lam.bitvector

This is my bitvector library. It was originally written in C for an undergraduate course, Programming in C. It's much different now.

Bitvector, a LAM library

Not much more than another bitvector at this time, although it is appreciably faster than std::bitset in my limited micro benchmarks (hopefully part of the 1% that matters). Here are some current features of the c++ module lam.bitvector:

• Modules-First: Built as a native C++23 module (import lam.bitvector;).
• Fully constexpr: Usable in compile-time contexts, including dynamic allocation.
• Allocator-Aware & Concepts-Driven: custom concept bitvector_allocator_c with full support for std::pmr and custom allocators.
• Memory Safety: extensive validation via AddressSanitizer (ASan) and UndefinedBehaviorSanitizer (UBSan).
• Modern IO: Built-in support for std::print and std::format.
• High Performance: Optimized for cache locality; faster than std::bitset in local parallel/chunked benchmarks.
• Core Guidelines Compliant: Clean, modern C++ codebase combed through by clang-tidy.

Currently this software is unlicensed. It is subject to change both in form and in license at any time; I plan to endow a license when the time is right. Before then I would like to test it with a few other projects.

Usage

import std;
import lam.bitvector;

int main() 
{
  lam::bitvector<> bv(64);
  bv.set(42);
  bv.toggle(10);
  
  std::println("Bit 42 is: {}", bv[42]);
  std::println("Population count: {}", bv.count());
}

Building and Testing

Requires CMake 3.31.6+ and a c++23 compiler that CMake can use to build import std, which these days is Clang 18+, GCC 15.2+.

# Build (Release)
cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Release
cmake --build build

# Run Tests
ctest --test-dir build

# Run Benchmarks
./build/benchmark

Configuration

You can enable sanitizers for debug builds using CMake options:

Option	Description	Default
LAM_ENABLE_ASAN	Enable AddressSanitizer (Memory safety)	OFF
LAM_ENABLE_UBSAN	Enable UndefinedBehaviorSanitizer	OFF
LAM_ENABLE_TSAN	Enable ThreadSanitizer (Race detection)	OFF

Example:

cmake -B build -DLAM_ENABLE_ASAN=ON -DLAM_ENABLE_UBSAN=ON

Performance

Results from Sieve of Eratosthenes (N=100,000,000) on local hardware:

Implementation	Time (ms)	Speedup
Sequential	~612 ms	1.0x
Parallel (8 threads)	~144 ms	4.2x
Chunked (L1 Cache)	~95 ms	6.4x

Measured with std::chrono::steady_clock in Release mode.