Until this commit, the SIB1's PDCCH_configCommon was set (via a pointer,
i.e., an indirection) to the one of the ServingCellConfigCommon (SCC).
Afterwards, the SIB1 code further populated this PDCCH configCommon,
indirectly populating the one of the the SCC, on which code later
depends.
Not only did this create possibilities for double-frees (since freeing
the memory of SIB1, then SCC would free the PDCCH configcommon twice),
but also it makes it harder to track where certain structures are
populated. Hence, this commit solves both issues:
- The SCC is correctly populated on initialization of the SCC
- The SIB1 makes a deep copy of the SCC' PDCCH configCommon
This resolves the two issues above.
The E1AP Setup Request contained the network configuration (IP address,
ports) as well as the actual E1AP Setup Request application data
(Supported PLMNs, ...). This has the drawbacks that
- The E1AP Setup Request is stored to retrieve IP addresses in the E1AP
module, which is confusing as the Setup Request, per standard, has no IP
info
- The CU-CP received an E1 Setup Request for configuration during start
up, but it did not actually receive such Setup Request, but merely the
IP configuration to set up the socket
This commit splits the E1AP Setup Request into a "real" Setup Request
for application data, and creates a new type e1ap_net_config_t to group
all IP configuration data. Further, a new ITTI message type
E1AP_REGISTER_REQ is introduced to group both types. What happens is
- RCconfig_NR_CU_E1() reads both E1AP application-level data and IP
configuration, as previously
- The data is sent to the CU-CP. It discards the E1AP Setup Request
data, and only uses the network configuration to set up the socket
- The data is sent to the CU-UP. It uses the network configuration to
connect to the CU-CP, and then sends the E1AP Setup Request to the
CU-CP.
Currently, the CU-CP still stores the Setup Request locally, which will
be changed in the next commit to send it to the RRC.
The DRBLists inside E1AP message are for E-UTRAN. We don't support that
with our CU-CP/UP yet, so remove it to reduce ambiguity and complexity.
For the same reason, we remove the CN Support, which for us is always
"NR".
!SIB1 is not correctly filled! This commit does not work with COTS UEs!
The next commit fixes this.
- Move radio config (e.g., minRXTXTIME, SIB1 TDA, do_SRS, etc) down to
MAC.
- Some parameters in MAC are redundant: delete them.
- Remove call to nr_mac_config_scc() from RRC and simulators, the MAC
initialization routine calls it implicitly
Previously, filling of F1 Setup Request depended on reading the RRC;
hence, for timing reasons, we only initialized and sent the F1 Setup
Request once the MAC and RRC are ready.
After moving SCC, SCD, MIB, and SIB1 to MAC, the MAC does not depend on
an RRC being present anymore, and hence, the F1 Setup Request does
neither. In this commit, we read the F1 Setup Request after initializing
the MAC.
The MasterInformationBlock is handled at the DU and sent to the CU in
the F1 Setup Response. Hence, move it down to the MAC. Furthermore:
- Change type because MIB_PDU_t is 4G type
- Simplify schedule_nr_mib()
Note: the reestablishment is broken, and a later commit fixes it.
The ServingCellConfigCommon is an inherently radio-related
configuration. As such, it should be handled by the DU, not the CU.
Therefore, move it "down" to the MAC.
- send F1 Setup using callback
- store f1 setup req locally for reference
- check F1 Setup Req against RRC data structures:
* if matches: send F1 Setup Response using callback
* if not matching: send F1 Setup Failure using callback
- don't send CU config update by default, we don't need this
- if the Setup Response does not contain a cell, don't activate. It is
possible the CU sends a gNB-CU configuration update, which will
initialize the structures
Use a struct to initialize the RRC at the beginning, and remove the
NRRRC_CONFIGURATION_REQ. This message is not necessary; the RRC/CU
should be initialized from the F1 Setup Request, coming from the DU.
Before this commit, we did not read the ServingCellConfigCommon (SCC)
completely at start; rather, we read most (but not all) of it, then also
the minRXTXTIME, and at RRC initialization in
openair_rrc_gNB_configuration(), we filled the
pusch-TimeDomainAllocationList into the SCC, depending on the
minRXTXTIME.
The reason might be that we read part of the SCC, use that to fill the
NRRRC_CONFIGURATION_REQ, read also the minRXTXTIME (part of the RRC
config), and then we have all we need to "finalize the SCC". Since we
move everything to the MAC, I avoid too much upfront code changes,
hardcode the minRXTXTIME (and check with an assert), and undo this in a
later commit when the "RRC config" relevant for radio parameters, e.g.,
minRXTXTIME, is moved to MAC.
I verified the equality of the SCC before (at the end
openair_rrc_gNB_configuration()) and after (at the end of
get_scc_config()) this commit manually using a printf, assuming that
after RRC initialization, we didn't do any further modifications to the
SCC. But in fact, I cannot know that we don't do it...
This branch series implements the usage of the F1 Setup Request
internally at the gNB, even if CU and DU run in the same process. Until
now, the F1 Setup Request contained struct fields to pass IP
configuration to the F1AP module; this information is not necessary in
monolithic, nor should it be relevant when passing the information in
F1 Setup Request.
Therefore, this commit moves all IP connection-related information out
of the F1 Setup Request and into a separate structure instead. In order
to give the F1AP module the necessary connection information, the
DU_REGISTER_REQ message is introduced that combines the F1 Setup Request
and IP configuration, such that the F1AP module can setup the SCTP
connection and then send the F1 Setup Request message.
In a later commit, there will be a callback for F1 Setup Request at MAC
to simply register the DU at the CU. The monolithic callback will simply
forward the F1 Setup Request The F1AP callback will look up the IP
configuration and create DU_REGISTER_REQ message containing IP
configuration and F1 Setup Request, which is then handled by the F1AP
module. To this end, a separate function in gnb_config.c allows to read
IP information from configuration.
- 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
Robert Schmidt