The compilation line was:
    ./build_oai --eNB -w USRP

The file openairinterface5g/cmake_targets/log/lte-softmodem.Rel10.txt
has been checked and all LOG_X (and 'msg') warnings have been fixed.

The memset was also too big. It should only clear
one subframe.

Only tested without abstraction.
Abstraction does not work for the moment anyway, but just to
keep that in mind.

Some of them might not be correctly handled.
I put /* TODO: XXX */ where I didn't know if
the fix was correct.

see issue #177

fix incorrect fss_pusch calculation
fix the incorrrect HARQ-PID checking
remove incorrect checking on downlink DCI HARQ PID value
fix incorrect downlink ACK/NACK feedback procedures in TDD
fix no activating PDSCH decoding when DL-DCI is indicating a downlink retransmission

see issue #176

           2- reset dataTx buff correctly
           3- pucch/pusch power control fix

see issue #166

enhanced wireshark logging. also logs BCH, PRACH and RAR. Tested with wireshark 2.2.2 on linux and windows

see issue #45

This reverts commit 5466562b16d70f2729f6103e1c0d3ee619470191.

fixed missing a slot_fep() processing on the 1st OFDM symbol of the first DL subframe after UL, causing no SI-RNTI DCI detected

           1-issue#35: miss dci detection
           2-issue#36: ack/nack on pusch are badly encoded
           3-issue#37: ack/nack are sent on pucch instead of pusch

           1- timing Advance fix
           2- harq UL fix

           1- implement max harq retrans
           2- bug in harqId > 1 for dci format 1A
           3- wrong G computation
           4- bug in harq UL for scheduling flag
           5- initial sync, limit freq offset to 150Hz

- add progess bar report
- reduce ssh max_tries to 10
- cleanOldPrograms -> change kill cmd line to be able to run autotest on same machine than lte-softmodem (thanks Rohit & Gabriel)
- Cleanning output prints
- add --skip-machine-preparation on cmd line
- add --skip-sanity-check on cmd line
- add HTML REPORT (no-S1 only)
- add XML detailled report (no-S1 only)