Jun 29, 2017 at 12:21 AM, yunchen chang <y2924uki at gmail.com>
> wrote:
>> Hello everyone,
>> I know BMv2 support any interface.
>> I have connected to the physical interface.
>> But when I use the firefox browser, it really slow to transmit the
>> request and response, almost failed after a while.
>> I know that disable logging can improve performance by "./configure
>> 'CXXFLAGS=-O2' --disable-logging-macros --disable-elogger" to BMv2.
>> After it, I find the log still work.
>> What's the possible mistake I made?
>> Thank you.
>> Best regards,
>> Abbie
>> _______________________________________________
>> P4-dev mailing list
>> P4-dev at lists.p4.org
>> http://lists.p4.org/mailman/listinfo/p4-dev_lists.p4.org
> --
> Antonin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.p4.org/pipermail/p4-dev_lists.p4.org/attachments/20170703/7525fb67/attachment-0002.html>

More information about the P4-dev mailing list