Eurecom
d1f7346c08
bug fix for missing wait sync
2018-01-05 17:02:49 +01:00
Wang Tsu-Han
cabb4561d5
seperating rx tx process and fixixng coding threads on different cores
2018-01-03 12:01:48 +01:00
Wang Tsu-Han
409eadc937
measuring time for coding worker
2017-12-14 17:17:31 +01:00
Wang Tsu-Han
7c27e09fcd
Bug fix for coding worker thread
2017-12-14 15:38:02 +01:00
Wang Tsu-Han
22e7b1e6da
extending dlsch_coding into four thread
2017-12-14 15:29:05 +01:00
Wang Tsu-Han
79b11d6b93
Bug fix for ulsch_decoding
2017-12-12 16:23:38 +01:00
Wang Tsu-Han
d82a0d900d
adding ulsch_decoding multiy threading
2017-12-12 15:24:00 +01:00
Wang Tsu-Han
7ac9d2ab81
Deadline scheduling for 40MHz
2017-12-11 14:27:26 +01:00
Wang Tsu-Han
d174e15ac8
Measure timing wait for FEP and FEPTX
2017-11-27 11:27:10 +01:00
Wang Tsu-Han
22911600a8
Adding measurment for waking up the thread
2017-11-24 14:31:19 +01:00
Wang Tsu-Han
6b99b1b1e4
emulated RF using seperate thread condition signal every time period
2017-11-21 14:41:36 +01:00
Wang Tsu-Han
69e90fa6cb
Creating Emulated FH
2017-11-17 11:11:24 +01:00
Wang Tsu-Han
3f6c5430a2
changing back the rxtx thread and add another worker thread for coding
2017-11-10 14:23:18 +01:00
Wang Tsu-Han
caef8c6e56
fixing the sleep delay problem for tx
2017-11-03 12:27:39 +01:00
Wang Tsu-Han
c84ddacd00
Seperate Front haul RX TX and process into three threads
2017-11-02 16:37:03 +01:00
Florian Kaltenberger
ecee5edcce
enabling new modulation
2017-10-27 16:16:05 +02:00
Wang Tsu-Han
85ce7de775
added thread for TX streaming
2017-10-27 10:33:30 +02:00
Wang Tsu-Han
23c54b8db4
added parallelization of fep and encoding
...
added numerology paramter
2017-10-23 14:47:26 +02:00
Raymond Knopp
ccfe5a0472
removed "short_offset" code from normal_prefix_mod. removed static declaration in PHY_ofdm_mod which was a cause for problems when parallelizing FFTs for FEP TX.
2017-10-17 09:41:27 +02:00
Raymond Knopp
5c60764408
testing FEP parrelization, default set back to single-thread. Reduced logging in MAC
2017-10-15 18:38:24 +02:00
Raymond Knopp
8420dce1aa
addition of worker thread for fep TX procedures in RU
2017-10-15 01:47:14 -07:00
Raymond Knopp
667b9f1a88
time profiling for RU and activation of feprx worker thread in RU if get_nprocs()>2
2017-10-15 08:31:10 +02:00
Raymond Knopp
75040edfdb
put pucch metric into uci->stat before calling harq_indication. Fixes an issue with PUCCH1 SNR measurements coming from HARQ.
2017-10-14 22:24:37 +02:00
Raymond Knopp
424c1b502a
added deactivation of ULSCH harq process after 3rd retransmission. Avoids problem of accumulating failed RA processes.
2017-10-14 21:14:05 +02:00
Raymond Knopp
549a0b61b4
Adjustment for received power calculation in eNB pusch procedures. Added factor to account for mcs (mimic the power control adjustment in UE).
...
activated periodic trace for PHR/PUSCH SNR/PUCCH SNR in MAC.
2017-10-14 11:17:27 +02:00
Cedric Roux
747372685a
bugfix: clear memory
2017-10-13 13:22:15 +02:00
Cedric Roux
00f45ec9f5
bugfix: modulo from frame
2017-10-13 12:19:12 +02:00
Cedric Roux
cf208d4192
bugfix: frame is bigger than 8 bits
2017-10-13 12:11:08 +02:00
Cedric Roux
dd44c82760
bugfix: fix assert
2017-10-13 11:52:34 +02:00
Cedric Roux
beaeea740d
bugfix: careful when setting harq_mask (didn't work for dci 1a sent to UE to do uplink)
2017-10-12 16:28:37 +02:00
Cedric Roux
f80f1f617d
bugfixes
...
- better ulsch harq management in phy (maybe not over)
- don't do schedule_CSI if UE not in RRC_CONNECTED (not sure if correct)
- don't call find_ulsch in fill_dci0
- some spaces cleaned
- log improved
2017-10-12 14:39:55 +02:00
Cedric Roux
df3ebacd88
add an assert
2017-10-12 10:01:38 +02:00
Cedric Roux
6abbf040be
bugfix: test existence before adding a new one
2017-10-11 17:45:53 +02:00
Cedric Roux
c407cdba2d
bugfix: use rnti 0 to mean "unused" instead of -1
...
-1 does not work
2017-10-11 16:44:27 +02:00
Cedric Roux
ac6f09eb2f
fix find_dlsch and find_ulsch
2017-10-11 12:08:28 +02:00
Cedric Roux
9b428e5e2f
bugfix: fix timing advance
...
- re-use ta_timer
- when ta_timer is running:
- no TA sent to UE
- no TA from UE (for UE to have time to apply previous TA command)
- decrease ta_timer every TTI
- be careful to use correct value of ta_update (31 means no TA)
Maybe not correct, to be checked.
2017-10-10 14:10:32 +02:00
Raymond Knopp
8e6cc01905
handling of PDCCH order. stoping UE-specific scheduler during RA procedure. checking for DCI format (resource allocation) to not overwrite resource_block_coding for format 1A DCIs.
2017-10-09 01:25:23 +02:00
Raymond Knopp
6bfc0d63b4
put 50 PRB back in default enb configuration. added handling of UE release in L1/L2. detection of stale ULSCH configuration
2017-10-08 13:04:08 +02:00
Raymond Knopp
b8235ff707
bugfix in dci generation (padding bits set to 0 in format 0 DCI)
...
removal of LOG_I put for debugging.
2017-10-07 08:22:02 +02:00
Cedric Roux
43d12970a2
bugfix: fix TX pdu_index usage (maybe still not correct)
2017-10-06 16:41:01 +02:00
Cedric Roux
1485cbda77
bugfix: fix handle_ulsch_harq_pdu
...
we can't use &ul_config_pdu->ulsch_harq_pdu.harq_information directly,
because that function is called for 2 cases and the substructure
harq_information is not as the same position in both cases.
2017-10-06 12:29:41 +02:00
Cedric Roux
2c9867ff5b
spacing cleanup
2017-10-06 12:18:19 +02:00
Raymond Knopp
09e683a3be
switching some LOG_I to LOG_D, fixes for RRU IF4p5. It was broken after changes in configuration procedures. Some configuration file changes.
2017-10-05 23:24:44 +02:00
Cedric Roux
10a63f3d05
add T trace
2017-10-05 11:57:31 +02:00
Cedric Roux
4e7f9312f8
bugfix: alignment for avx2
2017-10-05 09:50:59 +02:00
Cedric Roux
f2cb406bfb
fix some T traces
2017-10-04 17:58:30 +02:00
Cedric Roux
a993c9588f
T: fix trace
2017-10-04 17:27:08 +02:00
Cedric Roux
c97e054691
bugfix: fix CQI reporting
...
- don't use o_flip anymore in ulsch_decoding
- remove dl_cqi from eNB_UE_STATS
- put dl_cqi[NFAPI_CC_MAX] in UE_sched_ctrl
- adapt code for this dl_cqi change (from eNB_UE_STATS to UE_sched_ctrl)
- extract_pusch_csi has been fixed for CQI_ReportModeAperiodic_rm30,
the others need to be done. Program will exit if one uses them.
Not sure if setting sched_ctl->dl_cqi[CC_idP] has to be done in this
function...
2017-10-04 16:15:52 +02:00
Cedric Roux
0c6015cfa5
bugfix
2017-10-03 17:16:24 +02:00
Cedric Roux
b21b2dada3
bugfix: Msc_initial wrong
2017-10-03 15:31:16 +02:00
Eurecom