- 11 Apr, 2023 4 commits
- 
- 
Robert Schmidt authored
- 
Robert Schmidt authored
- 
Robert Schmidt authored- for updating cellGroup - for removing UE - not done for SI information and RRC timers 
- 
Robert Schmidt authored
 
- 
- 07 Apr, 2023 2 commits
- 
- 
Robert Schmidt authoredintegration_2023_w14 See merge request oai/openairinterface5g!2048 * !1861 NR UE RX Multi-threading * !2017 scheduling aperiodic SRS at UE * !2046 Correctly allocate GUTI in NAS UE structures * !2033 PBCH radio link failure detection at MAC * !1959 bugfix: change memory management of interface between PHY and MAC scheduler * !2041 hack: see "Bye." message when quitting gnb running with aw2s * !2047 NR RRC - RRCReestablishment - Fix MultiplePDU Regression 
- 
Florian Kaltenberger authored
 
- 
- 06 Apr, 2023 1 commit
- 
- 
luis_pereira87 authoredMerge remote-tracking branch 'origin/nr-rrcreestablishment-fix-multipdu-proper-commits' into integration_2023_w14 
 
- 
- 05 Apr, 2023 5 commits
- 
- 
Angelo Athanassopoulos authoredChanges 1. Passing function argument by reference. 2. Use a pointer for ue->established_drbs[drb_id - 1] in generateDRB. 
- 
Robert Schmidt authored
- 
Robert Schmidt authored
- 
Cedric Roux authoredThis commit fixes a bug caused by global variables in the gNB_MAC_INST structure used to transfer MAC scheduler's results to the PHY layer. When using several threads in the PHY layer, it may happen that the MAC scheduler is called a second time by the second PHY thread for the next slot before the content of the global variables in the gNB_MAC_INST structure used to store the result of the MAC scheduler for the current slot have been consumed by the first thread, leading to unpredictable results (all of them certainly wrong). The solution is to introduce a small memory management module for the interface between the PHY threads and the MAC scheduler. The MAC scheduler now fills the data structures passed to it, and not a global variable anymore. The PHY layer manages the structures and ensures that no data is overwritten before being fully consumed by the various threads of the PHY layer. Some care has to be taken because several PHY threads will access the same structures at the same time for a given slot (in read mode, so it's okay). A mechanism of reference counting has thus been introduced. Only one CC is supported. This restriction should not be too hard to remove if/when needed. (The code before this commit does not seem to work for more than one CC anyway.) 
- 
Angelo Athanassopoulos authored
 
- 
- 04 Apr, 2023 17 commits
- 
- 
francescomani authored
- 
luis_pereira87 authoredFlag integrityProtection work in inverse logic state, '0' means enabled which is the default the value, we need to set it to '1' to disable it 
- 
luis_pereira87 authoredFix SRB_configList2 and DRB_configList2 array position (new_xid) while creating RRCReconfiguration message after RRCReestablishmentComplete 
- 
Robert Schmidt authored
- 
Robert Schmidt authored
- 
Angelo Athanassopoulos authoredFixed a bug where the integrity protection was included in the ASN1 and caused significant throughput bottleneck. 
- 
Angelo Athanassopoulos authoredCosmetic changes, also changed the LOG message for updating the UEid, and moved the drb_id_to_setup_start and nb_drb_to_setup right before the function that they are used 
- 
Angelo Athanassopoulos authoredRemoved the unused variables, and changed the tenary operator conditions of drb_id_to_setup_start & nb_drb_to_setup, to the correct ones 
- 
Angelo Athanassopoulos authoredChanged the SRB2 and DRB at the do_RRCReconfiguration arguments, to pass the correct structures that are generated 
- 
Angelo Athanassopoulos authoredSimplify RRCReestablishmentComplete procedure with respective function, removed the block that creates the SRB2 and created a function, with doxygen 
- 
Angelo Athanassopoulos authoredSimplify RRCReestablishmentComplete procedure with respective function, removed the block that creates the ASN1 DRBs and created a function, with doxygen 
- 
Angelo Athanassopoulos authoredSimplify RRCReestablishmentComplete procedure with respective function, removed the block that updates NGU Tunnels and created a function, with doxygen 
- 
Angelo Athanassopoulos authoredSimplify RRCReestablishmentComplete procedure with respective function, removed the block that updates NAS PDUSession and created a function, with doxygen 
- 
Angelo Athanassopoulos authoredSimplify RRCReestablishmentComplete procedure with respective function, removed the block that frees all the NAS PDUs and created a function for this. 
- 
Angelo Athanassopoulos authoredCosmetic changes, added doxygen configuration, and changed the rrc_gNB_process_RRCReestablishmentComplete function arguments' alignment, so in the future if an argument is removed or added to be better visible in the commits 
- 
Robert Schmidt authored
- 
Robert Schmidt authored
 
- 
- 03 Apr, 2023 11 commits
- 
- 
Angelo Athanassopoulos authoredThe generateDRB function now creates a DRB and puts it in the struct member of gNB_RRC_UE_t, from this struct we can generate ASN1 messages by calling the generateDRB_ASN1 
- 
Angelo Athanassopoulos authoredAdded function to create ASN1 DRB-ToAddMod structs from, gNB_RRC_UE_t established_drbs 
- 
Angelo Athanassopoulos authoredAdded struct member to gNB_RRC_UE_t to store the established DRBs for a UE, this can be used to generate ASN1 structs, during RRCReconfiguration 
- 
Sakthivel Velumani authored
- 
Sakthivel Velumani authored
- 
Sakthivel Velumani authored
- 
Sakthivel Velumani authoredThe UE copies scope data not in a single buffer to avoid overwiring in case of multiple slots processing in parallel. But the scope upates data from single buffer thats assigned during selection. A getPlotValue() method is created to update the data pointer on every timerEvent. 
- 
Sakthivel Velumani authoredMutex for phy scope: Repeating this change again. My earlier change in commit 06ef3119 was erased by someone during merge of QT scope. 
- 
Sakthivel Velumani authored
- 
Sakthivel Velumani authored
- 
Sakthivel Velumani authoredTested with X300 and there was no real-time problems. Could get the same throughput as when NR_UE_CAPABILITY_SLOT_RX_TO_TX is 4. 
 
- 
