we have to lock when we dump to file in case of concurrent accesses

it's rare, but it happens

let's print an error in case of failure when trying to open it

it was 8 in the past
now it is 4
we should get the value from config, not hardcoded

Conflicts:
	openair2/LAYER2/MAC/eNB_scheduler_RA.c

This tracer takes the traces ENB_MAC_UE_DL_PDU_WITH_DATA
and ENB_MAC_UE_UL_PDU_WITH_DATA from a trace file
(as generated by the "record" tracer) and sends them
in UDP packets correctly formatted for wireshark to
dissect them.

One unresolved problem I face with my version of
wireshark (1.10.6) is that I don't get the frame number,
only subframe. See in the code to change the behavior
if you use a more recent version of wireshark.

You need to properly configure wireshark for it to work.
See $OPENAIR_DIR/openair2/UTIL/OPT/README.txt.

  - fix the run_enb_ue_virt_s1 script
  - change the Ki value back to the one used

fix gnutls download (closes #122)

Please merge this fix,

Regards,

Frédéric

See merge request !38

This reverts commit 3f20ce7b.

it was mistakenly pushed to develop

thanks to Supreeth Herle for reporting.

left click: enable autoscroll and go to end of list
right click: disable autoscroll

to scroll by only one line, hold CONTROL key and scroll with
mouse wheel

Conflicts:
	cmake_targets/build_oai
	openair1/PHY/LTE_TRANSPORT/dci.c
	openair1/PHY/LTE_TRANSPORT/proto.h
	openair1/SCHED/phy_procedures_lte_eNb.c
	openair1/SCHED/phy_procedures_lte_ue.c
	openair2/LAYER2/MAC/eNB_scheduler_dlsch.c
	openair2/LAYER2/MAC/eNB_scheduler_primitives.c
	openair2/LAYER2/MAC/eNB_scheduler_ulsch.c
	targets/SIMU/USER/oaisim.c
	targets/SIMU/USER/oaisim_functions.c

use --fapi-tracer with build_oai to activate
(basically set FAPI_TRACER to True in CMakeLists.txt)

the file /tmp/fapi.c will be produced at runtime

then cd openair2/LAYER2/MAC and compile fapi_replay.c to replay it

this is forbidden, libscheduler.a is not happy