| AArch64 | |
| ppc64le | |
| s390x | |
| x86-64 |
- Update to primesieve-12.4 - Move x86 CPUID code from cpuid.hpp to src/x86/cpuid.cpp - multiarch_x86_popcnt.cmake: Detect x86 POPCNT support - CMakeLists.txt: Use CMake list for all compile time definitions - CMakeLists.txt: Use CMake list for all link libraries
- Update to primesieve-12.3 - Add runtime POPCNT detection using CPUID for x86 CPUs - Improve GCC/Clang multiarch preprocessor logic - CMakeLists.txt: Remove POPCNT/BMI check for x86 CPUs
- Update to primesieve-12.1 - CMakeLists.txt: Fix undefined reference to pthread_create - Fix -ffast-math failure of --test option - Fix musl libc issue in unit tests - Improve status output
- Update to primesieve-12.0 - New -S and --stress-test[=MODE] command-line options. - New --timeout=SECS option for stress testing. - New -R && --RiemannR command line options. - New --RiemannR-inverse command line option. - Faster Riemann R function implementation. - Improve command-line option handling.
- Update to primesieve-11.2 - Fix CMake libatomic detection - Improved nth prime algorithm - Fix off by 1 error in OpenMP example in C_API.md - Fix off by 1 error in OpenMP example in CPP_API.md
- Update to primesieve-7.5
- Trim subjective parts from description.
- Fix review issues - Fix rpmlint warnings - Convert Fedora spec file to openSUSE format
- Update to primesieve-11.1 - Vectorized pre-sieving algorithm using x64 SSE2 and ARM NEON - Added POPCNT algorithm for x64 & AArch64 - Fix -Wstrict-prototypes warnings
- Update to primesieve-11.0 - primesieve version now match libprimesieve version (.so version) - Added new primesieve::iterator::jump_to() method (C++ API) - Added new primesieve_jump_to() function (C API) - Mark primesieve_skipto() as deprecated - Fix use after free in primesieve::iterator::clear() - Fix use after free in primesieve_iterator_clear() - Fix potential memory leak in malloc_vector.hpp
- Update to primesieve-8.0 - Added multiarch support (POPCNT, BMI2, AVX512) for x64 CPUs - Generating an array (or vector) of primes is up to 20% faster - Main CMakeLists.txt has been split up into multiple modules - Improved documentation of C & C++ APIs
- Update to primesieve-7.9 - Reduced memory usage and faster initialization - Reduced branch mispredictions by up to 30% - Improved nth prime approximation - Increase max sieve array size to 8 KiB
- Update to primesieve-7.8 - Improved pre-sieving, up to 10% speedup - Improved primesieve::iterator, up to 15% speedup - C++ refactoring: get rid of using namespace std