Previously, we would assume every downlink in
slot 7 was an RAR; now we assume every slot 7
that occurs before the CFRA procedure is done
is an RAR. Also changed the logic in scheduler
to allow us to get nr_ue_sdus in the MAC layer.
MR 1170 : 5G nFAPI Improvements
- Bugfixes in packing/unpacking
- Adds uplink indication functionality
- Changes the nFAPI system architecture to account for architecture changes in the develop branch
- Adds nFAPI system documentation
- Removes hardcodings
MR 1222 : Update RRC ASN1 file to version 16.4.1
MR 1223 : Fixes to NR UE MAC PDU generation
- included MAC CE sub-PDUs size in the computation of RLC data to transmit this fixes the error with residual mac pdu appearing at gNB side when transmitting higher amount of data
- got rid of the nr_generate_ulsch_pdu and moved the relevant UL-SCH MAC PDU generation procedure within the nr_ue_get_sdu and nr_ue_get_rach
- reduced memcpy calls and unnecessary memory allocations
- fetching RLC data and SDU size from the mac_rlc_data_req, removed the call to mac_rlc_get_buffer_occupancy_ind
- introduced a new function to generate the MAC CEs sub-PDUs
- removed unnecessary function input parameters
- minor fixes and cleanup
- enhanced logging
MR 1227 : removed duplicate structure, changed 'x' to 'r' and 'y' to 'i'
The ue_sched_mode is reset to 0 again when we
memset the ul_info and after the CFRA procedure
is complete, then uplink CRCs with the PUCCH
response (PUSCH) are not being sent because the
scheduler isnt being called. This makes the
gNB run out of uplink harq ids since its never
getting the proper responses. The change below
will ensure that we call the scheduler AFTER the
RACH is sent and all the time after that.
When running on some machines, the ue_sched_mode
will be initialized to zero. This will cause
the ul_info the go into the ONLY_PUSCH mode.
We do not want to ever call the scheduler and return.
Furthermore, we only want to call the scheduler after
the NRUE has sent the RACH_IND to the gNB.
Also, removed the check before calling the
nr_uescheduler that was associated with the
ra.state. Lastly, removed the DL_HARQ_PID
position change as that was incorrect.
Had to add a check for our NSA mode. If we are in
NSA mode, the DCI format options should be populated
with formats 1 and 7 exclusively. This fixed the
DLBWP calculation for the PDCCH PDUs.
How? Well, we had to properly allocate the
ul_config_req. We also had to update the
ue_get_rach function. There were several
places where the functionality would check
for a particular mode (ra, sa, phy_test, etc).
In some of these places we had to add NSA mode.
In the VNF, they start the scheduler after 3 sec,
this is sort of a hack, might want to revisit.
This commit also includes removal of printfs.
Raymond Knopp