Up to [cvs.NetBSD.org] / pkgsrc / math / volk
Request diff between arbitrary revisions
Keyword substitution: kv
Default branch: MAIN
volk: updated to 3.3.0 3.3.0 New kernels volk_16i_x2_add_saturated_16i volk_16u_x2_add_saturated_16u volk_32f_sincos_32f_x2.h volk_64f_x2_dot_prod_64f.h volk_8i_x2_add_saturated_8i.h volk_8u_x2_add_saturated_8u.h Improvements to a lot of kernels RiscV kernels are further improved and fixed RVV index_max/min kernels always return the correct (first) index now New AVX512 implementations for a lot of kernels Add more NEON kernels with better accuracy Documentation Working on auto-publishing latest docs More clarification on our software library dependencies policy Improved documentation on the underlying algorithms that are used Code quality cx-limited-range: Reduce scope of compile feature Fully rely on std::filesystem (we used to have a boost::filesystem fallback) Align CMake auto-format with GNU Radio Update to modern PIC enablement Fix NEON compile checks Update code style in more places tighter CI Add -Werror flag to CI for C compilation Remove obsolete CI, add new CI Fix obsolete MacOS Intel CI Tests Add specialized test suite for the rotator kernel Improved usability with gtest Tighter error bounds for a lot of implementations Usability new performance test output fastest implementation is marked with a star speed up vs. generic implementation is printed test "heat up" added
volk: updated to 3.2.0 3.2.0 New and improved kernels add RISC-V Vector extension (RVV) kernels New AVX512F implementation Improved and modernized CI ci: Add first native Linux ARM runners macos: Fix CI dependency error appveyor: Update to VS 2022/Python 3.12 Update android_build.yml Improved builds cmake: Fix 64bit host CPU detection cmake: Suppress invalid escape sequence warnings with Python 3.12 cmake/pkgconfig: use CMAKE_INSTALL_FULL_* variables cmake: Fix VOLK as a submodule build issue Adds toolchain file for Raspberry Pi 5 New and improved tests gtest: Start work on new test infrastructure tests: Add a log info print test gtest: Make gtest an install dependency gtest: Enable GTests in CI workflows tests: Beautify test output Documentation cpu_features: Update hints in README Code quality Add const to several args Usability features feature: add env variable kernel override
volk: update to 2.5.1 Now correctly detects external devel/cpu_features package, so remove patches for that. Also remove boost dependency since the package was changed to use C++17 instead of boost.
volk: update to 2.5.0
Kernels
volk_32f_stddev_and_mean_32f_x2: implemented Young and Cramer's algorithm
volk_32fc_accumulator_s32fc: Add new kernel
volk_16ic_x2_dot_prod_16ic_u_avx2: Fix Typo, was _axv2.
Remove _mm256_zeroupper() calls
Enforce consistent function prototypes
32fc_index_max: Improve speed of AVX2 version
conv_k7_r2: Disable broken AVX2 code
improve volk_8i_s32f_convert_32f for ARM NEON
Calculate cos in AVX512F
Calculate sin using AVX512F
Build
Fix python version detection
cmake: Check that 'distutils' is available
c11: Remove pre-C11 preprocessor instructions
volk: update to 2.4.0
pkgsrc: since the volk release tarball only provides the required cpu_features
library as a git submodule that has to be manually cloned, we instead depend on
devel/cpu_features from pkgsrc and patch the cmake files to accept the static
library as an external dependency.
Documentation
Update README to be more verbose and to improve usefulness.
Compilers
AppleClang: Treat AppleClang as Clang.
Paired with the cpu_features introduction, this enables us to use architecture
specific kernels on a broader set of platforms.
CMake
Enable to not install volk_modtool.
Remove "find_package_handle_standard_args" warning.
cpu_features
Use cpu_features v0.6.0 as a private submodule to detect available CPU
features.
Fix incorrect feature detection for newer AVX versions.
Circumvent platform specific feature detection.
Enable more architecture specific kernels on more platforms.
Kernels
Disable slow and broken SSE4.1 kernel in volk_32fc_x2_dot_prod_32fc
Adjust min/max for 32f_s32f_convert_8i kernel
Use INT8_* instead of CHAR_*
math/volk: import volk-2.3.0 The Vector-Optimized Library of Kernel is a library that contains kernels of hand-written SIMD code for different mathematical operations. Since each SIMD architecture can be very different and no compiler has yet come along to handle vectorization properly or highly efficiently, VOLK approaches the problem differently. VOLK is a sub-project of GNU Radio.