netmap-ipfw on em0 em1

[Evandro Nunes]

On Sat, Nov 8, 2014 at 5:26 AM, Mahnaz Talebi <mhnz.talebi at gmail.com> wrote:

> Hi Evandro.
> I've tested netmap-ipfw on real NICs.
> Use "
>
> ./kipfw -i netmap:em0 -i netmap:em1
>  " to run netmap-ipfw on em0 and em1. ipfw works as a bridge and copy
> incoming packets to em0 to em1 if they pass defined rules (and vice versa,
> from em1 to em0).
> If you still have problem with ipfw-netmap, please send your scenario for
> testing it.
>

dear mahaza, thank you for your suggestion

still didn't work, in fact the syntax you mentioned returns an error that
later turns out working just like if I had used ./kipfw netmap:em1
netmap:em2, see the output:

*** Global Sysctl Table entries = 41, total size = 2144 ***
[ 706.224574] session.c:do_server  [541] +++ listening tcp 127.0.0.1:5555
[ 706.224645] netmap_io.c:netmap_add_port [310] opening netmap device -i
[ 706.224666] netmap_io.c:netmap_add_port [320] error opening -i
[ 706.224681] netmap_io.c:netmap_add_port [310] opening netmap device
netmap:em1
[ 706.240897] netmap_io.c:netmap_add_port [326] --- mem_id 1
[ 706.240938] netmap_io.c:netmap_add_port [329] create sess 0x801449070
my_netmap_port 0x801429580
[ 706.240953] netmap_io.c:netmap_add_port [310] opening netmap device -i
[ 706.240964] netmap_io.c:netmap_add_port [320] error opening -i
[ 706.240976] netmap_io.c:netmap_add_port [310] opening netmap device
netmap:em2
[ 706.257132] netmap_io.c:netmap_add_port [326] --- mem_id 1
[ 706.257175] netmap_io.c:netmap_add_port [329] create sess 0x8014490a0
my_netmap_port 0x801429800
[ 706.257187] netmap_io.c:netmap_add_port [342] 0x801429800 em2 1 <->
0x801429580 em1 1 SWAP
[ 706.257455] missing.c:callout_run [378] running 0x61e9d0 due at 1 now 168
[ 706.257480] session.c:mainloop   [624] callouts 1 skipped 0
[ 707.000201] session.c:mainloop   [624] callouts 3213 skipped 0
[ 708.000200] session.c:mainloop   [624] callouts 7563 skipped 0
[ 709.000079] session.c:mainloop   [624] callouts 11896 skipped 0
[ 710.000044] session.c:mainloop   [624] callouts 16232 skipped 0
[ 711.000065] session.c:mainloop   [624] callouts 20567 skipped 0

so -i opt is considered a netmap port which is unable to be open and in the
end em2 and em1 and bridged, as it seems to be like on the line:

[ 706.257187] netmap_io.c:netmap_add_port [342] 0x801429800 em2 1 <->
0x801429580 em1 1 SWAP

so it's the same as ./kipfw netmap:em1 netmap:em2.

and therefore still have the same problem, packets count but I am
completely out of communication on both NICs.

my scenario is:

(Machine-A)<-->Machine-B<--->(MachineC)

Machine-A:
em0 172.16.251.3/24

Machine-B:
em1: 172.16.251.1/24
em2: 172.16.252.1/24
10.0-STABLE w/ latest netmap-ipfw and netmap code from google code
repository

Machine-C:
em0 172.16.252.3/24

without kipfw hooked, Machine-A and Machine-C reach each other.
but if ./kipfw netmap:em1 netmap:em2 is used, it turns I am completely out
of communication on both em1 and em2 NICs.














> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>