first_rx_set is polling-specific: it makes the polling loop wait to only
check the frame/slot counter once we set it for the first time. However,
that is equivalent to checking the frame/slot directly, which will also
be 0 in the beginning.
first_read_set seems to have the function of making frame/slot time
advance only if we started to read the samples in
xran_fh_rx_read_slot(). I interpret that as "only advance the time if
the gNB is reading from xran", which does not make sense to me, because
xran advances the time anyway. Also, polling would not notice, and
no-polling would simply warn about double messages. So the outcome is
likely the same, whether first_read_set is set or not.
I checked in multiple CI runs; it happens that first_read_set is always
set already when oai_xran_fh_rx_callback() is called anyway.
xran_fh_rx_read_slot() uses polling (a loop over cnt++) to wait for the
next frame/slot. The reasons I have been told is "it does not work
differently", or "there are real-time problems". None of those seem to
be true; by default disable polling, and use the queue instead.
It is possible to enable polling with the cmake option
OAI_FHI72_USE_POLLING=ON. Documentation has been updated to inform about
this change.
The queue L1_rx_out was in place to keep RX/TX processing relatively
"close" to each other (in terms of time, since in RFsim, they might
advance faster than realtime; see also parent commit). This lead to a
memory leak in radios that do not use ru->feprx (no front-end processing
for RX, e.g., PRACH), as the code to read the L1_rx_out messages was
inside a block of if (ru->feprx). Move it out so that the queue is
always read and messages are freed to eliminate the memory leak, and
remove the workaround for FHI72 to avoid this memory leak.
This also allows to run RFsim when compiling FHI72, which was not
possible before this commit.
Closes: #812
Some radios (e.g., RFsim) can run faster. A mechanism has been put in
place to keep RX/TX frame/slot numbers not too far from each other.
Refactor in a separate function with documentation to make this clearer.
38.413 section 8.2.1.4 says that if PDU session resource setup req
contains an existing PDU session, that one should be reported as failed.
That is rather complicated for us to do; simply reject all PDU sessions
and let the core deal with that.
If two PDU sessions are to be setup, the CU-CP sends two bearer context
setup requests. When running only the CU-UP, we inconditionally tried to
set up the CU-UP UE ID identifiers upon reception of the bearer context
setup request, leading to an assert. Check if the IDs are not already
set up, to avoid assertions in that case.
Integration: `2025.w06`
See merge request oai/openairinterface5g!3248
* !3202 Simplify usage of the old segment decoding libraries with the slot coding interface
* !3240 Fix typos in NR_SA_Tutorial_OAI_multi_UE.md
* !3242 Period based phytest bitmap
* !3238 Refactor tun_if.h
* !3000 Improvements for NR dlsim and ulsim
* !3239 Remove inexistant SIMD instruction
* !3246 Deadlock avoidance in rfsimulator
* !3251 nFAPI: make 4-layer on 100MHz work
* !3243 Reset E1 UE contexts after E1 Setup Response
* !3245 Added Support of 35Mhz,45Mhz,70Mhz Bandwidth
* !3219 E1AP enc/dec lib improvements
E1AP enc/dec lib improvements
This MR is including some fixes and improvements for the E1AP library,
focusing mostly on Bearer Context Management:
- struct reorganization: split setup and modification request messages
into distinct structures, ensuring compliance with 3GPP TS 38.463
specifications
- add encoding/decoding sub-functions for different IEs
- improvements to Bearer Context Management: Implemented various
improvements to the E1 Bearer Context Setup Response and Request,
including the adoption of encoding/decoding sub-functions for
different IEs
- use lib functions whenever need (e.g. cp_bearer_context_setup_request
in cucp_cuup_e1ap.c)
- handle optional struct members as pointers
- updated copy/equality check utility functions
EDIT: The MR is also including the changes from !3222 (closed):
> While testing E1 with security enabled for DRBs, it was found that
> data traffic was not functional.
>
> The problem was that security settings were modified in the PDCP
> reestablishment happening when receiving Bearer Context Modification
> Request.
>
> So some restructuring was done.
>
> The actual fix is in 18176297, the others are for support. Basically
> we introduce a boolean to check if the security settings are present in
> Bearer Context Modification Request and we change PDCP security settings
> only if they are.
>
> (Note that in OAI CU security settings are not sent in Bearer Context
> Modification Request, so we could also remove all this, but we need to
> be compatible with other cu-up and/or cu-cp.)
Added Support of 35Mhz,45Mhz,70Mhz Bandwidth
Added Support of 35Mhz,45Mhz,70Mhz Bandwidth for 30Khz Subcarrier
Spacing, Also added 35Mhz, 45Mhz bw for 15Khz Subcarrierspacing.
Referred Spec TS 38.101 v18.6.0 section 5.3.2, table 5.3.2-1
Reset E1 UE contexts after E1 Setup Response
38.463 sec. 8.2.3 says
> This procedure also re-initialises the E1AP UE-related contexts (if
> any) and erases all related signalling connections in the two nodes
> like a Reset procedure would do.
Hence, delete all contexts after reception of E1 Setup Response. This
minimizes a risk of receiving an E1 bearer setup req for an existing UE,
which currently leads to an assert [we trust the CU-CP does not send the
same UE ID twice, as we only "mirror" the existing UE ID]. After this
MR, this is what happens if these entities fail:
- CU-CP fails: all context lost implicitly, upon reconnection, both DU
and CU-UP reset all contexts (like "start from zero")
- DU fails: CU-CP triggers NGAP UE context release request immediately
(radio connection with UE lost), all UE contexts attached to this DU
cleared (like "start from zero" for affected DU)
- CU-UP fails: initially, nothing (UE has no user plane); upon
reconnection of CU-UP, CU-CP triggers NGAP UE context release request
(cause radio connection with UE lost) and sends F1 Reset to DU, all UE
contexts associated to this DU cleared (like "start from zero")
for the last case, a possible improvement could be to use the F1 Reset
message to only clear affected UE conetxts. I chose not to do this,
because this MR is very big now already, and possible "victim" UEs
that were not at that CU-UP will reconnect as well
In Bearer Context Modification Request, the security information
field is needed only if we change the PDCP algorithms/keys. Since we
don't, no need to pass it.
Let's keep the old code, maybe the current understanding of the
specifications is wrong.
If the Bearer Context Modification Request does not contain security
information, the PDCP entity shall keep the currently configured security
settings.
Passing -1 for ciphering_algorithm and integrity_algorithm is the way
to implement this logic.
Check if the IE is memory allocated to deal with presence/absence of this information
element in Bearer Context Modification Request and add encoding/decoding
in the code.
Note that for decoding the integrity settings may be absent, in which
case we consider NIA0 to be used (no integrity protection). To be refined
if needed.
And for encoding, we always encode integrity settings, even if it's NIA0.
May also be refined if needed.
This was not a big problem, but we were advertising wrong
algorithm and deriving keys using this wrong algorithm in case
DRB ciphering and/or integrity was not active. (Say SRBs
are configured with NIA2 but DRBs are configured without
integrity, we would advertise NIA2 and send the integrity
key derived for NIA2, instead of NIA0 for both.)
It was not a problem because in the PDU Session Resource To Setup
item, there is "securityIndication" which we use to effectively
activate/deactivate the ciphering and/or integrity.
But it did not look clean to see a SecurityInformation with
incorrect data when inspecting the message in wireshark.
So let's use the correct values.
We could also not include the SecurityInformation if both ciphering
and integrity are not used, and only include ciphering if integrity
is not used (because integrity settings are optional).
But then if only integrity is used, we still need to include
ciphering, setting algorithm to NEA0 (no ciphering), because
the ciphering settings are always included.
This logic is too complex, let's use the simple one to always
include SecurityInformation with NEA0 and/or NIA0 if ciphering
and/or integrity is not activated.
As in deba3ae066 ("Reset CU-UP E1 UE contexts after E1 Setup
Response"), the CU-CP should reset the E1 UE contexts after E1 Setup
Response. 38.463 sec. 8.2.3 says
> This procedure also re-initialises the E1AP UE-related contexts (if
> any) and erases all related signalling connections in the two nodes like
> a Reset procedure would do.
In the case of CU-CP, this is a bit more complicated. More precisely, we
need to wait for the CU-UP to be available again before informing the DU
about the loss (implemented via an F1 Reset message), because otherwise,
the UE might reconnect immediately, before the CU-UP is online again.
Hence, on connection loss, we mark affected UEs (by setting their E1
association ID to 0); as long as the CU-UP does not connect, nothing
will happen (but UEs don't have user plane). Upon CU-UP connecting
again, we request release of the UEs via NGAP (similarly to what is done
if the DU goes offline, see invalidate_du_connections()), free the UE
contexts locally, and send an F1 Reset message to the DU, because
38.473 sec 8.2.1.2.1 says
> a failure at the gNB-CU [...] has resulted in the loss of some or
> all transaction reference information [...] a RESET message shall be
> sent to the gNB-DU.
The core should accept and send an NGAP release command message; since
no UE contexts are to be found (already deleted), the CU-CP will just
send a release complete (together with some warnings).
It might be possible to more gracefully handle the situation (e.g., as
described in 38.401 section 8.9.5 "Change of gNB-CU-UP"), but
- we lack the PDCP SN exchange
- the current procedure has the advantage of being equivalent to
"restart of the entire gNB", with less downtown.
The next commit will use rrc_remove_ue() to delete a CU(-CP+-UP)
context. However, we don't send NGAP UE context release at that moment
(we will instead *request* the release); more generally, it's not clear
that every user of rrc_remove_ue() wants to actually send this message.
Hence, refactor the code by separately calling
rrc_gNB_send_NGAP_UE_CONTEXT_RELEASE_COMPLETE() where appropriate.
Bearer context setup request and modification request messages
differ in various IEs according to 9.2.2 of 3GPP TS 38.463, e.g.
the Bearer context setup request does not contain a list of
DRBs to modify and does not contain an UP TL configuration.
Thus, the goal of this commit is to split the setup request message
and the modification request into two different structs:
* adopt e1ap_bearer_mod_req_t where necessary (e.g. fill_e1_bearer_modif)
Also, reorganized existing struct members according to specs:
* remove irrelevant struct members from the setup request
* group members from the same IEs into structs
* set optional struct members IEs as pointers
* add and adopt enc/dec/cp/eq lib functions for single IEs
* add UP_TL_information_t struct with TEID and TL address info and adopt in stack
* use int instead of long for count types
* update the relevant copy/equality check utility functions
Extra:
* add a function to handle Security Information in RRC in the process
* add create_up_tl_info and adopt in unit tests
nFAPI: make 4-layer on 100MHz work
5G nFAPI headers specify 32 bits for the length field, so use it also
when segmenting messages. This should stabilize 4-layer MIMO operation
on 100MHz with nFAPI.
Deadlock avoidance in rfsimulator
This change introduces a countermeasure for deadlock in rfsimulator. The
deadlock happens when all entities are waiting for new data to come in,
and happens with 2+ clients, when a new client connects. I think this
issue is due to ordering of fullwrite calls, resulting in out-of-order
delivery of packets and eventually trashing the packets on the receiving
side. The out-of-order delivery warnings are printed just before the
system deadlocks but I have not found a better solution so far. The
workaround makes the server never lock up permanently by ignoring the
client failure to write on time after 10 tries.
This was tested locally for both UE as server and gNB as server and
works correctly, causing the deadlock to clear and the added log to be
printed several times when the deadlock is detected, after which the
system goes back to normal.
I have some gdb output of the executables during deadlock:
UE:
$7 = {conn_sock = 98, lastReceivedTS = 3226163740, headerMode = true, trashingPacket = false, th = {size = 13184, nbAnt = 1, timestamp = 3226150556, option_value = 0, option_flag = 0}, transferPtr = 0x7f6a500018a8 "\200\063", remainToTransfer = 24,
circularBufEnd = 0x7f6a503b3ac0 "", circularBuf = 0x7f6a501f1ac0, channel_model = 0x0}
(gdb) p t->buf[5]
$8 = {conn_sock = 97, lastReceivedTS = 0, headerMode = true, trashingPacket = false, th = {size = 0, nbAnt = 0, timestamp = 0, option_value = 0, option_flag = 0}, transferPtr = 0x7f6a50001900 "", remainToTransfer = 24, circularBufEnd = 0x7f6a50575ad0 "",
circularBuf = 0x7f6a503b3ad0, channel_model = 0x0}
nextRxTimestamp 3225937740
nsamps = 30720
gNB 1:
(gdb) p t->buf[0]
$4 = {conn_sock = 95, lastReceivedTS = 3226026876, headerMode = true, trashingPacket = false, th = {size = 1, nbAnt = 1, timestamp = 3226026875, option_value = 0, option_flag = 0},
transferPtr = 0x7f8dfc003ab8 "\001", remainToTransfer = 24, circularBufEnd = 0x7f8e1c3ff010 "", circularBuf = 0x7f8e1c23d010, channel_model = 0x0}
nextRxTimestamp 3225996956
gNB 2:
lastReceivedTS = 3226026875
$2 = {conn_sock = 95, lastReceivedTS = 3226026875, headerMode = true, trashingPacket = false, th = {size = 1, nbAnt = 1, timestamp = 3226026875, option_value = 0, option_flag = 0},
transferPtr = 0x744898003ab8 "\001", remainToTransfer = 24, circularBufEnd = 0x7448bc2e7010 "", circularBuf = 0x7448bc125010, channel_model = 0x0}
nextRxTimestamp 3226026875
As you can see all executables are in have_to_wait state.
38.463 sec. 8.2.3 says
> This procedure also re-initialises the E1AP UE-related contexts (if
> any) and erases all related signalling connections in the two nodes like
> a Reset procedure would do.
Hence, delete all contexts after reception of E1 Setup Response. This
minimizes a risk of receiving an E1 bearer setup req for an existing UE,
which currently leads to an assert [we trust the CU-CP does not send the
same UE ID twice, as we only "mirror" the existing UE ID].
Make the same change as in parent commit, i.e., write the full 32-bit
segment size.
I could not test this, as we do not reach these rx_data.indication
length. It's based on the fix in the parent commit, and to avoid future
bad surprises.
The 5G nFAPI message length is 32bits. In particular tx_data.requests
can be longer than 64kB. When segmenting, we should correctly write the
message of the current segment (across all 32bits), because the
length would interpreted wrongly otherwise.
This fixes a bug in which tx_data.requests were discarded for 4-layer DL
MIMO on 100 MHz with this message:
P7 unpack message length is greater than the message buffer
Further, increase the type of various (segment-related) variables to 32
bits. Currently, the maximum segment size is sxt to 65000 bytes (and in
will likely remain, because the maximum UDP size is 65536);
nevertheless, increase it in case we will ever go beyond this.
See also commit dee68e6319 ("nFAPI: increase maximum segment size to
65000")
Simplify usage of the old segment decoding libraries with the slot coding interface
This MR transforms the segment coding libraries into slot coding
libraries by statically linking the adaptation layer and the segment
coding libraries.
Therefore they can be used directly in ldpctest which uses the segment
coding interface as well as in all the other PHY simulators and the NR
softmodem which use the slot coding interface.
It is not anymore needed to use the
--nrLDPC_coding_segment.segment_shlibversion
flag. Instead a segment coding library can be directly chosen with
--loader.ldpc.shlibversion
since it implements the slot coding interface.
Integration: `2025.w05`
Closes#885
See merge request oai/openairinterface5g!3233
* !3102 Dockerized development environment
* !3185 Add IQ file recording and IQ file viewer to ImScope
* !3218 more layer1 cleaning
* !3224 Ensure execution of processSlotTX in order in NR UE
* !3231 Bugfix in frame and slot setting for ULSCH beam allocation
* !3229 Verify the integrity protection of the RRCReestablishment message
* !3230 FHI72: fix for single distributed antenna array for xran F release
* !2984 NAS 5GS refactor
* !3235 Parametrized JenkinsNode and JenkinsResource
* !2799 Changes to support multiple TDD patterns
* !3208 Downgrade gNB power limit LOG from "warning" to "debug"
Downgrade gNB power limit LOG from "warning" to "debug"
Downgrade the UE power limited LOG from LOG_W to LOG_D. This is to avoid
flooding stdout when UE in low coverage enters power limited state for a
longer period of time. The same information can be inferred from the
periodic UE print which also contains PH value (negative PH values
indicate that the UE is power limited)
Also, correct some typos in PH calculation.
Closes#885
Changes to support multiple TDD patterns
- Updated the configuration changes for 2 Patterns
- Update the TDD table configuration for NFAPI
- Added the tdd bitmap for the period
- Adapted the bitmap for UL/DL for the multi TDD pattern
- Updated the RACH procedure for multi TDD pattern
- Updated DL and UL scheduler for multi TDD pattern
Parametrized JenkinsNode and JenkinsResource
Parametrized variables in Jenkinsfile for automated tests on Colosseum
after update of OAI Jenkins server.
NAS 5GS refactor
The goal of this MR is to reorganize the NR NAS code into:
- enc/dec library functions (5GS, 5GMM, 5GSM, NR_NAS_defs.h,
nr_fgs_nas_lib.c)
- handlers and callbacks (nr_nas_msg.c)
The following changes are introduced:
1. migrate 5GS enc/dec lib (5GMM and 5GSM) to a dedicated folder
2. add a nested CMakeLists structure
3. refactor NR NAS enc/dec functions
4. refactor NR NAS messages struct definitions
5. remove most of dependency from NAS LTE
6. further refactor of PDU Session Establishment Accept processing to
separate enc/dec from handling
7. made naming of 5GS definitions consistent whenever necessary
8. cleanup unused code
9. remove redundant struct definitions
Also:
1. fix decoding of AMF Set ID in 5G-S-TMSI UE identity
2. replace Byte_t directly with uint8_t
- modify retrieve_ldpc_enc_op() to store encoded results in a packed format
- add timing measurements for T2 encoder
- add simde implementation for bit reversal
- Introduce new bit reversal implementation using simde_mm_gf2p8affine_epi64_epi8
instruction, based on https://wunkolo.github.io/post/2020/11/gf2p8affineqb-bit-reversal/.
This implementation requires GFNI and AVX512VL support on the target machine.
FHI72: fix for a distributed antenna array for xran F release
in the Rx callback I reset the number of section descriptions/number of
fragments to 0, but for each antenna up to max number of antennas per
RU, instead of up to max number of distributed antenna array
* Remove the ulsch_slot_bitmap and dlsch_slot_bitmap from nrmac
* Adopt the new is_dl_slot/is_ul_slot functions in the stack
to replace is_xlsch_in_slot: these functions are getting
the DL and UL slots as they are set in the TDD configuration
stored in tdd_slot_bitmap (frame_structure_t)
* Keep is_xlsch_in_slot in openair2/LAYER2/NR_MAC_gNB/gNB_scheduler_phytest.c:
this is necessary because the phy-test pre-processors are using the bitmaps
dlsch_slot_bitmap and ulsch_slot_bitmap to check which DL/UL slots to use
for transmission among those set in the TDD configuration
Co-authored-by: Guido Casati <guido.casati@firecell.io>
previously we had decodeRegistrationAccept acting as a handler and calling
decode_registration_accept to do the decoding. However the logic was mixed
between the two functions, the naming was confusing and the code was hard
understand. The goal of this commit is to simplify the code, improve
readability and maintainability, while adapting the relevant function
to the new OAI NAS libs structure.
To this purpose:
* move enc/dec logic to NAS lib openair3/NAS/NR_UE/5GS/5GMM/MSG/RegistrationAccept.c -
the logic of parse_allowed_nssai, get_allowed_nssai has been moved to
decode_registration_accept and decode_nssai_ie
* add decode_nssai_ie and use it for both Allowed/Configured NSSAIs list (IE handling was added)
* do processing in the handler handle_registration_accept
* simplify the code by removing nested calls
* improve handling of PDU length
Currently only Registration Result, 5GS Mobile Identity, Allowed/Configured NSSAI
are decoded. The other optional IEIs are skipped (only the length is processed).
Downgrade the UE power limited LOG from LOG_W to LOG_D. This is to avoid flooding
stdout when UE in low coverage enters power limited state for a longer period of
time. The same information can be inferred from the periodic UE print which also
contains PH value (negative PH values indicate that the UE is power limited)
- in the Rx callback I reset the number of section descriptions/number of fragments to 0,
but for each antenna up to max number of antennas per RU, instead of up to max number of a distributed antenna array
Ensure execution of processSlotTX in order in NR UE
Make sure that processSlotTX is executed in slot order.
There is a bug that causes decoding error when processSlotTX is executed
out of order. The root cause is not known at this point, but this change
avoids the bug altogether.
Add IQ file recording and IQ file viewer to ImScope
This changeset introduces the following changes to ImScope:
- Added the ability to record IQ data using the same mechanism as
ImScope.
- Added a new executable target imscope_iq_file_viewer for viewing IQ
files.
- Introduced a command line argument --imscope-record that enables data
recording mode.
- Refactored a lot of the scope code
Usage:
- To record IQ data, run a modem with --imscope-record flag. The
recording happens whenever a modem calls dumpScopeData, which
currently is done only in gNB on PUSCH DTX and PUSCH CRC NOK. There
is a limit of ~1GB of files written by the thread in a signle run to
avoid accidentally exhausting disk space. If a directory imscope-dump
is available the data will be recorded there in .imscope files.
Otherwise files are written to the directory the modem was run. This
is done so that the directory can be mapped in a docker container.
- To view recorded IQ files, use the new executable
imscope_iq_file_viewer. This can be done by running:
./imscope_iq_file_viewer <path_to_iq_file>
Replace <path_to_iq_file> with the path to the IQ file you want to view.
Dockerized development environment
This commit introduces a fully dockerized dev env you can start on any
machine that has docker without installing any direct OAI dependencies
on host.
This allows to migrate development environments between machines easily.
Several complementary scripts were added to enable smooth transition
between your docker environment and host environment. Your user and sudo
should work inside the environment.
In its current state, the environment can be used to compile
nr-softmodem with 7.2 FHI but it cannot run it due to missing
dependencies.
This commit introduces a fully dockerized dev env you can start on any
machine that has docker without installing any direct OAI dependencies
on host.
This allows to migrate development environments between machines easily. Several
complementary scripts were added to enable smooth transition between your docker
environment and host environment. Your user and sudo should work inside the
environment.
* the last 6 bits of the GUTI represent the AMF Pointer
* the AMF Set ID is 10 bits long
* the Registration Accept message received from the network contains the GUTI
which, when properly decoded, will be used in any possible following Registration Request
* use defines whenever possible
* introduce a switch to handle security type. Currently only integrity protected and ciphered
is handled, however the idea of the switch is to make the code ready for the future when
different security types will be handled and different actions should be taken based on that.
* define NAS_INTEGRITY_SIZE
* the goal of this commit is to refactor capture_pdu_session_establishment_accept_msg
into a decoding function
* PduSessionEstablishmentAccept.c is in the 5GSM library and shall only do encoding operations
* handling of PDU Session Establishment accept message is done in nr_nas_msg.c
* adopted new NAS decoding functions in the process
* improved QoS rules decoding
* convert getShort into a macro and add to NAS utils
as per legacy code, the NAS functions in nr_nas_msg.c were encoding
the header in the plain header contained in MM_msg
(e.g. plain -> mm_msg -> header in fgs_nas_message_t). The same
header was encoded in the mobility management message contained
in the same struct (e.g. plain -> mm_msg -> registration_request
-> protocoldiscriminator). As this was more complicated than
necessary, the goal of this commit is to simplify the code and
improve readability. Namely:
* Refactor MM_msg into fgmm_nas_message_plain_t to contain
(1) 5GMM Header
(2) 5GMM Payload union
* Use fgmm_nas_message_plain_t and fgmm_nas_msg_security_protected_t
in the 5GMM NAS procedures
* Remove NAS header IEs from 5GMM messages structs and use
the header in fgs_nas_message_plain_t.
This reduce also the repetitions of header members in each
of the 5GMM messages structs.
* Rename struct to distinguish between 5GMM and 5GSM
* Remove fgs_nas_message_t and nas_msg_header_t,
use directly basic 5GS struct definitions, simplify the code
* The callbacks in nr_nas_msg.c are now doing
(1) encode header with the common NAS lib
(2) encode contents with the 5GMM libs
* improve code readibility in the process
* replace encoding code with proper functions from the lib
* remove unnecessary header members in struct definition
this commit removes last enc/dec code in nr_nas_msg.c
* move enc/dec definitions to NR_NAS_defs.h and add new lib file nr_fgs_nas_lib.c
both used by nr_nas_msg.c
* move utility macros to fgs_nas_utils.h: this can be used in all the 5GMM, 5GSM lib files
* use nr_nas_msg.c for handlers/callbacks only, it should call the enc/dec
functions in nr_fgs_nas_lib and in the 5GMM, 5GSM libs
* 5GMM, 5GSM libs should only do the enc/dec of the NAS message contents
* move defines to openair3/NAS/NR_UE/nr_nas_msg.h as they are
only used in nr_nas_msg.c
* made naming of all message types consistent, using the same
prefix for all
* the only function used from nr_common.c is servingNetworkName which is
used in nr_nas_msg.c, so it was moved to the latter file and its scope
was reduced
Every NAS 5GS message can have 3 different headers: (1) security protected
(2) 5GMM plain (3) 5GS plain;
According to 3GPP TS 24.501:
> 4.2 Coordination between the protocols for 5GS mobility management and 5GS session management
> A 5GS session management (5GSM) message is piggybacked in specific 5GS mobility management (5GMM) transport
> messages. To this purpose, the 5GSM messages can be transmitted in an information element in the 5GMM transport
> messages.
A PDU Session Establishment accept is a 5GSM message and as such it is contained in a 5GMM
transport message. The code in capture_pdu_session_establishment_accept_msg is reflecting this.
The message is thus expected to have: (1) Security protected 5GMM header (2) Plain 5GMM message
(3) 5GSM Payload Container (4) Plain 5GSM message.
5.4.5 of 3GPP TS 24.501: also says:
> The type of the payload is identified by the Payload container type IE and includes one of the following: a) a single 5GSM message;
Thus, a 5GSM is not the only possible payload of a NAS transport message.
This commit is focusing on the refactoring of the structs definitions and adaptation of the CMakeLists.
Based on the above:
* Move NR_NAS_defs.h to 5GS subfolder in NR_UE
* Move NAS FGS structs and definitions to NR_NAS_defs.h
* Rename mm_msg_header_t to fgmm_msg_header_t
* Remove duplicated NAS struct definitions
(1) kept fgs_nas_message_security_header_t over security_protected_nas_5gs_msg_t
(2) Kept fgsm_msg_header_t over fgs_sm_nas_msg_header_t
* Remove unused definitions, structs, functions in the process
This commit is introducing the following macro changes:
(1) Migrate NAS 5GS lib files to a new 5GS subfolder in NR_UE
and organize NAS 5GS lib messages into 5GMM and 5GSM subfolders.
A nested CMake structure was implemented and CMakeLists.txt were
added whenever necessary. This is effectively separating 5GS
libraries from LTE libraries.
(2) In order to simplify the separation of 5GS libs from LTE libs,
remaining dependencies from NAS LTE (e.g. nas_ue_task.c, ue-idx-standalone)
were removed in the process
(3) ExtendedProtocolDiscriminator.c/.h and SpareHalfOctet.c/.h were deleted
and affected types were replaced with simple integer types
(4) Cleanup: unused definitions and libs (e.g. LIB_5GNAS_GNB)
were removed
(5) Update doc/Doxyfile
(6) clang-formatted the migrated files in the process
(7) Update includes of headers in affected files
This avoids the issue of compare_exchange_weak always failing. The issue
was that `store()` is not equal to `exchange()`. `exchange` uses memory
copy while `store()` copies the value representation of the argument, which
is not always the same thing.
- Added event parameter for recording which is reflected in the filename
- Added event in the UE for DLSCH NACK
- Added program name to the filename for ease of identification
- Reworked spinlock to use atomics
This commit introduces the following changes to ImScope:
- Added the ability to record IQ data using the same mechanism as ImScope.
- Added a new executable target `imscope_iq_file_viewer` for viewing IQ files.
- Introduced a command line argument `--imscope-record` that enables data recording mode.
- Refactored a lot of the scope code
Usage:
- To record IQ data, run a modem with `--imscope-record` flag.
The recording happens whenever a modem calls dumpScopeData, which currently is
done only in gNB on PUSCH DTX and PUSCH CRC NOK.
There is a limit of ~1GB of files written by the thread in a signle run to avoid
accidentally exhausting disk space.
If a directory `imscope-dump` is available the data will be recorded there in .imscope
files. Otherwise files are written to the directory the modem was run. This is done so that
the directory can be mapped in a docker container.
- To view recorded IQ files, use the new executable `imscope_iq_file_viewer`. This can be done by running:
```
./imscope_iq_file_viewer <path_to_iq_file>
```
Replace `<path_to_iq_file>` with the path to the IQ file you want to view.
* is_ul_slot: returns `true` for UL or mixed slots
* is_dl_slot: returns `true` for DL or mixed slots
* is_mixed_slot: returns `true` for mixed slots
These functions use the TDD slot bitmap for slot type checks
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* take into account pattern2 in nr_rrc_config_dl_tda
* use get_tdd_period_idx in nr_rrc_config_ul_tda
* use get_first_ul_slot_multi_tdd in set_SR_periodandoffset
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* refactor set_ideal_period to use frame_structure from MAC
* use get_ul_slot_offset to assign slot offset to nth UE and adopt in
updated config_csirs, configure_periodic_srs, set_csi_meas_periodicity
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* the function is returning the index of the first full UL slot in tdd_slot_bitmap
or the first UL slot (including mixed) if the "mixed" flag is provided
* the get_ul_slot_offset function calculates the offset of the nth uplink (UL) slot for a given UE index within a TDD period
* this function uses the frame structure bitmap to determine the UL slot positions
* allocate memory for pattern2 only if the IE is present (it is optional)
* fix scc (deallocate memory) when necessary
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* use config_frame_structure in config_common:
- fill frame_structure in MAC and NFAPI Config Request
- use get_tdd_period_idx to set periodicity index in Config Request
- use config_tdd_patterns to fill tdd_period_config
- use set_tdd_config_nr to complete NFAPI Config Request configuration
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* assuming num_dl_slots and num_ul_slots are including also
slots with dl/ul symbols respectively
* refactor set_tdd_config_nr
* moved function to MAC
* fill max_num_of_symbol_per_slot_list depending on slot type
* use new set_tdd_config_nr in stack
* minor refactor in config.c to improve readability (scs)
Co-authored-by: Guido Casati <guido.casati@firecell.io>
* set_tdd_bmap_period, config_tdd_patterns: to configure TDD period, including bitmap, in the frame structure
* get_tdd_periodicity: to retrieve the periodicity in milliseconds for the given TDD pattern
* get_tdd_period_idx: to get the TDD period index based on pattern periodicities
* get_ul_slots_per_period: get the number of UL slots (including mixed with UL symbols) in period
* get_full_ul_slots_per_period: get the number of full UL slots in period
* get_dl_slots_per_period: get the number of DL slots (including mixed with DL symbols) in period
* get_full_dl_slots_per_period: get the number of full DL slots in period
Co-authored-by: vijay chadachan <vijay.chadachan@firecell.io>
* allocate and configure NR_PUSCH_TimeDomainResourceAllocation in a separate function
* remove duplicated lines of code
* use in nr_rrc_config_ul_tda
* use calloc_or_fail
* clang-formatted
Integration: `2025.w04`
Closes#876, #896, #875, #893, and #878
See merge request oai/openairinterface5g!3226
* !3198 remove a set of compile options that don't work anyway
* !3171 Refactor SCTP Association Response Handling and E1 connection loss at CU-UP
* !3023 usrp constant and slow derive timestamp
* !3215 Remove unused code
* !3192 fix nr phy-test mode in ntn
* !3214 fix(nrLDPC_coding_t2): Miscellaneous fixes related to T2
* !3020 use ML receiver for 64 QAM UL-MIMO
* Demote error logs to debug for "PUCCH not processed"
* !3217 minor improvements in T hacks
* !3199 Add a small FAPI Hex parser utility
* !3209 Remove most of m64 type usage
* !3211 Avoid huge calloc by alloc'ing TBs independently
* !3212 F1AP IDs: Allow AddMod, fix concurrency bug
* !3152 E1 Setup enc/dec library and unit test
* !2999 Repair nFAPI in 5G
* !3210 FHI72: FDD support
* !3220 Improve RA logs
Improve RA logs
Modify existing LOG entries to include preamble index, timing advance and
estimated distance from the gNB. This makes mapping RA attempts to
physical/simulated UEs easier via sent/received preamble index or physical
distance from the gNB in scenarios with >1 UE.
FHI72: FDD support
- harmonize DL/UL PRB mapping in one get_xran_prb_map() function with support
of mixed slot and duplex mode
- do not assume TDD by default
Repair nFAPI in 5G
Add 5G nFAPI in OAI/Make it it work.
Many commits in this branch are basically bug fixes of things that don't work
properly, such as
- check for correct conditions (e.g. instance numbers in the RAN context)
- remove artificial limitation (e.g. only one PUCCH per TDD period in the PNF)
- improve performance (reduce big mallocs, make some static buffers such as a
global ring buffer for nFAPI messages in the PNF)
- fix bugs in nFAPI (e.g., increase maximum message size to ~64kB instead of
1500 bytes, because the latter is way too small for many TX_data.requests in
5G, and it will delay message arrival unduly)
- fix bugs in message enc/dec (e.g., handle FDD in config.request)
- adjust the L1 such that the condition "we run monolithic" is not necessery
(e.g., some message numbers in nFAPI struct where reset in MAC, instead of
L1, and this breaks when running the PNF)
There are instructions that explain how to set it up. Two CI tests have been
added, one with COTS UE and MIMO in DL+UL in TDD numerology 1 (30kHz), and a
second test with FDD mu=0 (15kHZ) in RFsimulator.
These were some old to-dos that have been addressed.
- test with COTS UE - 4x4 pipeline works good in DL, UL now also work
- test in RFsim
- make PNF status message
- make FDD in u0 work
This function packs P7 messages. At least RX_data.indication messages
might be much bigger than the global buffer that is used here. Allocate
the buffer on the stack, and make it bigger. Do not use the global
buffer, it's simply not necessary; increasing the global tx buffer size
might have negative effects elsewhere in the code.
That change might make the function reentrant. The mutex seems to only
guards the codec config. However, leave it to not introduce any
regressions as of now.
E1 Setup enc/dec library and unit test
This MR is adding a library for E1 Setup enc/dec functions:
- E1 CU-UP Setup Request
- E1 CU-UP Setup Response
- E1 Setup Failure
and relevant unit tests
F1AP IDs: Add update, fix concurrency bug
I suspect a concurrency bug: For instance, during reestablishment, the
CU needs to update the DU UE ID under which the UE is identified in the
DU. Previously, the CU would remove, then add the DU UE ID info. At the
same time, the PDCP thread might look up the information. This can lead
to asserts. Modify the code to do the update under (one) lock.
Also, refactor some code.
Similar to the parent commit, make the numerology in the VNF
configurable. Unlike for the PNF change, the VNF frame/slot info is in a
per-PNF connection structure to which the "oai_integration" code has no
access. So we need to hack the nfapi_vnf_p7_add_pnf() function signature
to take the numerology, to be able to save this on a per-PNF basis.
The fact that we store this on a per-PNF basis is not strictly required,
as the VNF cannot handle multiple numerologies as of now, either.
However, we might want to extend this in the future, so it seems prudent
to store this on a per-PNF basis, so that we could start handling
multiple numerologies at the L2 without the need to change the L1 or
nFAPI code itself.
Also, the numerology is not needed except for some code that deals with
timing over nFAPI. As of now, we don't use this code (nFAPI gets a
trigger every slot, as per the VNF request, see an earlier commit in
this series), but also here, let's not make the future more complicated
by storing the numerology for each PNF now (this could always be reverted).
For 4G, put numerology 0 (15kHZ SCS), which is the SCS that LTE uses.
Fill in the dynamically received numerology, and make the numerology at
the PNF configurable, following the changes in the parent commit.
The change for the reading is somewhat of a hack, because nFAPI, as of
now, does not really foresee to store the numerology, so we fill the
numerology during the start request, when the numerlogy has been
received in the config request, but at that time, the structure for the
time information (frame/slot) does not exist yet.
At both the PNF and the VNF, introduce a separate numerology field, and
pass the numerology when converting time information using macros from
nfapi_interface.h. The actually numerology is still hardcoded to 1 (as it
was before), but the follow-up commits will put a suitable numerology.
In both cases, but the numerology next to the frame/slot information.
Similar as the parent comment, remove time macros. In this particular
case, we can just pass the exact frame/slot information, instead of
squeezing it in a single integer.
The next commits will modify the time macros used in the 5G nFAPI code
to handle different numerologies. As a preparation, remove all the
usages of these macros where they really don't matter (e.g., don't
convert if we don't need to).
the L1 needs both a PDU (in TX_data.req) as well instructions for
the actual transmission (in DL_TTI.req) to encode a PDSCH message. In
nFAPI, it can happen that a DL_TTI.request message has been received
(configuring the PDSCH transmission), without the TX_data.request having
reached the PNF. The L1 assumes to have the PDU.
To avoid problems, ensure that only the pair of DL_TTI.req/TX_data.req
is delivered. Otherwise, drop either message.
nFAPI has a mechanism to segment messages that are too big for transport
over a given medium (see e.g. SCF 225, section 2.3.2).
The maximum segment size of 10000 makes that for larger payloads, e.g.
TX_data.request, many small segments are to be sent. This can create or
increase delays on the transport. On the other hand, the currently only
available transport mechanism, UDP, allows to transport packets of up to
almost 65535 bytes. Correspondingly, increase the maximum segment size
so that less segments are to be created, and potentially, less delay is
to be incurred. # Please enter the commit message for your changes.
Lines starting
Avoid delays in tx_data.request handling by avoiding big malloc()s and
copy operations. Reimplement function to (1) peek the frame/slot
numbers, (2) decide on buffer, and (3) unpack directly into
pre-allocated memory.
Co-authored-by: hsum <ming-hong.hsu@eurecom.fr>
Co-authored-by: chenyi <Yi-Quan.Chen@eurecom.fr>
Co-authored-by: Rúben Soares Silva <rsilva@allbesmart.pt>
The next commits refactor the rx handling of messages in the PNF. To
efficiently handle them, provide a "peek()" function that gives us the
frame/slot for which a message is destined. This allows to decide if a
message is on time and select a destination buffer, before unpacking it
completely, avoiding copies.
The time that a message between PNF/VNF to arrive depends mostly on the
transport. To allow for some delays, there is a slot_ahead time, during
which the VNF is allowed to schedule and send instructions to the PNF.
This can be multiple slots; the 1ms hitherto given might typically too
short. Increase to 10ms, to encompass a wider range of slot_ahead times.
Make the corresponding log message of when old (stale) message are
removed a bit clearer with respect to times.
This is the first commit in which nFAPI works. Follow-up commits improve
performance.
The usage of this flag in the L1 is simply wrong: either, the NFAPI mode
runs L1, or not. Correspondingly, add an assert to check this invariant.
In the NR_IF_module, the correct check to do is for VNF (implying no L1,
or "emulation of L1" [which is not true]), in which case we use
queues.
The L1 receives a pointer to the L1 payload to use for PDSCH generation.
This would overwrite an existing pre-allocated pointer. Remove this
preallocation. Instead, allocate it in the simulator (the only place
that actively uses this buffer allocation; in nr-softmodem, it's lost by
assigning a pointer). Allocate with an extra byte, because the payload
(in bits) might not be a multiple of 8 bits/1 byte, so the CRC is in the
last 3 bytes.
Also, improve the log with frame/slot information if the PDSCH payload
pointer is NULL.
The function now utilizes simpler variable names and logic to determine
if a given NR P7 request falls within the timing window.
The logic for determining if a request is within the timing window is as follows:
- The function calculates the absolute difference between the current
and received SFN slots, taking into account the possibility of
wraparound.
- If the absolute difference is greater than half of the maximum SFN
slot value, it subtracts this difference from the maximum SFN slot
value to get the actual difference.
- The function then checks if this difference is less than or equal to
the specified timing window. If it is, the request is considered to be
within the window.
Additionally, the commit updates the function signature to return a
boolean value for better readability and consistency.
Changes made:
- Simplified variable names for readability
- Improved logic for handling wraparound scenarios
- Updated function signature to return a boolean value
Co-authored-by: Rúben Soares Silva <rsilva@allbesmart.pt>
Since commit defc97ec16 ("Msg2 and Msg3 at gNB: check allocation before
committing to Msg2"), the scheduler calculate Msg2/Msg3 frame/slot
before allocation of Msg2, and not when receiving Msg1. It also respects
the RA response window correctly.
Correspondingly, we don't need to calculate here if the RACH.indication
is within the time window. The scheduler will do this.
If VNF and PNF are deployed through different processes, it's crucial to
maintain synchronization and ensure smooth operation. This commit
addresses the need to keep the VNF slot ahead of 2 (as required for
network deployment), while ensuring that PNFs wait appropriately to
avoid overtaking the slot. Specifically, this involves measuring slot
duration and implementing a wait mechanism to ensure that PNFs do not
exceed a slot duration of 300 microseconds. This ensures proper smooth
operation, especially in the case of RFsim.
Reset the counters. In monolithic, this happens in the scheduler; in the
PNF, without this change, it would trigger PDCCH again.
Co-authored-by: hsum <ming-hong.hsu@eurecom.fr>
Refactor pnf_phy_ul_dci_req(), pnf_phy_dl_tti_req(),
pnf_phy_tx_data_req(), pnf_phy_ul_tti_req(), to use L1 functions to
"load" L1 jobs to be executed, via L1 functions.
In monolithic, the sched_response mechanism ensures that FAPI messages
for one slot are not overwritten by scheduler instructions for the
following slots (i.e., it is like a memory allocation pool).
The allocation happens in nr_schedule_response(), not used by the PNF.
Hence, only deallocate when in monolithic.
The PNF uses a big ring buffer over all slots in a frame to ensure the
same.
This function applied an sf_ahead, bigger than the actual slot
indication message sending (so a message could NEVER arrive on time!).
Reduce to zero, because this is when it should arrive for us.
FAPI says these lists should be for MAX_TDD_PERIODICITY (which is maybe
a frame, or a TDD period? but likely not two frames). Reduce to one
frame.
Also, the nFAPI code assumes one entire frame, so use that (this commit
necessary to actually make the PNF work, otherwise, the
packing/unpacking of the FAPI message would need to do 2 frames).
Closes: #876
Upon reception of FAPI UL messages (e.g., RACH.indication), those
messages are put into queues, to be delivered in a single call,
NR_UL_indication(). Call the scheduler to trigger processing in UL.
Note that ideally, these queues should not exist at all, and messages
should just be delivered asynchronously, as this would lead to less
copying of messages. Currently, we fulfill the scheduler interface
instead.
Important! Trigger UL after having run DL scheduler and sent the
messages, to ensure short delay between Slot.indication, and the
response to those messages.
The previous designs seems to do:
loop {
poll_ind_queue()
if (msg)
scheduler;
pselect() on messages;
handle_message {
if slot_ind
put_ind_queue()
}
}
So basically, we artificially put a queue for slot indications in the
middle, but still handle it in the same thread(!). This for some reason
results in a big slow down if the PNF runs faster.
Now, we just do pselect(), waiting for messages. We handle the slot
indication immediately (the schedule should take some microseconds),
then return to pselect().
Use this function to implement distinct behavior for monolithic or PNF:
- monolithic: run the scheduler, as normal
- PNF: send a slot indication; the VNF will send an answer which the PNF
will handle elsewhere.
Add function declarations to make nr_ulsim compile, as NR_IF_Module.c is
part of the MAC sources. As of now, this is necessary; I think the right
way would be to take NR_IF_Module.c out of the MAC, and make this a
"glue" library that sticks L1 and L2 together, in monolithic (in VNF, in
that case, we should not link in L1, and for the PNF not the L2,
creating separate executables... Rome was not built in one day).
We are required to store multiple PUCCH (e.g., for multiple UEs).
However, nr_fill_pucch() automatically freed the previously active PUCCH
structure. Of course(!) this does not make sense.
However, the problem is that init_nr_transport() allocated only one
PUCCH structure. Also this does not make sense to me. The problem was
that slot_ahead was 0, which logically cannot happen in this function
(because slot_ahead is illogical, we must give the radio some time). The
reason is that it uses if_inst->sl_ahead, which, prior to this commit,
was not set.
The init_nr_transport() function is called normally from main() through
init_eNB_afterRU(). In this case, we initialize if_inst->sl_ahead to
what is set in main(). In the PNF case, however, we call
init_nr_transport() from the PNF thread, when receiving the start
request. In the main(), we wait for this to happen, before setting
if_inst->sl_ahead. Hence, initializing this variable (which we should
safely be able to do) before solves the problem.
The radio gives the clock. Therefore, we do not need to keep the time
separately in the PNF.
Also, the VNF should not run the ru_thread, as this is an L1 thread.
This is to align with monolithic/FAPI, where we get slot indications in
each slot. In the nFAPI, this is replaced by timing information, such
that the VNF ticks automatically, and gets indications from the PNF on
early/late messages. The timing information handling is not implemented
in the present nFAPI implementation. Instead, we adopt the FAPI behavior
and send timing information in every slot, abusing it to be a slot
indication.
There is no need to add the PNF in nr_start_resp_cb(), because it has
already been done in nr_param_resp_cb(). This might be wrong, because 4G
does it only in nr_start_resp_cb(), but for the moment, since it is also
used in nr_param_resp_cb(), avoid the double add.
The same function is called in nr_start_resp_cb(), why there and not here?
At reception of nr_param_response, we receive the remote PNF information
on PNF/VNF. We have to store it.
In nr_start_response, it might not be correct and/or can be taken out.
The next commit will remove it.
Remove most of m64 type usage
MMX instructions were first SIMD version, the registers management is complex
(_mm_empty()) so, it is better to remove it now and use more recent instructions
anyway, gcc/clang should replace it automatically as long as we enable sse2,
that should always be the case but for code understanding and for ARM porting
for example, it is better to explicitly remove it
Add a small FAPI Hex parser utility
Adds a small FAPI hex parser utility nfapi_hex_parser, that takes in a FAPI
message hexdump, either from the command line or a file, unpacks it and prints
the contents out to the console.
The functions to print the data to the console were added alongside the existing
utility functions, for both P5 and P7 messages.
This allows the user to quickly check if a message has the valid format to be
unpacked in OAI.
The file examples.md contains some examples of execution and output.
This utility is compiled alongside the NFAPI_LIB, as such, it is compiled
whenever we compile nr-softmodem, or any other target that uses the NFAPI_LIB.
Prior to this this commit, the structure NR_UE_info_t is 4881632 bytes.
On some machines, e.g., one of the CI systems, this can lead to realtime
problems when doing the initial calloc() of the data structure.
Commit e586efb29d ("Enable 32 HARQ
Processes in DL, UL") introduced the real-time problems by increasing
the number of HARQ processes.
The reason NR_UE_info_t is that big is that it contains buffers for DL
HARQ data that might need to be retransmitted (the L1 is stateless, it
cannot store this for the L2). To reduce the size, dynamically allocate
the DL HARQ buffers when they are needed. This also reduces the size of
NR_UE_info_t to 15840 bytes.
Encapsulate transportBlock in byte_array and accessory functions, and
allocate in big chunks of multiple of ~32kB.
Reported-By: Cedric Roux <cedric.roux@eurecom.fr>
Fixes: e586efb29dCloses: #875
ran_func_kpm.c uses cp_ba_to_str(), but it is not defined in OAI. Define
it so we can link properly, instead of relying on FlexRIC (where it was
defined).
Introduce function cu_update_f1_ue_data() which does remove+add under
one lock. This fixes a concurrency bug. For instance, during
reestablishment, the CU needs to update the DU UE ID under which the UE
is identified in the DU. Previously, the CU would remove, then add the
DU UE ID info. At the same time, the PDCP thread might look up the
information. This can lead to asserts. Use the above addmod
functionality to do the remove-add under a mutex.
Modify test to verify this works as expected.
use ML receiver for 64 QAM UL-MIMO
Change condition on MMSE/ML selection to include 64QAM for ML UL-MIMO receiver.
Only 256QAM will now use MMSE UL-MIMO receiver.
fix(nrLDPC_coding_t2): Miscellaneous fixes related to T2
1. HARQ combined offset:
The offset provided for each code block in the HARQ combined input and output
buffer shall not be above 512. A modulus was performed on the offset of the TB
in the buffer to avoid this situation. But the offset of the code block was
added afterward so that the total offset was sometimes going above 512.This
commit performs the modulus after adding the code block offset. Usually this
issue was triggering a segfault after less than 2 minutes of an iperf test on
some setup using T2. The same iperf test was successfully performed during 10
minutes on the same setup with this change.
2. Properly stop LDPC coding in gNB and UE:
LDPC coding was yet not properly stopped by using free_nrLDPC_coding_interface
in gNB and UE. This was not causing major issue until now. It was especially not
causing any issue with T2 LDPC coding using the PF. But using T2 VFs, if the
coding library is not properly stopped then the VF cannot be used anymore unless
the admin app is restarted. So this MR adds the missing free_nrLDPC_coding_interface.
fix nr phy-test mode in ntn
Following SIB19 MR !3019 (merged), the phy-test mode is no longer working
In this MR we update the handling NTN-Config from ServingCellConfigCommon
to be the same as when NTN-Config is received in SIB19
More details: !3019 (comment 145103)
usrp constant and slow derive timestamp
Sample write reorder function doesn't handle properly the clock shift.
It keeps the same time origin forever, so, if the clock sifts in one direction,
we finish to reach the maximum time shift that was designed for slot to slot
maximum shift.
We may also consider if this function is useful: gNB and UE currently ensure Tx
in order because there are other issues above if we let Tx preparation run out
of order in separate threads
Closes#893
- Refactor SCTP Association Response Handling
- Unified function for both CU-CP and CU-UP handling
- CU-CP: Notifies RRC of E1 connection loss with CU-UP
- CU-UP: Initiates a new SCTP association request to regain E1 connection
- this fix will prevent issue #878 from happening
- Send list of PDU Session Resources to Failed Setup in the (PDU Session
Resource/Initial Context) Setup Response
- depending on different failure scenarios
- e.g. (#878) when No CU-UP is associated and still the CU-CP receives a PDU
Session Setup Request
- add relevant functions to handle NGAP Setup Response messages with failed
list of items
- this is an extra failure handling to make #878 is handled if it ever happens
again
- NGAP Cause Refactoring
- Refactor struct for NGAP cause group and value
- Add extended radio network cause enumeration values
- Refactor redundant code into a new NGAP cause encoding function
- Minor logging, function descriptions improvements
Closes#878
Useful to debug realtime issues, for example when calling a periodical
routine, if it's not programmed carefully enough, it may be that the first
call takes way longer than the next ones (for example it uses some memory
that is not mapped in the process, leading to many page faults the first
time it's executed).
Having a periodic log for the timing of this routine with min/avg/max/count
will reveal that a call is way longer than average and will help in
finding where exactly in the code the time is spent, thanks to the ease
of use of the T tracer; you put T(T_XX, T_INT(1)) just before the code
you want to measure, you put T(T_XX, T_INT(0)) just after and time_meas
will show the delta time between those events. Then you cut the routine
in smaller pieces, measure with T(1)/T(0), and you finally reach the small
guilty part that takes too long.
In time_meas.c, when getting 0 for the event, we log delta time with
previous 1 for the event. "start_valid" was not cleared, so receiving
a second 0 without 1 in between may lead to wrong logging.
Should not happen in practice, but still, little bug.
(Note: we don't protect against streams of 1, but this case should also
not happen.)
Add utility functions to print the FAPI messages contents
Add dump functions for DL_TTI.request PDUs, remaining message structures to be finished
Add file examples.md with some examples on how to use nfapi_hex_parser
Now uses RC.nb_nr_L1_inst instead of RC.nb_nr_inst in order to configure the L1, since in PNF RC.nb_nr_inst is always 0, and so, it wouldn't configure the L1
chore(ci): Move to new Jenkins and Openshift Cluster
- Update helm-charts with new node roles
- Add resource definition in physim helm-charts to easily schedule them
on different cluster servers
- Update entitlement creation methodology
- Add namespace for aerial pipeline
Due to HW version differences between the CI RHEL build machine and
the FHI 7.2 test machine (cacofonix), FHI 7.2 test using RHEL image
failed, reporting an 'Illegal Instruction' error.
* this commit is making explicit the mandatory IE cn_support
in the E1 Setup Request message
* also, init cn_support in E1 Setup Request initial message preparation
* E1 CU-CP Setup Request is an E1 Interface Management message
whose type defition was missing, and it is introduced in this commit
(see 9.2.1.7 of 3GPP TS 38.463)
LDPC coding was yet not properly stopped by using free_nrLDPC_coding_interface in gNB and UE.
This was not causing major issue until now. It was especially not causing any issue with T2 LDPC coding using the PF.
But using T2 VFs, if the coding library is not properly stopped then the VF cannot be used anymore unless the admin app is restarted.
So this MR adds the missing free_nrLDPC_coding_interface.
The offset provided for each code block in the HARQ combined input and output buffer shall not be above 512.
A modulus was performed on the offset of the TB in the buffer to avoid this situation.
But the offset of the code block was added afterward so that the total offset was sometimes going above 512.
This commit performs the modulus after adding the code block offset.
- extend get_xran_prb_map() function to handle mixed slot
- properly fill PRB configuration based on DL/UL direction and duplex mode
- do not implicitly assume TDD
Modify xu_add_f1_ue_data() to be xu_addmod_f1_ue_data(), i.e., "addmod"
behavior in which it is possible to also update f1_ue_data. I do it this
way since an xu_updade_f1_ue_data() function would be almost the same as
the add function (only deleting existing data, whereas the original
add() was asserting in that case).
Modify tests to test for this behavior.
This fixes a concurrency bug. For instance, during reestablishment, the
CU needs to update the DU UE ID under which the UE is identified in the
DU. Previously, the CU would remove, then add the DU UE ID info.
At the same time, the PDCP thread might look up the information. This
can lead to asserts. Use the above addmod functionality to do the
remove-add under a mutex.
- the parameters used in DL/UL are exactly the same, except the direction and the number of PRBs;
therefore, passing the direction seems convenient for visibility purposes and reduces the code.
Note: Number of PRBs we store in fh_config for both DL/UL is num_rb_dl of openair0_config struct.
At the end, it is actually the same info.
Following SIB19 MR !3019, the phy-test mode is no longer working
In this MR we update the handling NTN-Config from ServingCellConfigCommon
to be the same as when NTN-Config is received in SIB19
Integration: `2025.w03`
See merge request oai/openairinterface5g!3206
* !3167 remove calls to clear SIMD registers, that was a historical issue
* !3179 UE statistics improvements
* !3193 Remove spinlock from threadpool task return
* !3195 UE MAC: in phy-test/do-ra, fill TB randomly
* !3201 T tracer: minor fixes for compilation
* !3204 CI: Increase number of test trials for timing tests
* !3034 Delay compensation for PUCCH2
* !3205 Fix calculation of RSSI in pucch_rx.c
Fix calculation of RSSI in pucch_rx.c
Fix two issues with RSSI calculation in PUCCH:
- RSSI was calculated using partial sum instead of average of squared samples.
- There was a missing parentheses pair which causes the RSSI value to underflow.
The proper way to calculate power in dBFS is
let signal_energy = average of squared samples of received signal over resource elements, antennas and symbols under measurement
then:
rssi[dBFS] = -10log10(max_sample^2 / signal_energy),
or (to avoid division):
rssi[dBFS] = -(10log10(max_sample^2) - 10log10(signal_energy))
CI: Increase number of test trials for timing tests
- Increasing a number of test trials aims to reduce variations and
improve reliability of the nr_ulsim and nr_dlsim test results.
- Disable CPU sleep states on the machine when running physims.
- Minor adjustment of timing threshold in 60 MHz phytest.
Delay compensation for PUCCH2
- Implementation of delay compensation for PUCCH2, which minimizes for
exampe logs "reported RSRP index XXX invalid"
- Adding a test to the CI to perform the delay compensation in PUCCH2
- Refactor code to use c16_t data type in PUCCH decoder
Remove spinlock from threadpool task return
The idea to use atomic aligned to cache line for counting finished jobs
is reused, but instead of polling a semaphore was added.
remove calls to clear SIMD registers, that was a historical issue
This works because gcc/clang replaces the MMX calls by SSE2 automatically.
There are some mm_empty() calls in bad places (places where we don't use
_m64) and it is lacking in places where we still use _m64. Given that
MMX calls are converted to SSE2, there is no problem, and we will remove
remaining _m64 variables.
Fix two issues with RSSI calculation in PUCCH:
- RSSI was calculated using partial sum instead of average of squared samples.
- There was a missing parentheses pair which causes the RSSI value to underflow.
The proper way to calculate power in dBFS is
let signal_energy = average of squared samples of received signal over resource
elements, antennas and symbols under measurement
then:
rssi[dBFS] = -10log10(max_sample^2 / signal_energy),
or (to avoid division):
rssi[dBFS] = -(10log10(max_sample^2) - 10log10(signal_energy))
* this function is used by both CU-CP and CU-UP
* it behaves differently depending on the type of E1 instance:
CU-CP: inform RRC of E1 connection loss with CU-UP
CU-UP: trigger a new SCTP association request by sending an ITTI to the CU-UP task
* after receiving the E1AP register request, the CU-UP will send a new association request to
SCTP until new E1 connection is achieved by sending a TIMER_HAS_EXPIRED message to the CU-UP
task. The function e1apHandleTimer() is basically triggering a new E1 association request
until new E1 connection is achieved
* print SCTP association state in e1ap.c whenever needed
Related to #878
* add function to check whether cu-up is connected
* add functions to send the relevant response messages with the failed list
e.g. to fill PDU Session Resource Failed to Setup items
* set return value of trigger_bearer_setup to bool and handle the
failure by sending setup response messages with list of PDU sessions failed to setup
i.e. send failed list of items when no CU-UP is associated to the CU-CP
note: it considers all PDU Sessions as failed to setup in the considered failure scenarios
* e.g. send a list of failed PDU session resources in the response message to reject
PDU Session Resource Setup in rrc_gNB_process_NGAP_PDUSESSION_SETUP_REQ
* the Initial Context Setup has a Failure message, but for the considered failure scenario
when no CU-UP is available the current design is just sending a list of failed to setup
items in the Response message
* transfer PDU Session Resource/Initial Context Setup Request NAS PDU to UE
closes#878
* new struct definition including both NGAP cause group and value
* add encoding function, temporarily in ngap_common.c
(in the long term to be place in a enc/dec NGAP library)
* extend Radio Network cause enum values
* add cause member to Setup Failure
Integration: `2025.w02`
Closes#866 and #887
See merge request oai/openairinterface5g!3197
* !3135 replace a set of #define by a C bit array, remove duplicated extern global variable declaration
* !3173 nrLDPC_coding: improvements
* !3181 Use UL actor for processSlotTx
* !3186 chore(ci): improved errors messages visibility
* !3101 NR gNB improvements for analog beam management
* !3130 NR gNB MSG3 TPC
* !3175 Improvements for PUSCH per symbol processing
* !3177 CI: change frequency for timing and B200-SABOX pipelines
* !3188 NR UE fix DCI mask size
* !3119 Config files improvements for PRACH
* !3196 Fix yaml example config file
* !3187 nrLDPC_coding: Fix naming, comments, superfluous variables and documentation typos in coding library interfaces
* !3076 move UL_INFO structure allocation local to a rx slot processing
* !3189 fhi_72: allow to provide dpdk_iova_mode to xran
* !3182 RSSI threshold for PUSCH & PUCCH power control
* !3103 Introduce ITTI queue in RRC-to-MAC direction
RSSI threshold for PUSCH & PUCCH power control
To prevent ADC railing or "clipping" when the received power is too
large, introduce RSSI threshold parameters for PUSCH and PUCCH, which
are now configured in units of 0.1 dB. Here is how you can configure and
use these thresholds:
1. Configuration:
- Set the pusch_RSSI_Threshold and pucch_RSSI_Threshold parameters in
your configuration file (e.g., gnb.conf) (MACRLC section).
- These parameters define the maximum acceptable RSSI values for
PUSCH and PUCCH, respectively. The unit is either dBm or dBFS,
depending on the RSSI reporting configuration.
2. Usage:
- The nr_limit_tpc function will automatically adjust the Transmit
Power Control (TPC) commands based on the configured RSSI
thresholds.
- If the RSSI exceeds the threshold, the TPC command will be limited
to prevent further increase in power, thereby avoiding ADC
clipping.
Introduce ITTI queue in RRC-to-MAC direction
resolve ue rcc to mac race conditions by a itti queue, but it doesn't
resolve the problem of exact slot we should apply the new cell
configurtion
fhi_72: allow to provide dpdk_iova_mode to xran
- Introduce a gNB configuration option that enables users to select the
desired IOVA mode for DPDK ("PA" or "VA") with dpdk_iova_mode
parameters in fhi_72 configuration section. If not set, the default
remains PA mode to ensure backward compatibility with existing
configurations. Detailed information about IOVA modes, including their
advantages and disadvantages, can be found in the DPDK documentation:
Memory in DPDK Part 2: Deep Dive into IOVA
- Using IOVA VA mode allows DPDK to run without requiring SYS_ADMIN
capability for the docker container.
- Switch to IOVA VA mode in the FHI7.2 pipeline and reduce the required
capabilities for the docker container.
Move UL_INFO structure allocation local to a rx slot processing
UL_INFO is allocated a gNB, so require mutex and obscur information
passing function to function. Instead, make a local UL_INFO per slot,
removing the mutex and making the code easier to understand.
Use UL_INFO on the stack, to simplify the code. move in the same struct
the associated rach, pdu, crc, srs arrays that where pointed from
directly in gNB structure Keep in nfapi the existing logic that make
malloc for theses rach, pdu, crc, srs array lists
Pass only relevant fields, where appropriate.
Co-authored-by: Robert Schmidt <robert.schmidt@openairinterface.org>
nrLDPC_coding: Fix naming, comments, superfluous variables and documentation typos in coding library interfaces
This MR is clarifying the interfaces to the LDPC coding libraries. It
aims to improve the readability of the interfaces and compliance of
naming with specification 38.212.
Fix yaml example config file
amf_ip_address is decoded as a paramlist_def_t and not as a paramdef_t,
so in the config file it should be a list of ip addresses
In current develop just by trying to run this config file with:
sudo ./nr-softmodem -O ../../../targets/PROJECTS/GENERIC-NR-5GC/CONF/gnb.sa.band78.fr1.106PRB.usrpb210.yaml
results in a segfault.
Config files improvements for PRACH
Some PRACH parameters in configuration files are not properly set.
- number of SSB per RACH occasion should be 1 when we have only 1 SSB
(it would still work with 1/2 but it makes little sense and might
limit UE selection of ROs).
- some PRACH configuration indices would generate PRACH conflicting with
PUCCH and therefore reducing scheduling capabilities
Further cleanup of some config files still using deprecated fields
Switching to IOVA VA mode in the FHI7.2 pipeline reduces the required
capabilities for the Docker container. This change allows the removal
of the SYS_ADMIN capability.
In DPDK, IO Virtual Addresses (IOVA) refers to the memory addressing
scheme used for IO operations. The two IOVA modes supported by DPDK are:
- IOVA as Physical Address (PA): IO memory addresses correspond directly
to physical addresses, no address translation required.
- IOVA as Virtual Address (VA): IO memory addresses align with user-space
virtual addresses, requiring IOMMU to remap physical memory.
Previously, this was hardcoded to 0, to use IOVA as PA mode, with xRAN's
DPDK EAL initialization using the "--iova-mode=pa" flag.
This commit introduces a gNB configuration option to allow users to select
the desired IOVA mode. The default remains PA mode to
ensure backward compatibility.
Using IOVA VA mode allows DPDK to run without requiring SYS_ADMIN
capability in the docker container.
NR UE fix DCI mask size
Previous maximum size was 15 but in practice it could be at least up to
18. Also moved assertion on non-contiguous type0 DLSCH allocation to L1
(to remove blocking point for L2 implementation).
CI: change frequency for timing and B200-SABOX pipelines
Change center frequency of
- RAN-SA-B200-Module-SABOX-Container - 20 MHz TDD SA scenario from
3309.48 MHz to 3609.12 MHz
- RAN-gNB-N300-Timing-Phytest-LDPC - Timing phytest 40 MHz scenario from
3619.08 MHz to 3319.68 Mhz
to avoid interference between the SABOX and timing pipeline.
Improvements for PUSCH per symbol processing
- Improved the assert in task_ans.c
- Fixed and issue with PUSCH per-symbol processing that could cause an
issue in case num_pusch_symbols_per_thread is not 1.
NR gNB MSG3 TPC
This MR improves the handling of MSG3 TPC commands at MAC gNB.
TODO: Compute TPC when PRACH sends a valid power measurements (currently
not the case for OAI L1, to be verified with Aerial L1)
Closes#866
NR gNB improvements for analog beam management
- Improvements at gNB MAC layer to properly handle analog beam
management for all channels
- Simple beam switching mechanism based on UE RSRP report
- Introduced new configuration parameters `pusch_RSSI_Threshold` and `pucch_RSSI_Threshold`
(MACRLC section)
- Added `nr_limit_tpc` function in `gNB_scheduler_primitives.c` to limit TPC based on RSSI
threshold.
- Integrated `nr_limit_tpc` in PUCCH and PUSCH power control
Example Configuration
puxch_RSSI_Threshold takes value between -1280 and 0 and is mapped to range -128.0 dBm/dBFS
to 0.0 dBm/dBFS in 0.1 dB increments. Unit depends on RSSI reporting config.
pucch_RSSI_Threshold = -500; # Set the PUSCH RSSI threshold to -50.0dBm/-50dBFS
pucch_RSSI_Threshold = -600; # Set the PUCCH RSSI threshold to -60.0dBm/-60dBFS
chore(ci): improved errors messages visibility
1. There was a trailing && before $BUILD_OPTIONS in the command line
2. Added leading lines and more pattern to the grep command that
displays errors/warnings messages during docker build since we don't
have access to all.txt if build fails
nrLDPC_coding: improvements
One improvement in this changeset: Functions nr_interleaving_ldpc,
nr_deinterleaving_ldpc, nr_rate_matching_ldpc and
nr_rate_matching_ldpc_rx that were previously linked with the PHY layer
in the executables are now linked with the LDPC coding implementations
where they are called.
The parent commit cleans up and groups oai_exit in one place,
nr-softmodem-common.h. Therefore, this 5G file is included in the 4G
defs_common.h (at least it's mostly 4G), which is not nice. Instead,
define oai_exit also in 4G softmodem-common.h, and include this in
defs_common.h. Remove the cyclic inclusion of defs_common.h in
softmodem-common.h, and resolve all subsequent errors around unknown
types and extern definition mismatches.
Commit d641a76196 ("fix BSR report malformed, add SHORT BSR when it
can") removed code to fill the remainder of a TB, if it is not filled
with data from RLC, with random data when using phy-test or do-ra modes.
This random data filling is intentional, as it might prove useful for
demos or when debugging (e.g., to see used spectrum instead of 0, etc.).
Reintroduce the functionality, but unlike the original code, make it
dependent on the mode, not RFsim.
See also discussion in #888.
Fixup: d641a76196
* size of fgs_nas_message_security_header_t is platform-depending
while protocol_discriminator must be 1 byte
* fgs_protocol_discriminator_t stays as enum type
Reduce race conditions through a itti queue, such that RRC-initiated MAC
memory modifications happen at the same time as MAC thread writes.
However, it doesn't resolve the problem of exact slot we should apply
the new cell configurtion.
This assert checks if the structure is correctly initialized. The structure
is one-time use so the assert is valid, but the value of the atomic was added
to the output to improve debugging.
Integration: `2024.w51`
Closes#879
See merge request oai/openairinterface5g!3180
* !3155 UL BLER vs SNR plot
* !3170 Replace AssertFatal with static_assert for cmdline arguments check
* !3172 A script to run CI tests locally.
* !3151 Optimize PHY_ofdm_mod CYCLIC_PREFIX in case of incidentally aligned pointers
* !3164 Fix and refactor channel average
* !3154 Fix TPMI for UL retransmissions
Fix TPMI for UL retransmissions
If the ul_ri of srs_feedback changes between a UL transmission and a
retransmission, we use the TPMI of transmission.
Closes#879
Assertion ((*nrOfLayers==1 && srs_feedback->tpmi <= 1) || (*nrOfLayers==2 && srs_feedback->tpmi == 0)) failed!
In compute_precoding_information() /home/user/openairinterface5g/openair2/LAYER2/NR_MAC_COMMON/nr_mac_common.c:2896
TPMI 1 is invalid!
Fix and refactor channel average
- Fixing missing init average variable with zero and refactor the code #86b11ab7
Fixes just the missing avg1=0
- Refactor the code to a common SSE/AVX2 inline function using also horizontal add.
- We use int64_t result from now on.
- We decided to fix length to x12 instead of old x9 openair1/PHY/NR_UE_TRANSPORT/dci_nr.c
//compute average channel_level on each (TX,RX) antenna pair
void nr_pdcch_channel_level(int32_t rx_size,
c16_t dl_ch_estimates_ext[][rx_size],
NR_DL_FRAME_PARMS *frame_parms,
int32_t *avg,
int symbol,
int nb_rb)
{
for (int aarx = 0; aarx < frame_parms->nb_antennas_rx; aarx++) {
//clear average level
simde__m128i avg128P = simde_mm_setzero_si128();
simde__m128i *dl_ch128 = (simde__m128i *)&dl_ch_estimates_ext[aarx][symbol * nb_rb * 12];
for (int rb = 0; rb < (nb_rb * 12) >> 2; rb++) { <-------------- here
avg128P = simde_mm_add_epi32(avg128P, simde_mm_madd_epi16(dl_ch128[rb], dl_ch128[rb]));
}
DevAssert(nb_rb);
avg[aarx] = 0;
for (int i = 0; i < 4; i++)
avg[aarx] += ((int32_t *)&avg128P)[i] / (nb_rb * 12); <-------------- and here
LOG_DDD("Channel level : %d\n",avg[aarx]);
}
}
Optimize PHY_ofdm_mod CYCLIC_PREFIX in case of incidentally aligned pointers
It seems that the cyclic prefix is in most cases a multiple of 512 samples.
This means that in most cases the idft output pointer is already aligned and
there is no need to perform an extra memcpy. This saves the memcpy time in
most cyclic prefix insertion cases.
A script to run CI tests locally.
Added ./run_locally.sh script. The script takes one argument being the testcase
that one wants to run. The testcase is run locally. This means that some
testcases that utilize hardware resources will not run. However most if not all
rfsimulator testcases should run as in CI.
To this end, the CI python scripting framework was updated:
- override and <srv_node> elements from XML so all commands are executed locally
in case --local flag is passed
- Allow RemoteCmd class to run on localhost
Replace AssertFatal with static_assert for cmdline arguments check
Replace AssertFatal with static_assert for command line arguments definition
length check. The static assert will not allow the code to compile in case
the size is incorrect.
This could save some time in the future since its easy to forget to extend
CMDLINE_PARAMS_CHECK_DESC.
Added ./run_locally.sh script. The script takes one argument being the testcase
that one wants to run. The testcase is run locally. This means that some testcases that
utilize hardware resources will not run. However most if not all rfsimulator testcases
should run as in CI.
To this end, the CI python scripting framework was updated, adding --local flag which changes
the script behavior as follows:
- overrides <node> and <srv_node> XML elements to 'localhost' so all commands are executed locally
- Avoid running image pull, image cleanup and workspace creation steps of the scripts:
user is responsible for that
Replace AssertFatal with static_assert for command line arguments definition
length check. The static assert will not allow the code to compile in case
the size is incorrect.
This could save some time in the future since its easy to forget to extend
CMDLINE_PARAMS_CHECK_DESC.
Alignment of temporary IDFT buffer was changed to 64 bytes. This is in preparation
for possible use of different IDFT implementation which could use AVX-512, requiring
64-byte alignment.
Integration: `2024.w50`
Closes#753
See merge request oai/openairinterface5g!3166
* !3159 CI: Add FHI7.2 test with Metanoia RU
* !3111 Enable USRP Tx thread in nrUE
* !3144 UE data race fix
* !3149 Fix several issues in ulsch_mmse_2layers
* !3158 GTP: use direct call instead of ITTI
* !3160 FHI-72: initialize mask, find correct libxran
* !3019 SIB19 utilization on UE
* !3162 Add Rocky 9.5 to list of distributions
* !3163 NGAP-Initial Context Setup Failure
* !2952 New LDPC coding library interface for decoding and encoding complete slots in one call to the library
* !3125 Add support for O1 in O-DU via telnet
* !3161 Cmd line option for sync in frames for iqrecorder
Cmd line option for sync in frames for iqrecorder
B200 with high clock drift will go out of sync within 280ms which was
the default duration to sync in with iqrecorder. Now this value can be
specified via cmd line so the user can change it according to needs.
Add support for O1 in O-DU via telnet
This change set adds a simple telnet interface to nr-softmodem. Together
with oai/o1-adapter>, this implements a part of the O1 interface, for
the DU only. Currently supported uses cases:
- some stats are send to an SMO (such as bandwidth, PLMN, frequency, ...)
- re-configuration of bandwidth, frequency
- stop/start of the softmodem, while keeping the CU and L2 running
- the o1-adapter implements two alarms, load overload and base station
offline, using this interface.
It is possible to use the telnet interface by itself (i.e., "o1" without
the o1-adapter). To do this, compile with telnet support (./build_oai
--build-lib telnetsrv) and start the nr-softmodem with O1 module
(./nr-softmodem --telnetsrv --telnetsrv.shrmod o1). You can then do the
following:
- see some statistics as exposed to o1-adapter, in JSON: echo o1 stats | nc 127.0.0.1 9090 && echo
- stop the L1: echo o1 stop_modem | nc 127.0.0.1 9090 && echo
- reconfigure bandwidth: e.g., echo o1 bwconfig 20 | nc 127.0.0.1 9090 && echo (also supported: 40, 60, 100)
- start the L1: echo o1 start_modem | nc 127.0.0.1 9090 && echo
Note that the reconfiguration works with split-8 radios. Notably, it
does not work yet out-of-the-box with 7.2 radios, because they have to
be manually reconfigured on bandwith/frequency switch. The planned
M-plane should allow such use case, though.
To help with the above, this branch also includes fixes (for more
information, refer to the corresponding commits):
- fixes for stopping the L1, notably stop threads in right order, stop
thread pool of RU
- collect some general MAC metrics, fix some bugs in MAC
- fix problems in F1AP gNB-DU configuration update message, update test
- fix some problems in RRC
The consumer of this function just passes a pointer to something to be
sent. This function should not free this memory, it's simply not it's
task. This is the same bug as above in
"CU_handle_gNB_DU_CONFIGURATION_UPDATE(): don't free the message of RRC"
This adds a basic telnet module to support the O1 interface at the DU,
to be used together with the oai/o1-adapter>. Concretely, this first
version supports to
- read/modify cell parameters
- stop/start the L1
Implements the re-start of the L1 after it has been stopped through
stop_L1(). It takes into account some changes, such as bandwidth, or
frequency, when restarting the L1. To this end, it also triggers a
gNB-DU configuration update message on each restart to inform the CU
about any changes in the DU. FDD has not been tested.
Refactor L1 stopping code into a function. The function is basically the
same as what we were doing at the end of main() of nr-softmodem.c.
This will be reused to support the "soft-restart" functionality in the
next commits.
Need to proper forward declaration, as we otherwise need to make
telnet_o1 module dependent on F1. Make functions publicly accessible,
as the F1 functions are used in an upcoming commit to start the L1.
Need to keep track of number of RUs, as the RU initialization increments
the number of RUs. Therefore, subsequent initializations would fail if
we don't decrement first. There is no better place (or I did not find
it) to decrement the number of RUs than in nr_phy_free_RU(). For this
reason, we also have to manually set the number of RUs in nr_prachsim to
1, otherwise the assert will trigger.
Split out these functions, to be reused in a follow-up commit to
correctly have gNB-DU configuration update message working.
Change the return type copy_f1ap_served_cell_info(), to return the copy,
which IMO is more intuitive then passing target and source in
parentheses.
For the "restart" feature, earlier versions where creating multiple UL
TDA. This is not a problem (anymore), but to avoid that we call this
multiple times, add this assert.
The existing code prevents double initialization if e.g., both gNB&UE
call these functions (like in a simulator). However, in the case of a
double-initialization such as stopping and re-starting the L1, the
memory will be freed, but not freed again. Basically, this works:
generate()
generate() # not initialized again
but this leads to problems:
generate()
free()
generate() # not initialized again, but should be!
Reimplement the check based on the data to be initialized, such that an
intermediate free() will be handled properly.
These are the "main" worker threads (apart the ru_thread). If the
thread-pool stops, they might get stuck. E.g., in PUSCH channel
estimation, it uses a local response queue, and if the thread pool is
terminated, it seems that the L1_rx_thread can get stuck in there.
To limit these problems, attempt to stop these threads first; stop the
queue from which they read and make them terminate; then, stop the
thread pool.
Also, fix the termination of respPuschSymb queue.
A later commit implements the "restart" of the L1. The L1 reads
configuration options. First, allow to re-read configurations by
increasing the amount of memory the config module might allocate. Then,
avoid an out-of-bound write by checking that we still have memory
available (and assert if not possible).
New LDPC coding library interface for decoding and encoding complete slots in one call to the library
Previous attempts to integrate LDPC decoding relying on offloading show
the limitation of the original code segment coding interface for the
library implementing LDPC coding. In order to perform the offload and
therefore the decoding in an optimal fashion, it is relevant to provide
the implementation with the broader amount of data possible per call,
which is a slot. A previous MR related to T2 offload managed to offer
an interface for decoding transport blocks by modifying only
nr_ulsch_decoding.c. But this may not be sufficient for performing an
efficient decoding in any situation. Slots could be shared between
multiple transport blocks in case multiple UEs are connected to the RAN,
which makes the transport block size shrink and therefore degrades the
efficiency of the offloading.
This MR offers a broader interface with a gNB decoding entry point in
phy_procedures_nr_gNB.c. The file nr_ulsch_decoding.c now lets the
decoding library implement deinterleaving, rate-unmatching and decoding
for the slot. In gNB encoding and UE decoding and encoding as well,
tasks are partitioned in a similar fashion.
As it allows to feed the implementation with the broadest amount of data
possible, it allows to fit any coding solution in a library. This will
avoid further modification to OAI source code and addition of options to
select an implementation when integrating new decoding solutions. An
implementation of this interface will therefore be offered for every
existing decoding solution in OAI and solutions added in the future will
be provided as independent module code without any modification to the
core of OAI code. The way to choose the decoding solution and to
configure it was also unified using configmodule. The library can be
chosen through the option --loader.ldpc.shlibversion. The slot coding
solutions shipped with this MR can be configured through configmodule
(e.g., T2 address and isolated cores).
This MR includes three coding implementations:
- A wrapper around segment decoding libraries for test purpose: libldpc.so
- A one shot slot coding and encoding with the T2 board: libldpc_t2.so
- A decoder on FPGA using the XDMA driver: libldpc_xdma.so
Improvements brought by this MR:
- Easy integration of a wider variety of LDPC coding solution at gNB and UE
- Easy selection of the LDPC coding solution (option --ldpc-offload-enable is kept for now but it can be ultimately removed)
- T2:
* More efficient when many UEs are connected
* Decoding and encoding in nr-uesoftmodem
NGAP-Initial Context Setup Failure
NGAP Failure message - Initial Context Setup Failure. When an Initial
Context Setup Request is received with the UE context as null, it
triggers Initial Context Setup failure
SIB19 utilization on UE
Following SIB19 initialization and scheduling on gNB (!2899 - merged)
and Receive SIB19 on nrUE (!2920 - merged), this merge request aims to
introduce the actual use of the SIB19 provided parameters by gNB in UE.
Changes needed for utilization include: initialization of position
coordinates parameters for UE (which are needed to calculate propagation
delay between UE and SAT), propagation delay calculations and timing
advance adjusts. Moreover, this change removes the now unused arguments
for ntn_koffset, ntn_ta_common and ntn_ta_commondrift
GTP: use direct call instead of ITTI
Using ITTI calls for user plane implies (1) a memory allocation, (2)
mutexes, and (3) queueing messages for each user plane packet, which is
heavy. Use a direct API call instead to reduce overhead.
It seems that the cyclic prefix is in most cases a multiple of 512 samples.
This means that in most cases the idft output pointer is already aligned and
there is no need to perform an extra memcpy. This saves the memcpy time in
most cyclic prefix insertion cases.
More complex implementations will be achieved using the new reworked interface
- t_nrLDPC_dec_params: remove F, Qm, rv and perCB
- encoder_implemparams_t: remove Qm, Tbslbrm, G, rv and perCB
- Remove all variables (like OPENAIR1), they serve no purpose and make
unnecessarily long lines
- Remove linking to T, it does not seem to be necessary
- Avoid intermediate variables to keep sources, and declare libraries
directly, making for shorter code
NGAP Failure message - Initial Context Setup Failure.
When an Initial Context Setup Request is received with the UE context as null, it triggers Initial Context Setup failure.
Fix several issues in ulsch_mmse_2layers
Fix incorrect array type of determ_fin which was set to int32_t while the
array holds the results of simde_mm_abs_epi32 which produces unsigned
values.
Remove assert on right-shifting negative integers as it is impossible for
determ_fin to hold negative values.
Change type of sum_det to uint32_t to prevent possible overflow. Fix
the type cast (int -> uint32_t) to prevent casting before shift in
nr_ulsch_comp_muli_sum which might have resulted in right shifting
negative integers.
Closes: #753
UE data race fix
When a new sync request is sent to PHY, we clear harq buffer and start
synchronization process. This could lead to unexpected behavior or segv
because DL actors might be using the buffer we just cleared. To avoid
this, the DL actors are shutdown and re initiated before clearing harq
info.
Some users copy libxran.so to a system directory (e.g., /usr/local/lib).
At the same time, we require xran_LOCATION to be set to find the build
directory. Use HINTS to prefer the location passed through
xran_LOCATION.
Also, use NO_DEFAULT_PATH to outright disallow global scope, to prevent
"accidentally" finding libxran.so in global scope (this is for
multi-user systems that are used in development, and where people have
different version of libxran for testing purposes). To still use
packages in e.g., /usr/local/, it would be possible to set xran_LOCATION
to this directory. (there is a problem that the version information will
likely not be read correctly, so this might not be enough)
B200 with high clock drift will go out of sync within 280ms which was
the default duration to sync in with iqrecorder. Now this value can be
specified via cmd line so the user can change it according to needs.
Set the number of barriers to 512 and implement proper slot and frame
based selection. Now the dynamic barriers should not be overwritten in
case the rx-to-tx offset is high. The maximum offset supported is 512.
- Removed ntn_koffset, ntn_ta_common, ntn_ta_commondrift arguments in order to be used directly from SIB19
- Removed global NTN_UE_Koffset variable and it's use
- Introduced getter functions for ntn related parameters
Using ITTI calls for user plane implies (1) a memory allocation, (2)
mutexes, and (3) queueing messages for each user plane packet, which is
heavy. Use a direct API call instead to reduce overhead.
- Added new structure to hold timing advance parameters
- Made SIB19 related calculation to save to this struct
- Modified in ra to get ta struct values
Move handling of sync request from MAC from UE_dl_preprocessing(). This
will process the sync request before receiving the samples of current
slot from radio so we don't 'lose' the slot.
Integration: `2024.w49`
See merge request oai/openairinterface5g!3148
* !3065 NAS Service Request enc/dec implementation
* !3129 Refactor UL MAC PDU decoding
* !3136 Use ninja -k option in CI
* !3137 T tracer: add macpdu traces for the NR UE
* !2392 Documentation: add multi-UE deployment with RFsimulator
* !3145 a lib has been added but missed dependancies
* !3050 doc: Add instructions for 5G NR gNB frequency configuration
* !3141 CI: Allow restart of the container if deployment fails
* !3138 Align YAML config getlist with libconfig implementation
* !3140 remove un-necessary global variables
* !3146 Updates to match more fapi-compliant tx_data.request
* !3025 New threadpool, fixed in NR UE
* CI: Remove USRP mgmt_addr from LTE-2x2 conf files
* !3147 Fronthaul xran library update to F release (2. and final step towards M-plane integration)
* !3142 Add CI test to trigger RLF with high pathloss in channelmod
Upon restoration to ploss, set to 20; when setting to 0, the UE is not
able to sync to the cell anymore.
Co-authored-by: Robert Schmidt <robert.schmidt@openairinterface.org>
Fronthaul xran library update to F release (2. and final step towards M-plane integration)
- xran library support -> both E and F releases
- RU delay profile removed -> not used in neither of the releases
- DU delay profile updated
- one-way delay measurements support is improved/fixed in the F release.
However, not implemented here as it depends on the RU capabilities
The "mgmt_addr" parameter is not required for USRP N310 initialization. The
critical parameter for successful operation is the "addr", which specifies
the IP address of the SFP port used for data transfer.
Including both "mgmt_addr" and "addr" can lead to initialization failures if
one of the specified addresses is unavailable. Since the "mgmt_addr" is
unnecessary for this context, it has been removed to avoid potential conflicts
and simplify the configuration.
Fix incorrect array type of determ_fin which was set to int32_t while the
array holds the results of simde_mm_abs_epi32 which produces unsigned
values.
Remove assert on right-shifting negative integers as it is impossible for
determ_fin to hold negative values.
Change type of sum_det to uint32_t to prevent possible overflow.
Fix the type cast (int -> uint32_t) to prevent casting before shift in
nr_ulsch_comp_muli_sum which might have resulted in right shifting negative
integers.
Updates to match more fapi-compliant tx_data.request
The packing of PDUs as required in aerial for the Tx_data.Request wasn't
complaint with FAPI. That was fixed in Aerial 24-2. This change makes L2
pack the payloads in the way that Aerial now requires.
remove un-necessary global variables
Remove global variables that can be removed by a trivial way:
- simply dead globals
- or reference is a single source, so can be static
setting them static reduces the scope, that is a progress, but of course as
for all static variables, the code is still not re-entrant
Major improvements:
1) bbu_offload - enable packet handling on BBU cores
2) dlCpProcBurst - DL CP processing on one or more symbols
3) SRS CP processing - depends on the RU capabilities that can be
retreived via M-plane (if RU is CAT B)
4) nSecDesc - the number of section descriptors per symbol; it was hardcoded to 0 in E release;
depends on the RU capabilities that can be retreived via M-plane
5) RunSlotPrbMapBySymbolEnable - enable PRB mapping by symbol with multisection;
depends on the RU capabilities that can be retreived via M-plane
6) DSS (Dynamic Spectrum Sharing) support - depends on the RU capabilities that can be retreived via M-plane
Minor fixes in F that exist in E:
1) prach_config->nPrachFilterIdx - explained in commit ID f3045396
2) T1a_min_cp_dl - explained in the commit under title "Clarification on DU delay profile"
3) fh_config->nULRBs - explained in commit under title "Further clarification on xran E release parameters"
Removed const for prb_conf argument in oran_allocate_cplane_buffers()
function, as F release requires the not const struct for
xran_init_PrbMap_by_symbol_from_cfg() and xran_init_PrbMap_from_cfg()
even though the struct is not modified inside of these functions.
- remove xran parameter io_cfg->port logging => this parameter
is filled within xran library in both E and F releases
- PRACH offset explanation; not limited to >= max(Ntx, Nrx);
workaround done in xran to support PRACH eAxC IDs same as PUSCH eAxC IDs
- DL PRB not used in xran; UL PRB used in xran F release, therefore
removed from the patch
- these fixes should have been done in the commit ID f3045396
Align YAML config getlist with libconfig implementation
This aligns the behavior of params_yaml with params_config library:
- when a mapping is read using getlist it returns the number of elements in
the mapping instead of 0
- when a mapping is read using get it returns the number of parameters used
for input
CI: Allow restart of the container if deployment fails
In CI we sometimes encounter fail of the gNB/UE deployment (and fail of the test
scenario) caused by unsuccessful initialization of the USRP N310.
This MR enables to optionally restart the gNB or UE container, if the deployment
fails on the health check during start up. By default, restarts of the container
are not allowed, but we can enable them by setting num_attemps > 1 for a given
deployment in the XML file.
Logs from failed deployment attempts are collected.
This MR aims to avoid known CI failure caused by "USRP N310 Initialization Failure"
mentioned in #871.
Documentation: add multi-UE deployment with RFsimulator
Extended doc/NR_SA_Tutorial_OAI_nrUE.md with section about how to run multiple UEs
- multiple UEs with namespaces (provided new script)
- multiple UEs with Docker UEs
- updated rfsimulator and channel simulation tutorials
- updated README.md
This change makes it so the fapi_nr_p7_message_pack is used to pack TX_DATA.request for Aerial, reducing the amount of repeated code.
Add the capability to pack TLVs with TAG=2 (offset from first address)
Packs the TLV.length differently according to whether or not Aerial is in use ( for Aerial TLV.length is 16-bit, for OAI is 32-bit )
In CI we sometimes encouter fail of the gNB/UE deployment (and fail of
the pipeline) caused by unsuccsessful initialization of the USRP N310.
Restart the container if the health check fails during initialization.
Introduce a configurable parameter (from XML) to set max number of
attempts for the container deployment.
Store logs from failed deployment attempts.
Rework GetContainerHealth() to work with docker compose and services,
instead of using docker and containers. Return health and message about
deployment status (used for console and HTML logging) for a given service.
T tracer: add macpdu traces for the NR UE
MIB, SIB1, random access and regular scheduling are traced.
At the beginning of a connection (rrc setup request and rrc setup), the
RNTI is reported as 0. Might be fixed, somehow, if absolutely needed (not
sure).
Refactor UL MAC PDU decoding
Refactor the UL MAC PDU decoding: A single function is created to decode the
entire MAC subheader
uint8_t decode_ul_mac_sub_pdu_header(uint8_t *pduP, uint8_t *lcid, uint16_t *length).
It returns the length of the MAC subheader and outputs the LCID and Length.
This new function is called from 2 functions: lcid_crnti_lookahead() and
nr_process_mac_pdu().
After calling it in nr_process_mac_pdu(), the validation of the PDU can be
made for all the MAC subPDUs in a single place, allowing for less and more
clean code (as opposed with the previous version where the validation was
made multiple times inside the LCID switch-case).
Additionally, this MR updates the list of LCIDs as in 3GPP TS 38.321 v18.3.0
Table 6.2.1-2: Values of LCID for UL-SCH when the LX field is not present or
is set to 0.
- removed some whitespace noise.
- cleaned up CMakeLists.txt
- reintegrated task_manager.c into thread-pool.c
- cleaned up some unnecessary code
Co-authored-by: Cedric Roux <cedric.roux@eurecom.fr>
Co-authored-by: Mikel Irazabal <mikel.irazabal@openairinterface.com>
Integration: `2024.w48`
See merge request oai/openairinterface5g!3134
* !2951 Integrate Tracy with NR UE
* !2985 E1AP Bearer Context Setup encoding/decoding library and unit tests
* !3115 T2: minor fix for 4x4 processing
* !3117 F1AP lib: minor fixes
* !3126 NR MSG3 scheduling improvements
* !3128 NR UE RRC T304 in phy-test mode
* !3132 Run the last job of gNB pusch channel estimation inline
* !3133 NR UL HARQ handling improvements
* !3091 Update NR UE threading model
* !3030 Imscope updates
* !3131 FAPI P7 Refactoring and unitary test creation
* !3033 Increase the number of UE supported at gNB by having more than 1 PUCCH2 frequency occasion per slot
MIB, SIB1, random access and regular scheduling are traced.
At the beginning of a connection (rrc setup request and rrc setup), the
RNTI is reported as 0. Might be fixed, somehow, if absolutely needed (not
sure).
This aligns the behavior of params_yaml with params_config library so that
when a mapping is read using getlist it returns the number of elements in
the mapping instead of 0.
FAPI P7 Refactoring and unitary test creation
This MR is a follow-up to the work done in !2714 this time separating
the P7 FAPI messages into its own library (nr_fapi_p7).
In the same way, utility functions to copy, free and compare have been
added for each message, as well as an addition function to get the
allocated size for a message, which is used for the messages that can be
too big to fit in a buffer that's allocated in the stack.
Unitary tests have been added for each of the messages, testing
pack/unpack and the utility functions added.
Some users of the OAI CI complain that since we build with -Werror, it
is difficult to see all warnings/errors and fix them, necessitating
multiple attempts at times to see all warnings. To allow them to see
more warnings, use ninja option -k10 to keep going after errors, and
show more errors at once.
There are command line options that one might pass to ninja/make, but
for which cmake does not provide an interface. For instance, make/ninja
support -k/-kN to "keep going" after errors. Cmake does not provide an
option, but it is still possible to pass this. Make a build_oai option
to allow this, used in the next commit.
Imscope updates
- Using imgui docking branch - allows window docking.
- Moved scopes to different windows to allow docking.
- Disabled scatterplot for time domain samples - issues with amount of
vertices caused segfaults.
- Disabled IQ heatmaps for RX IQ samples on gNB and UE - this was
incorrect and currently I don't know how to fix it. This will be
reenabled once it is.
- added a simple menu and disabled demo windows by default.
Update NR UE threading model
This updates the NR UE threading model by preventing calling threadpool
from threadpool and fixes deadlock issues related to processSlotTX. By
making processSlotTX run to completion instead of starting and waiting
for other threads to finish it prevents locking thread pool cores,
therefore prevening deadlocks.
There is a slight difference here compared to the previous version where
I do not use the UL actor as I believe it is not necessary,
processSlotTX satisfies the run-to-completion requirement now. If in the
future any UL procedure would be parallelized using the thread pool it
should either be done in a run-to-completion model or the processSlotTX
function should be taken off the thread pool.
The documentation has been updated.
- Using imgui docking branch - allows window docking.
- Moved scopes to different windows to allow docking.
- Disabled scatterplot for time domain samples - issues with amount of vertices caused segfaults.
- Disabled IQ heatmaps for RX IQ samples on gNB and UE - this was incorrect and currently I don't know
how to fix it. This will be reenabled once it is.
- added a simple menu and disabled demo windows by default.
4G/5G FAPI code is located in the same file. Since 5G code calls 5G
specific functions, these functions need to be defined in 4G.
The solution would be to separate the FAPI changes into 4G and 5G.
Add nr_fapi_p7 library to NFAPI_USER_LIB for access to utility functions
Change message handlers to use the utility functions to copy and properly free the messages
Change the SRS handler to call the handler for the measurements instead of trying to copy to RC.gNB[0], which does not exist
Change nfapi_nr_p7_message_unpack to not allocate unneeded memory
Changed parameter in unpack function to be inline with other unpack functions.
Changed nfapi_p7_allocate call to calloc to avoid dependency to NFAPI_LIB.
Added unitary test for RACH.indication( test pack/unpack, free, copy and compare )
Added constant NFAPI_NR_RACH_IND_MAX_PDU to follow constants related to other P7 messages limiting the maximum number of PDUs to use.
Removed now unneeded dependency to NFAPI_LIB and NFAPI_USER_LIB
Changed parameter in unpack_nr_srs_indication to void * to be inline with other unpack procedures.
Added memory allocation for pdu_list inside unpack function.
Add unitary test for SRS.indication( test pack/unpack, free, copy and compare )
Added rands16 to provide random int16_t values, needed for SRS.indication unitary test
Added constant NFAPI_NR_SRS_IND_MAX_PDU to follow constants related to other P7 messages limiting the maximum number of PDUs to use.
This adds the equality, free and copy functions for SRS.indication message as described in the Table 3–129 of SCF222.10.04, it does not add the functions for the report type structures, as these are always sent/received inside the uint32_t array.
The utility functions for these will be added at a later commit, and subsequently tested in the upcoming unitary test for SRS.indication.
Extra utility function to calculate allocated memory added.
Apply clang-format to pack/unpack procedures.
Add comments to improve readability of iqSize usage.
Fix unpacking return value to not return 0 on success.
Renames nfapi_nr_srs_reported_symbol_t prgs into *reported_symbol_list for two reasons:
So that the name is more representative of what the parameter represents.
For it to actually be a list, and not assume only 1 reported symbol is used.
Allocate reported_symbol_list in unpack_nr_srs_beamforming_report.
Updated all references to prior prgs to point to the first position of reported_symbol_list, still only effectively using the first reported_symbol.
Rename fill_srs_reported_symbol_list into fill_srs_reported_symbol and update input parameter name, since it only fills the information related to one reported symbol
Reordered PDU pack/unpack function to follow PDU value order
Changed nfapi_p7_allocate calls to calloc to avoid dependency from FAPI lib to nFAPI lib.
Add unitary test for UCI.indication( test pack/unpack, free, copy and compare )
Fix PUCCH Format2/3/4 value to be indexed at 0, not 2.
This is according to Table 3-66 of SCF222.10.02
Add utility functions related to UCI.indication
Extra utility function to calculate allocated memory added.
Changed parameter of `unpack_nr_crc_indication` to void* to stay inline with other unpack procedures.
Changed nfapi_p7_allocate call to calloc to avoid dependency from FAPI lib to nFAPI lib.
Add unitary test for CRC.indication( test pack/unpack, free, copy and compare )
Now allocates memory for `uint8_t* cb_crc_status` before pulling in unpack procedure.
Add utility functions related to CRC.indication.
Extra utility function to calculate allocated memory added.
Changed parameter of unpack_nr_rx_data_indication to void* to be inline with other unpack procedures.
Changed nfapi_p7_allocate calls to calloc to avoid dependency from FAPI lib to nFAPI lib.
Add unitary test for RX_DATA.indication( test pack/unpack, free, copy and compare )
Changes pack/unpack procedure to expect pdu_length of 32 bits instead of 16.
This change was first added for compatibility with Aerial L1.
Add utility functions related to RX_DATA.indication
Extra utility function to calculate allocated memory added.
Allocate memory for TX_DATA.request TLV value pointer on tag == 1
In the past, this wasn't noticed because tag == 1 is not used
Add utility functions related to TX_DATA.request
Add extra function to calculate allocated memory
Create separate function to pack/unpack rx_beamforming PDU
Reorder PDU handling order to follow from 0 to 3
Changed PUSCH pdu_bitmap switch to separate if to handle multiple optional sub-pdus
Fix bit_len_harq size to 16 bit
Add utility functions related to UL_TTI.request
Add pack/unpack functionality for parameters
added in previous commit
Add utility functions related to DL_TTI.request,
creating 2 news files to contain these functions
Reorder switch to follow PDU Type order
This makes it so the functions are available to both FAPI and nFAPI libraries, and to more easily move them later to the FAPI library, while being available for nFAPI
Update NR P5 and P7 nFAPI headers to have 32-bit message_length
In the NR functions, this causes the header parameters following message_length to have their position shifted by 2 bytes, due to the increase in message_length size.
Change needed to accommodate SCF222.10.02 Table 3-3
Separates nfapi_p4_p5_message_header_t and nfapi_nr_p4_p5_message_header_t
Change the callbacks signature to accept/return void* intead of expecting a particular header type, being the NR/LTE distinction done inside the callback functions.
Create NR related functions for nFAPI transmit timestamp and checksum calculation.
Move nfapi_p7_codec_config_t and nfapi_p4_p5_codec_config_t to separate header to be shared between LTE and NR VNF/PNF
Move nfapi_nr_timing_info_t into NR header.
Move NR specific function declarations to NR header.
Fix Assert condition to properly check that body was packed successfully
Fix retval value when performing bitwise AND with push16
Remove commented code
Remove dummy_subframe variable, which is initialized once in
start_request() and then copied to memory on the heap. It is not needed.
In nr_start_request(), it should not be necessary, but we keep it for
the moment to not make more changes than necessary.
This avoids a segfault for me, but I am not exactly sure why.
The previous code fills random number in the TDD config, which leads to
unrealistic TDD patterns. Filling wtih a specific, realistic TDD config
allowed to fix the bug in the next commit.
Testing showed that some messages could be at least slightly bigger than
the previous maximum size of 8192. Given that the maximum transport
block could be larger, increase this value to 32768.
unpack_dl_tti_ssb_pdu_rel15_value() reads two additional bytes,
seemingly for RX power information. However, table 3-79 does not
contain such info, so I remove to match the corresponding pack function
pack_dl_tti_ssb_pdu_rel15_value().
TODO: tx power info relevant?
Since tag 2024.w45/commit f54ca5714e,
standalone mode/--sa is the default and does not exist. Correspondingly,
it has been removed from the documentation.
For users reading recent documentation, but running an older version of
OAI, there might be a mismatch in that --sa is necessary to run OAI, but
not given in the documentation. Provide explanatory text to make users
aware of this.
Run the last job of gNB pusch channel estimation inline
In case of 1 antenna this prevents the usage of threadpool which
eliminates the thread pool overhead. In case of >1 antenna the
threadpool contention should be reduced providing smaller benefits.
NR UE RRC T304 in phy-test mode
RRC timer T304 is stopped upon completion of the associated RA handover
procedure. There is no RA procedure in phy-test mode so T304 should not
be started to avoid unwanted expiration.
NR MSG3 scheduling improvements
- fix wrong condition on MSG3 allocation assertion message
- remove assertion altogether and handle properly the case where MSG3
can't be allocated
T2: minor fix for 4x4 processing
- correct max number of LDPC segments for T2 processing
- use pointer arithmetics in retrieve_ldpc_enc_op() in T2 lib to
optimize loop processing
Implement run-to-completion for processSlotTX. This is achieved by using a new
dynamic_barrier_t which allows to specify callback after the first join has been
made.
This commit introduces dynamic_barrier_t. Its a thread barrier that counts joins
but allows callback to be specified later. See std::barrier for a basic barrier
description.
In case of 1 antenna this prevents the usage of threadpool which eliminates
the thread pool overhead.
In case of >1 antenna the threadpool contention should be reduced providing
smaller benefits.
* tac is optional IE, and it was assumed as always present
* is now allocated in DU container by copy_f1ap_served_cell_info
when present in the updated cell info and missing in the
DU container
Integration: `2024.w47`
See merge request oai/openairinterface5g!3120
* !2972 Parallelize PUSCH channel estimation
* !3105 Fix a small bug with total_res in gNB PUSCH symbol processing
* !2996 Ue monitoring stats
* !3122 Avoid assert on non-implemented SM write callback
* !3110 Fronthaul 7.2 improvements (1. step towards M-plane integration)
* !2966 Refactor of includes in gNB/UE for improved header management
* !3123 Reduce double-usage of CI XML files
* !3029 NR UE RRC reconfiguration improvements
* !2670 chore(ci): adding Flexric-RAN integration test
* !3124 CI: use OC CN in aerial pipeline
chore(ci): adding Flexric-RAN integration test
- Triggering a standalone FLEXRIC + 5G-RAN pipeline: OAI-FLEXRIC-RAN-Integration-Test
- will be triggered by the RAN-Container-Parent
- but also by the OAI-FLEXRIC pipeline (to be done once integrated)
- Added an new python option: --FlexRicTag= than if not used, won't have any effect
- Nice addition:
- if --FlexRicTag=develop or --ranCommitID=develop --> will use the latest commit
on either dev or develop branch.
- Scenario-wise
- Using the cn5g deployment file from Luis in documentation, at least we will use
it to make sure it is working at all time
- We can add more test and especially checkers on the flexric calls
Linked to !3085 (merged)
Occasionally, the CN deployment takes longer than 60 seconds to complete
successfully. As a result, the CI reports failure, even when all other test
stages pass (e.g., gNB attaches to the CN, UE connects, and iperf tests succeed).
Replace current docker deployment of OAI CN by deployment on OC. With change
of the CN, we also need to change the routing for UE module - change the
mbim script for controling the UE.
Reduce double-usage of CI XML files
Some pipelines use the same XML files. For instance,
RAN-NSA-B200-Module-LTEBOX-Container and RAN-LTE-{TDD,FDD}-LTEBOX-Container
use both fr1_epc_{closure,start}.xml, but use the core network on different
hosts, by means of different pipeline configuration (EPCtype, etc).
A future MR will remove these additional pipeline configuration, and store
all the CN config in the repository. To simplify the transition, duplicate
some XMLs (i.e., remove duplicate usage of them).
After this MR, the only "double usage" of XML files for core (un-)deployment
should be through pipelines RAN-LTE-{TDD,FDD}-LTEBOX-Container.
Refactor of includes in gNB/UE for improved header management
This merge request implements changes suggested by the Include What You Use
(IWYU) open source tool to streamline the project’s header file usage.
The adjustments ensure that each file includes only the necessary headers for
the symbols it uses, making the code more efficient and maintainable.
Unnecessary includes have been removed, and missing headers have been added
where required. This improves modularity and reduces inter-dependencies between
files, helping to avoid future build issues caused by unused or missing includes.
Each output has been manually reviewed.
Followed a similar approach as in !2953 (merged).
Affected folders/files:
- NR_MAC_gNB/main.c
- nr-softmodem.c
- unitary simulators: nr_dlsim, nr_ulsim, prachsim
- openair2/RRC/NR
- openair3/NGAP
- openair2/LAYER2/nr_pdcp
- openair2/SDAP/nr_sdap
Fronthaul 7.2 improvements (1. step towards M-plane integration)
- added explanations for each parameter used for XRAN library
- added new config files for VVDN, Benetel 550 and Metanoia RUs
- changed config file for Benetel 650
- deleted 8x8 650-550, and added 8x8 650-650 config file
- updated FHI README
- setting one or more VFs; compatibility with Mellanox NIC as well
- fhi_72 section
Avoid assert on non-implemented SM write callback
The asserts might be used to intentionally stop the running process.
Further, it might be an annoyance if it happen unintentionally. Just
print an error message instead.
If we are pulling FlexRIC, overwrite the tag to use with what has been
provided on the command line. This is because we hardcode the tag schema
"branchname-commitId[0:8]" everywhere else, but this does not apply for
FlexRIC, and we have to overwrite this here.
- pass parameter directly instead of indirectly via class
- harmonize to svr_id to read through common code in the next commits
- pass images to express which images will be deleted
- refactor function to use "with" for cls_cmd, and return success (or
not)
- reuse cls_containerize.Containerize.PullImage()
- pass parameter directly instead of indirectly via class
- always read node name and use this instead of "sometimes" looking it
up from a class member. With "sometimes" I mean that if no node name
is given, it will take it from self.eNBIPAddress, which is confusing
at best if we can give the node name directly
- consistently put the node name in the xml file
- pass parameter directly instead of indirectly via class
- call it simply "images", to harmonize with clean-function later
- harmonize to svr_id as in the previous commit
- rewrite function to use with-statement with cls_cmd
- optionally pass a tag to use (which will be used in a test, and can be
used to pull flexric)
- Introduce Pull_Image() which does the actual pulling, can pull from
any registry, and will be reused in the case of pulling from cluster.
* ideally the result of these macros shall be relevant
only in unit tests, in order to validate enc/dec NAS functions
* use a local print macro
* remove dependency from LOG module
* simplified the code
* cleanup TLVDecoder.c (kept it for LTE dependencies)
* FGMM enc/dec library files
* Plain message encoding + logic for security protected message
* added fill_fgstmsi function
* add checks on length in 5G-S-TMSI encoder function
* encoding of NAS Service Request and relevant mandatory IEs
Ue monitoring stats
- one commit to add UE mac stats similar to gNB one and remove a partial
implementation at phy level of some counters
- second commit only rename the two UE persisted HARQ tyepdef names for DL
and for UL (one was saying "UE" but not the direction and the second the
direction but not "UE")
The asserts might be used to intentionally stop the running process.
Further, it might be an annoyance if it happen unintentionally. Just
print an error message instead, and return to FlexRIC that nothing is to
be sent.
The CI pipeline RAN-NSA-B200-Module-LTEBOX-Container uses the same core
network (ltebox) as the pipelines RAN-LTE-{TDD,FDD}-LTEBOX-Container,
triggered through the same XML files. However, the core networks are on
different hosts, configured through (Jenkins) pipeline configuration. A
future MR will move the entire core network configuration into the XML
files [1]. To facilitate this, create separate XML files for the NSA
pipeline.
[1] The advantage is that it will be easier to reason with only the XML
files what is going to be run where and how.
Fix a small bug with total_res in gNB PUSCH symbol processing
Replaced total_res with res_per_task when deciding whether to start a threadpool
task or not. This does not have big impact on the software as the
nr_pusch_symbol_processing will recheck ul_valid_re_per_slot anyway but it may
reduce the number of thread pool tasks created.
- identification of each parameter in fh_init and fh_config structs
- identification of unused parameters in xran library
- identification of parameters used for CAT A O-RU
=> CAT B not supported in this release
- pass parameter directly instead of indirectly via class
- harmonize to svr_id to harmonize parameter with parameters in the next
commit (and towards a future harmonization of directly taking node
names instead of "IDs")
The queue for finished jobs of PUSCH channel estimation is only used inside
nr_pusch_channel_estimation(). Since initialization is fast, it can be
declared on the stack in the function where it is used.
This commit enables the optional use of Tracy for profiling NR UE & gNB.
See https://github.com/wolfpld/tracy/releases/tag/v0.11.1 for details.
How to compile:
use -DTRACY_ENABLE=ON to enable instrumentation and sampling
For gNB currently only sampling is implemented.
How to collect data:
- Windows hosts: tracy-profiler.exe in link above
- Linux hosts: Compile tracy-profiler from source
Enabling collection from docker:
- open port 8086
- to collect cpu data
--privileged
--mount "type=bind,source=/sys/kernel/debug,target=/sys/kernel/debug,readonly"
--user 0:0
--pid=host
How to add instrumentation:
- See examples in this commit or manual in the link above.
Fix error reported by the T2 card when running nr_dlsim with 4 DL layers.
Error can be reproduced by running:
./nr_dlsim -n100 -s30 -e25 -b273 -R273 -x4 -z4 -y4 -P --ldpc_offload.dpdk_dev 41:00.0 -c
So far we set NR_LDPC_MAX_NUM_CB to 72, which is sufficient for max 2 DL layers.
Set max possible number of LDPC segments based on the max number of layers
and max number of DLSCH segments (currently it is 144).
Integration: `2024.w46`
See merge request oai/openairinterface5g!3106
* !3095 UE: Set default SSB config
* !2991 fix BSR report malformed, add SHORT BSR when it can (instead of LONG BSR)...
* !3104 Trigger deregistration only in SA mode → small fixup?
* !3107 Clip the input for -32768 because this will make different result in...
* !3109 T tracer: support float types in traces
* !2777 NR UE trigger for re-establishment
* !3108 Prevent segfaults in gNB stack
* !3059 Test for init_RA
* CI: increase UE thread pool size
* !3036 Further improvements in analog beam management for CSI-RS
* !3093 Added NTN FDD FR1 bands defined in 3GPP TS 38.101-5
Added NTN FDD FR1 bands defined in 3GPP TS 38.101-5
Added support for NTN FDD FR1 bands 254, 255, 256 as specified in 3GPP
TS 38.101-5. Example Conf files added for band254 for 15Khz, 5Mhz and
30Khz, 10Mhz configurations. Tested NTN GEO configuration on band 254
using these conf files.
There seem to be randomly occurring deadlocks when running the UE in the
CI. MR !3091 should fix this. While waiting for this, apply a workaround
by increasing the number of worker pools in the UE to 12.
Since tag 2024.w45/commit f54ca5714e,
standalone mode/--sa is the default and does not exist. Correspondingly,
it has been removed from the documentation.
For users reading recent documentation, but running an older version of
OAI, there might be a mismatch in that --sa is necessary to run OAI, but
not given in the documentation. Provide explanatory text to make users
aware of this.
Test for init_RA
A testcase for init_RA function.
The testcase redefines the following functions as stubs:
- nr_mac_rrc_ra_ind
- nr_write_ce_ulsch_pdu
- mac_rlc_data_req
- lockGet_ul_config
- release_ul_config
- remove_ul_config_last_item
- nr_ue_configure_pucch
These stubs can be used to check init_RA behavior
NR UE trigger for re-establishment
The implementation in this MR triggers RRC re-establishment at the UE in case
of radio link failure either for out of sync messages from lower layers or in
case maximum number of RLC retransmissions is reached.
/opt/oai-gnb/bin/nr-softmodem -O /opt/oai-gnb/etc/gnb.conf --sa --log_config.global_log_options level,nocolor,time --rfsimulator.options chanmod --telnetsrv --telnetsrv.listenaddr 192.168.71.150 --telnetsrv.shrmod ci
=================================================================
==7==ERROR: AddressSanitizer: odr-violation (0x7fce78b85be0):
[1] size=4 'log_mem_write_side' /oai-ran/common/utils/LOG/log.c:78:14
[2] size=4 'log_mem_write_side' /oai-ran/common/utils/LOG/log.c:78:14
These globals were registered at these points:
[1]:
\#0 0x7fce7b691658 in __asan_register_globals ../../../../src/libsanitizer/asan/asan_globals.cpp:341
\#1 0x7fce7bd0c47d in call_init elf/dl-init.c:70
[2]:
\#0 0x7fce7b691658 in __asan_register_globals ../../../../src/libsanitizer/asan/asan_globals.cpp:341
\#1 0x7fce7a5e2eba in call_init ../csu/libc-start.c:145
\#2 0x7fce7a5e2eba in __libc_start_main_impl ../csu/libc-start.c:379
==7==HINT: if you don't care about these errors you may set ASAN_OPTIONS=detect_odr_violation=0
SUMMARY: AddressSanitizer: odr-violation: global 'log_mem_write_side' at /oai-ran/common/utils/LOG/log.c:78:14
==7==ABORTING
[INFO tini (1)] Spawned child process '/opt/oai-gnb/bin/entrypoint.sh' with pid '7'
[INFO tini (1)] Main child exited normally (with status '1')
* trigger re-establshemnt on DU: reset RLC counters on DRB 1
for OAI UE, since SRB 1 does not work for lack of periodic
measurements reports
* how is the re-establishment triggered?
* a transmission on DRB 1 occurs
* detection of a mismatch of RLC counters at the UE side
* max number of RLC retransmissions reached
* RLF detection
* verify re-establishment on CU: we ping the core from the UE; this
sends packets on DRB, and because the sequence number mismatches, will
trigger a reestablishment. We accept in the ping that a couple of
packets might be lost; after ping is over, reestablishment must be
done as well (else the ping will already fail).
Clip the input for -32768 because this will make different result in...
Clip the input for -32768 because this will make different result in
mult_complex_vectors() compared to C version
Trigger deregistration only in SA mode
Upon press of ^C, the UE triggers a NAS deregistration. This does not work
in do-ra or phy-test modes, because there is no NAS. Therefore, in non-SA
modes, just stop the UE directly.
This can potentially improve the 5G RFsim CI test case, because together with
a high RFsim timeout, the chances of the last HARQ status line being without
retransmissions is much higher. If that does not help, we can further increase
the retx threshold.
fix BSR report malformed, add SHORT BSR when it can (instead of LONG BSR)...
fix BSR report malformed, add SHORT BSR when it can (instead of LONG BSR)
simplify the code and make the code more explicit, but the may structure
remain (nr_write_ce_ulsch_pdu interface is complex, merging it into the
called would make simpler and more interstandable code), fix some asserts
related to this part of code
Added support for NTN FDD FR1 bands 254, 255, 256 as specified in 3GPP TS 38.101-5
Conf files added for 15Khz, 5Mhz and 30Khz, 10Mhz configurations added for these bands
In some runs of a multi-UE deployment, we see segfaults because
cell_group->spCellConfig is NULL. We should not segfault in that case,
so check for spCellConfig.
When the UE list is full, the iterator will run through all
MAX_MOBILES_PER_GNB UEs. It expects that eventually the pointer is NULL,
which is not the case if we have MAX_MOBILES_PER_GNB (or it simply
depends on the following memory). So give one array entry more, to
ensure that the last one is NULL.
- add new data structures for antenna processing.
- create arrays that keep track of certain values (max_ch, noise_amp2, nest_count and delay) per each antenna. These values are processed in order to determine one final output.
- add time processing measurements for each antenna, and prints them in nr_ulsim.
- make num_antennas_per_thread configurable in nr_ulsim and the gNB config file.
Co-authored-by: Nada Bouknana <bouknana@eurecom.fr>
Co-authored-by: Brice Robert <brobert@valiha.com>
UE: Set default SSB config
After initial sync the UE sets the ssb index bitmap in nrUE_config with
index obtained in initial sync. This allows UE to continue decoding PBCH
and stay synched with the gNB till SIB1 is received.
Replaced total_res with res_per_task when deciding whether to start a threadpool task or not.
This does not have big impact on the software as the nr_pusch_symbol_processing will recheck
ul_valid_re_per_slot anyway but it may reduce the number of thread pool tasks created.
Upon press of ^C, the UE triggers a NAS deregistration. This does not
work in do-ra or phy-test modes, because there is no NAS. Therefore, in
non-SA modes, just stop the UE directly.
If the RLC cannot retransmit a package, 3GPP specs say that we should
declare radio link failure (RLF). Up to this commit, the code did
nothing though. In this commit, add a per-UE RLF callback; the RLC will
call into this function to signal that RLF has been detected through
maximum retransmissions on a particular RLC bearer.
Integration: `2024.w45`
See merge request oai/openairinterface5g!3097
* !3048 Cleanup and refactoring at gNB initialization
* !3077 Address most thread safety issues in NR UE
* !3089 Improve cross-compile documentation
* !3092 Remove Interop-F1/Accelleran-CU testing pipeline
* !3096 Remove x86-specific instructions from telnet server
* !2949 Better polar decode
* !3070 CI: Add test with 4 DL layers, 60 MHz BW to SA-2x2 pipeline
* !3098 nr rlc: hotfix: test allocation valid before use
* !3090 Add UL PRBs Blacklist section to NR_SA_Tutorial_COTS_UE.md
* !3082 fix blocking bugs in command nr_dlsim -n100 -e27 -s35 -b 273 -R273 -x4 -y4 -z4
* !3100 CI: Remove 5g_l2sim files
* !3024 Set SA mode as default
* !3087 Remove some usage of the old CI sshConnection class
* !3074 CI: cleanup of cls_physim.py
Extract nr_ra_procedures.c into a separate library and add a testcase for init_RA function.
The testcase redefines the following functions as stubs:
- nr_mac_rrc_ra_ind
- nr_write_ce_ulsch_pdu
- mac_rlc_data_req
- *lockGet_ul_config
- release_ul_config
- remove_ul_config_last_item
- nr_ue_configure_pucch
These stubs can be used to check init_RA behavior
CI: Cleanup of cls_physim.py
- Remove unused code and file
- Rework functions for CUDA and T2 log analysis
- Check if test was successful for nr_ulsim and nr_dlsim in
RAN-gNB-N300-Timing-Phytest-LDPC pipeline
Set SA mode as default
- Set standalone (SA) mode as the default.
- Adjust commands in Docker Compose files by removing --sa flags and
adding --nsa flags where applicable.
- Add a function to verify that only one mode is being used at a time.
- Update all relevant documentation.
In the last commits, SSHconnection class has been removed from
cls_containerize.py. We therefore don't need the username and password,
as cls_cmd uses certificates only. Remove the usage of the username and
the password, and harmonize the use of GetCredentials().
Recent refactoring replaced the CreateWorkspace function call with a
dedicated XML step (for clarity, and because it avoids needless
recreation of workspaces). In the proxy function, the same
CreateWorkspace xml step has been added, cloning the OAI RAN repo, when
we actually need the repository for the proxy. Hence, fix this by
calling the CreateWorkspace() function manually, but only when
necessary, to build the proxy. (It would only be a problem if the
pipeline would rebuild the proxy, which did not happen because not much
work is happening on the proxy repo).
Further, refactoring and simplication of the proxy build function were
done:
- never tag proxy with develop, that makes no sense
- overwrite CreateWorkspace() info after having created and used the
workspace (the "if" to build the proxy)
- simplify copying of logs
- add simple unit test
- remove check for ubuntu; the dependcy is "docker". Simplify the code
to not obscure with a class variable.
The first thing Build_Proxy does is to remove the repository. Further,
Build_Proxy does not use the OAI RAN repo, but the proxy repo. The
Create_Workspace() is therefore completely useless.
Remove CheckAndAddRoute(), as it is complicated to read, uses a
multitude of hostname/IP/password combinations. Replace with scripts on
the machines, which, as of this commit, are:
- porcepix: ~/after-reboot.sh
- ofqot: ~/after-reboot.sh
- nepes: ~/after-reboot.sh
- obelix: ~/how-to-resetup-obelix-out-of-reboot.sh
It is clearer in CI logs to use absolute paths, and that is also a
"preference" of the new SSH class (cls_cmd.py) in the sense that some
functions don't handle "cd" gracefully (because of underlying Python
library paramiko).
For zip, though, we should change to have nice relative paths in the
archives, so use cd <path> && zip ... to achieve that, instead of
changing to a directory. This way, the path on which we operate is still
visible in the logs in each line.
The clang build fails with
/oai-ran/openair1/PHY/MODULATION/gen_75KHz.cpp:16:23: error: variable length arrays in C++ are a Clang extension [-Werror,-Wvla-cxx-extension]
16 | complex<double> t[len];
| ^~~
/oai-ran/openair1/PHY/MODULATION/gen_75KHz.cpp:16:23: note: function parameter 'len' with unknown value cannot be used in a constant expression
So ignore this warning (and make the build pass), since OAI uses VLAs in
various places.
The previous version would put the user-defined options first. However,
the default options include e.g., -Wall. If a user tries to override
(see for instance the next commit) -Wall or other options might override
this and not honor the user choice.
To always have the user have the last work, put user-defined compile
options last.
THere are four possibilities, for the three options (phy_test, do_ra,
nsa):
- None is set -> SA
- Only one is set of each -> corresponding mode
Introduce a function softmodem_verify_mode() to check for these.
- Introduce "Run_Physim" function for running ldpctest (with CUDA) and
nr_dlsim/nr_ulsim (with T2) in RAN-gNB-N300-Timing-Phytest-LDPC test
pipeline.
- Create "cls_analysis.py" and introduce "analyze_physim" for
log analysis of physim tests.
- Define patterns to be searched in the physim logs.
- Extract and print information about total gNB RX/TX PHY processing time
from nr_ulsim/nr_dlsim logs.
- Remove "cls_physim.py" as it is not used anymore, cleanup "main.py".
fix blocking bugs in command nr_dlsim -n100 -e27 -s35 -b 273 -R273 -x4 -y4 -z4
errors in main OAI code:
- one stack overflow
- one variable array of size 0
- one direct usage of unaligned SIMD
and small fixes inside the tests tool nr_dlsim to handle this large configuration (4layers, 100MHz)
use a better complex vector multiplication function
Since commit 041b2b2d7a, the CI main loop
iterates over all steps, even if there has been a problem during an
earlier XML step. In all cases, UndeployObject() is marked to always
run, as we always need to make sure that all deployed services are
terminated. UndeployObject() always recovers logs. Hence, it's not
necessary to do it here as well; it leads to duplicated logs.
This is visible in all pipeline with a failed deployment: the logs
appear twice.
create_workspace.sh used to check out develop by default. However, this
is unnecessary because the script later checks out a specific reference
(which can be a commit ID, which is not accepted by git clone, the
reason we don't check out during clone); the repository will simply be
empty in the beginning, which should not be a problem.
Further, it is problematic because we use this script for the build of
the L2sim proxy, and the repository does not have develop; thus, without
this modification, L2sim proxy build will always fail.
- simplify the code
- prevent to use "beta" tables from one decoder run to the next one (to see it
is the case, don't initialize beta, you will see it doesn't decode)
- decrease memory footprint of the decoder by x10 or more
- save about 15% of cpu in polartest unitary test tool, and more if we run in
realistic use case
- still this decoder performs much worse than the (very) slow OAI version by 3dB
telnetsrv uses the Intel/AMD-specific "cpuid" instruction, which is not
available on ARM machines. Remove the call.
The previous code was checking for the number of cores, which is
retained by a call to get_nprocs(). I did not find an equivalent for the
number of threads in C. In C++, there would be
https://en.cppreference.com/w/cpp/thread/thread/hardware_concurrency, in
case it's relevant for anybody.
Also, add it to the cross-compilation pipeline
csiim_vars/csirs_vars used to be global array for each UE. Now its present
in phy_data which is in dynamically allocated memory.
MAC config was done without mutex while DL and UL threads could be running
at the same time.
get_ue_nas_info was allocating memory possibly from more than one thread.
Moved it into new init function for UE NAS layer.
Addreses most issues in #869
The goal of this MR is to improve readability, maintainability and consistency
of gNB initialization functions.
This MR is:
- cleaning up redundant and unused code
- refactoring init_gNB, l1_north_init_gNB, RCconfig_NR_L1, NRRCConfig
- removing redundant memory allocation for gNB RAN context and L1 instances
- cleaning up redundant log lines
telnetsrv uses the Intel/AMD-specific "cpuid" instruction, which is not
available on ARM machines. Remove the call.
The previous code was checking for the number of cores, which is
retained by a call to get_nprocs(). I did not find an equivalent for the
number of threads in C. In C++, there would be
https://en.cppreference.com/w/cpp/thread/thread/hardware_concurrency, in
case it's relevant for anybody.
the version in develop is difficult to understand, and surprizing design
it loops only when avx2mod == 0 because when this test is false, the
size is always small i also spent time to understand, but i think i have
not changed it by not looping over the array.
Use non-recursive polar decoder after traversing the tree once. Maximum number
of nodes is set to 600 (unsure of the maximum). This is more performant
than the recursive version.
- Ensure packed struct through bitfield
- decoder tree buffer increase: i don't find evidence in the 3GPP
standard of the max size of polar coded PDU. even with this max size,
the way this tree is built is not trivial. so, it is hard to predict,
i just put 8 system pages (32KB) that looks enough [1]
- a lot of reformatting
[1] https://arxiv.org/pdf/1804.04389
according to this paper, the largest polar code in 5G is in UL, so
likely the biggest problem will be in gNB, not the UE. in DL, the max
output coded block is 512, in UL 1024. probably the tree we discuss is
related to this size.
"frozen bits" means only bits we don't encode/decode because polar code
size is only by power of 2.
We see this in OAI code: NR_POLAR_PUCCH_N_MAX is 10, so 2^10 is 1024,
that is the maximum size for 5G polar code length. in the code, we
assert if N > 512: "N = %d, not done yet\n",
polarParams->N". so, the today OAI code can't do PUCCH large blocks.
using current oai limitation of 2⁹ max code size, we can create up to
2^9+2⁸... nodes, so 1023 nodes in the tree. the problem then is that
each node have variable size, depending on the 'level' (the power of 2
we are currently). the algo provision large alpha and beta to be able
to store all lower levels in the tree (int leaf_sz = 1 << level)
therefore upper tree nodes are large (the top level for n=9 uses about
1.5KB) and the bottom nodes uses 64 bytes. if all nodes are populated,
the size will be very large: 1.5KB + 2 * 800 + .... but, we create
nodes in the tree only for the "information bits", so the payload, that
is limited to 128 bits (else we assert, the code is not "finished").
1.5KB * 128 is >> 8KB i allocated. but, we can't create more than one
mode at top level, so there is maximum 1.5KB + 2 * 800, + 4 * 400 + 8 *
200 + 16 * 100 + remaining of 128-32 nodes of 64 bytes. about 14KB.
i allocated 1024*sizeof(__m256i) bytes = 32KB that looks good.
* the goal of this function is to initialize and configure the RAN Context
* the goal of get_options is to get command line options
* this commit aims at making the implementation more consistent
* move initialization of RC.nb_nr_CC to this function
* cleanup redundant initializations of RC.nb_nr_macrlc_inst outside NRRCConfig
* cleanup redundant initization of RC.nb_nr_macrlc_inst from getter function (get_node_type)
* cleanup RC initialization in RCconfig_NR_L1
* removed redundant log lines (print number of instances in NRRCConfig only)
* do RC init after getting CL options and configuration
* get_node_type shall access configuration only, not RC context
* NRRCConfig shall be called only once: made it thread-safe
* nb_nr_inst represents the number of gNB instances and the number of RAN contexts
* it used also to count the number of RRC instances in the node
* it is set from config file in NRRCConfig, therefore redundant checks after
this function have been removed
* redundant global variable NB_gNB_INST, set after nb_nr_inst, was removed
* gNB and L1 init is done in init_gNB and RCconfig_NR_L1
* moved RCconfig_nr_prs out of create_gNB_tasks for consistency
* do RCconfig_nr_prs only for L1 instances
* removed initialization of gNB
* separate function to get blacklisted UL PRBs
* call the getter in both L1 and MAC/RLC configuration
* other minor improvements
Integration: `2024.w44`
See merge request oai/openairinterface5g!3081
* !2988 Add functional DFT test
* !3072 snow3g: fix array size
* !3073 small fix for 3 digit mnc usage in sib1
* !3040 Update pbchsim to work with other numerologies
* !3078 Simplify logs in gNB&UE
* !3080 Add Fedora 41, build_oai cleanup
* !3061 RLC optimization
* !3008 Fix regression 4g
* !3084 Fix a typo in nrue.uicc.yaml file
* !3085 chore(ci): creating a develop tag on top of develop-commit-id tag for the local private registry
chore(ci): creating a develop tag on top of develop-commit-id tag for the local private registry
Create a 2nd tag of develop for all images on the local registry. We
are doing this in order to simplify cross-repository testing between
FlexRIC and RAN.
Fix two issues with softmodem yaml files
- Fix a typo that prevents casting max_chan to integer in nrue.uicc.yaml
- Remove unused field ssb_PositionsInBurst_PR from gNB example config
yaml file
Fix Regression 4G
- Fix various bugs and/or regressions in 4G code
- Also, the CI does not test RFsim 4G with --sanitize, enables this to
detect problems.
Add Fedora 41, build_oai cleanup
- Add Fedora 41 to list of supported distributions
- Remove some dependencies in RHEL that are not necessary
- Cleanup some build_oai documentation of non-supported options
Simplify logs in gNB&UE
- reduce the amount of printf() that outputs useless info
- in MAC, PHY, NGAP, RRC, mark some "important" UE events with LOG_A for
better visibility -> can put log_config.global_log_level analysis to
only see important stuff
- in RRC, use some macro to consistently print logs
- track the cellID of a UE in RRC
small fix for 3 digit mnc usage in sib1
When 3 digit MNC is used in PLMN, during preparation of SIB1, first
digit of MNC was not taking into account. UE derives Serving Network
Name (SNN) using PLMN info from sib1. Because of 5GCore and UE has
different SNNs, this leads to XRES*, RES* mismatch during Authentication
Procedure between 5GCore and UE. (SNN is used for derivation of these
keys)
Add functional DFT test
- clean up code on (i)dft calls
- add DFT test for some DFT sizes against trivial implementation: it
tests a simple square signal in only real values
- correctly scale DFT-16
Moved some functions around so that instead of a function pointer a function
call is used. This enables inlining of functions which reduces number of function calls.
During preparation of SIB1, first digit of MNC was not taking into account.
This leads to XRES*, RES* mismatch during Authentication Procedure between 5GCore and UE because of derived serving network names.
This commit removes many logs that just show memory addresses, indices,
etc., which, I presume, do not mean anything to the "average user" and
developers alike.
Integration: `2024.w43`
Closes#769 and #841
See merge request oai/openairinterface5g!3069
* !2994 Refactor file descriptor handling in rfsimulator
* !3010 Fix heap overflow in websrv
* !3043 CI: Update RAN-gNB-N300-Timing-Phytest-LDPC test cases
* !3060 process registration reject in UE
* !3062 fix nr rlc tests
* !3067 Bugfix: Fill QFI for N3 GTP messages
* !3057 Bugfix in ue pusch mac pdu creation
* !3068 getopt() returns int
* !3056 F1AP Enc/Dec Library follow-up
* !3064 T: don't check argument in optimized version of T IDs
* !2786 Add F1 Handover procedures
* !2980 Refactor NAS 5GS PDU Session Establishment Accept handling and other NAS improvements
* !3063 Update yaml-cpp dependency handling
* !2989 NR gNB scheduler improvements for BWP operation
This MR removes limitations on BWP scheduler operation for which only
the information from the first UE of the list was taken into account. To
do so, I moved the VRB map check for DL and UL scheduler out of the
preprocessor and in the scheduler itself.
Closes: #769
Update yaml-cpp dependency handling
- Added CPM fallback in case package is not found automatically.
- Using yaml-cpp::yaml-cpp target instead of yaml-cpp to silence
deprecation warnings on newer cmake versions.
By omitting the variable CPM_SOURCE_CACHE as CACHE'd, it is actually
given to CPM, but does not appear when viewing it's content, e.g.,
through ccmake. Mark is as CACHE so that it will also appear.
- Added CPM fallback in case package is not found automatically.
- Using yaml-cpp::yaml-cpp target instead of yaml-cpp to silence deprecation
warnings on newer cmake versions.
Add F1 Handover procedures
- Clean up at RRC and MAC
- Add DU handover procedures, completely handle CellGroupConfig at DU
- Add CU handover procedures
- Maintain reestablishment working between/after handover
- Changes at UE to resync on different frequency DUs
Instructions: see doc/handover-tutorial.md
Things not implemented/open/missing: see #855
When receiving a measurement report including an A3 event, trigger
handover towards the indicated cell.
Note that currently, we are still limited to one Cell by DU, hence
looking up a cell comes down to looking up a DU, as is done here.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: Luis Pereira <lpereira@allbesmart.pt>
Co-authored-by: Robert Schmidt <robert.schmidt@openairinterface.org>
This commit implements reestablishment during handover when occurring at
the source DU, either as a handover Failure or "normal" reestablishment
(something else went wrong). In this case, the CU will trigger HO
cancel, which in F1, means to cancel on the target DU. Also, undo any
previous ID changes. Retain the old CellGroupConfig (before handover)
for the case of reestablishment at source. If not needed, it is freed
after the reconfiguration complete.
Reestablishment on target cell during handover cannot be handled by the
gNB (yet). It is the same as with a normal reestablishment (not during
handover) on a different cell: that cell has no notion of the old UE, so
we cannot simply indicate the old UE to reestablish; at least, we would
need to cleanup the CellGroupConfig, and I am not sure this works out of
the box, either.
After handover, the last CellGroupConfig received is the one for
handover, containing a reconfigurationWithSync. Resending this
CellGroupConfig to the UE after reestablishment would trigger a
handover. Free the reconfigurationWithSync to have the UE complete
reestablishment at the current cell.
The CU implementation of reestablishment modifies the CellGroupConfig.
It should not do this, but does as of now. The specification (38.473 Sec
8.7) foresees an optional F1 UE context modification procedure (which
can be used to receive the CellGroupConfig). This should be done instead
to retrieve the correct CellGroupConfig.
Note that the F1 UE context setup procedure, similar to the UE context
modification procedure, would also allow to perform reestablishment on a
DU that is not the original one.
If we release a request from a DU during, handle as follows:
- from the source: Handover is still ongoing, so ignore it by
acknowledging the release, and marking handover as done. In this case,
we did not receive an acknowledgement from the target, but since the
source DU does not keep the UE context, we could not cancel the
handover anymore either. The UE might come back on any DU through
reestablishment (or simply be connected on the target)
- from the target: consider this to be a genuine release, and forward to
AMF.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
The HO is completed (from a pure RAN perspective) when receiving
reconfiguration complete. In that case, remove the HO context, and
signal successful handover through the callback. For the F1 handover,
this means releasing the UE at the source DU.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
We expect the UE's RRC Reconfiguration Complete to come through the
target DU, as the "new UE" at the target CU. Also, function
CU_handle_UL_RRC_MESSAGE_TRANSFER() checks that the UL RRC Message,
through which we will receive RRC Reconfiguration Complete, matches the
DU UE ID stored in RRC. Since the DU UE ID will be the one at the target
DU, while up to now, we stored the DU UE ID at the source DU.
Correspondingly, update the DU UE ID to the one at the target DU; also
update the corresponding assoc_id.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
In this commit, handle the reception of a UE context setup response, in
the case of handover. In that case, store the gNB-DU UE ID and C-RNTI of
the UE at the target DU, and trigger a reconfiguration towards the UE at
the source DU via the callback.
To ensure that the CU-UP's PDCP is SN-aligned with the UE, use
cuup_notify_reestablishment() to reset the bearers with PDCP
reestablishment, which has already been implemented in the last commit.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
This is preparatory work to handle the UE context setup response from
the target DU. There are two cases:
- "Initial" UE context setup response (during connection setup): we need
to inform the CU-UP via an E1 message of the new bearers (GTP tunnel
update)
- F1 handover: we need to inform the CU-UP via an E1 message that there
are DRBs to reestablish
There are two different functions, e1_send_bearer_updates() and
cuup_notify_reestablishment(), to do this. Harmonize the code, before
implementing the actual handover logic in the next commit.
Add a function to trigger an F1 handover, including the corresponding
telnet function implementation. The function looks up source&target DU,
creates a HO context, and reads PDU information to send a UE context
setup request to the target DU towards which to handover.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
Functions get_qos_char_from_qos_flow_param() and
get_qos_characteristics() will be used in rrc_gNB_mobility.c to read PDU
information for handover. Move the definitions to a header to reuse
later.
HandoverPreparationInformation will be used in the follow-up commit to
trigger a handover.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
This F1 message sends us the CellGroupConfig of the UE. Store it in case
we need it. For instance, at least in OAI, it is possible to make a
handover without a DRB (this might or might not be standard compliant).
However, in this case, we did never receive a UE context setup request,
which stores the CellGroupConfig. Hence, store the first CellGroupConfig
we receive, to be able to do e.g., handover.
RETURN_IF_INVALID_ASSOC_ID() checks if an assoc_id (of a DU) is valid.
It will be also used in the mobility RRC module later. Move it to the DU
header (where it logically belongs) to move it from there.
Prior to this commit, after RA failure, the DU would send a UE context
release complete, without actually requesting one. This could lead to
inconsistent state between DU and CU, because the CU might just discard
this (unrequested) message, and not actually trigger a release [1].
Thus, refactor the release request into a function, and use it to
request release of a UE for which RA has not been completed. If the CU
does not know the UE, release the UE immediately.
[1] One possibility would be that during reestablishment, the CU would
request to look up an old DU UE ID when it has been released, resulting
in
Assertion (oldUE) failed!
In dl_rrc_message_transfer() /home/oaicicd/robert/openairinterface5g/openair2/LAYER2/NR_MAC_gNB/mac_rrc_dl_handler.c:860
CU claims we should know UE bc7f, but we don't
In handover, we might need to remove a UE that we knew under it's
C-RNTI, for which we also have the RA process. Hence, when removing a UE
by RNTI, remove it also from RA.
* DL NAS Transport messages can piggy-back 5GSM messages
* the relevant decoding functions can be called in the handler (e.g. capture_pdu_session_establishment_accept_msg)
* the capture_pdu_session_establishment_accept_msg function is already handling
the code in the case FGS_PDU_SESSION_ESTABLISHMENT_ACC of NAS_DOWNLINK_DATA_IND
Closes: #841
Bugfix in ue pusch mac pdu creation
When more than one sdu with equal priority is present, there could be
buffer left after serving the lcids and also not enough to equally serve
all lcids. This commit checks for this condition. Before this fix the ue
would send malformed mac pdus when serving multiple lcids.
See getopt(3): returns int, so store it in an int variable.
I went through all occurrences, project-wide. All return values of
getopt() should be stored in ints now.
process registration reject in UE
when the 5GC rejects the ue, the message was not processed (information
element not declared)
this MR process it to print it and stop the UE
a more complete code should be done when the UE will be able to select
another cells than the most powerful PBCH, but it require larger
enhancement
CI: Update RAN-gNB-N300-Timing-Phytest-LDPC test cases
- update enc/decoder tests in RAN-gNB-N300-Timing-Phytest-LDPC - test
only for given SNR value (check of the test results to be added in the
following MR)
- add cores to Tpool in the nr_dlsim test cases, currently we test with
only one core
- always execute all the nr_ulsim/nr_dlsim tests, even if one of them
fails
- add new nr_dlsim test cases - test with 4 antennas, 2 layers
Refactor file descriptor handling in rfsimulator
- add hashtable iterator to hashtable.c
- use hashtable for file descriptor to buffer_t mapping in rfsimulator
This enables the use of file descriptors above 250. The linux file
descriptors can reach the value of FD_SETSIZE which could be e.g. 1024.
This is not an issue in most cases but if the UE or gNB process opens a
lot of file descriptors for other purposes we might see rfsimulator file
descriptors reach > 250 and cause a segfault.
Also added tests for new hashtable functions.
Instead of hardcoding to 0, instruct the GTP module to send N3 GTP
packets with the right GTP.
We currently only support one QoS Flow, so check that it is really only
one.
When more than one sdu with equal priority is present, there could be
buffer left after serving the lcids and also not enough for equaly
serving all lcids. This commit checks for this condition. Before this
fix the ue would send malformed mac pdus when serving multiple lcids.
After RA, in the case of HO, we need to make sure to use the right radio
configuration (MCS table, etc). This commit is important to make HO
actually work.
Our initial CellGroupConfig always contains LCID1, because in normal
connection setup, the DU has to automatically set it up anyway. In
handover/UE Context Setup, the CU requests to set up SRB1 (which we
assume to be LCID1), but we reuse the initial CellGroupConfig. To not
have SRB1 added twice, skip in this case.
This commit implements adding a new UE on the target DU in case of
handover. We identify this case by the fact that the gNB-CU did not send
a gNB-DU UE ID (which is optional in F1AP, but not in our
implementation), such that we know there is no corresponding UE, which
therefore has to be created.
If it happens that there is a handover preparation information, we read
the UE capabilities from this information, and subsequently trigger the
creation of a new UE.
Another change is that for all bearers that we are asked to setup, we
add reestablishRLC so that the UE does a reestablishment during the
reconfiguration.
As we have to remove the reconfigurationWithSync and reestablishRLC IE
after application of the CellGroupConfig (because we might reuse the
CellGroupConfig later), free these IEs. As there are now multiple fields
to free, introduce a separate nr_mac_clean_cellgroup() function that
handles cleaning of the CellGroupConfig.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
The next commit will modify the UE context setup request procedure to
(possibly) receive UE capabilities through Handover Preparation
Information, which extracts the UE capabilities for the UE to be setup.
To avoid code duplication, extract the code that will be needed in UE
capability extraction from Handover Preparation Information.
Implement the Transmission Action Indicator F1AP forwarding logic, and
add it to the F1 internal representation structure. After the
transmission interrupt timer (during which SRB1 can send messages), the
UE will be out-of-sync (we assume it's not there anymore).
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
Implement the HandoverPreparationInformation F1AP forwarding logic, and
add it to the F1 internal representation structure.
No logic is implemented in this commit.
Co-authored-by: batuhanduyuler <batuhan.duyuler@firecell.io>
Co-authored-by: rmagueta <rmagueta@allbesmart.pt>
Refactor the RRC processing timer to be a generic "transmission
interrupt" timer, and use NR_timer_t for it. Also,
depending on the kind of RRC processing (see 38.331 table 12.1-1),
timing requirements might be different. However, the DU does not always
know which RRC processing is ongoing, and does not get this information
via F1. Hence, introduce a "worst case" delay time, which aligns with
the spec (it says these times are "minima") and is accepted by COTS UE.
The gNB MAC might need to stop transmission either for some time (RRC
processing), or continuously (transmission stop requested by CU). Those
can be handled in one common timer implementation. The continuous case
will be used in a follow-up commit.
Co-authored-by: Luis Pereira <lpereira@allbesmart.pt>
CI: remove unused code, clean up imports
* Remove pyshark, matplotlib imports, as they are not used
* Remove stats_monitor.py, as the CI does not use it
* Clean up many warnings shown by pyflakes regarding includes that we don't use
Adapt as code above offending line to not overflow array. Avoids UBSan
error
openair1/SCHED_UE/phy_procedures_lte_ue.c:2983:36: runtime error: index 8 out of bounds for type 'int [8][1]'
The IES_DECODE_U16 macro loads a 16 bit integer from an arbitrary
address; this can lead to reads from misaligned addresses, and undefined
behavior sanitizer warns, like, e.g., so:
openair3/NAS/COMMON/IES/ProtocolConfigurationOptions.c:62:5: runtime error: load of misaligned address 0x7f3c24032c01 for type 'uint16_t', which requires 2 byte alignment
To fix this, first use memcpy() to load to a 16-bit aligned address,
then do the actual decoding.
- rename as_nas_info_t member data to nas_data to have a uniq name to
trace between complex malloc/free of this pointer
- fix, maybe not perfect, these pais od malloc/free
- remove typedef name changes across transmission of these as_nas_info_t
over itti, as name changing makes it even more confusing
- remove globals that are likely possible races
- in as_message_decode() fix memcpy size, it is hard to understand the
initial idea, likely the author beleives a struct of struct makes a
intermediate pointer
- in network_api.c remove globals (static but used globally)
- remove casts that are there to hide gcc warnings on probable unaligned
access
- remain a weird mix of packed/unpacked versions of the same typedef,
but gcc is happy
pyflakes warns about various errors, addressed here. For cls_cluster.py,
remove (unknown) sys.exit and replace through Exception, as sys.exit()
has the unwanted effect that the CI script would stop. Instead, raise
an Exception that would show the problem in the HTML, and ensures that
the script runs until the end.
These are the warnings flagged by pyflakes.
cls_cluster.py:297:4: undefined name 'sys'
cls_containerize.py:39:1: 'subprocess' imported but unused
cls_containerize.py:41:1: 'threading' imported but unused
cls_containerize.py:43:1: 'multiprocessing.Process' imported but unused
cls_containerize.py:43:1: 'multiprocessing.Lock' imported but unused
cls_containerize.py:43:1: 'multiprocessing.SimpleQueue' imported but unused
cls_containerize.py:49:1: 'cls_cluster as OC' imported but unused
cls_containerize.py:50:1: redefinition of unused 'cls_cmd' from line 42
cls_module.py:28:1: 'os' imported but unused
cls_module.py:29:1: 'sys' imported but unused
cls_module.py:34:1: 'subprocess' imported but unused
cls_module.py:35:1: 'datetime.datetime' imported but unused
cls_module.py:211:11: undefined name 'sshconnection'
cls_oaicitest.py:37:1: 'pexpect' imported but unused
cls_oaicitest.py:47:1: 'cls_cluster as OC' imported but unused
cls_oaicitest.py:54:1: 'matplotlib.pyplot as plt' imported but unused
cls_oaicitest.py:55:1: 'numpy as np' imported but unused
cls_oai_html.py:34:1: 'sys' imported but unused
cls_oai_html.py:41:1: 'multiprocessing.Process' imported but unused
cls_oai_html.py:41:1: 'multiprocessing.Lock' imported but unused
cls_oai_html.py:41:1: 'multiprocessing.SimpleQueue' imported but unused
cls_physim.py:37:1: 'multiprocessing.SimpleQueue' imported but unused
cls_static_code_analysis.py:39:1: 'time' imported but unused
cls_static_code_analysis.py:40:1: 'multiprocessing.Process' imported but unused
cls_static_code_analysis.py:40:1: 'multiprocessing.Lock' imported but unused
cls_static_code_analysis.py:40:1: 'multiprocessing.SimpleQueue' imported but unused
epc.py:41:1: 'multiprocessing.Process' imported but unused
epc.py:41:1: 'multiprocessing.Lock' imported but unused
epc.py:41:1: 'multiprocessing.SimpleQueue' imported but unused
main.py:60:1: 'pexpect' imported but unused
main.py:66:1: 'datetime' imported but unused
main.py:68:1: redefinition of unused 'subprocess' from line 63
main.py:70:1: 'multiprocessing.Process' imported but unused
main.py:70:1: 'multiprocessing.Lock' imported but unused
main.py:70:1: 'multiprocessing.SimpleQueue' imported but unused
provideUniqueImageTag.py:2:1: 'os' imported but unused
Integration: `2024.w42`
Closes#547, #853, #854, #858, #859, #862, #844, and #746
See merge request oai/openairinterface5g!3053
* !3047 Resolve compilation warnings on ARM
* !3052 Optimize T_ID() macro.
* !3045 CI: Reduce tested DL throughput on Aerial
* !3039 fix for Other SI reception when BWP start != 0
* !2997 websrv regression
* !3041 Update NR SA Tutorials
* !3042 Update TESTBenches.md to reflect current testbench
* !3004 Hotfix to remove an unwanted TX power reduction when using multiple TX antenna ports.
* reduce ssPBCH_BlockPower in gNB config of FHI7.2 pipeline
* !3046 A testsuite and a benchmark for RLC AM mode
* !3038 Various bugfixes for gNB stability, cleanup
* !3051 PUCCH2 DTX detection
* nr_psbchsim: fix segv by providing correct get_nrUE_params()
* !3049 Improve PUSCH and PUCCH SNR stability
* !3054 Allocate and add sD to DRB NSSAI in case sD != 0xffffff
Currently, sD has not been allocated, so the octet string is empty. Fix the
allocation to correctly write&read NSSAIs in UE context setup/modification
request messages.
When running, nr_psbchsim, we get this UBsan error:
openair1/PHY/NR_UE_ESTIMATION/nr_adjust_synch_ue.c:79:16: runtime error: member access within misaligned address 0x0ff4278c791c for type 'struct nrUE_params_t', which requires 8 byte alignment
The problem is that the corresponding definition of
get_nrUE_params() in psbchsim.c does not actually return the UE params,
but void, so we might or might not get something that resembles valid
data in nr_adjust_synch_ue.c. Fix this by including nr-uesoftmodem.h
(that provides the correct declaration) and define it correctly. Due to
type mismatches, also adapt downlink_frequency, uplink_frequency_offset,
and openair_cfg (which might result in the same type of bugs).
This adds DTX detection in PUCCH 2 receiver. This is detected by MAC
already which will drop PUCCH 2 indications that are detected as DTX.
For the moment this uses the same pucch0_dtx_threshold value indicated
in L1 section of the configuration file. The feature firstly reduces
unnecessary logging on stdout and the use of invalid CSI measurements in
the MAC scheduler.
Closes: #547
- Fixes#853, #854, #858, #859, #862
- handle repeated NGAP Initial context setup requests, thereby fixing assert:
Assertion (!ue_p->as_security_active) failed!
In rrc_gNB_generate_SecurityModeCommand() /home/eurecom/raymond/openairinterface5g/openair2/RRC/NR/rrc_gNB.c:2590
logic error: security already activ
- A lot of cleanup: remove almost all use of protocol_ctxt_t in 5G RRC and use
existing pointers instead
Introduce two tests for NR RLC AM entity:
- test_nr_rlc_am_entity
- benchmark_nr_rlc_am_entity
The benchmark can be used to verify processing time improvements for NR RLC AM
entity. The benchmark creates two entities and creates an unbalanced
communication channel between them with 10% loss packet loss and 100:1 PDU sent
ratio. This allows the lists in RLC to build up and triggers RLC AM code
related to looking through the segment/pdu lists.
Example benchmark output:
ubuntu@zeus:~/bpodrygajlo/openairinterface5g/cmake_targets/build$ ./openair2/LAYER2/nr_rlc/tests/benchmark_nr_rlc_am_entity
[LOG] init aborted, configuration couldn't be performed
log init done
CMDLINE: "./openair2/LAYER2/nr_rlc/tests/benchmark_nr_rlc_am_entity"
[CONFIG] debug flags: 0x00400000
2024-10-16T14:58:20+02:00
Running ./openair2/LAYER2/nr_rlc/tests/benchmark_nr_rlc_am_entity
Run on (8 X 4700 MHz CPU s)
CPU Caches:
L1 Data 48 KiB (x4)
L1 Instruction 32 KiB (x4)
L2 Unified 1280 KiB (x4)
L3 Unified 12288 KiB (x1)
Load Average: 0.48, 0.27, 0.20
***WARNING*** CPU scaling is enabled, the benchmark real time measurements may be noisy and will incur extra overhead.
--------------------------------------------------------------------
Benchmark Time CPU Iterations
--------------------------------------------------------------------
BM_nr_rlc_am_entity/100 50096 ns 50097 ns 13101
BM_nr_rlc_am_entity/256 140386 ns 140387 ns 4868
BM_nr_rlc_am_entity/1024 536323 ns 536311 ns 1309
BM_nr_rlc_am_entity/4096 2084464 ns 2084385 ns 338
BM_nr_rlc_am_entity/16384 8309318 ns 8309458 ns 81
BM_nr_rlc_am_entity/20000 9843792 ns 9843774 ns 68
This hotfix removes division by num_antenna_ports in PDSCH precoding matrix generation.
These matrices are defined for CSI reporting in this way but here we are using them for
transmission and this results in an unwanted 10*log10(num_antenna_ports) reduction in
transmit power (3dB for 2 antenna ports, 6dB for 4 antenna ports, etc.)
- Update links for the latest Ubuntu 24.04 LTS iso file
- Update docker install to the latest instructions on the official website
- Update UHD to v4.7.0.0 and update dependencies for Ubuntu 24.04
Test configuration: Tested SIB19 reception on SCS120 Khz + 50Mhz bw
config file used:
targets/PROJECTS/GENERIC-NR-5GC/CONF/gnb.sa.band257.u3.32prb.usrpx410.conf
+ NTN config commands used
sudo ./nr-softmodem -O ../../../targets/PROJECTS/GENERIC-NR-5GC/CONF/gnb.sa.band257.u3.32prb.usrpx410.conf --sa --rfsim -d
sudo ./nr-uesoftmodem --band 261 -C 27533280000 -r 32 --numerology 3 --ssb 72 --sa --rfsim -d --ue-scan-carrier
The Issue is that SIB19 is not being received because of CRC error, only
SIB1 was being received.
This changes the way T_IDs are defined. The T_ID macro was using a
integer to pointer cast in order to provide type safety. This trick
disables some compiler optimizations, as explained here:
https://clang.llvm.org/extra/clang-tidy/checks/performance/no-int-to-ptr.html.
Removing the type cast reenables the compiler optimizations.
This MR aims to address warnings raised when compiling for ARM ( tested
with cross compilation, not directly on ARM hardware ). The addressed
warnings are as follows:
- An unused variable in file oai_dfts_neon.c
- Wrong printf format specifier related to oai_cputime_t
- Wrong parameter type in qam64_qam64 call
This MR closes issue #844
Also, correctly set the pkg-config search path for libraries for the
target system. Also closes: #746
A UE might send a ULInformationTransfer without the actual content present
(intentionally or not). Check this and handle appropriately to avoid
a segfault when reading the empty NAS message.
Closes: #862
Pass UE context pointer directly. In the switch, do not return to not
create memory leaks (after the switch, memory is freed). Move the
function up to make it appear before the use of the function.
Pass UE context, without protocol_ctxt_t, for cleanup. Remove the return
from within the switch, as this would lead to memory leaks; nothing is
executed after the switch statement (except for freeing memory).
The existing check for the initialization of GTP is wrong; GTP returns
-1 if it could not initialize. Correct this, such that the DU will stop
if GTP cannot be initialize.
If we continue, the DU is requested to set up a GTP tunnel; because GTP
is not initialized, it assumes it is in monolithic/no F1, and does not
set up a tunnel. The F1 UE Context setup response encoder then fails
with this assertion:
Assertion (tnl->teid > 0) failed!
In DU_send_UE_CONTEXT_SETUP_RESPONSE() ../../../openair2/F1AP/f1ap_du_ue_context_management.c:415
Exiting execution
../../../openair2/F1AP/f1ap_du_ue_context_management.c:415 DU_send_UE_CONTEXT_SETUP_RESPONSE() Exiting OAI softmodem: _Assert_Exit_
Aborted
Closes: #853
A UE might send a MeasurementReport without the actual IEs present
(intentionally or not). Check this and handle appropriately to avoid
assertion
Assertion (measurementReport->criticalExtensions.present == NR_MeasurementReport__criticalExtensions_PR_measurementReport && measurementReport->criticalExtensions.choice.measurementReport != ((void*)0)) failed!
In rrc_gNB_process_MeasurementReport() ../../../openair2/RRC/NR/rrc_gNB.c
Closes: #859
A UE might send a UE Capability Information without unsolicitatedly
(intentionally or not). Check this and handle appropriately to avoid
assertion
Assertion (UE->xids[xid] == RRC_UECAPABILITY_ENQUIRY) failed!
In handle_ueCapabilityInformation() ../../../openair2/RRC/NR/rrc_gNB.c
Closes: #858
A UE might send a rrcReestablishmentComplete without the actual IEs
present (intentionally or not). Check this and handle appropriately to
avoid assertion
Assertion (reestablishment_complete->criticalExtensions.present == NR_RRCReestablishmentComplete__criticalExtensions_PR_rrcReestablishmentComplete) failed!
In handle_rrcReestablishmentComplete() ../../../openair2/RRC/NR/rrc_gNB.c:1364
Closes: #854
Prior to this commit, if we receive an Initial Context Setup Request for
a UE that has AS security set up, the gNB asserts with
Assertion (!ue_p->as_security_active) failed!
In rrc_gNB_generate_SecurityModeCommand() openairinterface5g/openair2/RRC/NR/rrc_gNB.c:2590
logic error: security already activ
However, TS 38.413 says:
> The AMF may initiate the Initial Context Setup procedure if a
> UE-associated logical NG-connection exists for the UE
(I interpret this as "it might happen anytime")
> store the received Security Key in the UE context and, if the NG-RAN
> node is required to activate security for the UE, take this security key
> into use.
(I interpret this as "do not set up AS security again")
In this commit, handle if we receive an Initial Context Setup Request
for an existing UE context. As said in the spec, we store the keys. If
security is not active, we trigger RRC Security Mode Command. If it is
active, we skip Security, and directly set up a PDU session. If there is
also no PDU session, we simply forward the NAS message, and acknowledge.
If there was no AS security, after security, the gNB would trigger the
setup of PDU sessions (if any), and then forward any NAS PDUs, which is
stated in the if block if AS security is already active.
This changes the way T_IDs are defined. The T_ID macro was using a integer to pointer cast in
order to provide type safety. This trick disables some compiler optimizations, as explained
here: https://clang.llvm.org/extra/clang-tidy/checks/performance/no-int-to-ptr.html.
Removing the type cast reenables the compiler optimizations.
Prior to this commit, the include path for some libraries found through
pkg-config when cross compiling is not correct:
$ grep include CMakeCache.txt
blas_INCLUDEDIR:INTERNAL=/usr/include/x86_64-linux-gnu
...
This is because we did not tell pkg-config to search in the right path.
Since for cross-compilation, we install packages in parallel in the
system, override the primary directory where the .pc files located, in
the cross-compilation toolchain file for cmake.
Closes: #746
Since long long is the bigger type ( 8 bytes ) of the two used ( long long and uint32_t ),
and long long is also available on ARM, change this parameter to always have this type.
This allows to always use format specifier '%lld' in the logs that use TICK_TO_US, which would otherwise return long long or uint32_t depending on architecture.
Set a maximum SNR value using the "-S" option to ensure that nr_dlsim
or nr_ulsim is executed only for a given SNR. This approach allows us
to verify the success of the encoding/decoding process at the specified
SNR. Check if the test succeeds is not yet implemented (currently we
only retrieve an information about encoding/decoding processing time).
To be implemented in the separate MR.
Always execute all the nr_dlsim/nr_ulsim tests even if one of them
fails. Each test takes around 5s, so we can execute them all to get
results for all the test scenarios.
which will drop PUCCH 2 indications that are detected as DTX. For the
moment this uses the same pucch0_dtx_threshold value indicated in L1
section of the configuration file.
rrc_gNB_process_security() does not only set (process) the security
algorithms, but also checks for those. However, we don't use the check.
Simplify the function, to reintroduce the check later. Rename to
set_UE_security_algos() to better reflect what the functions do.
- remove Sphex UE - not used anymore
- remove RAN-L2-Sim-Test-5G from a list of pipelines - currently disabled
- 4G RFSim was moved to cacofonix
- minor cleanup
Integration: `2024.w41`
See merge request oai/openairinterface5g!3032
* !3006 Fix some sizeof and switch-case issues
* !3028 Add CMakePresets.json
* !3011 Aerial: cleanup unused code, avoid superfluous memcpy(), avoid L1 structures
* !3022 NR CSI-RS L1 refactoring → check CI
* !2935 Improvements in handling the beamforming procedure via FAPI
* !3026 NR ULSCH scheduler fix when using transform precoding
* !3031 CI: Increase test bitrates for 100 MHz scenario in SA-2x2 pipeline
* !3017 Fix SRS
* !2731 Implementation of 2-Step RA
* !3027 Add dockerized include what you use tool
3GPP 38.331: If absent, the UE applies the SCS as derived from the prach-ConfigurationIndex in RACH-ConfigGeneric (see tables Table 6.3.3.1-1, Table 6.3.3.1-2, Table 6.3.3.2-2 and Table 6.3.3.2-3, TS 38.211 [16]).
This procedure for msg1_SubcarrierSpacing NULL is not implemented, so we aplly the same value of UL SCS.
Introduce CMakePresests.json which is a simple way to perform incremental build using cmake
New configure resets were added:
- default: Configure compilation with default options
- tests: Same as above but ENABLE_TESTS and SANITIZE_ADDRESS is ON
New build presets were added:
- 5gdefault: Build the software for NR rfsimulator test
- default: same as 5gdefault
- 4gdefault: Build the software for LTE rfsimulator test
- tests: build all unit tests
To configure using configuration preset:
`cmake --preset <preset_name>`
To build using a build preset:
`cmake --build --preset <preset_name>`
Modify the UL/DL UDP iperf test cases to use a higher requested bitrate
for the scenario with 100 MHz bandwidth in the SA-2x2 pipeline. Adjust
gNB config file to allow for higher UL throughput.
Resolve error:
nfapi/oai_integration/aerial/fapi_nvIPC.h:56:16: error: unknown type name ‘nvipc_params_t’
56 | int nvIPC_Init(nvipc_params_t nvipc_params_s);
By including the header. Also, move the definiton of the struct to the
MAC, where it is actually stored, and out of the radio common_lib.h,
since this is not a radio (split 7/8) parameter.
When compiling aerial, we compile without CUDA support (see e.g.,
docker/Dockerfile.gNB.aerial.ubuntu22). Hence, do not check for it in
aerial for the moment, we won't use it for the foreseeable future.
In Aerial bindings (FAPI-Aerial), the NFAPI mode must be
NFAPI_MODE_AERIAL (otherwise, we would not be executing this code).
Hence, the switches don't make sense. Remove them.
The check can more easily be checked with
git show <commit> --ignore-all-space
As is visible at the top of the patch, for compiling aerial_lib (the
binding FAPI-Aerial), we always add ENABLE_AERIAL. Thus, all the code
within ENABLE_AERIAL will be compiled; we can also simply take it out.
We can directly read from buffers, so the memcpy() only takes time.
Also, limit the scope of the variables for the unpacking of
RX_DATA.indication.
Co-authored-by: Rúben Soares Silva <rsilva@allbesmart.pt>
These functions look up the numerology, and code can be harmonized:
- ssb_index_from_prach()
- find_SSB_and_RO_available()
- schedule_nr_prach()
The next commits also need to look up the numerology from the
UplinkConfigCommon. This provides the common implementation to be used
in follow-up commits.
Integration: `2024.w40b`
See merge request oai/openairinterface5g!3021
* !2890 F1AP encoding/decoding library
* !2969 Time domain samples scope for NR UE & gNB
* !2998 warmup RF board in UE
* !3009 Fix the ratio of PDSCH EPRE to DM-RS EPRE
* !2965 Refactoring of prepare_scc in gNB config
* !3002 CI: Change frequency used by AW2S pipeline
* !3007 Enable gNB configuration through YAML file
* !3001 Minor ue fixes
* !2971 Use getaddrinfo() when connecting to rfsimulator
the code creates a sink to read all traced packets and drop it. but
this works only if we send to a local address and if there is not
another sink (like if we trace both UE and gNB in the same machine).
this sink's purpose is only to prevent to get automatic packets back
from the kernel saying "no listener". So, instead of killing the trace
when we can't create the sink (as done previously), we now just let the
trace in operation.
Integration: `2024.w40`
Closes#756 and #849
See merge request oai/openairinterface5g!3014
* !2886 (doc): Added a new doc on hardware requirements
* !2955 dockerized clang-format error detection tool
* !3005 Fix symbol size in nr_dlsch_mmse
* !2930 SRS tx power
* SA-B200-Module-SABOX pipeline: improve radio
* SA-AW2S-CN5G: accept 5% ping packet loss
* !3003 Resolve "SEGFAULT in L1_rx_thread" - CI
* !3016 fix tests for log2_approx and log2_approx64
* patch to move RFsim-4G to cacofonix
It can happen that L1 gives us a TB that we interpret as receiving UL
MAC TB with LCID "UL_SCH_LCID_CCCH1". This might or might not be
correct, but currently, we observe segfaults when trying to encode the
cellgroup in send_initial_ul_rrc_message() (encode_cellGroupConfig()).
This fix prevents a segfault by not delivering a MAC SDU to RLC if CCCH
is already present. Note that in some cases, this could still happen,
e.g. Resume Request. However, we don't support this message, so ignore
it for the moment.
This fix enables a gNB to operate continuously for extended periods
(days), which is essential for real deployment scenarios.
Closes: #756
The test was actually never executed correctly (unsigned -10 is never <= 10), and log2_approx64_ref() was not accurate enough.
Therefore replace it by directly comparing one value above and one value below each rounding boundary.
To skip the test, we use the gtest command line parameter "--gtest_filter" to filter out the unwanted test.
This parameter is described as follows:
--gtest_filter=POSITIVE_PATTERNS[-NEGATIVE_PATTERNS]
Run only the tests whose name matches one of the positive patterns but
none of the negative patterns. '?' matches any single character; '*'
matches any substring; ':' separates two patterns.
In this pipeline, we frequently see losses of a single packet; we did
not find the true reason yet.
Accept 5% losses, assuming that serious packet loss would be catched by
(1) higher ping losses (2) iperf would be affected.
This commit refactors startServer to use getaddrinfo() to
prepare the socket creation. Doing so, it also adds support for
IPv6 by using AF_UNSPEC as the address family.
This commits uses getaddrinfo(3) when connecting as client to the
rfsimulator. This enables name lookups, through DNS or a hosts file.
This change makes it easier to use the simulator in a dynamic
environment like Docker or Kubernetes by only targeting the name of
the container or service.
Move h_b_f_c (or SRS power control adjustment state) to UL BWP structure. This is a per
bwp parameter so it makes sense to make its lifetime the same as BWP, even if we dont support
>1 UL BWPs.
the create_workspace.sh script checks if a branch should be merged into
what we are trying to checkount (i.e., typically develop into
feature-branch). The problem is that it uses a short-hand for an if
branch.
[ -n "${merge}"] && git merge...
If we don't request a merge, [ -n XX ] will evalute to 1, and not do the
git merge. Since it's the last command in the script, this makes the
script return 1 (failure), when it actually succeeded.
Add exit 0 to ensure we return 0 after successful end of the script. If
there is a failure, the script will terminate early. It can be tested by
e.g., giving a wrong commit ID.
Reviewed-by: Jaroslava Fiedlerova <Jaroslava.Fiedlerova@openairinterface.org>
Integration: `2024.w39`
Closes#846
See merge request oai/openairinterface5g!2990
* !2718 Added Jenkins job for Colosseum testing integration
* !2901 Fix log2_approx
* !2983 fix(docker): making tini download platform specific to support ARM64 native builds
* !2919 Add E2 SM in the target dockerfiles
* !2981 NR gNB fix PMI report evaluation
* !2979 NR-UE: Merging different UE NR SLOT FEP procedures into one.
* !2899 SIB19 initialization and scheduling on gNB
* !2953 Remove unnecessary c includes
* !2926 Fix SRS FAPI at gNB
* !2987 fix for variable power in PDSCH REs in case of 25 PRBs Bandwidth
* !2993 remove DRX from UE capabilities: not supported yet
* !2711 NR UE UL time alignment handling improvements
* !2995 some small fixes for 5G NR
* !2992 cleaner faster nr_pdcch_demapping_deinterleaving() inner loop
* !2986 NR PDCP ticks for F1 split
* !2982 CI framework cleanup
Add a new scope data source (ueTimeDomainSamples) which provides time domain samples
for antenna 0 to the scope. Add graphical representation of the data in imscope.
Removed C-define from ENABLE_IMSCOPE cmake option, the #define was unused and this way
code does not have to be recompiled when imscope enable state changes.
Currently, NR PDCP ticks are active only with monolithic gNB, triggered by the MAC scheduler in gNB_scheduler.c.
In F1 split mode, as there is no MAC, there’s no source to advance the NR PDCP timer via "nr_pdcp_wakeup_timer_thread()".
This commit introduces a dedicated thread for NR PDCP ticks in gNB-CU and CU-UP mode.
Without it, packet losses at the F1 uplink would cause indefinite buffering, as t-Reordering won't expire due to the lack of NR PDCP timer ticks.
Use a script, which can be tested independently of CI code (by running
it on the host on which it will be used), and print the directory that
has been prepared in the HTML. Also, in the description, use the
infinitive/imperative(?) of the verb to harmonize with other steps.
A lot of CI code is python mixed with bash, e.g.,
ssh = getConnection(host)
ssh.run('ls')
ssh.run('echo')
At least some of this CI code would benefit if it was written in a
simple bash script, returning error codes and potentially other
information either through stdout/stderr or files, to the calling Python
code:
ssh = runScript(host, script)
# script does: ls; echo
This commit introduces the possibility to run entire scripts. The idea
is that the executor has a script (on localhost), which is either
executed locally or on a remote host. For the remote host, the script is
not copied but piped into a remotely executed bash. In both cases,
output is either returned like the Cmd.run() function with returncode
and mixed stdout/stderr, or optionally redirected into a file on the
(remote) host, which can be treated further by the Python code in later
steps.
This reimplements the CI deployment functions, and makes them a bit more
robust. Concretely:
- Introduce a generic "deployment tag" option that can be used to
override the tag to use for specific images. By default, it is the
current "branch-commitID[0:8]" tag name (requires the change from
function ImageTagToUse() to CreateTag(), impacting also pull/push
image functions)
- Avoid sed for image names, and use an .env file that docker-compose
picks up automatically; the deployment analyzes a potentially existing
.env file and updates instead of simply overriding. For instance, some
pipelines might use -asan images for the gNB and "normal" (non-asan)
images for UE, and a simple overwriting might make the -asan image
name tag disappear for the gNB, resulting in deployment failures).
Finally, undeployment removes the .env file, so that there are no
modifications in the repository after undeployment.
- Redo the previous behavior of autodetecting asan, and use that (the
current function always uses asan, no matter what)
- Remove deployKind/displayedNewTags globals, as they are not necessary
- Make the usedImage output in HTML slimmer
- On undeployment, print service names next to undeploy analysis, and
return success/failure.
- Make the functions generally less verbose and easier to read
Note that as of and only in this commit, deployment does not work, as
all the YAML files have not been updated to work with this updated
version. The next commit adds tests for the new deployment, and updates
one YAML file (also being used in the tests). The follow-up commit then
modifies all YAML files.
The BuildeNB steps either does things that are already implemented
(created workspace), or things that we don't use (separate build_wait),
and is generally quite convoluted when all it does is calling build_oai.
Reimplement by reusing Create_Workspace, and make build_oai in a single
step. To clean up, make a new module in which this function does not
need a class instantiation. Instead, we give all necessary parameters
directly.
The next commit will use "echo -e". The problem with that is that bash
and sh differ:
- in bash: this results in "a\nb" (a+newline+b)
- in sh: this results in -e a\nb (-e a+newline+b)
The problem is that by default, commands are executed through sh, which
nobody expects. Change to bash, which is likely more aligned with what
people want to use.
After the previous commit, the CI will always do an undeployment step
even if the CI failed. Hence, these XMLs are not necessary anymore. Some
XMLs are updated to do what their "cleanup XML" counterparts did, such
as removing additional images.
Use the previous commit on returning success from each task execution
function, and implement handling of errors in the main loop. Add a
script with which this is visible using Custom_Commands.
For reasons I don't understand, the XML option reading and actual test
case execution are at different places in the code. This is problematic,
because what this led to is that options are not passed as function
arguments, but indirectly through class members. This led to a lot of
"indirect code"; it's a total mess
To improve this, move them together, so that parameters can be passed
directly as function arguments later on.
Because we reference the globals, mark them inside
ExecuteActionWithParam(). That is a Python antipattern, but necessary at
this moment. I believe that we can remove the following globals over
time:
- SSH (with new cls_cmd, which we don't need to pass)
- EPC (replaced through a class similar to ue_module, i.e., refer to
core networks on specific hosts through an ID, instead of a global)
- HTML (should use templating, where individual functions give back a
dictionary of what they want to print, and a final pass assembles the
entire HTML file)
- RAN/SCA/PHYSIM/CLUSTER/ldpc (can be replaced with stateless scripts
build/run eNB/gNB on a specific node)
Update the regex for both sender and receiver data to allow optional
bitrate unit prefixes - accepts also no prefix. Modify the patterns for
lost and total packets to accept negative values (`-?\d+`).
Add some basic unit tests to verify that Iperf/Ping work. Harmonize the
logging by only print in the "main" iperf/ping function, to avoid that
if we test with multiple UEs, the output intermixes. Also, do one common
mkdir before starting multiple pings/iperfs. Finally, harmonize the
paths so that Iperf and Ping log their files in the same place.
* added missing Transaction ID to te F1 Setup Failure message struct
* fill F1 Setup Failure transaction ID in RRC before calling the F1 Setup Failure function
Experiments did show that the time sync is less stable in terrestrial mode if the integral part of the PI controller is used.
So set it to 0.0 by default. For NTN LEO use case we can still use it by providing the command line parameter --time-sync-I 0.2 .
fixes#846
* _F1_EQ_CHECK_GENERIC can be used to check the equality of two different members of F1AP structs
and return `false` if the condition is not met.
When the condition fails, the macros will print a debug message to the standard error stream,
including the function name, line number, and the provided format string and arguments.
Add a new target f1ap_lib that groups F1AP ASN.1 encoding&decoding for
messages defined f1ap_messages_types.h. The objective is to, for each
F1AP message,
- add encoder&decoder (similar to what already exists, but without
mandatory message sending afterwards)
- eq_X() function that checks two messages for equality (to be used
mostly in tests)
- cp_X() function that deep-copies a message, e.g., to be used when
sending ITTI messages, instead of manual copying done currently
- free_X() function that frees the memory associated to a memory, to
avoid memory leaks.
Each message will be tested, notably that encoding&decoding as well as
copying yield identical messages.
The actual message encoding & decoding, tests, and their use will be
done in the forthcoming commits.
The currently used type does not cover all possibilities. While we don't
encode&decode them, it might still be good to be able to express it, so
introduce the new type.
Co-authored-by: Guido Casati <hello@guidocasati.com>
* num_SI is the size of the list of gNB-CU System Information messages
* the new struct member is an array, sized after the maximum nb
of SIB types, which contains the SI container, the container length
and the SIB type
The previously used net_ip_address_t type is limited to some bytes, but
the CU name might be longer. Use char * to allow longer DNS names.
Also, this type separates IPv4/v6, which is not necessary, as it is now
handled by the same C API (getaddrinfo()), since !2635.
Co-authored-by: Guido Casati <hello@guidocasati.com>
Integration: `2024.w38b`
Closes#839
See merge request oai/openairinterface5g!2977
* !2704 nr pdcp: add support for nea1
* !2800 UE improvements
* !2920 Receive SIB19 on nrUE
* !2975 Remove hardcoding of CPU measurements (respect option `-q`)
* !2976 Update latest SIMDE version
* !2957 T: increase max number of args
* !2916 NR UE PUSCH resource mapping optimization
* !2787 CI: increase max number of concurrent workers to 64
* !2974 fhi_72: allow to provide dpdk_mem_size to xran
* !2964 CI: Replacing the UE on Aerial setup
* !2973 CI: remove unused code and files
See e.g., egrep(1) [https://linux.die.net/man/1/egrep]:
> two variant programs egrep and fgrep are available. egrep is the same
> as grep -E. fgrep is the same as grep -F. Direct invocation as either
> egrep or fgrep is deprecated
See also the rationale of the POSIX description of grep
(https://pubs.opengroup.org/onlinepubs/9699919799/utilities/grep.html):
> This grep has been enhanced in an upwards-compatible way to provide
> the exact functionality of the historical egrep and fgrep commands as
> well. It was the clear intention of the standard developers to
> consolidate the three greps into a single command.
Closes: #839
As of now, these seem to be the Jenkinsfiles used on Jenkins
ci-scripts/Jenkinsfile-cluster
ci-scripts/Jenkinsfile-colosseum
ci-scripts/Jenkinsfile-GitLab-Container
ci-scripts/Jenkinsfile-GitLab-COTS-UE-Test
ci-scripts/Jenkinsfile-nrue
ci-scripts/Jenkinsfile-physim-deploy
ci-scripts/Jenkinsfile-push-registry
ci-scripts/Jenkinsfile-tmp-full-ran
ci-scripts/Jenkinsfile-tmp-multi-enb
ci-scripts/Jenkinsfile-tmp-multi-enb-nsa
ci-scripts/Jenkinsfile-tmp-ran
(gather all files in a file, then cat <file> | sort | uniq)
Remove all others.
The default values set by xran for DPDK socket memory/limit is 8192. It
is changed when the variable for the memory size, given down to xran, is
non-zero (cf also previous patch). Since the last patch, if nothing is
given in OAI configuration, OAI will pass down a size of 8192. However,
a configuration value of 0 is a valid choice (it is a pre-allocation,
DPDK will allocate dynamically), so modify the default in xran to be 0
so that all sizes, including 0, can be given to DPDK.
Note that the currently used implementation of xran will always try to
allocate this memory on NUMA node 0. If you want to use another one,
this field has to be set to 0 (no pre-allocation), and DPDK will
allocate it dynamically.
Co-authored-by: Sagar Arora <sagar.arora@openairinterface.org>
xran has a mechanism to request DPDK to allocate a configurable amount
of bytes for huge pages. We set it to 0 until now; expose this variable
through configuration, and provide a meaningful default 8192 MiB.
Provide documentation.
- refactor nr_fill_nfapi_dl_SIB_pdu to use it on both SIB1 and OtherSI
- extract tbs related calculations into function get_tbs to avoid duplicates
- moved OtherSI search space initializations to nr_mac_config_scc
Mapping of data REs in DMRS symbol depends only on number of CDM group
no data. So the special case function that depended on delta is removed
and a single function is used.
Build on ARM fails in the previous version, while the current master
(1a09d3bc) works. So let's update to a newer version to avoid problems
on ARM, and assuming x86 works.
- Better trace DLSCH decoding result
- Rename the decode result variable to separate with later ack/nack
encode of this decoding result from 3GPP (not trivial true=1 coding)
- Cleanup decodeResult usage
- Remove unused defines
Co-authored-by: Francesco Mani <email@francescomani.it>
./run_tests.sh was not functional anymore because of the introduction
of AssertFatal() in nr_rlc_entity.c
Including the .h and providing exit_function() solves the problem.
Some people needed that.
Let's stop to 19 for the time being. It's tricky enough.
We can increase later if needed. (A previous commit
made the tracers to accept up to 32 variables.)
T_MAX_ARGS was 16 and some people start to use the T machinery with
more than 16 values. So be it. Let tracers be friendly and accept 32.
(More commits to follow for full support.)
Also why is T_MAX_ARGS in T_defs.h when it's only used by event.h?
Hum?
The F1AP UE context release procedure can happen in at least two cases:
1. core requested a release of a UE
2. CU requests DU a release of a UE without a core release (e.g.,
handover)
In case 1., the CU has to fully release the UE in the CU, then ack the
NGAP message from the core, whilst in 2., nothing of that should happen
after the DU acknowledges the F1AP UE context release cmd.
cuup_notify_reestablishment() used
&ue_p->established_drbs[drb_id]
to look up a bearer, which is wrong, because it should be drb_id - 1
as in the top of the loop. Also, it used the dedicated get_drb()
function; harmonize to use only this function.
A follow-up commit refactors the filling of information of DRBs to set
up in a separate function (to be reused for handover later). Hence, we
cannot store the DRB TEID/address in the same functions, because it is
a logically separate step, and might not always be available.
In e1_send_bearer_updates(), we send the DL (DU) TEID and address to the
CU-UP. Therefore, correct the variable names to reflect that the F1-U
information pertains to the DU.
Also, move the tunnel info update function f1u_dl_gtp_update() to a more
prominent place, and take the pointer to the struct containing the
"interesting" information.
A maximum RRC buffer size of 1024 might be very small. We should
increase it, but then the 5G RRC_BUF_SIZE does conflict with other
definitions. To make it unique, also rename to NR_RRC_BUF_SIZE.
Always set the firstActive and defaultBWP in the initial
cellGroupConfig. This is necessary to make the Quectel accept the
handover RRCReconfiguration; it seems that this also works reliably in
normal connection set up, so we can add it.
Pass consistently the NR_timer_t object through a pointer. Since we have
to modify all occurrences of is_nr_timer_active(), we use that occasion
and rename that function to nr_timer_is_active() to make it consistent
with all other functions, starting with nr_timer.
Reorganize the sequence of checks done in reestablishment to make it
more logical:
- check conditions of request (correct C-RNTI)
- check conditions of DU (has MIB/SIB1, MTC)
- check the UE, and from which DU it comes
This will also help to extend it towards reestablishment coming from
different DUs in the case of handover.
Before this bug fix, when using F1, we get this warning:
Constraint validation failed: LCID: constraint failed (/home/richie/openairinterface5g/cmake_targets/ran_build/build/openair2/F1AP/MESSAGES/F1AP_LCID.c:30)
This is because (1) we don't actually store the LCID, then (2) we did
not put it into the F1 ASN.1 structures. Fix both, to make this warning
disappear.
To actually print the full error message, increase the error buffer
size. Also, for the future, to avoid guessing, print ASN.1 XML when we
encounter a failed constraint validation.
It can happen that SCTP communication is simply lost, e.g.
[SCTP] SCTP_ASSOC_CHANGE to SCTP_COMM_LOST
[SCTP] sctp_recvmsg (fd 104, len -1 ): Connection reset by peer:104
The previous code only handled the SHUTDOWN event, but in abnormal
situations, we should also handle the connection state change and signal
to RRC that the endpoint is dead, which is now implemented.
User of the SCTP module use the assoc_id for connection identification.
The previous code set the assoc_id to -1, so there is no way of telling
which connection broke. Use the assoc_id instead
The failure mode observed here was that the CU sent a release command
for a UE that the DU does not know. Given, that the outcome of the
release command (no UE) is equal to the current state (no UE), from
the DU point of view, we can safely ack the command.
This is a temporary workaround for issue with UE PHY not applying TX power
instructions from MAC. To be reverted once implementation of TX gain control
is finished.
This commit adds yaml-cpp library to required docker images and build commands. There is one workaround
in cross-compilation for apt behavior for installing libyaml-cpp-dev. Apt will uninistall previous
versions of the package whenever another one is installed. This means that installing the arm version
breaks amd version and vice versa. This makes it impossible to have both libs installed at the same
time which is required by the cross-compilation build command (it calls cmake twice, once with native
and once with target architecture)
This commit introduces a substitute module library for setting up softmodems using YAML instead of libconfig.
The softmodems should work the same with both types of files. Extra care was put to make sure the behavior
of both modules is the same.
Example yaml config files were added for the UE and gNB
pdschRxdataF_comp data source was handled incorrectly. Each symbol entire slot data was copied. This
caused pdschRxdataF_comp to mostly contain zeros. The call to UEscopeCopy was moved from symbol loop
into last symbol processing. In case imscope is used padding and unused REs are not copied to the
sink.
Added checks inside scope functions to ensure correct functionality.
A new phy scope based on ImGui and ImPlot. This scope uses a different concurrency model than previous
scopes.
The PHY thread writing the data first checks if the data is ready to be written. If its not, nothing is copied.
The GUI thread reads data if available and marks it ready to write. This makes sure that the PHY threads are not
busy copying data that would never be displayed.
Some of the scopes also have a freeze functionality that further limit the amount of data that needs to be copied
from PHY threads. If a scope is "frozen" it still allows the user to explore the data using plots zoom/pan functions
but doesn't cause PHY threads to perform extra writes on the displayed data.
A compile option was added to enable/disable the scope. Use cmake -DENABLE_IMSCOPE=ON to enable the scope.
Update CXX standard to 17; it is required by some libraries (e.g., dear imgui). The oldest gcc version in
use by a distribution supported is Ubuntu 20, which has gcc 9.4 with c++17 support.
According to the ThreadPoolExecutor documentation [1], the default
number of maximum workers min(32, os.cpu_count() + 4). The problem is
that we execute many pipelines from nano, on which os.cpu_count()
reports 4, so we can only have up to 8 concurrent workers by default.
This leads to the fact that on pipelines with many UEs (e.g.,
AW2S-AmariUE), we don't ping/iperf on all UEs at the same time.
Fix this by increasing the number of workers to 64. This should be fine,
as there is almost no real work being done in the Workers (only opening
a shell and execute programs such as ping/iperf).
[1] https://docs.python.org/3/library/concurrent.futures.html#threadpoolexecutor
Create_Workspace is used used in XML files instead of directly being
executed in DeployObject(). This can be problematic if we deploy
multiple services on the same host, since a previous workspace will be
deleted. Also, it removes futile git clones.
There are other XML steps (CppCheckAnalysis, LicenceAndFormattingCheck,
BuildImage, BuildProxy, BuildClusterImage) that implicitly called
Create_Workspace. Those calls have been removed and care has been taken
to insert the corresponding Create_Workspace steps in the XML.
There was concern that someone in the future might call the function
with the wrong inputs. So, based on the discussion on the MR below
https://gitlab.eurecom.fr/oai/openairinterface5g/-/merge_requests/2916#note_132506
the SIMD for multiplying DMRS vector with amplitude is removed and is
now done with plain C (no SIMD).
The execution time for the function dmrs_amp_mult() in this commit is
10x higher that the previous commit.
It is better to write to memory that are grouped together to reduce
probability of cache miss. This simple change improved the function's
execution speed by 70%.
Removed the unnecessary memset for the buffer.
Integration: `2024.w35`
Closes#825
See merge request oai/openairinterface5g!2943
* !2181 ULSCH power computation
* !2915 benchmark for rotate_cpx_vector
* !2936 Avoid complete rebuild on cmake run with git info change
* !2855 NR UE replace phy_config_request_sent
* !2937 small fixes for rfsim
* !2925 CI: Increase tested throughput in SA 2x2 pipeline
* !2934 Run optimized parity check in ldpc encoder only on supported Zc values
* !2912 changes for multi-RU support in fhi72
* !2928 Distribution check: check for F40, remove old U/RHEL releases
* !2913 NR gNB cleanup
* !2927 Update Aerial docs to reference current version
* !2940 remove double define with nr_common.h
At least in Eurecom, there are no RHEL 7/CentOS 7+8 servers. Existing
installations will work, but we don't want to encourage people to use
these old distributions, so remove it. Also, they reached EOL
Add CentOS stream as an alternative.
In the past, it seemed that we had to add a trailing slash (/) after the
OpenShift Registry URI. Now, at least on the CI machine matix, this
makes it fail:
INFO: oc whoami -t | docker login -u oaicicd --password-stdin default-route-openshift-image-registry.apps.oai.cs.eurecom.fr/
INFO: docker pull default-route-openshift-image-registry.apps.oai.cs.eurecom.fr/oaicicd-ran/oai-gnb:develop-46a1d2a6
WARNING: command "docker pull default-route-openshift-image-registry.apps.oai.cs.eurecom.fr/oaicicd-ran/oai-gnb:develop-46a1d2a6" returned non-zero returncode 1: output:
Error response from daemon: Head "https://default-route-openshift-image-registry.apps.oai.cs.eurecom.fr/v2/oaicicd-ran/oai-gnb/manifests/develop-46a1d2a6": unauthorized: authentication required
Remove the trailing slash makes it work, like so:
INFO: oc whoami -t | docker login -u oaicicd --password-stdin default-route-openshift-image-registry.apps.oai.cs.eurecom.fr
INFO: docker pull default-route-openshift-image-registry.apps.oai.cs.eurecom.fr/oaicicd-ran/oai-gnb:develop-46a1d2a6
The docker pull succeeds.
>Direct leak of 7 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4887 in __interceptor_malloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:145
>1 0x55555819ab3c in generatePduSessionEstablishRequest /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:938
>2 0x5555581a8a9f in nas_nrue /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1295
>3 0x5555581a3078 in nas_nrue_task /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1215
>4 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
> Direct leak of 4 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x55555819c025 in generatePduSessionEstablishRequest /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:993
>2 0x5555581a8a9f in nas_nrue /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1295
>3 0x5555581a3078 in nas_nrue_task /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1215
>4 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
> Direct leak of 1 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x55555819c8f7 in generatePduSessionEstablishRequest /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:987
>2 0x5555581a9b75 in nas_nrue /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1299
>3 0x5555581a414e in nas_nrue_task /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1219
>4 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
> Direct leak of 16 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x55555818ef1c in generateAuthenticationResp /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:588
>2 0x5555581b0fca in nas_nrue /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1405
>3 0x5555581a3069 in nas_nrue_task /home/guido/repo/openairinterface5g/fix-ue-memory-leaks/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1213
>4 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* Data buffer for NAS Registration Complete and Authentication Response is malloced in
the NAS thread and transfered with ITTI message to RRC, where needs to be freed
Direct leak of 24 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4887 in __interceptor_malloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:145
>1 0x5555572b0607 in generateAuthenticationResp /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:608
>2 0x5555572b0607 in nas_nrue /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1402
>3 0x5555572baef6 in nas_nrue_task /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1198
>4 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
Direct leak of 10 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4887 in __interceptor_malloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:145
>1 0x55555729754e in generateRegistrationComplete /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:782
>2 0x555557299ad3 in handle_registration_accept /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1212
>3 0x5555572b06a0 in nas_nrue /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1411
>4 0x5555572baef6 in nas_nrue_task /repo/openairinterface5g/develop/openair3/NAS/NR_UE/nr_nas_msg_sim.c:1198
>5 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
It is very possible that the operating system returns a file
descriptor bigger than the size of bridge->buf (MAX_FD_RFSIMU, which
is 250 as of now). As a minimum, let's fail if the socket file
descriptor is out of bounds.
A proper solution may be implemented later, if needed.
Updated links to latest available L1 documentation
Removal of unneeded environment variable in aerial_l1_entrypoint.sh
Removal of unneeded checkout command from doc
Reword Aerial Entrypoint script paragraph to separate in multiple sentences
New UE has been tested with Aerial setup
Change bash blocks to reflect that cuBB container no longer uses root
Add alternative instructions to obtain nvipc sources from container
Add note in Prepare the L1 image section specifying default config file and location
Co-Authored by: Reem Bahsoun <reem.bahsoun@openairinterface.org>
Before this commit, the following will trigger a complete rebuild:
- create another git name (git checkout -b test)
- run cmake again manually (cmake ..)
- run ninja again
The top-level CMakeLists.txt uses add_definitions() to add git
information to all targets; if some git information changes, this will
mark all command lines as dirty, triggering the build.
There is an easy solution: instead of passing this information on the
command line, pass it in a header. This commit uses configure_file() to
generate a file oai_version.h in the build directory with the same
information. Various sources files, e.g., main executables, use this to
output the same information as before. It is then also not necessary to
use header guards, as cmake takes care to create this file.
If cmake is re-run, it will update the header, and trigger only files
including this header will be rebuilt (which can be shown with the above
stpes). The information passed is the same as before, and updates the
information when the old method would have updated it.
This commit introduces google benchmark library and one benchmark for
rotate_cpx_vector. Also extracted common tools used in testcases for
generation and allocation of byte aligned vectors in phy_test_tools.hpp
Integration: `2024.w34`
See merge request oai/openairinterface5g!2923
* !2906 mod: Improve signal processing speed of rotate_cpx_vector function
* !2909 NR UE SIB1 free fix
* !2917 Cleanup NR UE NAS instance handling
* !2924 Fix misalignment issues in TLVEncoder macros ENCODE_U16, ENCODE_U24
* !2907 CI: add new testcases to RFSim-5G pipeline, reduce ping and iperf test time
* !2893 LEO satellite delay and Doppler simulation and adjusted NR_UE time sync
The correct order for xran seems to be first Uplane, then C plane. In
the past, we used to switch (for reasons I don't know), but let's
reverse it to the proper order.
* place all NR UE CL optional options in the same section, in a table
* add a section for common CL options
* add subsection for UE capabilities
* removed redundant text
Bypass UE log analysis when testing sidelink, as it fails due to the absence
of expected information and patterns (not present in sidelink communication).
Instead, a custom command is used to verify the success of sidelink tests by
checking for the presence of "PSBCH RX:OK" in the log of the second UE.
As we don't have SIB19 with continuously updated ta-Common and ta-CommonDrift values yet, the UE can only adapt it's UL timing according to it's DL drift.
This is required for the UE to keep UL sync with fast time drift and relatively long RTT (e.g. LEO satellite).
We should replace this with the standard compliant way as soon as we have SIB19 with continuously updated contents.
This commit resolves undefined behaviour runtime errors related
to misaligned memory access. E.g.
> openair3/NAS/COMMON/IES/FGSMobileIdentity.c:193:3:
> runtime error: store to misaligned address 0x608000021e2b for type 'uint16_t',
> which requires 2 byte alignment 0x608000021e2b: note: pointer points here 00 f1 ...
* ENCODE_U24: safely encode 24-bit values into a buffer
by using memcpy to copy 3 bytes in network byte order (big-endian).
* ENCODE_U16: safely encodes a 16-bit value into a buffer, handling
misalignment by memcpy 2 bytes to buffer in network byte order (big-endian).
- Save nas instance id in nr_ue_nas_t and use that for instance determination.
- Use correct field in itti message (destinationInstance instead of originInstance) to determine instance
- Make sure both destinationInstance and originInstance fields are correct in NAS interface functions.
While we don't have SIB19 support yet, UE gets this parameter from command line, just like --ntn-koffset and --ntn-ta-common.
We use this value to initialize the UE's time-tracking PI controller accumulator.
This allows to correctly track the time even if the delay is changing significantly (e.g. due to LEO satellite movement).
We also use it to compensate for the time drift that happenes while performing the initial sync.
This now allows to indicate "there was a new sampleShift calculation, and the actual value is 0"
additionally to "there was no new sampleShift calculation".
Integration: `2024.w33`
See merge request oai/openairinterface5g!2911
- !2784 Reworking beam structures at NR MAC gNB
- !2889 Fix delta for PUSCH DMRS type 2
- !2908 NR MAC UE CSI improvements
- !2896 Fix for llr functions reading/writing outside buffers
- !2902 nr rlc: reduce "SDU rejected" logging
- !2884 NR UE: Cleanup global variables used for command line parameter storage
- !2910 NR gNB DMRS symbol race
- CI: Update phytest-timing test threshold
This fixes out-of-bounds access: nr_ulsch_16qam_llr, nr_ulsch_64qam_llr and nr_ulsch_256qam_llr.
For 256qam this also fixes incorrect llr calculation on arm: the existing AVX code for 2 REs
case produced results not in line with the rest of the code. A testcase check_2_res_256_qam was
added to visualise the differences which can later be used to revive AVX acceleration for 2REs case.
Integration: `2024.w32`
See merge request oai/openairinterface5g!2900
* !2810 add harqid in wireshark traces, refactor the trace_pdu parameters
* !2897 nr rlc: add a lock to protect access to time variables
* !2903 remove some openair1 headers included in opeinair2 files
* !2876 Enable 32 HARQ Processes in DL, UL
* !2849 Fix UE SRS crash
* !2898 Fix memory leaks in OAI gNB and OAI UE
* In LTE there is logic that checks the thread configurtion with the
getters `get_thread_parallel_conf()` and `get_thread_worker_conf()`
in NR these getters are not used, only setters, which have no impact
* buffer is allocated but not freed after use by PDCP
> Direct leak of 192 byte(s) in 3 object(s) allocated from:
> 0 0x7ffff74b4c38 in __interceptor_realloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:164
> 1 0x55555775ee74 in encode_dyn_cb /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/per_encoder.c:23
> 2 0x55555774776e in _uper_encode_flush_outp /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_encoder.c:126
> 3 0x55555774776e in uper_encode /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_encoder.c:36
> 4 0x555557747c91 in uper_encode_to_new_buffer /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_encoder.c:85
> 5 0x5555576c5172 in do_NR_ULInformationTransfer /openair2/RRC/NR/MESSAGES/asn1_msg.c:965
> 6 0x555556fba691 in rrc_nrue /openair2/RRC/NR_UE/rrc_UE.c:1819
> 7 0x555556fc4d2e in rrc_nrue_task /openair2/RRC/NR_UE/rrc_UE.c:1717
> 8 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* bcch_message is malloced by ASN.1 decoder and needs to be always freed after use along with its contents
> Direct leak of 64 byte(s) in 1 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x555557737ecd in SEQUENCE_decode_uper /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/constr_SEQUENCE_uper.c:37
>2 0x555557746f81 in uper_decode /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_decoder.c:83
>3 0x5555577471fe in uper_decode_complete /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_decoder.c:16
>4 0x555556f8b1c9 in nr_rrc_ue_decode_NR_BCCH_DL_SCH_Message /openair2/RRC/NR_UE/rrc_UE.c:732
>5 0x555556fbc365 in rrc_nrue /openair2/RRC/NR_UE/rrc_UE.c:1775
>6 0x555556fc4c2e in rrc_nrue_task /openair2/RRC/NR_UE/rrc_UE.c:1718
>7 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* strdup is returning a malloced string that needs to freed after use
> Direct leak of 90 byte(s) in 1 object(s) allocated from:
>0 0x7ffff745b9a7 in __interceptor_strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:454
>1 0x555556c7b341 in main /executables/nr-uesoftmodem.c:461
>2 0x7ffff5e29d8f in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
* local dl_dcch_msg shall be freed after use
> Direct leak of 384 byte(s) in 6 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x555557737ced in SEQUENCE_decode_uper /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/constr_SEQUENCE_uper.c:37
>2 0x555557746da1 in uper_decode /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/uper_decoder.c:83
>3 0x555556fa4b31 in nr_rrc_ue_decode_dcch /openair2/RRC/NR_UE/rrc_UE.c:1612
>4 0x555556fbb25c in rrc_nrue /openair2/RRC/NR_UE/rrc_UE.c:1792
>5 0x555556fc4a4e in rrc_nrue_task /openair2/RRC/NR_UE/rrc_UE.c:1718
>6 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* The RRC task is freeing the ITTI message but the message members that are allocated separetely still need to be freed
> Direct leak of 470 byte(s) in 6 object(s) allocated from:
>0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
>1 0x55555721f117 in itti_malloc /common/utils/ocp_itti/intertask_interface.cpp:64
>2 0x555556f2356c in deliver_sdu_srb /openair2/LAYER2/nr_pdcp/nr_pdcp_oai_api.c:775
>3 0x555556f4e10e in nr_pdcp_entity_recv_pdu /openair2/LAYER2/nr_pdcp/nr_pdcp_entity.c:184
>4 0x555556f278ad in do_pdcp_data_ind /openair2/LAYER2/nr_pdcp/nr_pdcp_oai_api.c:285
>5 0x555556f278ad in pdcp_data_ind_thread /openair2/LAYER2/nr_pdcp/nr_pdcp_oai_api.c:307
>6 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* strdup is returning a malloced string which needs to be freed after use
> Direct leak of 90 byte(s) in 1 object(s) allocated from:
> 0 0x7ffff745b9a7 in __interceptor_strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:454
> 1 0x5555591598a3 in main /executables/nr-softmodem.c:667
> 2 0x7ffff5e29d8f in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
* allocate memory in heap for the struct ue_CapabilityRAT_Container used locally for buffer encoding
* free the struct members only for the ul_dcch_msg allocated on the stack
> Direct leak of 40 byte(s) in 1 object(s) allocated from:
> 0 0x7ffff74b4a57 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:154
> 1 0x555556f7a89f in nr_rrc_ue_process_ueCapabilityEnquiry /openair2/RRC/NR_UE/rrc_UE.c:1861
> 2 0x555556fa4d85 in nr_rrc_ue_decode_dcch /openair2/RRC/NR_UE/rrc_UE.c:1650
> 3 0x555556fbb33c in rrc_nrue /openair2/RRC/NR_UE/rrc_UE.c:1790
> 4 0x555556fc4b2e in rrc_nrue_task /openair2/RRC/NR_UE/rrc_UE.c:1716
> 5 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
* strdup() function is returning a malloced string
which must be freed after use
* replace the strdup with direct use of strings to avoid heap allocation
> Direct leak of 23 byte(s) in 1 object(s) allocated from:
> 0 0x7ffff745b9a7 in __interceptor_strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:454
> 1 0x555559e36f87 in RCconfig_nr_parallel /openair2/GNB_APP/gnb_config.c:2054
> 2 0x555559e38ea3 in NRRCConfig /openair2/GNB_APP/gnb_config.c:2097
> 3 0x555559155542 in get_options /executables/nr-softmodem.c:423
> 4 0x555559155542 in main /executables/nr-softmodem.c:623
> 5 0x7ffff5e29d8f in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
> Direct leak of 14 byte(s) in 1 object(s) allocated from:
> 0 0x7ffff745b9a7 in __interceptor_strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:454
> 1 0x555559e370c5 in RCconfig_nr_parallel /openair2/GNB_APP/gnb_config.c:2063
> 2 0x555559e38ea3 in NRRCConfig /openair2/GNB_APP/gnb_config.c:2097
> 3 0x555559155542 in get_options /executables/nr-softmodem.c:423
> 4 0x555559155542 in main /executables/nr-softmodem.c:623
> 5 0x7ffff5e29d8f in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
- num DL/UL harq processes, bits for harq id in DCIs, set according to IEs configured in RRC messages
- removed going through all the harq processes before triggering nr_ue_ulsch_procedures in phy_procedures_nr_ue.c
- did not remove usage of MAX_HARQ_PROCESSES in init phase
- Enabled configuration for number of dl, ul harq processes in conf file
- set "num_dlharq = 32; num_ulharq = 32;" under "gNBs" section in the conf file
- default values when not configured is 16 for DL, UL harq processes
- IEs in the RRC messages set according to the number of DL, UL harq processes configured.
- If value 32 is configured , then IEs related to num harq processes and dci bits also configured in Bandwidth parts.
- MAX HARQ PROCESSES set to 32
- for PDSCH - default setting in OAI is 16, according to 3GPP rel17 38.331 these are the possible configs
- if no IES are defined then 8
- if nrofHARQ-ProcessesForPDSCH is defined then these values- {n2, n4, n6, n10, n12, n16}
- if nrofHARQ-ProcessesForPDSCH-v1700 is defined then 32
- for harq bits in DCI
- DCI 1_0 - 4 bits spec 38.212, section 7.3.1.2.1
- DCI 1_1 - 4/5 bits according to IE harq-ProcessNumberSizeDCI-1-1, spec 38.212, section 7.3.1.2.2
- for PUSCH - default setting in OAI is 16, according to 3GPP rel17 38.331 these are possible configs
- if no IES are defined then 16
- if nrofHARQ-ProcessesForPUSCH-v1700 is defined then 32
- for harq bits in DCI
- DCI 0_0 - 4 bits spec 38.212, section 7.3.1.1.1
- DCI 0_1 - 4/5 bits according to IE harq-ProcessNumberSizeDCI-0-1, spec 38.212, section 7.3.1.1.2
- FOR HARQ32 , DCI 1_1, DCI 0_1 should be configured in UE dedicated search space. In OAI it is set.
Integration: `2024.w31`
See merge request oai/openairinterface5g!2894
* !2812 NR UE MAC PRACH scheduling fixes
* !2885 Fix signal_energy_nodc for lengths that are not multiples of 4
* !2888 nr rlc: fix decode_poll_byte()
* !2892 Dockerfiles: correct ENV format
* !2891 UE cmd line info: print threequarter-sampling param if used
* !2837 replacing assertions with LOG_E in UE UL locking mechanism
* !2879 style(docker): newer versions of docker require keyword as to be of same case as from
* !2832 Enable PHR reporting for NR UE
* !2880 CI: Improve error handling in Iperf3 JSON analysis functions
* !2813 Rewrite time_stats_t counters for DLSCH in NR UE.
* !2882 Automatic scaling to PUCCH2 receiver
* !2691 NR UE data inactivity timer
* !2848 Refactor network interface handling
* !2865 CI: Create FHI 7.2 pipeline
- Modify CN deployment and termination functions to include namespace and path parameters.
- Update `Module_UE` class with new attributes for namespace and CN path.
- Integrate `cnID` for OC CN deployment and termination in `epc.py`.
- Extract `cn_id` from XML configurations for CN operations.
- Update XML files to include `cn_id` tag for 5G Core initialization and
termination. (Used in OAIUE and AW2S pipeline, for OC CN deployment. This change
will not affect other CN deployments.)
Create directories for gNB deployment with VVDN and Benetel550 RUs.
Each directory includes scripts for setting up and cleaning up the
host machine running the gNB.
Handle IPv6 (and IPv4v6) PDU sessions by correctly instructing
nas_config() to set up the interface with provided IP addresses.
Note that as of now, IPv6 PDU sessions have not been fully tested.
Remove a useless assert at the gNB; to my understanding the gNB does not
care about the IP session type (at least for IP), as it just
transparently tunnels the IP packets from/to GTP.
I added a comment in the code where to change to request a PDU session
different from IPv4. Currently it is hardcoded to IPv4, and it seems
that a default OAI 5GC config would reject a PDU session type different
from IPv4.
- Read interface flags before setting status to not inadvertently
overwrite existing flags
- Set interface flag point-to-point, remove multicast (I think a UE is
not supposed to send anywhere else than to the UPF, so it seems to not
make sense to declare the interface as multicast)
These rules & route are necessary to properly send packets on oaitun_ue1
interface:
1. This forces the packets coming from a subnet different than the UE's
subnet to go back through oaitun_ue1 rather than via what the default
route defined on the system (e.g., important if ping from the
internet arrives).
2. On machines setting net.ipv4.conf.oaitun_ue1.rp_filter=1 (e.g.
RHEL), this prevents that source filtering for packets coming back is
applied and those packets be dropped. By default, many hosts have
rp_filter=2, so no strict source filtering is applied, and it would
work; on others, this rule prevents source filter dropping.
Prior to this commit, there was a global variable baseNetAddress that
could be set independently through a setter and also through the
configuration module. This baseNetAddress (16 bits IPv4) would then be
complemented with two more bytes in nas_config().
However, this is counter-productive, as not only we have a global
variable that is avoidable (we can give the entire address to
nas_config() directly), but it also would not work with IPv6. Hence,
modify to give nas_config() the entire address. A follow-up commit will
add IPv6 support.
- refactor functions to make them shorter
- provide doxygen strings
- open socket for operation once instead of in each function
- use named constant (from enum) to differentiate interface UP/DOWN
- Linux kernel defines name length as IFNAM_SIZE, so use that for
interface name length. Also, it seems no null-byte is needed.
I initially planned to entirely remove the netmask configuration.
However, in specific configurations, e.g. in 4G Rfsim with feMBMS,
routing does not seem to work anymore using these commands:
iperf -B 10.0.2.2 -s -u -i1
iperf -c 10.0.2.2 -B 10.0.2.1 -u -t 30 -b 2.00M -i1
Therefore, we simplify by hardcoding the netmask to /24, which allows
the above to work.
Remove the broadcast address from the UE, as logically, a UE has nothing
to broadcast: it can only directly talk to the core. Additionally, at
least as of now, the UE interface is shown as "POINTOPOINT", so
broadcasting is not possible.
Does the same change in the _mbms() functions as in the previous commit.
Also, simplify logic in netlink_init_mbms_tun(): directly use the id in
the interface. This requires a slight ID change in the users, which is
also done in this commit.
Instead of implicitly using a hardcoded prefix "oaitun_" interface
name, move that out to the callers of these functions to make it clear.
The ID still remains outside.
The "sister functions" nas_config_mbms()/netlink_init_tun_mbms() will do
the same change in the next commit.
Comparing both functions, it seems that nas_config_mbms_s1() sets the
same parameters, assuming that in nas_config_mbms(), we set
- (UE_NAS_USE_TUN || ENB_NAS_USE_TUN) yielding true -> should yield the
same interface name
- netmask is hardcoded to 255.255.255.0
- baseNetAddress is hardcoded 10.0
So replace the call with nas_config_mbms() by setting parameters
accordingly.
- Remove unused NAS_config(): It does most of what the highly similar
nas_config() does.
- Remove unused set_gateway()
- Remove unused blocking_NAS_config()
- Remove unused functions getNetMask()
- Remove unused broadcast functionality: functions setBroadcastAddress()
and getBroadcastAddress(), as well as global variable broadcastAddr
are not used in consumers. Remove it.
- Make functions static
I don't know about any user of the kernel modules. It is effectively
legacy code we carry around, for no good reason. Remove it for the
moment. If we ever need it, we can revert.
Remove netlink code, and the PDCP_USE_NETLINK macro, as there is no
known use, it makes the code more complicated, and is simply "technical
debt".
In 4G PDCP, a subtle change has been introduced, in pdcp_module_init()
of pdcp.c. Looking at the diff of this commit, it can be seen that there
is now a new "else-if (ENB_NAS_USE_TUN_W_MBMS_BIT)" branch that is under
what was the "if (PDCP_USE_NETLINK)" branch, while the code fragment was
actually inside the "else" branch of PDCP_USE_NETLINK. This seems to be
necessary: previously, if we did not use noS1 mode and we don't use any
TUN (i.e., the eNB), the else block would set up this interface. It is
important that this interface is brought up, because if we omit it,
RFsim will not work. It seems to me that this interface will regularly
create some traffic; without it, RFsim will be blocked, and the UE
cannot connect.
In short, it seems to be important to have the MBMS interface, which
injects traffic in the RFsim side, which will unblock RFsim. The change
in pdcp_module_init() preserves the total changes of the if-else
(PDCP_USE_NETLINK).
Tests showed that we sometimes exceed the response window. In that case,
the UE cannot connect anymore, as do-ra uses contention-free RA, and
prefills a RA process. Avoid this by giving more time in the CI test.
Remove function pointers that were used in the past to implement the
F1(-U) interface in 4G, i.e., to have different functions to handle
split/monolithic operations. As we don't have F1 in 4G anymore (see
commit 9dc2282249), we don't need the
function pointers anymore.
On that occasion, rename the 5G functions by prepending nr_ to them;
this allows to refactor the interface in 5G further (which is not the
focus of this changeset).
Integration: `2024.w30`
See merge request oai/openairinterface5g!2877
* !2830 NR feature set update
* !2868 Fix support for USRP UHD Versions => 4.5.0.0
* !2870 fix(docker): base image for target oai-gnb-aw2s shall be jammy
* !2874 Fix memory leaks in gNB PDU Session Setup Request processing
* !2866 RRC: reject PDU sessions if security is not active
* !2829 Automatic update of USRP RX gain during initial search using --agc option
* !2878 USRP: add a T trace to record the TX IQ data on antenna 0
* !2633 Handle F1 Reset at DU, and ensure state re-init on F1Setup
* !2864 Fix an assert in power control
* !2754 Introduce gold sequence cache
38.473 requires to reset UE-related contexts in F1 Setup Request, so do
it.
However, avoid resetting UE states in phy_test: it has preconfigured UE
which we should not reset. Note that if NSA/phy-test used F1, we could
likely reset the UE state here, as the F1 UE context setup would come
afterwards.
Add gold sequence cache with dynamic size, private to each thread. It
periodically reorders by call rate. To free the gold sequences after end
of life time of a thread, use the pthread API.
This assert and assumption is incorrect, according to 38.101-1 UE channel is defined only by the BWP size
and BWP size can be any value supported by the RRC parameters defining the BWP.
To determine the UE channel the smallest bandwidth supporting the BWP is selected. This is an assumption based
on references in the same document to UE channel bandwidth being equal to one of the values in set
{5, 10, 15, 20, 25, 30, 35, 40, 50, 60, 70, 80, 90, 100}.
Also:
- calculated P_CMIN during BWP setup since the parameter doesn't depend on DCI.
- initiialize uniqCfg structure in test_nr_ue_power_procedures main function which enables LOG prints during tests
The next commit implements the release of all UEs following the F1 Reset
message. It uses nr_mac_release_ue(), which as of now, triggered a
release complete message automatically, which we don't want in the case
of F1 Reset.
This commit refactors the code to split UE release at MAC/RLC and
release complete message sending to allow to release UEs without sending
the complete. Corresponding users are updated.
* aper_decode allocates memory for for local pointer `pdusessionTransfer`, in `decodePDUSessionResourceSetup`
* this memory seems not to be freed but its members are by:
`ASN_STRUCT_FREE_CONTENTS_ONLY(asn_DEF_NGAP_PDUSessionResourceSetupRequestTransfer,pdusessionTransfer);`
which btw in its header says: "AVOID using it in the application code; Use a safer ASN_STRUCT_RESET() instead"
* the following mem leak was detected by ASAN:
```
Direct leak of 72 byte(s) in 1 object(s) allocated from:
*0 0x7ffff74b4a57 in __interceptor_calloc /src/libsanitizer/asan/asan_malloc_linux.cpp:154
*1 0x55555cb306df in SEQUENCE_decode_aper /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/constr_SEQUENCE_aper.c:36
*2 0x55555ca9fa28 in aper_decode /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/aper_decoder.c:78
*3 0x55555b99bae6 in decodePDUSessionResourceSetup /openair2/RRC/NR/rrc_gNB_NGAP.c:273
*4 0x55555b99fec2 in trigger_bearer_setup /openair2/RRC/NR/rrc_gNB_NGAP.c:357
*5 0x55555b9b54bf in rrc_gNB_process_NGAP_PDUSESSION_SETUP_REQ /openair2/RRC/NR/rrc_gNB_NGAP.c:830
*6 0x55555b936871 in rrc_gnb_task /openair2/RRC/NR/rrc_gNB.c:2428
*7 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
```
* This commit replaces ASN_STRUCT_FREE_CONTENTS_ONLY with ASN_STRUCT_FREE, which will free the memory allocated for the entire struct and its members
* fill_DRB_configList_e1 is filling `DRB_configList->list` then passing to PDCP
* the contents of the struct are allocated but seem not to be freed
* the following mem leak was detected by ASAN
```
Direct leak of 32 byte(s) in 1 object(s) allocated from:
*0 0x7ffff74b4c38 in __interceptor_realloc /src/libsanitizer/asan/asan_malloc_linux.cpp:164
*1 0x55555caa6a3e in asn_set_add /cmake_targets/ran_build/build/openair2/RRC/LTE/MESSAGES/asn_SET_OF.c:27
*2 0x55555c8b8bb6 in fill_DRB_configList_e1 /openair2/LAYER2/nr_pdcp/cucp_cuup_handler.c:42
*3 0x55555c8be6eb in e1_bearer_context_setup /openair2/LAYER2/nr_pdcp/cucp_cuup_handler.c:189
*4 0x55555d329668 in cucp_cuup_bearer_context_setup_direct /openair2/RRC/NR/cucp_cuup_direct.c:31
*5 0x55555b9a2c37 in trigger_bearer_setup /openair2/RRC/NR/rrc_gNB_NGAP.c:437
*6 0x55555b9b54bf in rrc_gNB_process_NGAP_PDUSESSION_SETUP_REQ /openair2/RRC/NR/rrc_gNB_NGAP.c:830
*7 0x55555b936871 in rrc_gnb_task /openair2/RRC/NR/rrc_gNB.c:2428
*8 0x7ffff5e94ac2 in start_thread nptl/pthread_create.c:442
```
* using ASN_STRUCT_RESET to free the memory used by the members of the structure
without freeing the structure pointer which is allocated on the stack
- if search fails, keeps increasing the rx gain (set by --ue-rxgain) by
3 db until USRP max rx gain
- if initial sync is success, sets the rxgain to the value of
(TARGET_RX_POWER (50) - pbch digital power in dB/RE)
OAI 5GC seems to have a bug, and occasionally might send a PDU session
setup request without having sent a UE context setup request, resulting
in no security context activated for a UE. This lead to an assertion in
rrc_gNB_process_e1_bearer_context_setup_resp().
To prevent the gNB from stopping, check if security is active before
triggering PDU sessions. If not, the gNB will reject PDU sessions.
Integration: `2024.w29`
See merge request oai/openairinterface5g!2867
* !2854 Fix NSA test
* !2715 Aerial multi l2
* !2798 CI: Create Iperf2 unidirectional function
* !2859 Switch to BWP where RA is configured when Msg3 contains MAC CE for C-RNTI
* !2861 7.2 xran: notification generation code in ru_thread excluded from compilation
* !2853 Cleanup executables focusing on NR UE
* !2860 Fix segfault in SRS modulation when ue-nb-ant-tx (fp) does not match UE capabilities
* !2863 Add configuration for CBRS RU
* !2749 immediately reuse UL/DL HARQ ID and don't wait for feedback if disable_harq == 1
* the mismatch between UE capabilities, passed to PHY by NFAPI PDU, and --ue-nb-ant-tx
causes a segfault in SRS modulation
* issue with indexing txdataF[p_index], which fails due to an out-of-bounds access
since first dimension of txdataF is sized based on ue->frame_parms.nb_antennas_tx
* N_ap used in SRS modulation is based on the number of antenna ports
in the SRS configuration (srs_config_pdu->num_ant_ports)
sched_ul_bytes is reduced already in case the last HARQ round fails through handle_nr_ul_harq() -> abort_nr_ul_harq().
Therefore, the reduction in nr_rx_sdu() is wrong, and btw. it wrongly assumes the max. UL HARQ rounds to 4.
So simply remove this peace of code in nr_rx_sdu().
sched_ul_bytes tracks the volume of in-flight data.
When data is successfully received, sched_ul_bytes is decreased in nr_rx_sdu().
In case of reusing HARQ processes, the tb_size stored there is not correct.
Therefore use the actual SDU size instead.
Don't break after first occurence of status template (interval "0.0- 1.0 sec").
With this modification we get final result, which is at the last line of log
(for interval "0.0-30.2 sec").
Remove code related to "rfsim4g_enb_fembms" testcase from Iperf_Module() -
replaced by Iperf2_Unidir(). Use copyin() for both remote and local
commands - no need to check if running on localhost anymore.
Create directory for log collection on executor node -
used for log collection in pipelines running on localhost.
Copy the iperf3 report to the executor machine for log collection.
The report needs to be copied to two different directories to ensure
proper log collection across all pipelines. Using absolute paths
ensures compatibility with both remote and local commands.
making aerial shm prefix configurable (so we can use muliptle VNFs per PNF)
making the nvipc thread affinity configurable and using threadCreate.
removing aerial_vnf_nr_aerial_p7_start_thread
In NSA, the RA process contains the RNTI of the UE to add. The previous
implementation would not remove the RNTI in NSA/for CFRA, which is
unclear (why should it stay? RA is over). Also, we did not clear all
contents, which masked a bug, as we were reusing an RA process for a
subsequent attach, leading to a segfault on reattach. That is fixed in
this commit.
Also, we were using the RA structure after having cleared it in
_nr_rx_sdu(), which semantically makes no sense.
To harmonize handling of the preambles: completely reset the RA
processes inside nr_clear_ra_proc(). For SA, also add the initialization
of allowed preambles.
Finally, this commit fixes a memory leak: we always (also in SA) fill a
list of preambles in a dynamically allocated preamble list. Since the
total list is only up to 64 preambles of 1B each, put it directly into
the struct instead of dynamically allocating it.
Remove the secondaryCellGroup variable in RRC, as it is not used there.
Note that in the case of F1, the secondaryCellGroup entirely belongs to
the DU, i.e., MAC. Also, we don't do anything with it in RRC.
NSA is quite verbose, some messages are large and it is not very
"interesting" for the average user. The logs can be enabled on demand,
or directly seen in Wireshark instead.
The RRC UE context has the RRC UE ID. Use that instead of hardcoding
something that might be wrong (because the RRC UE ID can be reused,
whereas the counter just blindly increases).
The next commit will use that RRC UE ID to free the PDCP context for the
UE. It might be wrong on reattach otherwise, because the RRC UE ID could
again be 1.
Variable spCellConfig is only used in function rrc_add_nsa_user(). Make
it local to this function to prevent possible bugs (limit the possible
scope as much as possible).
Variable reconfig is only used in function rrc_add_nsa_user(). Make it
local to this function to prevent possible bugs (limit the possible
scope as much as possible).
Simplify code by returning an allocated object to a CG_Config. It leads
to less code and is easier to reason about, because each call will
return a new CG_Config, which can be freed, irrespective of other calls
to fill_default_reconfig().
Simplify code by returning an allocated object to an RRCReconfiguration.
It leads to less code and is easier to reason about, because each call
will return a new RRCReconfiguration, which can be freed, irrespective
of other calls to fill_default_reconfig().
get_default_secondaryCellGroup() would "sanitize" the servingCellConfig,
i.e., remove BWP that are invalid. However, it is difficult to
understand (we can do sanitization once at the beginning) and reason
(why the servingCellConfig is modified in
get_default_secondaryCellGroup(), because all we want is a new
CellGroupConfig, without modifying other data structures).
Instead, do sanitization once after reading the configuration in
gnb_config.c. Also, modify physical simulators nr_dlsim and nr_ulsim to
apply this sanitization before a call to
get_default_secondaryCellGroup(), as the latter would have done the
sanitization.
This reverts commit 4c658cdd50.
This commit was introduce to signal a known segfault. The bug has been
fixed in the last commit, so this commit can be reverted as the
AssertFatal is not necessary anymore.
Prior to this commit, upon a release of a UE at the gNB in NSA, the gNB
fails with a segv. This is because the servingCellConfigCommon is used
in the reconfigurationWithSync (through a simple pointer). At the
release, the entire secondaryCellGroup is freed, leading to a free of
the servingCellConfigCommon. The gNB then tries to use it (e.g., for
looking up the numerology), and leading to a segv.
To avoid this problem, make a "deep copy" of the ServingCellConfigCommon
so that we can safely free the UEs secondaryCellGroup, without freeing
the initial ServingCellConfigCommon.
Switch to BWP where RA is configured, typically in the InitialBWP
At this point, MAC CE for C-RNTI received at gNB, UE already switched and triggered RA in that BWP, need to do BWP switching also at gNB for that C-RNTI
Integration: `2024.w28`
See merge request oai/openairinterface5g!2856
* !2847 UE setting wrong frequency in 5G rfsimulator SA test
* !2850 NR UE MAC DCI 00 BWP handling
* !2852 \[E2 agent\] Move UE RRC_CONNECTED state signaling
* !2851 Minor code cleanup and fixes
* !2842 UE assertion for msg1_SubcarrierSpacing in RA
* !2831 PUSCH power control state
* !2838 NR UE UL FAPI PDU length fix
* !2723 add initial NTN support for NR UE
* !2764 Adding 7.2 containerfiles and docker-compose
* !2827 Increase RX_DATA.indication PDU Length to 32 bits
Correct the array by taking out a verification entry in the verification
array that led to runtime error
Assertion (numparams == (sizeof(cmdline_CheckParams)/sizeof(*(cmdline_CheckParams)))) failed!
In get_common_options() /home/richie/openairinterface5g/executables/softmodem-common.c:109
Error in arrays size (41!=42)
- Add dedicated target docker files for 7.2, both RHEL and Ubuntu
- in existing dockerfiles, update to also build E2 SMs
- for UBI images, only the entitlements are necessary, the rest can be
omitted (it is superfluous)
Signed-off-by: arora <sagar.arora@eurecom.fr>
Co-authored-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>
The actual propagation delay should not be smaller than the one specified.
A too short propagation delay might result in too early reception of Msg1, and with that in a mis-detection of the RAPID.
Until commit d74b2bd19f, the PSS detection step size was 4, anyways.
The PSS detection step size is a compromise between detection duration and accuracy.
The most accurate detection step size would be 1, but this would lead to even longer initial sync time.
- E2SM UE ID requires guami -> therefore RRC_CONNECTED state signaling
to RIC is now done after receiving Initial Context Setup Request
from AMF, which contains the guami info
According to the RRC specification, cellSpecificKoffset is:
Scheduling offset used for the timing relationships that are modified for NTN (see TS 38.213 [13]).
The unit of the field K_offset is number of slots for a given subcarrier spacing of 15 kHz.
If the field is absent UE assumes value 0
This parameter `cellSpecificKoffset_r17` can be set via UE command line parameter `--ntn-koffset`.
Integration: `2024.w27`
See merge request oai/openairinterface5g!2839
* !2781 Fix several ASN.1 message related memory leaks in NR
* !2790 docker compose override files for testing with locally modified executables.
* !2835 Demote logs for NR band and duplex spacing
* !2822 security in re-establishment
* !2834 Fix CN tutorial SD value
* !2843 Delete assertion for PLMN in RRCSetupComplete
* !2771 NR UE PDSCH-CSI overlap
* !2721 T2 card: LDPC encoder/decoder TB processing
* !2828 Bugfix to TX_DATA.request PDU_Length computation, change input parameter to uint32_t
* !2841 Reduce number of LOG prints in NR UE log
* !2844 phy-test mode ULSCH schdeuler hotfix
* !2765 Fix issues blocking multiple UE in one process
* !2826 NR gNB ULSCH fixes
* !2845 Another phy-test ULSCH fix (missing TDA assignment)
* !2840 chore(build): removing dependency to libatlas3
address sanitizer fails with this error upon starting the DU:
==7==ERROR: AddressSanitizer: odr-violation (0x7f8ed2594a00):
[1] size=4 'log_mem_write_side' /oai-ran/common/utils/LOG/log.c:78:14
[2] size=4 'log_mem_write_side' /oai-ran/common/utils/LOG/log.c:78:14
These globals were registered at these points:
[1]:
#0 0x7f8ed5092658 in __asan_register_globals ../../../../src/libsanitizer/asan/asan_globals.cpp:341
#1 0x7f8ed570d47d in call_init elf/dl-init.c:70
[2]:
#0 0x7f8ed5092658 in __asan_register_globals ../../../../src/libsanitizer/asan/asan_globals.cpp:341
#1 0x7f8ed3fe3eba in call_init ../csu/libc-start.c:145
#2 0x7f8ed3fe3eba in __libc_start_main_impl ../csu/libc-start.c:379
==7==HINT: if you don't care about these errors you may set ASAN_OPTIONS=detect_odr_violation=0
SUMMARY: AddressSanitizer: odr-violation: global 'log_mem_write_side' at /oai-ran/common/utils/LOG/log.c:78:14
So we do as suggested and disable ODR violations to make the test pass.
A proper solution would be to link this function only once.
The total_length parameter in compute_PDU_Length being uint16_t causes
the passed value to be possibly truncated, which can cause the
PDU_Length to be incorrectly calculated.
This change prevents that truncation, since it now matches the size
of the values passed into the function where it is being called.
- r_amf->plmn_Identity was interpreted incorrectly, therefore remove it
- selected_plmn_identity selects one of the list of PLMNs; at the
moment, OAI supports only 1 PLMN ID; print a warning if it is
different, and proceed with index 0
More information in issue #801
- make it work for SRB or DRB
- adapt callers of this function
- cleanup key derivation: the User Plane keys are generated at some places
but are not used, remove the generation
Integration: `2024.w26` (part 2)
See merge request oai/openairinterface5g!2823
* !2794 Power control changes
* !2816 Fix test asan errors
* !2820 fix(T): some missing dependencies on T
* !2714 FAPI P5 Refactoring and unitary test creation
* !2819 Prevent unnecessary assertions for invalid DCI at NR UE
* !2825 Do not extract CSI report if PUCCH format 2 CRC check fails in polar decoder
* !2805 GUAMI decoding and storing in RRC UE context fixed
This fix prevents the scheduler to process a wrong CSI feedback message, for example, it avoids the scheduler to compute a wrong MCS from CQI or a wrong RI from CQI
- Introduced command line and configuration options to set up PCI address, CPU cores used by DPDK, and DPDK file prefix for T2 offload.
- Made the dpdk_dev parameter (for setting the PCI address of the T2) mandatory.
- Configure dpdk_dev in the CI test cases.
- Replaced bbdev TEST_ASSERTs with the OAI macro AssertFatal() to harmonize error handling.
- Removed unused functions and macros from nrLDPC_offload.h.
- Reworked nrLDPC_decoder_offload.c to process all segments in a TB with a single call to LDPCdecoder() or LDPCencoder().
- Perform CRC checks on the T2.
- Modified nr_ulsch_decoding (gNB decoder) to:
- Count processed segments.
- Properly perform CRC checks.
- Created a constant NR_LDPC_MAX_NUM_CB to define the maximum number of codeblocks.
- Conditional library loading:
- On gNB: Load ldpc_t2 library if offload is enabled using the --ldpc-offload-enable flag.
- On UE: Load both ldpc_t2 and ldpc libraries since only encoder offload is supported.
- General cleanup of nrLDPC_decoder_offload.c for improved readability and maintainability.
- Modified the structure of LDPC encoder/decoder/offload parameters:
- Introduced a structure for code block (CB) related parameters.
- Removed parameter E from the encoder and offload parameter structures for clarity.
- Replaced E with perCB->E_cb in the UE encoder code.
This reverts a change in commit
ccba7c875d. There is no real reason to
show this warning (the gNB will work just fine, simply not schedule
since no free HARQ) and it makes users unsure.
- come straight to point: how to clone (don't talk about git config)
- refer to contributing documentation where necessary instead of
repeating information
- CellGroupConfig is handled through DU (hence MAC)
- mention number of DUs/CU-UPs handled per CU-CP
- mention number of UEs handled
- take out BWPs, as we don't support it properly
- typos
Update instructions and CI for cross-compilation on Ubuntu 22.
Note that for the dockerfile, we use egrep -L which has changed and does
not exit with code 0 on no match. Instead, use -vqz to invert return
code (0 if nothing found, otherwise 1).
iperf2 2.0.13, used in Ubuntu 20, does not need a "back channel"; we can
therefore use it in feMBMS (which does a broadcast, without back
channel). Ubuntu 22 comes with 2.1.5, which seems to require a back
channel, or at least, we cannot use it reliably in feMBMS. Hence,
downgrade to older iperf2 version.
Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>
This updates all images in the CI to Ubuntu 22. It also updates
corresponding documentation. Remove python(2) from the image, as it is
not needed. Update documentation.
It seems that the gcc coming in Ubuntu 22 does not like
stack-protector-strong, so remove it as well.
Since we compile Flexric, which does not support Ubuntu 22's default
gcc-11 (it hits a compile bug), switch to gcc-12 manually.
Since we update the compiler, we also need to be careful to use the right
address sanitizer version (which is bundled with the compiler version).
To avoid a reinstallation of the wrong asan version in target images, we
modify Ubuntu 22 targets to copy libasan from the base image. This is necessary
since we use update-alternative in the base image, but we don't want to
install compilers in the target images to use update-alternatives to
install the right asan version... This should be undone when switching
to Ubuntu 24 in the future, as it will have a newer default gcc version,
such that we don't need to update the compile in the first place.
Upgrade libpython in lteUE for U22 upgrade to make it compatible with
UHD.
A number of warnings appear when compiling flexric with gcc-12, which we
use in the next commit. Those have been fixed upstream. Switch to the
updated version to avoid warnings.
It seems libatlas-base-dev provided an `enum CBLAS_TRANSPOSE`, but we
don't install it anymore. Now, in Ubuntu, there is only a typedef'd
version in cblas.h; for RHEL(-derivatives), the same typedef'd version
is in cblas/cblas.h.
Ubuntu 18.04 is now in "Extended Security Maintenance", but otherwise
gets no updates. Remove it from supported distributions (regarding
installing packages); users can of course still use it.
Users are encouraged to upgrade to Ubuntu 20+.
Apart from the obvious change of adding ubuntu24.04 to supported
distributions, we need to remove liblapacke-dev as it is incompatible
with libatlas3-base. Installing both aborts with:
The following packages have unmet dependencies:
liblapacke : Breaks: libatlas3-base (< 3.10.3-14) but 3.10.3-13ubuntu1 is to be installed
It seems that we don't need it though, as liblapacke-dev installs the
right atlas library as a dependency.
Fix small issue noted when packing tlv value with tag 1 in which it would pack length*4 bytes, now calculated how many "segments" of 32 bits it needs to pack, same logic for unpacking tag value 1
Added missing support for packing/unpacking tag 0x001C (supportedMaxModulationOrderDl)
Now properly handles tag 0x0005 (NumConfigTLVsToReport) and subsequent TLVs that come after this value ( Table 3-9 SCF 222.10.02 )
Implement FAPI P5 message unpack, as well as message length check function procedure in FAPI P5 library.
Update check_nr_unpack_length in nfapi_p5.c to reference FAPI P5 message length check function.
Will be reused from other files in later commits.
Added packing/unpacking procedures for STOP.indication and ERROR.indication.
Separate a packing function for STOP.request for use in the NR FAPI library.
Expose needed (n)FAPI procedures to aerial_lib ( both P5 and P7 ) via a
header file.
Move (n)FAPI P7 packing/unpacking function interface into separate header file
Include needed headers in aerial_lib
- Phy config update and Sidelink frame parameters initialisation
- PSBCH scheduler to trigger TX PSBCH/RX PSBCH actions
- Sidelink indication with rx ind to trigger send SL-MIB to MAC
- Enable Scope for PSBCH use -d or --dqt
- Enable system testing of PSBCH for example using RFSIMULATOR:
- UE1 is a SYNCRONISATION SOURCE UE which transmits PSBCH
started using command -
sudo ./nr-uesoftmodem --sl-mode 2 --sync-ref 4 --rfsim --rfsimulator.serveraddr server
- UE2 syncs onto UE1 during sidelink search, gets the timing and continues to receive PSBCH
started using command -
sudo ./nr-uesoftmodem --sl-mode 2 --rfsim --rfsimulator.serveraddr 127.0.0.1
- In the default use case 2 Sidelink SSBs sent over 16 frames.
- In case SL-SSB configuration needs to be changed use -O option in the above command like
-O ../../../targets/PROJECTS/NR-SIDELINK/CONF/sidelink_preconfig_1txpool.conf
Integration: `2024.w24`
See merge request oai/openairinterface5g!2801
* !2789 CI: Increase iperf/iperf3 command timeout
* !2762 improve nrUE NAS security: accept NEA1,NEA2,NIA1,NIA2
* !2791 fix regressions for large physical cell id values
* !2792 Add support for MIMO 2x2 with USRP B210 at 20 MHz BW and 3/4 sampling rate
* !2748 Small UE fixes
* !2745 Add google test testcases for nr_get_Pcmax
* !2785 Update FHI 7.2 interface, test bandwidths, fix integer overflows
* the goal is to make the documentation more procedural and
descriptive and easier to understand
* added missing info in RLC-NR about reassembling of AM frames
Avoid UBSan error:
openair1/PHY/NR_TRANSPORT/nr_ulsch_demodulation.c:1524:41: runtime error: signed integer overflow: 737285888 + 1459370880 cannot be represented in type 'int'
Avoid UBSan error:
openair1/SCHED_NR/phy_procedures_nr_gNB.c:867:37: runtime error: signed integer overflow: 1744750952 + 412859840 cannot be represented in type 'int'
Avoid UBSan error:
openair1/PHY/NR_ESTIMATION/nr_measurements_gNB.c:246:18: runtime error: signed integer overflow: 1457321472 + 922889664 cannot be represented in type 'int'
Avoid UBSan error:
openair1/SCHED_NR/phy_procedures_nr_gNB.c:869:43: runtime error: signed integer overflow: 1546663455 + 1635930955 cannot be represented in type 'int'
Integration: `2024.w23`
See merge request oai/openairinterface5g!2788
* !2510 NR UE MAC UCI on PUSCH and aperiodic CSI reporting
* !2597 LTTng logger (CTF)
* !2751 synch the USRP time accross devices using the host clock
* !2775 Remove some global variables
* !2779 remove rb_mask_ul from permanent storage for stack local usage
* !2782 USRP: add a T trace to record the RX IQ data on antenna 0
* !2783 build_oai: fix bad options' names
* !2753 Pucch power control
* IP address of VNF shall be aerial1's one
* scripts within cubb container shall be run with sudo now
Signed-off-by: Raphael Defosseux <raphael.defosseux@eurecom.fr>
In some testcases, especially when iperf test duration is short (10s),
client does not report on time and report collection is interrupted.
This commit increase timeout for iperf/iperf3 commands.
Avoid UBSan error:
openair1/PHY/NR_ESTIMATION/nr_measurements_gNB.c:183:22: runtime error: signed integer overflow: 2122747784 + 27403601 cannot be represented in type 'int'
avoid UBsan error:
openair1/PHY/NR_TRANSPORT/pucch_rx.c:266:21: runtime error: signed integer overflow: 2113190528 + 117251752 cannot be represented in type 'int'
avoid UBsan error:
openair1/PHY/NR_TRANSPORT/nr_ulsch_demodulation.c:459:77: runtime error: signed integer overflo w: 2009540296 + 668087372 cannot be represented in type 'int'
xran's xran_bm_init() uses rte_pktmbuf_pool_create() to allocate DPDK
buffers. For a 1x1 configuration, the number of buffers might be too
small, and rte_pktmbuf_pool_create() might fail. This commit ensures
that we increase the number of buffers to have enough even in a 1x1
configuration. This works by chosing the next power of two, which is
recommended as per DPDK documentation.
Closes-Bug: #750
Neighbor configuration can be given by gNB.conf
get_MeasConfig function is extended as it will prepare A2 / A3 Report
Configs. Single Meas Obj is used for all.
rrc_gNB_process_MeasurementReport function is extended to process A2 /
A3 measurement events. helper functions are written to fetch neighbour
cell informations from sequence containers.
(get_neighbour_cell_information, get_neighbour_config)
- neighbour_config.conf file includes neighbour and measurement related parameters
Neighbour List:
- nr_cellid: the identity of existing cell who has neighbour
- neighbour_cell_configuration(list): includes the identity parameters
of the neighbour cell such:
- gNBID (incase of N2 Handover / to be identified from AMF)
- nr_cell_id (target gNB needs to consider)
- physical_cellId (needed for measurement reporting)
- absoluteFrequencySSB (needed for measurement reporting)
- subcarrierSpacing (needed for measurement reporting)
- plmn / tracking area code (incase of N2 Handover / needed for AMF
to identify / allow HO)
NR Measurement Configuration
- Periodical
- enable (default is 0, 1 enables the reporting)
- includeBeamMeasurements, maxNrOfRS_IndexesToReport (configurable
reporting parameters)
- A2
- enable (default is 0, 1 enables the reporting)
- threshold , timeToTrigger (configurable reporting parameters)
- A3
- cell_id (the nr_cellid of neighbour cell. So this A3 parameters
will be used for the neighbour
- [-1] means default. so it can be applied to any neighbour if
there is no specific config
- Neighbour specific A3 Threshold can be used or default can be
selected depending to the neighbour planning
- The neighbour-config.conf specifies, apply default A3 for
neighbours 3 and 4 | apply specific one for neighbour 2
- offset, hysteresis, timeToTrigger (configurable reporting parameters)
the neighbour_config.conf can be included in the gNB.conf file. Example
is given: gnb.sa.band78.106prb.rfsim.neighbour.conf
A sequence array container is used for neighbour cell configuration. RRC
structure has this container. A Measurement configuration structure is
defined. This structure has Periodical / A2 Event structures and also a
sequence container for A3 Event List.
These structures are filled during the initialization. While F1 Setup
Response is coming from DU to CU, the intra frequency / inter frequency
neighbour decision is given for the existing cell. Also new neighbours
could be added here if needed.
With the capabilities now given as part of this MR, it happens that when
checking the "docker compose config", the container_name might not be
close to the service name. Something like
service:
cap_drop:
...
cap_add:
...
container_name:
This is inconvenient, as the previous grep for the service name + 3
lines might not cover the service name. To circumvent this, give the
service name directly in "docker compose config". (no clue why it has
not been done like this from the start)
- sa_b200_gnb
- sa_aw2s_gnb
- sa_e1_b200
- sa_f1_b200
- sa_sc_b200_gnb
- nsa_b200_gnb
By default, the container has no access to /dev. Mount all USB devices
into the container to ease usability: by default, USB devices are
mounted under /dev/bus/usb using the bus ID, which is different on every
system and might change on reboots or with (un-)plugging USB devices.
Hence, mount all USB; the process has not many capabilities anyway, and
thus might not do much harm.
A proper solution would be to write a udev rule to mount a device using
its vendor and product ID, and only mount the corresponding device into
the container.
- SA 2x2 tests on 60 and 100 MHz
- N310-based OAI gNB + UE
Give SYS_NICE and IPC_LOCK for performance. On the UE side, give
NET_ADMIN (for interface bringup) and NET_RAW (for ping an TUN).
It might be desirable to run the UE without superuser rights. More
concretely, it might run without NET_ADMIN. In this case, various
corresponding operations on the TUN socket fd might fail. In this
commit,
- improve error reports to make them more clear
- do not automatically exit(1), but return from the function to ensure
continuity of the softmodem
Use common threadCreate() function to create P7 thread with high(er)
priority. Remove old calls to set_priority(), as this is already done
with threadCreate().
Use common threadCreate() function to create P7 thread with high(er)
priority. Remove old calls to set_priority(), as this is already done
with threadCreate().
Use common threadCreate() function to create P7 thread with high(er)
priority. Remove old calls to set_priority(), as this is already done
with threadCreate().
The main thread of nr-uesoftmodem (executing main()) is mostly working;
it is thus not necessary to set high priority. Instead, the threads
created with threadCreate() will receive higher priority, if possible
(as determined by the availability of SYS_NICE capability).
To inform if SYS_NICE is not present, print a warning in main().
SYS_NICE is a capability that allows a process to set thread affinity
and priority, among other things (see capabilities(7) for more info).
In this commit, add a function that allows to determine if the process
has this capability, and try to change the thread affinity and priority,
if requested. If the capability does not exist, the function will simply
not attempt to change the corresponding thread attributes.
To determine if the process has SYS_NICE, libcap can be used. However,
it might not be installed by default. To avoid requiring another
dependency, if we detect that libcap is not present, use a workaround by
try to set a real-time scheduling policy; if it's present, or can be
changed, we assume that the process has SYS_NICE (and clean up, if
relevant).
Simplify reading of capabilities
set_latency_target() attempts to set keep "low latency" by
- writing a specific latency to /dev/cpu_dma_latency
- setting manually the minimum CPU processor frequency to be the maximum
There is no functionality to undo this after stopping the softmodem(s),
and most users are probably not even aware that OAI does this. It is
generally preferable to set this beforehand using a governor or by
disabling sleep states (as outlined in the tutorial, and in a follow-up
commit in the performance tuning docs).
The previous mlockall() call, to lock memory to RAM, is retained in a
new function. There were additional mlockall() calls, which have been
replaced with lock_memory_to_ram(), where necessary.
Print a clear error if a config file does not exist (unclear previously,
when the user would get a confusing, unspecific, "file I/O error" on
line 0)
I tried to free the memory that had been allocated at that point, and
verified that using the address sanitizer. Nevertheless, the function
does too much, and is too complicated for refactoring.
Use correct baseTag for Unit Tests build in execution:
- if build_helper or othe files changed, which would trigger rebuild of
ran-base, use the corresponding ci-temp tag
- use that tag consistently afterwards
The RTT for GEO satellites is more than 500 ms.
And as the TA update interval must be greater than the RTT, we increase it to 1 second instead of 100 ms.
Should also do no harm for terrestrial networks.
According to the RRC specification, cellSpecificKoffset is:
Scheduling offset used for the timing relationships that are modified for NTN (see TS 38.213 [13]).
The unit of the field K_offset is number of slots for a given subcarrier spacing of 15 kHz.
If the field is absent UE assumes value 0
This parameter `cellSpecificKoffset_r17` can be set in the gNB conf file, in the section `servingCellConfigCommon`
A particular GCC version (v13?) used by CI machines potentially has a
bug that issues a mismatch bound warning on function array arguments.
This commit should prevent it from happening.
There is already a command line option called 'ue-scan-carrier' which
seemed to blindly look for SSB by changing the center frequency by
+/-100Hz if the previous attempt failed. This is probably copied from
LTE and is not suitable for NR. This commit uses this option to either
1. Perform initial sync on one GSCN if the SSB position if known
2. Or scan the all GSCN (if the SSB postion is unknown)
Also, the following is done for re-synchronization
1. Scan all GSCN when UE comes back from IDLE state
2. Scan one GSCN when UE does re-estabishment
Removed dependency of lte-softmodem.h in nr-softmodem.c.
CONFIG_HLP_UESCAN is declared also in lte-softmodem.h.
Changed Info to Debug because the logs are too much if large number of
GSCN scans run concurrently. Print PSS correlation peak and average
power of detected cells.
Till this commit, the UE gets the position of SSB in frequency from
command line and does the syncronization by locating the SSB in time. It
does this by generating a PSS time signal during initialization with the
SSB offset given via command line. Then PSS detection is done by
correlation.
In this commit, the UE computes a list of GSCN in the current NR band
that falls within the current bandwidth for the set center frequency.
For each GSCN in the list, the UE starts a thread and performs the
following:
1. Generate PSS time signal with SSB offset for current GSCN.
2. Detect PSS, SSS and decode PBCH (this step remains the same).
After all threads finishes, the results are collected.
For now, only the first detected PBCH is used to proceed further. So
this PBCH is indicated to MAC and UE exits initial sync.
Future work: The cell search can be standalone feature where the UE
could scan for SSB in the entire NR band. This would require changing
radio's frequency on the fly as the UE's bandwidth would not cover an
entire NR band. The cell detection can futher be imporved by looking
for SIB1 for each decoded PBCH.
Integration: `2024.w22` (Part 1)
See merge request oai/openairinterface5g!2773
* !2698 Handle division by 0 gracefully in average channel level calculation
* !2743 Demystifying 10log10(2^30) in signal energy calculations.
* !2767 NR UE imeisv fix
* !2768 Over-estimate short and long BSR for better latency and TCP throughput
* !2769 maxMIMOLayers parameter in configuration file
38.321 6.1.3.1 says: "The size of the RLC headers and MAC subheaders are
not considered in the buffer size computation" [for BSR]. However, prior
to this commit, we take the size signaled of the BSR and schedule the UE
with that amount of data, making it impossible for the UE to transmit
all its data at once (because it is the size signaled + headers).
In this commit, over-estimate the buffer status size. As we cannot know
the exact number and size of headers, we simply use "BSR size + n", as
done in function overestim_bsr_index().
* nb_re_pdsch is directly linked to nb_rb_pdsch, if the latter is 0 the
former is also 0
* compute average channel level only when number of RE for PDSCH is not 0
the average channel level will be set to 0 otherwise
* Added a log message to handle this special case, which is consequence
of other issues in the RX chain
* with this commit, the inconsistency in the RX chain will be handled
in later in nr_ue_pdsch_procedures with a demodulation failure
other minor changes:
* Improved clarity by using NR_NB_SC_PER_RB instead of hard-coded value 12 in nb of RBs calculation
* cleaned up unused variable and outdated comments
Compiling the nrUE with -fsanitize=undefined leads to a crash
because the function nas_stream_encrypt_eia1() accesses the data
as uint32_t but the data is not aligned.
Actually this function wants 64 bits (big-endian) of data, so let's
introduce a simple function to read a big-endian 64 bits value.
It may seem bad (unoptimized), but looking at the output of gcc
without -fsanitize=undefined this will be translated as just one
movbeq instruction (on the machine where I did the test), which is
actually less instructions than the previous version which was doing
too movbel plus orq.
Moreover, after the main loop we need to process the last remaining
bytes (so less than 8). The code was wrong, reading past the input
data (and also not caring about alignment). It may work, it may fail,
it depends on many things. But it's plain wrong. So this was replaced
for something better, simpler. No need for mask32bit() anymore, only
one call to U64() is needed.
And now -fsanitize=undefined is happy, so all is good.
We also restrict the input length to be multiple of 8 bits. I don't
think it's a problem. To be refined if I'm wrong. (I think RRC and
NAS messages are 8-bits aligned, so it shouldn't be a problem.)
It should not be > 0xffffff.
Normally the core network would prevent this case from happening.
Plus to reach this value means to send a lot of NAS messages, which
is very doubtful in practice.
Anyway, it's better to deal with the case.
We exit if that happens. To be refined if needed (very doubtful).
This function was very wrong, now it's just wrong.
(It does not sanitize the input properly. Some bad input may have
unexpected effects on the nrUE.)
It needs a serious rewrite.
I just wanted to "fix" it so that it processes correctly some
correct input data.
OAI core network is sending Deregistation Accept in clear, even
after security context activation.
The standards is not clear to me, so let's accept it.
To be removed if incorrect with respect to the standards.
(The core network will need to be fixed then.)
The version before this commit is wrong, taking wrong byte for msg_type
in some cases.
I don't have much time to implement a proper NAS PDU parser, so let's go
with hardcoded values.
Some later work will be done to have a cleaner solution. As of now, it's
too much work.
Get the security algorithms from SecurityModeCommand.
Restructure the code a bit: add the function handle_security_mode_command()
and adapt the code to use it.
Derive also knas_enc (not used yet).
Do the derivation for knas_enc and knas_int only after reception
of SecurityModeCommand. We need the algorithms configured by core
network to derive those keys.
Integration: `2024.w21`
See merge request oai/openairinterface5g!2757
* !2733 Remove duplicated function to reverse N bits
* !2712 rfsimulator: add support to simulate a long propagation delay
* !2695 improve polar init and polar readability
* !2709 Fix cmdline configmodule in phy simulators
* !2756 remove some enb dependencies in NR files
* !2687 bug fix ue UL payload filling randomly done in retransmissions (pointer not set)
* !2543 power/PHR normalizations for deltaMCS power adaptation
* !2719 Refactor LC handling at MAC
* !2736 Fix debug for ongoing transactions
* !2747 Fix aerial memory leak
* !2761 Use timeout command to kill hanging iperf3
* !2752 fix-dci11-type0-bitmap-setting
* !2759 Fix invalid MSG2 HARQ feedback attempt and remove accumulated_delta_PUCCH field from UE FAPI IF
* !2740 bugfix/cleanup: remove uint8_t to encode size plus some cleanup
* !2760 fix 5G NR SA FDD configs for 24 and 25 PRBs
* !2520 Documentation on coding style, workflow, and review instructions
It can happen that in the current DL slot, we cannot send an allocation
for Msg3 in a later UL slot. Up to this commit, MAC printed a warning.
However, this is not really a warning, as it will likely work in the
next or a later slot. So demote this to debug.
The starting point was the function do_NR_DLInformationTransfer() which
may deal with more than 256 bytes.
Then why not cleanup all the functions in the same file.
(And do a bit of formattting fixing while we're here.)
Then cleanup the callers of those functions.
Then, because of do_SIB23_NR() let's also cleanup in some
structures.
Not all the codebase was cleaned up, there surely remain some uint8_t
(and char/short/uint16_t) which are wrong.
In some pipelines, it happens that iperf3 is hanging on 100% CPU load.
In these cases, a simple TERM signal does not seem to be enough to kill
it; KILL is necessary.
While e.g., RemoteCmd's driver paramiko has a timeout command, the
documentation [1] is not really clear how it is trying to stop commands.
Since it effectively does not stop, we have to assume it is only sending
TERM.
To remedy this, use the timeout command to start iperf3. We use option
-v for diagnose timeouts on stderr, and send KILL after an additional 3
seconds (-k3) if TERM does not succeed. The timeout is sent to the same
duration as the overall iperf timeout.
[1] https://docs.paramiko.org/en/3.4/api/channel.html#paramiko.channel.Channel.settimeout
make sure the SSB is offset by 4 PRBs from the BW start.
Therefore, the ARFCN for the SSB center must be offset by 14 PRB from the ARFCN for the BW start.
A to-be-added UE's RA is added through nr_mac_add_test_ue() (the name is
misleading, it is not a "test UE") through its CellGroupConfig. Using
the process_CellGroupConfig() is unfortunate, because it is not really
clear what part of the CellGroupConfig gets evaluated (it is certainly
not all). Move the code for adding a new RA directly in the user-add
function, and refactor this one.
Introduce new LC configuration "priority", as specified in 38.321, and
use to sort LCs by priority. We set the priority of an SRB according to
38.331 Section 9.2.1, and for a DRB to be the minimum priority of the
QoS flows associated to this DRB.
Introduce new functions to add/release LCs without needing to go through
the intermediate CellGroupConfig. This harmonizes LC handling at the "DU
handler" to keep it consistent with RLC bearer handling.
These functions allow to group LC configuration handling, and we can
remove functions for setting NSSAIs and QoS by handling in one central
place (while adding LCs).
nr_mac_add_test_ue() is used by simulators and also in NSA mode to add a
new UE. Ideally, these users should also call the ue context setup
request handler to add a UE context; for the moment, respect the passed
list of LCs to add. Similarly, for do-ra, we require a dedicated
function to still parse the CellGroupConfig, so we keep that part of
functionality.
On reestablishment, since we do not process the CellGroupConfig from the
old UE, we need to add LCs for the new UE based on the onces from old
UE.
Use a common structure nr_lc_config_t to group LC configuration for one
logical channel. Use seq_arr for simple lookup, adding, removing of
elements without manual indexing logic.
This commit retains the previous logic of adding/releasing LCs through
the process_CellGroupConfig(), although we know while building the
CellGroupConfig if we add/release LCs. The next commit fixes this.
If we do not make this change, with the next commit, we get the
following error from gcc:
openair2/E2AP/RAN_FUNCTION/O-RAN/ran_func_kpm.c:299:46: error: assignment of read-only location '*(arr_ue_id.ue_info_list + (sizetype)(arr_ue_id.sz * 2444864))'
| arr_ue_id.ue_info_list[arr_ue_id.sz] = *ue;
In a later commit, we will use find_if in L2. In the case of building
with E2 agent, we link e2_agent which comes with its own implementation
of find_if_arr(), which results in a name clash. Rename the OAI
implementation, as nothing uses this, anyway.
- addition of control of deltaMCS if enabled in configuration file.
- introduce separate parameter to disable 256QAM in UL
- put min_grant_prb configuration back in pf_ul() as default prb size.
Add the option `--rfsimulator.prop_delay` to specify the simulated receive-path (gNB: UL, UE: DL) propagation delay in ms.
This option can be used e.g. to simulate a simple GEO satellite channel by specifying `--rfsimulator.prop_delay 238.74` to both, the gNB and the UE.
This is part of the work developed by Fraunhofer IIS within the ESA-funded project 5G-GOA (https://connectivity.esa.int/projects/5ggoa).
@@ -5,22 +5,26 @@ We want to make contributing to this project as easy and transparent as possible
Please refer to the steps described on our website: [How to contribute to OAI](https://www.openairinterface.org/?page_id=112).
1. Sign and return a Contributor License Agreement to OAI team.
2. Register on [Eurecom GitLab Server](https://gitlab.eurecom.fr/users/sign_up) if you do not have any.
2. Register on [Eurecom GitLab Server](https://gitlab.eurecom.fr/users/sign_up)
if you do not have any.
- We recommend that you register with a professional or student email address.
- If your email domain (`@domain.com`) is not whitelisted, please contact us (mailto:contact@openairinterface.org).
- If your email domain (`@domain.com`) is not whitelisted, please contact us
(mailto:contact@openairinterface.org).
- Eurecom GitLab does NOT accept public email domains.
3. Provide the OAI team with the **username** of this account to (mailto:contact@openairinterface.org) ; we will give you the developer rights on this repository.
4. The policies are described in these wiki pages: [OAI Policies](https://gitlab.eurecom.fr/oai/openairinterface5g/wikis/oai-policies-home).
- PLEASE DO NOT FORK the OAI repository on your own Eurecom GitLab account. It just eats up space on our servers.
- You can fork onto another hosting system. But we will NOT accept a merge request from a forked repository.
3. Provide the OAI team with the **username** of this account to
(mailto:contact@openairinterface.org) ; we will give you the developer
rights on this repository.
4. The contributing policies are described in the [corresponding documentation
page](doc/code-style-contrib.md).
- PLEASE DO NOT FORK the OAI repository on your own Eurecom GitLab account.
It just eats up space on our servers.
- You can fork onto another hosting system. But we will NOT accept a merge
request from a forked repository.
* This decision was made for the license reasons.
* The Continuous Integration will reject your merge request.
- All merge requests SHALL have `develop` branch as target branch.
## Coding Styles ##
# License #
There are described [here](https://gitlab.eurecom.fr/oai/openairinterface5g/wikis/guidelines/guidelines-home).
## License ##
By contributing to OpenAirInterface, you agree that your contributions will be licensed under the LICENSE file in the root directory of this source tree.
By contributing to OpenAirInterface, you agree that your contributions will be
licensed under the license described in the file [`LICENSE`](./LICENSE) in the
# It is not a good way of assigning pods to the nodes: this way we bypass the scheduler. At the moment we don't provide the resource information of these pods.
# Therefore, Openshift assigns the pods to the same node because it thinks the pods don't consume much resources. This isn't the case, they consume a lot of resources.
nodeName:
dlsim100rbtm2:acamas
dlsimbasic:acamas
ldpctest:acamas
nrdlschsim:acamas
nrdlsimbasic:acamas
nrdlsimdmrsptrs:acamas
nrdlsimmcsmimo:acamas
nrdlsimoffset:dedale
nrpbschsim106rb:dedale
nrpbchsim217rb:dedale
nrpbchsim273rb:dedale
nrpsbchsim:dedale
nrprachsim:dedale
nrpucchsim:dedale
nrulschsim:demophon
nrulsim3gpp:demophon
nrulsimmimo:demophon
nrulsimmisc:demophon
nrulsimscfdma:demophon
polartest:demophon
smallblocktest:demophon
ulsim:demophon
# will place on two nodes intel 3rd gen and 5th gen with RT kernel
mySSH.command('for pod in $(oc get pods | tail -n +2 | awk \'{print $1}\'); do oc describe pod $pod >> cmake_targets/log/physim_pods_summary.txt; done','\$',10)
These scripts are used by a Jenkins [job](../Jenkinsfile-colosseum) to trigger automated testing of OpenAirInterface (OAI) gNB and softUE on the [Colosseum](https://www.northeastern.edu/colosseum/) Open RAN digital twin.
Once a test is triggered, a new OAI LXC container at the specified OAI version will be built on Colosseum (if not already present), and gNB and softUE will perform TCP uplink and downlink connectivity test via the iPerf [tool](https://iperf.fr/).
The OAI branch to build and test can be specified through the `eNB_Branch` parameter passed through Jenkins (`eNB_TargetBranch`, which defaults to the `develop` branch, is used if `eNB_Branch` is not specified).
The Colosseum network scenario to test is specified through the `Colosseum_Rf_Scenario` Jenkins parameter, which defaults to a base Colosseum scenario without artificially added channel effects (e.g., only hardware impariments of software-defined radios, cables, and channel emulator).
Once the test ends, results are analyzed through the OAI automated test report generation tool available [here](https://github.com/ztouchnetworks/openairinterface-automated-test-reports), which builds a test report from the iPerf and OAI logs, and marks the test as successful or unsuccessful.
Results from successful tests are saved in history files and used to compare more recent tests.
A test is considered successful if the downlink throughput achieved during the test is greather than or equal to the average of the test history, which spans successful test executed since May 2024.
This is used to identify possible regressions of OAI runs executed on the Colosseum testbed.
#three config for level of parallelism "PARALLEL_SINGLE_THREAD", "PARALLEL_RU_L1_SPLIT", or "PARALLEL_RU_L1_TRX_SPLIT"
parallel_config = "PARALLEL_RU_L1_TRX_SPLIT";
#two option for worker "WORKER_DISABLE" or "WORKER_ENABLE"
worker_config = "WORKER_ENABLE";
}
);
security = {
# preferred ciphering algorithms
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
