1
0
mirror of https://github.com/gnss-sdr/gnss-sdr synced 2024-06-16 18:19:58 +00:00
Commit Graph

191 Commits

Author SHA1 Message Date
Carles Fernandez
9c95aeb07f
Avoid deprecation warnings when VOLK >= 3.1
Bump CMake max version to 3.28
Do not leak build system paths when cross-compiling. Fixes OpenEmbedded QA warning
volk_gnsssdr: Fix 64-bit integer testing
CI: Fix for Python 3.12
2023-12-13 22:12:29 +01:00
Carles Fernandez
4999d38194
Improve portability
Improve support for RISC-V arch in volk_gnsssdr via cpu_features (updated to c919e9a).
Set BLA_PREFER_PKGCONFIG if not defined by the user, helps to find lapack in some archs (e.g., riscv64)
Fix for Fedora/OpenSUSE, which do not install cpu_features along volk.
Remove volk_gnsssdr_8ic_s8ic_multiply_8ic kernel since it was problematic in some archs
Improve Python detection
Fixes for GCC 13
Remove ENABLE_GENERIC_ARCH configuration option and old ARM detection mechanism
Update gtest to 1.13.0
Add CI jobs for volk_gnsssdr and gnss-sdr in non-x86 archs
2023-01-26 18:39:25 +01:00
Carles Fernandez
e509354751
Fix for Android 2022-09-05 12:55:07 +02:00
Carles Fernandez
ea386ece93
Add include for cpowf 2022-09-05 12:30:12 +02:00
Carles Fernandez
8982e4d7ad
Enable NEON kernels in aarch64 architecture 2021-08-11 13:01:25 +02:00
Carles Fernandez
5f1268959e
Workaround for volk-gnsssdr-windows CI job 2021-04-17 14:10:14 +02:00
Carles Fernandez
68fdedb224
Remove _mm256_zeroupper() calls
Manual usage of _mm256_zeroupper() is not required and leads to miscompilation with GCC 10.2 and optimization level -O3
2021-01-04 14:07:56 +01:00
Carles Fernandez
7308745f05
Apply more concise file header format
Re-license CMake scripts with BSD-3-Clause
2020-12-30 13:35:06 +01:00
Carles Fernandez
7e1f0176f4
Make SIMD instructions work on MS Windows 2020-11-20 16:44:06 +01:00
Carles Fernandez
499de7a9f1
Update file headers 2020-07-28 16:57:15 +02:00
Carles Fernandez
0c4bdfac48
Use INT8_MIN/MAX instead of SCHAR_MIN/MAX for consistency and portability 2020-06-28 13:25:23 +02:00
Carles Fernandez
c470d43432
Fix shadowed variables
See MISRA C++:2008, 2-10-2 - Identifiers declared in an inner scope shall not hide an identifier declared in an outer scope
and https://rules.sonarsource.com/cpp/RSPEC-1117\?search\=shadow
2020-02-26 22:40:00 +01:00
Carles Fernandez
2ded19342f
Improve headers 2020-02-12 00:04:29 +01:00
Carles Fernandez
4d0d263280
Make the software package compliant with the REUSE Specification v3.0 (see https://reuse.software/spec/)
Update license headers to SPDX format (see https://spdx.org/)
Add license to all files
Add CI job in GitHub Actions to ensure compliance
2020-02-08 01:20:02 +01:00
Carles Fernandez
d6ec2a7b2c
Update links (mainly from http to https, fix broken links) 2020-02-05 21:24:46 +01:00
Carles Fernandez
bc7edede7c
Fixes for MSVC 2019-11-25 21:48:52 +01:00
Carles Fernandez
0520d400b3 Apply clang-tidy 2019-09-11 22:31:34 +00:00
Carles Fernandez
c4db5476fb Always have a space between // and comment 2019-08-19 10:49:37 +02:00
Carles Fernandez
a4d6800e22
Update copyright year
Applied on files changed since last release
clang-format applied
2019-07-26 12:38:20 +02:00
Carles Fernandez
41036c1538
Fix kernel 2019-03-24 19:53:51 +01:00
Carles Fernandez
441ff637eb
Fix AVX implementation
Samples that are remanent from num_points / 16 were not adecuately integrated
2019-03-24 13:50:12 +01:00
Carles Fernandez
33403dedea Fix high_dynamics correlator 2019-03-21 21:10:56 +01:00
Carles Fernandez
5f79053d65
Sort includes automatically with clang-format
Order:
1.- Main header
2.- Local headers
3.- Third-party library headers
4.- System headers

each category sorted lexicographically
2018-12-09 22:00:09 +01:00
Antonio Ramos
4375d0a051 Clean code 2018-09-17 19:19:19 +02:00
Antonio Ramos
2903e48e39 Dump carrier_phase_rate and modify matlab file 2018-09-08 23:20:29 +02:00
Antonio Ramos
b66540b776 Add moving average smoother to carrier phase rate 2018-09-07 19:18:03 +02:00
Antonio Ramos
47e5ef7f39 Improve generic kernel 2018-09-06 17:46:55 +02:00
Carles Fernandez
41527fb921
Rename volk_gnsssdr_32f_xn_fast_resampler_32f_xn to volk_gnsssdr_32f_xn_high_dynamics_resampler_32f_xnn 2018-08-23 19:39:41 +02:00
Carles Fernandez
d29befa7e3
Fix int vs unsigned int comparisons 2018-08-12 11:37:21 +02:00
Carles Fernandez
66bfbffe89
Add AVX implementation 2018-08-09 22:00:22 +02:00
Carles Fernandez
c5f10cd56c
Add sse4_1 implementation 2018-08-09 21:08:58 +02:00
Carles Fernandez
727cc115c4
Fix kernel 2018-08-08 18:20:09 +02:00
Carles Fernandez
0b5c827eda
Add Doppler rate in fast_resampler kernel. Still not used 2018-08-08 12:03:58 +02:00
Javier Arribas
a23231479f Adding new gnss-sdr volk kernel for a faster local signal replica generation 2018-08-03 11:02:01 +02:00
Carles Fernandez
927f4d6b21
Add AVX2 implementation 2018-07-20 09:39:37 +02:00
Carles Fernandez
0d60d97c24
Use https instead of http in links when available
Update copyright year in headers
2018-05-13 22:49:11 +02:00
Carles Fernandez
a92e6de3d9
Add support for aarch64 2018-04-29 01:19:48 +02:00
Carles Fernandez
48f65b067b Fix typos discovered with codespell 2018-03-25 13:43:12 +02:00
Carles Fernandez
891478cf2c Apply automated code formatting to volk-gnsssdr
See http://gnss-sdr.org/coding-style/#use-tools-for-automated-code-formatting
2018-03-03 12:09:45 +01:00
Carles Fernandez
2a811e96b7 Get rid of program_options and unit_test_framework boost libraries dependencies 2018-02-09 14:32:13 +01:00
Carles Fernandez
a6e6aef331 Fix / Add missing headers 2017-11-22 19:55:32 +01:00
Carles Fernandez
fe241c8b2d clean kernel 2017-11-16 20:54:04 +01:00
Carles Fernandez
eecfa6c582 Add AVX2 2017-11-16 19:51:15 +01:00
Carles Fernandez
6d216fa5a1 Add volk_gnsssdr_16ic_conjugate_16ic kernel 2017-11-16 19:45:15 +01:00
Carles Fernandez
27eece55da Add aligned kernel for AVX 2017-11-15 02:10:25 +01:00
Cillian O'Driscoll
676c1506da Updated volk_gnsssdr_module for real codes
Added 16i and 32f resamplers and 32fc_32f and 16ic_16i rotator dot product to
enable use of real (rather than complex) local code replicas
2017-09-11 15:15:27 +01:00
Damian Miralles
f4c221609c avx2: Adds support for AVX2 intrinsic to some kernels
Adds AVX2 intrinsic to several kernels ranging from multiply and add to
more detailed operations such as convert. New peotokernels also may take
advantage of the fused multiply add (fma)
2017-08-21 12:05:39 +02:00
Carles Fernandez
77eb09c2a7 Remove uneeded stdio.h 2017-08-13 10:01:10 +02:00
Carles Fernandez
d8fa3ff14f Replace 255 by 0xFF to apease the clang compiler
This avoids hundreds of warnings of type:

warning: implicit conversion from 'int' to 'char' changes value from 255
to -1 [-Wconstant-conversion]
mask_imag = _mm_set_epi8(255, 255, 0, 0, 255, 255, 0, 0, 255...
            ~~~~~~~~~~~~ ^~~
2017-04-01 11:34:46 +02:00
Carles Fernandez
67105d62b7 Avoid macro redefinitions 2016-10-16 13:01:53 +02:00