The absslot_tx overflow issue observed during integration testing of AW2S
RU with GPS as the source for the clock synchronization. Causes the gNB
not to start the TX thread and stops the processing.
This commit introduces a separate thread (l1_tx_thread) that processes
gNB DL/TX slots. It receives a message from the ru_thread when a new
slot started, and starts processing.
The DL part of the scheduler is run in the l1_tx_thread. Therefore,
only call UL indication (for scheduler) in UL slots. The UL indication
previously triggered the DL scheduler -- hence, it had to be called in
every slot. Now, since the DL scheduler is moved into the DL thread, we
don't have to call the scheduler in every slot anymore.
The reorder thread is removed, as reordering with this scheme is not
necessary anymore.
The main advantage of this version is that the TX can be scheduled
earlier (sl_ahead). Further, there can no longer be race conditions in
the TX path, since the scheduler/L2, TX L1 processing and the RU tx
function are all execute in the same thread (L1_tX_thread).
Also, the scheduler prepares PRACH/PUSCH/PUCCH FAPI PDUs. As of this
commit, the scheduler runs in the TX processing chain, whereas these
PDUs are destined for RX processing. Thus, too avoid data races,
instead of triggering the RX processing at the same time as TX
processing in the RU thread, this commit changes the code to trigger RX
processing after the scheduler has been run in TX processing to avoid
data races.
Finally, we synchronize RU and L1 threads. This is important for
rfsimulator, since unlike most radios, rfsimulator can run slower or
faster, depending on I/O and CPU (it does not have a stable "tick").
Co-authored-by: kiran <saikiran@iitj.ac.in>
Force the CPU affinity for ru_thread explicitly to ru_thread_core. It is
already done via threadCreate(), but force again as per change request.
Delete commented 7.2 code that did the equivalent. We cannot set it in
the driver, as we don't have access to variable ru->ru_thread_core.
Closes: #732 (suggestion 2)
Suggested-by: Andrew Sergeev <andrew.sergeev@adtran.com>
For the initialization of the xran library (split 7.2), we need some
parameters such as PRACH or frame config. However, to date,
initialization of split at level 7 use the split 8 openair0_config
structure, which does not have the necessary information.
Hence, add a single structure that contains that information.
* make a single call to various function
* make fill_rf_config() static, it is only used in this file
* do not initialize the number of antennas in openair0_cfg in
init_NR_RU(), as it is already done in fill_rf_config()
Harmonize the call to openair0_transport_load(), by putting it in a
single place instead of multiple calls
Further, delay the call to ru_thread, for two reasons:
1) it is somewhat "symmetric" with local RF, which is done there as well
2) for the xran library, we need to read some parameters such as PRACH,
frame format, etc. At the RU, the corresponding FAPI config request
parameter structure is copied from the gNB to RU at the beginning of
the ru_thread, so we have to wait for that to happen.
For FR2 and including analog beamforming, we control the beams via GPIO
on radio units.
In this commit, introduce a configuration option to switch the GPIO
control for different devices: "generic" is the current GPIO control,
"interdigital" configures the GPIO configuration to be used with an
InterDigital device.
- move following radios: AW2S, BladeRF, IRIS, LMSSDR, RFsimulator, USRP
- move following transports: Ethernet, benetel4g, benetel5g
- for each: create independent cmake variable to switch compilation
on/off, in order: OAI_AW2S, OAI_BLADERF, OAI_IRIS, OAI_LMSSDR,
OAI_SIMU, OAI_USRP, OAI_ETHERNET, OAI_BENETEL4G, OAI_BENETEL5G
- Remove RF_BOARD variable, allows to compile multiple boards at once
- where applicable: remove USERSPACE/LIB in path, i.e., move files up to
directories, as all are userspace libraries
Note that with the above, it is now mandatory to enable a radio
implementation before compiling it, e.g., for USRP:
./build_oai -w USRP
cmake .. -DOAI_USRP=ON && ninja oai_usrpdevif
reorder_thread_disable we use the native message queue and L1_rx_thread instead of the thread-pool for the main L1 thread. There is a single TX thread (no slot parallelization)
For some (unclear) reason, nr-softmodem with aw2s device does not print
"Bye." when stopping it with control+c, which is problematic in the CI.
This little hack solves the problem.
(Maybe not the right solution, to be changed if needed.)
Laurent THOMAS