Allow the user to disable libdnet check through --without-libdnet.
This option is useful to cross-compile tcpreplay to avoid build failures
if libdnet is installed on host but not on target, see
https://bugs.buildroot.org/show_bug.cgi?id=12096

Signed-off-by: Fabrice Fontaine <[email protected]>

Depending on compilaiton flags, UNUSED() was declaring variables with
different names.
Meaning the following can be either expanded to:
  UNUSED(var) -> var
  UNUSED(var) -> UNUSED_var __attribute__((unused))
depending on the compilation flags.

The code using such a variable would need to also be conditional to the
same compilation flags ... and that's to heavy.
Let's declare variables by the name they are given instead.
  UNUSED(var) -> var /* with/without the attribute, but always "var" */

Fixes #612

Signed-off-by: Gabriel Ganne <[email protected]>

The test logic on datalen was inverted.

Processing truncated packats should now raise a warning like the
following:
  Warning: <pcap> was captured using a snaplen of 4 bytes.  This may mean you have truncated packets.

Fixes #616 #617

Signed-off-by: Gabriel Ganne <[email protected]>

There is a small typo in libopts/autoopts/options.h, libopts/configfile.c, libopts/restore.c.

Should read `hierarchically` rather than `hierarcicaly`.

Prior to this change, time between packets was calculated as deltas
and over time errors would quickly add up. The work here switches
to use the wall time since the first packet was sent as the reference
which is compared to the PCAP packet timestamp relative to the first
packet. In this way there are no errors and clock drift is avoided.
As a result tcpreplay can now be used to play back very time sensitive
captures such as video streams with PCR.

Bug #634 fix gcc 8.3.0 complier warnings

Fix UNUSED Macro declaration

Also did some fixes to Juniper Ethernet protocols to fix some bugs
and support various types of Juniper Ethernet protocol types. Used
Wireshark sources to figure out all the different packet types that
Juniper uses.

Unable to test all types because of lack of JNPER DLT pcaps.

Also applied a fix for DLT_RAW to prevent similar issues.

Bug #616 Add checks for datalen for DLT_JUNIPER_ETHER

Bug #616 Bug_#617_CVE-2020-24265 includes updates that fixes
this issue. See PR #637 for details.

Bug #617 CVE-2020-24266 fix tcpprep get_l2len()

…t_packet

Bug #619 buffer overflow get_next_packet()

Add safety and failure reporting for packet captures with
caplen too small.

…_PF_INET6

Bug #624 DLT_NULL/DLT_LOOP support for multi-platform PF_INET6 values

Centos 7 uses automake version 1.13.4 wich is unable to generate
proper [include directives](https://www.gnu.org/software/automake/manual/html_node/Include.html).

To build on Cento 7 use [release version](https:/appneta/tcpreplay/releases/latest).

…pile

Bug #553 Centos 7 disable maintainer compile

Change calculation after high packet sent count. The updated
calculation will not work well for something like `--pps=0.1`
but that would take many years to overflow, even on 32-bit
machines.

Bug #629 Compensate for overflow on high pps tests

fixed simple typo, to short -> too short

docs: fix simple typo, hierarcicaly -> hierarchically

fix heap-buffer-overflow when DLT_JUNIPER_ETHER

comply with Tcpreplay code style

send_packet: Avoid clock drift by using time since first packet

Bug #632 fix pps invalid value protection

Bug #630 apply updates made in PR #631 into send_dual_packets()

configure.ac: fix --without-libdnet

This issue became more evident with #630. First file has proper
deltas, second file sends with no delay.

Tested with 2 ping.pcap files and obverved proper delays.

…ile_works

Bug #472 Reset timestamps for multiplier between files

Code says `jnpr_eth`, but documentation says `jnpr_ether`.

Signed-off-by: Vincent Bernat <[email protected]>

clean up string manipulation in error exit

tcprewrite: fix DLT name for DLT_C_JNPR_ETHER in documentation

Bug #658 fix libopts warnings

remove duplicate #include "lib/strlcpy.h"

replace memset followed by strncpy with strlcpy

add myself to CREDIT

Bug #411 #651 - don't create tap0 if it already exists

Bug #662 ASAN detected memory leaks

…dlt_conversion

Bug #582 calculate L2 length based on exporter DLT