yagoda
69dc16c4c8
changes to fix arm compilation
2018-11-27 14:27:59 +01:00
Xavier Arteaga
8c3a0153b9
Added missing AVX512 intrinsics and flags. Fixes #291 .
2018-11-21 18:09:09 +01:00
Andre Puschmann
d981f129e0
Merge branch 'master' into next
2018-09-20 12:54:05 +02:00
IgnasJ
67c8bf1368
Fixes for srsLTE on ARM ( #229 )
...
* Fix ARM NEON code compilation
Fix LV_HAVE_NEON defintion incorrecly used instead of HAVE_NEON in some places
Replace vqabsq_s32 with vabsq_f32 as vqabsq_s32 requires int type (fails to compile)
Fix missing NEON code path in mat.h in srslte_mat_2x2_mmse_csi_simd()
* Fix timestamp overflow issue on 32-bit systems with Soapy driver
'time_t secs' can be 32-bit on some systems. This causes calculation:
'secs * 1000000000;' to overflow.
2018-09-19 17:34:59 +02:00
Ismael Gomez
bc9d342959
New optimization on the PHY for both UE and eNodeB ( #251 )
...
* New parallel Turbodecoder implementation in SSE/AVX 16-bit and 8-bit
* Optimised UL Interleaver
* Include TB CRC calculation in FEC encoder
* New threading priorities
2018-09-04 17:51:35 +02:00
Pedro Alvarez
8017b792cf
Changing SRSLTE_PHY_DEBUG back to SRSLTE_DEBUG.
2018-07-13 11:59:03 +01:00
Pedro Alvarez
7aaa9a1789
Moved srslte_debug_handle_crash to common/debug.c
2018-07-09 14:33:50 +01:00
Pedro Alvarez
96786e96da
Splitting the lib debug into phy and common debug.
2018-07-09 13:15:35 +01:00
Xavier Arteaga
f01f7b4945
Added Vector max abs SIMD function
2018-06-28 09:28:17 +02:00
Ismael Gomez
e18ba937dc
Limit uplink signal normalization to avoid clipping
2018-06-27 16:29:40 +02:00
Xavier Arteaga
681b98ae50
Added vector CFO
2018-05-25 16:06:32 +02:00
Xavier Arteaga
0bc3be7abb
Added DL CSI decoding to TM2 and TM3
2018-04-20 11:27:39 +02:00
Ismael Gomez
384e0f8649
Fixed UL interleaver (missing SIMD deinterleaver)
2018-04-17 19:16:55 +02:00
Andre Puschmann
e7a268d79b
fix SIMD compile issue on NEON
2018-04-04 12:20:32 +02:00
David Rupprecht
9d71bec7b6
Unified include guards
2018-03-31 19:04:04 +02:00
Ismael Gomez
321a750f56
Added features to ringbuffer
2018-03-16 11:23:37 +01:00
Andre Puschmann
57e0c01fc4
check max buffer length in hex print
2018-03-07 21:23:57 +01:00
Xavier Arteaga
ec901373d4
Correction ofo simd.h for AVX512
2018-03-01 14:01:36 +01:00
Xavier Arteaga
6fc9c96c58
Added CSI softbits weightening for Single antenna transmission
2018-02-28 12:07:31 +01:00
Xavier Arteaga
ff5ac85c7d
Added int16 to float SIMD vector function
2018-02-21 14:51:09 +01:00
Ismael Gomez
a279ab47f0
Improved neighbour cell accuracy. Changed RRC to avoid segfault when neighbour cell addition
2018-02-02 19:31:22 +01:00
Xavier Arteaga
c4247c7aca
Fix coverity for complex 16bit
2018-01-31 10:44:52 +01:00
Ismael Gomez
e16839d7a7
Merge branch 'next' into 16bit_avx_viterbi
2018-01-10 16:06:49 +01:00
yagoda
d749ee66f4
introducing 16 bit viterbi support
2018-01-08 17:05:23 +00:00
Xavier Arteaga
a01c5ea08f
Fixes #119 : channel estimation subframe averaging
2018-01-08 12:05:31 +01:00
Ismael Gomez
be880e16f4
Do not include srslte.h in debug
2017-12-29 00:25:31 +01:00
Ismael Gomez
a799098784
Merge branch 'next' into merge_patch
2017-12-21 11:28:18 +01:00
yagoda
23afd44a7d
modifications to enable phy lib logging
2017-12-20 14:57:22 +00:00
Ismael Gomez
821d5595c5
Fixed AVX compilation
2017-12-20 12:35:28 +01:00
yagoda
1b1301101c
adding support for phy(C level) logging to file
2017-12-18 19:04:47 +00:00
Ismael Gomez
142ee30029
Merge branch 'next_novolk' into next
2017-12-13 10:28:08 +01:00
Ismael Gomez
b0a2fa3b0b
Merge branch 'enbmimo' into mob_crypto_mimo
2017-12-04 12:10:46 -06:00
Ismael Gomez
f1c6bcd252
Merge branch 'next' into mobility
2017-11-30 20:13:06 +01:00
Ismael Gomez
a3a1d268b7
Improved CFO estimation/correction by filtering central 6 PRB. Cleaned ue_sync/sync/pss objects. Used const attr in vector and other objects
2017-11-29 18:30:21 +01:00
yagoda
38903de07c
adding simd xor functionality
2017-11-27 11:10:50 +00:00
Ismael Gomez
d428760b20
Merge branch 'next' into mobility
2017-11-24 12:02:41 +01:00
Xavier Arteaga
176dae270c
Fixes #105 . Added backtrace print in 'srsLTE.backtrace.crash'
2017-11-22 16:00:07 +01:00
Xavier Arteaga
91e706d4d1
Fixes #107 : srslte_simd_f_sqrt NaN solved if input is 0
2017-11-22 12:23:46 +01:00
Ismael Gomez
12d8b373c7
srsUE takes and reports intra-frequency measurements correctly
2017-11-20 10:02:39 +01:00
Xavier Arteaga
75957d41d7
Added DCI formats 2 and 2a inScheduler. Improved UE PHCH, corrected some MIMO bugs on both sides.
2017-11-08 17:10:16 +01:00
Ismael Gomez
641eceb328
Merge branch 'next' into mobility
2017-11-07 17:13:57 -08:00
Xavier Arteaga
15e75b2c65
Created Bit interleaver object and SSE optimised interleaver
2017-10-23 14:23:03 +02:00
yagoda
3292f9c269
simd.h tidy up & small fix for eMBMS
2017-10-19 16:38:58 +00:00
Xavier Arteaga
0504e7a51b
Fixed test for abs value. Solved compilation Neon warnings and SSE errors
2017-10-18 12:49:43 +02:00
yagoda
f4b9e7311a
adding neon support to new vector structure
2017-10-17 15:51:27 +00:00
Ismael Gomez
c4c1c62360
stable connection. Runs out of samples in ringbuffer if clock differences
2017-10-13 17:59:12 +02:00
yagoda
2019ca31ef
adding neon support for new kernel structure
2017-10-13 15:35:48 +01:00
Ismael Gomez
f4e883d24f
Divided phch_recv in subclasses
2017-10-10 16:42:24 +02:00
Xavier Arteaga
1a5cf45dda
Solved compilation error for SSE (Tested in Atom)
2017-10-09 16:30:32 +02:00
Ismael Gomez
dbae016b00
Removed unused vector functions
2017-10-02 18:16:03 +01:00
Xavier Arteaga
94a06867a3
Optimized SIMD includes and solved AVX512 bugs
2017-09-29 16:42:46 +02:00
Xavier Arteaga
9e5f999666
Added more functions
2017-09-28 11:04:26 +02:00
Xavier Arteaga
c41ad5453c
Solved bugs and compilation error in simd and vector_simd
2017-09-25 17:08:11 +02:00
Xavier Arteaga
0c52c5651d
Merge branch 'next_novolk' of https://github.com/softwareradiosystems/srsLTE into next_novolk
2017-09-25 13:23:50 +02:00
Xavier Arteaga
c9f6bfccd4
Refactored vector library with SIMD independent architecture inline functions test-benchmark
2017-09-25 13:19:34 +02:00
Xavier Arteaga
1c3b5552be
added c16 type and architecture independent inline SIMD calls
2017-09-25 13:15:59 +02:00
Xavier Arteaga
8078238cb5
Removed test macros from mat.h
2017-09-25 13:08:38 +02:00
Ismael Gomez
bf79d463d4
Merge branch 'next' into next_novolk
2017-09-13 17:10:22 +02:00
Ismael Gomez
7aa33c1f27
Merge branch 'next' into next_novolk
2017-09-13 13:05:28 +02:00
Ismael Gomez
b0639ab394
resolved merge conflicts
2017-09-05 15:26:36 +02:00
Ismael Gomez
9dffa10550
fixed compiling issues
2017-09-04 17:56:58 +02:00
Ismael Gomez
a51b6b09dd
Merged with next. Attach successful
2017-09-01 19:32:57 +02:00
Ismael Gomez
e5feec1934
Temporal commit before merge. Changed sequence objects to avoid reallocating memory. new NAS/RRC working with initial attachment
2017-09-01 13:29:11 +02:00
Ismael Gomez
408bb7da66
Merge branch 'next' into next_novolk
2017-08-31 14:54:06 +02:00
Xavier Arteaga
d933f1b817
Renamed module algebra (it is now named mat) and moved usefl math SIMD macros to simd.h
2017-08-31 12:24:26 +02:00
Xavier Arteaga
0947173fc1
Merge branch 'next' into next_mimo
2017-08-29 17:30:40 +02:00
Xavier Arteaga
bbf7ffffa2
Added Condition number calculation for Rank Indicator in TM3
...
(cherry picked from commit f8a4472)
2017-08-29 17:11:49 +02:00
Ismael Gomez
616e18c570
fixed PUCCH correlation estimator
2017-08-24 15:16:13 +02:00
Xavier Arteaga
2c07a16189
DL grant parameter are now vectors of MAX_CODEWORDS
2017-08-21 16:53:12 +02:00
Xavier Arteaga
94325d13d2
Scalable PDSCH with MAX_CODEWORDS PDSCH plus decoding threads for bitrate increase
2017-08-17 10:23:03 +02:00
Xavier Arteaga
48d508aeba
Added srs_lte_cpy for aligned copy which improves a bit performance for aligned data
2017-08-17 10:19:19 +02:00
Xavier Arteaga
d1709e06af
Added algebra SSE, AVX (+FMA) implementatiokns with test
2017-08-17 10:14:37 +02:00
Xavier Arteaga
1d00e1acaa
Merge with next_with_matlab
2017-08-03 13:56:57 +02:00
Xavier Arteaga
6142a5f9e5
Merge to next_with_matlab
2017-07-25 13:17:36 +02:00
Ismael Gomez
f629e10fcf
fixed bug in sub_sse() and added couple of avx functions
2017-07-07 18:44:17 +02:00
Ismael Gomez
fd2a6f13e2
Merge branch 'next' into next_novolk
2017-07-07 16:34:05 +02:00
Ismael Gomez
12cbf02606
changed name of vector_simd function
2017-06-22 18:52:12 +02:00
Ismael Gomez
681a2455d9
Merge branch 'next' into next_novolk
2017-06-22 18:08:14 +02:00
Ismael Gomez
215dac6662
fixed some issues with AVX machines
2017-06-09 13:03:47 +02:00
yagoda
e0fb5d5cf6
updating avx vectors and viterbi
2017-06-01 12:30:37 +02:00
yagoda
0fe981e608
adding avx viterbi and avx vectors
2017-06-01 12:30:15 +02:00
Andre Puschmann
d079d25b2c
rename srslte folder and src subfolder
2017-05-30 13:05:04 +02:00