Switch to SpeakEasy.net DSL

The Modular Manual Browser

Home Page
Manual: (FreeBSD-5.4-RELEASE)
Apropos / Subsearch:
optional field

NG_BPF(4)                BSD Kernel Interfaces Manual                NG_BPF(4)

     ng_bpf -- Berkeley packet filter netgraph node type

     #include <&lt;net/bpf.h>&gt;
     #include <&lt;netgraph/ng_bpf.h>&gt;

     The bpf node type allows Berkeley Packet Filter (see bpf(4)) filters to
     be applied to data travelling through a Netgraph network.  Each node
     allows an arbitrary number of connections to arbitrarily named hooks.
     With each hook is associated a bpf(4) filter program which is applied to
     incoming data only, a destination hook for matching packets, a destina-
     tion hook for non-matching packets, and various statistics counters.

     A bpf(4) program returns an unsigned integer, which is normally inter-
     preted as the length of the prefix of the packet to return.  In the con-
     text of this node type, returning zero is considered a non-match, in
     which case the entire packet is delivered out the non-match destination
     hook.  Returning a value greater than zero causes the packet to be trun-
     cated to that length and delivered out the match destination hook.
     Either or both destination hooks may be the empty string, or may not
     exist, in which case the packet is dropped.

     New hooks are initially configured to drop all packets.  A new filter
     program may be installed using the NGM_BPF_SET_PROGRAM control message.

     This node type supports any number of hooks having arbitrary names.

     This node type supports the generic control messages, plus the following:

          This command sets the filter program that will be applied to incom-
          ing data on a hook.  The following structure must be supplied as an

              struct ng_bpf_hookprog {
                char            thisHook[NG_HOOKSIZ];     /* name of hook */
                char            ifMatch[NG_HOOKSIZ];      /* match dest hook */
                char            ifNotMatch[NG_HOOKSIZ];   /* !match dest hook */
                int32_t         bpf_prog_len;             /* #isns in program */
                struct bpf_insn bpf_prog[0];              /* bpf program */

          The hook to be updated is specified in thisHook.  The BPF program is
          the sequence of instructions in the bpf_prog array; there must be
          bpf_prog_len of them.  Matching and non-matching incoming packets
          are delivered out the hooks named ifMatch and ifNotMatch, respec-
          tively.  The program must be a valid bpf(4) program or else EINVAL
          is returned.

          This command takes an ASCII string argument, the hook name, and
          returns the corresponding struct ng_bpf_hookprog as shown above.

          This command takes an ASCII string argument, the hook name, and
          returns the statistics associated with the hook as a struct

          This command takes an ASCII string argument, the hook name, and
          clears the statistics associated with the hook.

          This command is identical to NGM_BPF_GET_STATS, except that the sta-
          tistics are also atomically cleared.

     This node shuts down upon receipt of a NGM_SHUTDOWN control message, or
     when all hooks have been disconnected.

     It is possible to configure a node from the command line, using
     tcpdump(1) to generate raw BPF instructions which are then fed into an
     awk(1) script to create the ASCII form of a NGM_BPF_SET_PROGRAM control
     message, as demonstrated here:


         PATTERN="tcp dst port 80"

         cat > /tmp/bpf.awk << xxENDxx
           if (!init) {
             printf "bpf_prog_len=%d bpf_prog=[", \$1;
           } else {
             printf " { code=%d jt=%d jf=%d k=%d }", \$1, \$2, \$3, \$4;
         END {
           print " ]"

         BPFPROG=`tcpdump -s 8192 -ddd ${PATTERN} | awk -f /tmp/bpf.awk`

         ngctl msg ${NODEPATH} setprogram { thisHook=\"${INHOOK}\" \
           ifMatch=\"${MATCHHOOK}\" \
           ifNotMatch=\"${NOTMATCHHOOK}\" \
           ${BPFPROG} } }

     When built as a loadable kernel module, this module includes the file
     net/bpf_filter.c.  Although loading the module should fail if
     net/bpf_filter.c already exists in the kernel, currently it does not, and
     the duplicate copies of the file do not interfere.  However, this may
     change in the future.

     The ng_bpf node type was implemented in FreeBSD 4.0.

     bpf(4), netgraph(4), ngctl(8)

     Archie Cobbs <archieATFreeBSD.org>

BSD                            December 2, 1999                            BSD