8851 Commits (feec3c5905e4a7df3b291630840ebff38bc06299)
 

Author SHA1 Message Date
Andre Puschmann 6dc7841369 epoll_helper: move code to block OS signals into own helper function
before epoll can handle signals, the calling thread needs to block
them. the blocking needs to happen before any other threads get
spawned so they inherit the mask.
4 years ago
Andre Puschmann d30822784f enb: refactor MAC PDU logging
* remove single line MAC PDU subheader logging in favor
  of to_string() helper that prints entire MAC header in single line
* log MAC PDU after parsing and packing
4 years ago
Andre Puschmann 33c050791f pdu: add missing to_string() content for MAC SCH PDU subheaders
* add both scell activation CE
* add rnti to CRNTI and CONT RES CE
4 years ago
Ismael Gomez b397878d25 Fix comments in ue.conf 4 years ago
Andre Puschmann 0ec49a7b19 update readme and bump version of 20.04.2 4 years ago
Andre Puschmann 9e7d881bc8 srsenb: remove QAM256 related options in RRC Connection Reconfig
this part of the RRC connection reconfig has caused issue for some
newer phones like the Iphone X, One Plus 6T.
The phone would receive the message but not respond with a reconfig
complete.

The patch fixes this issue by disabling QAM-256 support altogether.
4 years ago
faluco c178e362b8 - Rewrite log management so that a log sink can be swapped for each testcase. 4 years ago
Xavier Arteaga 1d71cf8557 Avoid zero division 4 years ago
faluco 5e8a4c898d
Bring back the circular array object. (#1712) 4 years ago
Ismael Gomez 59b6301cc8 Remove legacy arbritrary 1s timeout for eNB requests 4 years ago
Andre Puschmann 6a56476b2a rlc: increase RLC Tx SDU queue
the Tx SDU has been set to 128 PDUs because this gives approx.
150kByte UL buffer state (max value for Rel 8 BSR).

Previously, however, we've always reported too much pending UL
data because we weren't subtracting the transmitted data from
the reported data.

with this newer BSR reporting, however, even with full buffers we would
never report full buffers because the queues size is too short.

This seems to cause issues for iperf TCP sessions not resulting
in full DL rate, see isse #1703. It seems that the TCP bandwitdh
probing works better with larger buffers (in our case this can
only be achieved with more PDUs in the pool).

This should fix #1703
4 years ago
Andre Puschmann 569d8ee860 rlc: remove unused macro
this seems to be a legacy copy of the RLC_TX_QUEUE_LEN macro
in rlc_interfaces_types.h
4 years ago
faluco 1860006203 - Fixed leaks in prach module spotted by asan. 4 years ago
Andre Puschmann 87d4a5dc9c enb: use circular array to access TTI and HARQ based data structures
this makes use of the new circular array to remove the need
to apply the modulo operation when safely accessing the underlying
array.
4 years ago
Andre Puschmann e80aa0a553 add circular array class to simplify access to truly circular arrays 4 years ago
yagoda 24574caf44 cleaning up the PRACH time offset calculation 4 years ago
yagoda b5a8d82058 -adding more prach tests to make test, removing some variables from stack
-PRACH time offset to TA unit conversion
4 years ago
yagoda 59082770d3 moving some arrays to the heap, fixing formatting, setting dft norm 4 years ago
yagoda f6ea431555 -adopting new PRACH equalization approach for successive cancellation
-adding frequency domain time offset detection
-adding to testing of PRACH
4 years ago
yagoda ec7873e7cc refactoring PRACH, adding phase correction to successive cancellation 4 years ago
yagoda 4d8888aae6 - adding simple successive cancellation to PRACH detection
- adding testing for same and for offset detection
4 years ago
Ismael Gomez 1bd3e76f98 Move unncessary buffer in radio 4 years ago
Ismael Gomez c4080e35cc Set different timestamp behaviour for X300 and B200 4 years ago
Pedro Alvarez 08a0d6bc0a Fix compilation warning with unecessary comparison. 4 years ago
Pedro Alvarez b59eca1b27 Wait until reestablish complete is finished to swap rnti in s1ap and gtpu. 4 years ago
Pedro Alvarez 6ec26ea627 Make sure that last_submitted_rx_sn is not 0 after handover. 4 years ago
Xavier Arteaga 52355024e7 SRSUE PHY: move mutex lock to cc worker and unlock for stack calls 4 years ago
Ismael Gomez 4c1d3ef7e8
Fix implementation of Random Access problem (#1689)
* Fix implementation of Random Access problem

* Apply AS activated condition on RLF before calling Reestablishment
4 years ago
Ismael Gomez f49fbd6d02
Fix the order of HO/reest measurement actions (#1696) 4 years ago
Andre Puschmann 33bb8388a5 run_lte: fix UE/eNB args
* remove PHY thread limitation
* set rx_gain for UE to avoid starting AGC thread for ZMQ
4 years ago
Andre Puschmann 655c7ae8ae enb: fix warning when removing user
when removing a user from the eNB we iterated over all possible LCIDs
and set the buffer state to zero in the scheduler.

this resulted in following log entries:

13:57:23.856334 [RRC ] [I] Received Release Complete rnti=0x46
13:57:23.856352 [MAC ] [W] [ 5149] The provided lcid=6 is not valid
13:57:23.856362 [MAC ] [W] [ 5149] The provided lcid=7 is not valid
13:57:23.856368 [MAC ] [W] [ 5149] The provided lcid=8 is not valid
13:57:23.856371 [MAC ] [W] [ 5149] The provided lcid=9 is not valid
13:57:23.856376 [MAC ] [W] [ 5149] The provided lcid=10 is not valid

we now check if the bearer exits at RLC and only report those to the MAC.
4 years ago
Andre Puschmann c9bf4d14bd run_lte: fix check for apps being shutdown correctly
since the new logging system doesn't tag the log file after closing
we now use the stdout to verify the UE has been shut down correctly.
4 years ago
Andre Puschmann c16ac6c890 ue_stack: only push sync event to event queue is stack is still running
stopping the UE in ZMQ mode caused a dead-lock because the Stack was
stopped before the PHY causing the sync queue to overflow. Since we
use a queue-length of 1 in ZMQ, mode, the PHY sync thread was blocking to
push a new sync event while the stack thread was already stopped.

this patch makes sure no new sync events are queued after the stack has
been terminated.
4 years ago
Andre Puschmann aec18a93d1 mux,bsr: refactor UL buffer status reporting
this patch fixes the UL BSR as per TS 36.321, it includes following
main changes:

* report UL buffer state to reflect the UEs transmit buffer after
  the MAC UL PDU containing the BSR has been built.
  In other words, if the UE, for example, can transmit all outstanding
  data in an UL grant, it will not report any pending data to transmit.

* refactor MUX routines and subheader space calculation
4 years ago
Andre Puschmann d48bc8837e pdu: refactor MAC PDU class
this patch refactors the MAC PDU class, the main changes are:

* add to_string() method to pretty print PDU and subheaders
  This allows to have a single log entry per MAC PDU with all its
  contents.
  It removes the C-style fprint() method

* Simplity payload vs. w_payload_ce
  Before we've used payload when reading the PDU and w_payload_ce
  as a buffer when writing. In all getters we needed to differentiate
  between both. Now payload points to w_payload_ce initially and is
  only updated when parsing a new PDU.

* add various helpers, e.g. to get subheader index, update a BSR, ..

* fix PDU test to use new to_string() method
4 years ago
Andre Puschmann 92d76a5eb6 mac_test: update tests involving BSR handling
* split existing BSR test into dedicated test for short, long, and trunc
  BSR tests

* update MAC tests that include packing UL BSR.
  (After checking TS 36.321 again it seems we have reported old UL buffer states)
4 years ago
Ismael Gomez 54eed11e1d Refactor decision for extended CSI/SRS 4 years ago
Ismael Gomez c9daf1f61a Make const args 4 years ago
Ismael Gomez e2154d2213 Renamed constants to SRSLTE and use SRSLTE_CFI macros 4 years ago
Ismael Gomez 38e2ffe414 Use Format1A when using CA and 15 PRB due to DCI size ambiguity 4 years ago
Ismael Gomez 3ee15191e7 Allow inactive cells to report CQI too during Reconfiguration 4 years ago
Ismael Gomez 51521ad8e4 Improved PDCCH blind search and fixed a few issues with ambiguous DCI size with Release 10 4 years ago
Andre Puschmann 1d1d52f3e6 enb: fix stopping of input thread
since we've used a blocking read of stdin (with std::getline())
we had a race when the eNB was stopped and the user was still making
keyboard inputs. this is because the we didn't wait until the input
thread was terminated until we stopped the eNB.

we know use poll to query stdin (getline has no timeout mechanism).
4 years ago
faluco db03275337 - Fix compiler errors when trying to build the project without SIMD support. 4 years ago
faluco 12e82a947f
LTO compiler and linker fixes (#1676)
* Guard the LTO feature in a CMAKE option and document possible issues of using it.
4 years ago
Andre Puschmann 14289ebcc6 dl_harq: lower log level when MCS changes in DL grant to debug
fixes #1489

this is normal behavior and should not result in a warning message.
4 years ago
Ismael Gomez 06a5bdaab2 Process all calls to RLF in the main RRC task 4 years ago
Xavier Arteaga af8898943f SRSENB: implement console cell_gain command
SRSENB: fix 5GNR compilation

SRSENB: fix override methods
4 years ago
Paul Sutton 96835a3b92
Merge pull request #1679 from softwareradiosystems/csv_out_of_order
Fix order of discription in user_db.csv.example.
4 years ago
Pedro Alvarez aafd2cd81c Fix order of discription in user_db.csv.example. 4 years ago