- introduce loader_reset() that frees all memory
- resolve memory leak in loader_format_shlibpath()
- correctly handle library reload: check if a function has been
retrieved already. Don't leak memory (strdup()), and handle multiple
reloads of different functions from the same SO
- install libasan in physim containers
- run physim containers with SYS_CAP_PTRACE capability
- remove static-libasan: not available in RHEL8:
The physims are tested in a container environment on top of RHEL8. To
my knowledge, no package provides the static ASan library there.
Therefore, remove it here to be able to run these tests.
Functions generate_missing() and missing_size() had different
computations of sn_count. Not sure if it is a problem or not,
but let's do the same in both functions.
Hopefully the computation is correct.
integration_2022_wk07_c
See merge request oai/openairinterface5g!1451
This merges the following MRs:
- MR !1436 fairRR LTE scheduler fixes: should make the connection more stable, less UL retransmissions
- MR !1386 phy sync improvements: the 5G NR UE should sync better
- MR !1361 FDD support (no traffic): adds FDD support for the scheduler, but heavy traffic has not been tested
- MR !1447 using uid for PUCCH RRC conf: use the RRC UID to decide on PUCCH resources, should in particular avoid the "PUCCH straddles DC carrier" assert
- fix for 5G F1 RFsim test fail
Previously, we thought the memset in the nr_ue_dl_indication()
was causing issues / exposing a race. This is not the case. The
memset is not problematic, the issue is that in the emulate_l1
mode we were not always configuring the DCI nbits before handling
the DCI when we are pushing high (15 Mbps) throughput traffic.
Overwriting sched_pusch information at that stage would in particular
lose information on the current frame/slot. Overwriting sched_pusch with
the retInfo structure is already done at the end of
allocate_ul_retransmission(), and is therefore not necessary at this
stage.
We compute the preferred DL TDAs when receiving the cellGroup from the
RRC. However, in F1 and the RFsim, it seems that the scheduler tries to
access the preferred DL TDAs before they have been set after receiving
the cellGroup. In this commit, add a simple check to verify the
preferred TDA has been set up.
Instead of conditionally calling the pucch_scheduler
we now only call the functions we need from the stub
function directly. This is better because we are not
modifying the OAI MAC layer code and ensuring that we
are not calling the pucch_scheduler twice for the same
sfn/slot
Robert Schmidt