This reverts commit 85e1c34c.

This reverts commit 0280c083.

1. Changed FIFO types to threadsafe to ensure order of execution
2. Changed RB_NB_TH to 1 because PUCCH have some issues when running 2 slots in parallel (race). Dont know exactly where it happens.

Created separate job for PDSCH processing so that its exectued in the right order in rfsimulator mode.

1. RX PBCH slot n
2. RX PDCCH slot n
3. Start TX PUSCH in parallel for slot n+6
4. RX PDSCH slot n
5. Join PUSCH
6. TX PUCCH slot n+6

when signal quality drops. when cmd line option used ue goes back to sync mode

thread1: RX
thread2: TX

CI: removing 1 VM in the pipeline

See merge request oai/openairinterface5g!1078

Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>

Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>

Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>

Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>

MR1029 : Benetel driver update  
This MR fixes an issue with the driver for the 4G Benetel RRUs.
Power levels on the 5G side were adapted to avoid saturation of the signals. 
5G performance is now the same as with USRP. 
Support for 2 antennas for both 4G and 5G was added

MR1062 : fixing the RE offset for each Rx antenna in pusch channel estimation

MR1046 : Add support for NR UL SC-FDMA up to 100 MHz

MR1053Nr : pdcp nea2 security

MR1049 : improve rfsim
-Fix a regression 
Earlier parameter reading was moved to an external thread,  
Eventually the check at the end of the main was too early, 
Declaration of  some extra parameters are now on command line
-Enhancement: 
use the rfsim as a server on UE side, so we can connect two xNB to one UE
-Simplification: 
remove rfsim flags that have been made for convergence with replay function in usrp driver, but this is useless as they changed their code
-Fix a bug in ubuntu 20.04 (now the code is ready in whole OAI)


MR1056 : Bugfix: NR BSR calculation
Fixes a bug in the scheduler for BSR calculation. 
Before, we might wrongly track the BSR of a UE and not schedule it anymore although it has data. 
Should be fixed now and improve UL throughput. 

MR963 : Nr mac multi rach global edge
-Handling of Multiple Users triggering RACH request in different RACH Occasions in same slot
-Providing Random Access Response according to RACH request

fixing the RE offset for each Rx antenna in pusch channel estimation

MR978 : 
We have incorporated 5G NR nFAPI into the develop branch
P5 and P7 interfaces for NR have been implemented. 
With this MR, downlink transmission through nFAPI will be possible. 
We have also made sure that 4G nFAPI can be used from within the develop branch.

reducing UL target power for Benetel 5G RRU and adding missing parameter for PUSCH multi thread processing

1) Only count new transmission as scheduled bytes (which is then
compared against the BSR)

2) When subtracting scheduled bytes after successful reception, subtract
TBsize from correct HARQ process

go figure...

Dirty!

Dirty.