Ismael Gomez
c9c43e44e9
Fixed PDU buffer size
6 years ago
Andre Puschmann
2ae35e4ce0
make sure status PDU doens't exceed grant size
6 years ago
Pedro Alvarez
9597cc4c14
Clang-formated liblte_*
6 years ago
Xavier Arteaga
5e49aca835
Merge pull request #570 from softwareradiosystems/feature_256qam_fix
...
Resource allocation extended tables and PDSCH table 2 for 256QAM
6 years ago
Ismael Gomez
6d2c6ec58d
Do not change master clock rate in UHD if dynamic rate is supported
6 years ago
Guillem Foreman
2a8ec5c9b9
More fixes 256QAM
6 years ago
Guillem Foreman
eab73829f6
Changes requested
6 years ago
Ismael Gomez
538f38cf54
Let PDCP Control PDUs be processed
6 years ago
Pedro Alvarez
9b6aaac1b5
Ignore PDCP status PDUs, instead of passing them to the GW. Check on the GW if the packet is large enough to hold an IPv4 header.
6 years ago
Ismael Gomez
11d42acb3a
Fix using deallocated PDU
6 years ago
Ismael Gomez
8c41625599
Add RLC suspend and resume
6 years ago
Andre Puschmann
db196cc052
add RLC AM test for status PDU generation with small and big MAC grants
6 years ago
Andre Puschmann
db9f093b63
add debug log when handling reordering timer
6 years ago
Andre Puschmann
849ad2655c
fix creation of malformed RLC AM status PDU
...
the issue was caused by an unneeded length check during the
assembly of the status PDU. This length check caused an too early
break from the while-loop that adds all missing SNs to the status PDU
leaving the PDU struct in an invalid state where ack_sn was equal to
one of the NACK_SNs.
The fix is to simply remove the length check during the generation to
fully generate the status PDU. If, for whatever reason, MAC provides
a too small grant to fully fit the status PDU, the PDU will not be
generated, i.e. the calling function checks the return value and does
not pass the (then too big PDU) down the stack.
6 years ago
Andre Puschmann
9f3a947c02
modify RLC AM retx test to provide exactly a MAC opportunity of the size
...
that is reported as buffer state
6 years ago
Xavier Arteaga
aa6652155c
Phy: added ue_ul normalization mode parameter
6 years ago
Xavier Arteaga
20823e191d
SRSUE: Added UL force amplitude optional parameter
6 years ago
Andre Puschmann
4e86b2f6b2
introduce layerd architecture we use for UE to eNB
...
- this allows flexible use of different PHYs and radios
- use common radio_multi (moved to lib)
6 years ago
Xavier Arteaga
2530894374
Channel emulator: added input pointer checks
6 years ago
Ismael Gomez
d595c4d882
Fixes issue #545 and unlocking of tti_sync condvar before closing causing deadlock at stop
6 years ago
Andre Puschmann
aff26abc89
fix memleak in Soapy driver when no device was found
6 years ago
Ismael Gomez
773c2a9721
Fixed multiple issues in Random Access procedure and extended MAC test
6 years ago
Andre Puschmann
6b2961fcee
add MAC DL-SCH unpacking test
6 years ago
Andre Puschmann
92f0d58856
mac: move further init/config into ctor to allow early logging, etc.
6 years ago
Francisco Paisana
92b4b168fe
added default rlc um and am in the interface as factories
6 years ago
Francisco Paisana
b903c1a52c
made default init for rlc um test
6 years ago
Francisco Paisana
58e2e51044
removed rlc_interfaces. The rlc_cnfg types are now part of the rrc_interface_types. Removed implicit RB initialization in the RLC.
6 years ago
Francisco Paisana
c5509caece
added extra operators and remove rrc_asn1 in some cmake targets
6 years ago
Francisco Paisana
b936629221
removed rrc asn1 header from bcd_helpers, from rlc_interface, and rlc tests
6 years ago
Francisco Paisana
bb0dbf0f1c
created a asn1 utils with helper functions and types related to the asn1. These types/functions will help isolate the rrc_asn1 lib from most of the layers, reducing this way the compile times, and leading to better isolation. This will also make the transition to NR smoother. I removed the asn1 types from the NAS and USIM interface so far.
6 years ago
Francisco Paisana
416f9896ec
fixes to include the findings from PR #550
6 years ago
Guillem Foreman
46724c336d
Fixed errors for phy_dl_test and pdsch test
6 years ago
Guillem Foreman
4627737c87
Revert "Disable 256QAM tests failing in CentOS"
...
This reverts commit 833ce5224a
.
6 years ago
Guillem Foreman
c71763dc93
Fixed minor message in modem test
6 years ago
Guillem Foreman
db4127ca64
Fixes 256QAM, added pdsch table 2, fixed warnings
6 years ago
Guillem Foreman
535325bc37
srsLTE: added resource allocation extended tables for 256QAM and integration with PDSCH test
6 years ago
Andre Puschmann
62a4fa1c93
fix various potential buffer overruns
...
this is based on suggested fixes by Dirk Grabinski and David who
have fuzzed various parts of srsLTE using afl
6 years ago
Andre Puschmann
701db61fde
rlc: extend stress tester to simulate MAC PDU corruption
6 years ago
Andre Puschmann
ecb6a0f99b
fixing RLC AM bug where corrupted PDU causes segfault
...
the underlaying issue is that a corrupted PDU is passed to RLC
which claims to have segments whose total size exceed the size
of the entire PDU. Those PDUs are not ignored.
6 years ago
Xavier Arteaga
a417d3a70a
Added DCI trace if cqi is requested
6 years ago
Xavier Arteaga
87433c83da
SRSUE: 64QAM R8 compatible
6 years ago
Ismael Gomez
833ce5224a
Disable 256QAM tests failing in CentOS
6 years ago
Pedro Alvarez
498f1b8e4b
Small PDCP fixes:
...
- Moved most the initialization of the pdcp_entity to the header.
- Initilize some variables in gtpu_ntoa.
- Removed debug print.
- Format eNB GTP-u debugging code.
6 years ago
Pedro Alvarez
1c51d2ca9a
Small clang format changes.
6 years ago
Pedro Alvarez
5d8dcbe4cb
Make sure that PDCP SN is initialized properly in the eNB. UE attaches again when using encription.
6 years ago
Pedro Alvarez
53d2220aab
Resolved conflicts with unique pointers after rebase in PDCP entity.
6 years ago
Pedro Alvarez
2643b9070f
Fixing mistake in setting PDCP last_submitted_pdcp_rx_sn
6 years ago
Pedro Alvarez
6d5dbdedfa
Fixed mistake in comparison in pdcp_entity.
6 years ago
Pedro Alvarez
693338da07
Added TX HFN print to make it easier to debug HFN mis-match
6 years ago
Pedro Alvarez
6b59cc0760
Debug print for PDCP Reestablish.
6 years ago