PentHertz
/
srsLTE
mirror of https://github.com/PentHertz/srsLTE.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
7,699 Commits 3 Branches 23 Tags 67 MiB
Commit Graph

7699 Commits

Author SHA1 Message Date
Francisco 8fe27a72d0 enb - decreased default enb.conf mac.max_nof_ues to 8 to avoid large memory pool preallocations 2021-04-09 12:54:22 +01:00
Francisco 71f1f1b556 enb - log warning when rnti-specific memory block is full 2021-04-09 12:54:22 +01:00
Francisco a890a22ccd enb - rnti memory pool now fallbacks to new/delete if rnti-specific memory block is full 2021-04-09 12:54:22 +01:00
Francisco a780a6a7b6 fix obj pool interface 2021-04-09 12:54:22 +01:00
Francisco 344bdf3165 allocation of UE rlc and pdcp classes in the ue dedicated memory pool 2021-04-09 12:54:22 +01:00
Francisco 60d5c6df7b implementation of memory pool with individual memory blocks per UE that get accessed via their rnti 2021-04-09 12:54:22 +01:00
Andre Puschmann db496df1e6 mac_nr: add support for periodic BSR reporting 
building on the previous refactor this patch now adds support
for peridoic BSR reporting (using short BSR). It furthermore does
the following changes:

* add BSR packing
* add proc_bsr_nr unit test
* move mac_nr test code into test folder under src (needs to be done with other test code too)
 2021-04-09 12:28:05 +02:00
Andre Puschmann 7188603f2e mac_nr: don't push MAC CE LCIDs to RLC when unpacking DL MAC PDU 
this should fix log warnings then demux was trying to feed MAC CEs
to RLC:

10:50:54.845116 [MAC ] [I] [ 1153] Handling subPDU 1/2: rnti=0x0 lcid=61, sdu_len=1
10:50:54.845117 [RLC ] [E] Radio bearer id must be in [0:11] - 61
10:50:54.845118 [RLC ] [W] LCID 61 doesn't exist. Dropping PDU.
10:50:54.845119 [MAC ] [I] [ 1153] Handling subPDU 2/2: rnti=0x0 lcid=63, sdu_len=33
10:50:54.845119 [RLC ] [E] Radio bearer id must be in [0:11] - 63
 2021-04-09 09:49:14 +02:00
Andre Puschmann 6dd8c21073 mac_sch_pdu_nr: add unpacking of TA CE in DL-SCH 
also add basic test to verify correct unpacking
 2021-04-09 09:49:14 +02:00
Francisco 9b7060a098 check ptr value before dereferencing 2021-04-08 21:44:42 +01:00
Francisco dc763b5045 rlc am - fix bugs in buffered_pdcp_pdu_list 
The count and container elements were not being correctly updated in the events
of SN overwrite, clear(), clear_pdcp_sdu(sn) of sn that doesnt exist
 2021-04-08 20:11:15 +01:00
Francisco ebbb74e639 rlc am - use of circular_map container in RLC AM window 
The window elements are not being correctly cleaned up when clear() is called
or when overwritten by newer SN. Furthermore, the window count member is not being
correctly updated when the insertion of a newer SN overwrites the previous one.

I used the circular_map container to avoid this sort of bugs
 2021-04-08 20:11:15 +01:00
Ismael Gomez c4069b58af Call HARQ new tti only for the current PID 2021-04-08 20:24:21 +02:00
David Rupprecht 82c9636156 Refactor pdcp ctor seperate the configure function 2021-04-08 19:08:17 +02:00
Andre Puschmann 1a42e0b1d1 epc,nas: add error handling for all make_byte_buffer() calls 
not handling the case where allocation was failing was causing a
segfault. It seems that when doing high rate UDP iperf and
disconnecting the UE causes the buffer pool to become empty due
to the PDU buffering. Thus, causing a crash when attaching the UE again.
 2021-04-08 18:20:44 +02:00
Andre Puschmann cbe7980d7d enb,mac_controller: apply new MAC config with disabled DRB 
fix regression in RRC that updated the local MAC config to
disable DRB scheduling after maxRetx, but didn't actually apply the config
 2021-04-08 18:19:55 +02:00
David Rupprecht 56e947bf27 Added functionality for releasing a NR bearer 2021-04-08 17:37:36 +02:00
David Rupprecht cc6d8128ad Added gw interface if it is running 2021-04-08 17:37:36 +02:00
David Rupprecht 7a481c40ab Added mac prach nr proc error handling 2021-04-08 17:37:36 +02:00
David Rupprecht f9ea02b6bf Added functions for rrc reestablishing procedure if nr reconfig failes 2021-04-08 17:37:36 +02:00
David Rupprecht 43ec714ce1 Added functionality for printing nr metrics 2021-04-08 15:55:28 +02:00
Xavier Arteaga 58a9ef6e7f Minor NR PHY tweaks and aesthetic 2021-04-08 11:39:55 +02:00
Xavier Arteaga db1a1c059c NR PHY workers TTI synchronization between DL and UL work 2021-04-08 11:39:55 +02:00
Xavier Arteaga 991c6e7016 Refactored NR HARQ-ACK feedback enqueue 2021-04-08 11:39:55 +02:00
Xavier Arteaga bd11b66b6c Fix unitialised NR PHY arguments 2021-04-08 11:39:55 +02:00
Xavier Arteaga 22239b29c5 Disable smoothing filter for NR PDCCH DMRS 2021-04-08 11:39:55 +02:00
Xavier Arteaga 9f54c6ae6d Fix NR PUCCH info string and memory leak 2021-04-08 11:39:55 +02:00
Francisco bb9601c13b gtpu - fix gptu manager members destruction order 2021-04-07 22:26:42 +01:00
Francisco b7d4a72ea1 adt - fix circular map clear() and erase(it) functions 2021-04-07 22:26:42 +01:00
Francisco 643c73acee gtpu - fix removal of gtpu tunnels when rnti is released 2021-04-07 22:26:42 +01:00
Andre Puschmann 7459a65a8e enb,rrc_ue: remove unused variable 2021-04-07 22:19:08 +02:00
Andre Puschmann 0fdcea2924 rlc_am: count every segment retx as a full retx for this SN 
previously we were only counting retx if we retx the start of a segment.
this could lead to unwanted behaviour, i.e., not counting retx
correctly and thus not triggering the maxretx attempt, if the receive
always sends NACKs with a SO_start.

The RLC spec is not clear on how this should be handled correctly but
IMHO using an integer number of retx is reasonable, even for segments
that might be retransmitted more often.

The alternative of using a fractional retx counter that may be increamented
proportional to the segment size that is retx is another alternative
but considered too complex to implement (and test correctly).
 2021-04-07 22:18:22 +02:00
Andre Puschmann c74661ae95 rlc_am: add log message when received NACK SN is already on retx queue 2021-04-07 22:18:22 +02:00
Francisco d02cc51e33 use mutex or atomic in running flag to signal the interruption of a thread 2021-04-07 16:11:31 +01:00
Francisco 89628b691f add thread sanitizer cmake flag 2021-04-07 16:11:31 +01:00
Xavier Arteaga 49731ab5a1 Initial NR-DCI refactor 
Added NR DCI 1_1 size

Several DCI NR fixes
 2021-04-07 16:57:12 +02:00
Xavier Arteaga dee2ec30cd SRSUE: use new MAC interface for NR SR procedure 2021-04-07 14:16:46 +02:00
Xavier Arteaga 01990c2e1d SRSUE: Add PHY-MAC SR related interfaces 2021-04-07 14:16:46 +02:00
Xavier Arteaga 64acf03158 Fix sch_nr_test 2021-04-07 13:56:06 +02:00
Xavier Arteaga f4ae2d73aa SRSUE: report positive ACK for retx 2021-04-07 13:56:06 +02:00
Xavier Arteaga 20a4c39fa4 SRSUE: Disable PDSCH retransmissions in NR 2021-04-07 13:56:06 +02:00
Xavier Arteaga 1dcd75a6c7 Ignore CRC for RV!=0 in phy_dl_nr_test 2021-04-07 13:56:06 +02:00
Xavier Arteaga 3504bfcf5a Fix NR-SCH test 2021-04-07 13:56:06 +02:00
Xavier Arteaga 39e97cbcf4 Fix unbounded memory access 2021-04-07 13:56:06 +02:00
Xavier Arteaga 29c6adad93 Avoid unused LDPC memory in AVX2 decoder 2021-04-07 13:56:06 +02:00
Xavier Arteaga 392a8cc1b6 Fix LDPC RM number of useful iterations 2021-04-07 13:56:06 +02:00
Xavier Arteaga 3568f633c3 Use LDPC RM number of useful LLR and rename functions 2021-04-07 13:56:06 +02:00
Xavier Arteaga ecd2c1cdcc Improved LDPC decoded performance and other NR aesthetical changes 2021-04-07 13:56:06 +02:00
David Rupprecht 506adca930 Set eutra nr band info based config value 2021-04-07 13:24:26 +02:00
Francisco cdf72248f3 Reimplement batch-based background object/memory pool 
Main changes:
- addition of pool utilities
- The node size/alignment and batch allocation threshold are now runtime arguments
- object pool and memory pool are not anymore based on the same class.
  The object pool cannot use intrusive free list because it would overwrite the object
  memory
 2021-04-07 12:02:22 +01:00