Expand description
Β§la-stack
Fast, stack-allocated linear algebra for fixed dimensions in Rust.
This crate grew from the need to support delaunay with fast, stack-allocated linear algebra primitives and algorithms
while keeping the API intentionally small and explicit.
Β§π Introduction
la-stack provides a handful of const-generic, stack-backed building blocks:
Vector<const D: usize>for fixed-length vectors ([f64; D]today)Matrix<const D: usize>for fixed-size square matrices ([[f64; D]; D]today)Lu<const D: usize>for LU factorization with partial pivoting (solve + det)Ldlt<const D: usize>for LDLT factorization without pivoting (solve + det; symmetric SPD/PSD)
Β§β¨ Design goals
- β
Copytypes where possible - β Const-generic dimensions (no dynamic sizes)
- β
const fnwhere possible (compile-time evaluation of determinants, dot products, etc.) - β Explicit algorithms (LU, solve, determinant)
- β
Robust geometric predicates via optional exact arithmetic (
det_sign_exact,det_errbound) - β
Exact linear system solve via optional arbitrary-precision arithmetic (
solve_exact,solve_exact_f64) - β No runtime dependencies by default (optional features may add deps)
- β Stack storage only (no heap allocation in core types)
- β
unsafeforbidden
See CHANGELOG.md for details.
Β§π« Anti-goals
- Bare-metal performance: see
blas-src,lapack-src,openblas-src - Comprehensive: use
nalgebraif you need a full-featured library - Large matrices/dimensions with parallelism: use
faerif you need this
Β§π’ Scalar types
The core types use f64. When f64 precision is insufficient (e.g. near-degenerate
geometric configurations), the optional "exact" feature provides arbitrary-precision
arithmetic via BigRational (see below).
Β§π Quickstart
Add this to your Cargo.toml:
[dependencies]
la-stack = "0.4.1"Solve a 5Γ5 system via LU:
use la_stack::prelude::*;
// This system requires pivoting (a[0][0] = 0), so it's a good LU demo.
// A = J - I: zeros on diagonal, ones elsewhere.
let a = Matrix::<5>::from_rows([
[0.0, 1.0, 1.0, 1.0, 1.0],
[1.0, 0.0, 1.0, 1.0, 1.0],
[1.0, 1.0, 0.0, 1.0, 1.0],
[1.0, 1.0, 1.0, 0.0, 1.0],
[1.0, 1.0, 1.0, 1.0, 0.0],
]);
let b = Vector::<5>::new([14.0, 13.0, 12.0, 11.0, 10.0]);
let lu = a.lu(DEFAULT_PIVOT_TOL).unwrap();
let x = lu.solve_vec(b).unwrap().into_array();
// Floating-point rounding is expected; compare with a tolerance.
let expected = [1.0, 2.0, 3.0, 4.0, 5.0];
for (x_i, e_i) in x.iter().zip(expected.iter()) {
assert!((*x_i - *e_i).abs() <= 1e-12);
}Compute a determinant for a symmetric SPD matrix via LDLT (no pivoting).
For symmetric positive-definite matrices, LDL^T is essentially a square-root-free form of the Cholesky decomposition
(you can recover a Cholesky factor by absorbing sqrt(D) into L):
use la_stack::prelude::*;
// This matrix is symmetric positive-definite (A = L*L^T) so LDLT works without pivoting.
let a = Matrix::<5>::from_rows([
[1.0, 1.0, 0.0, 0.0, 0.0],
[1.0, 2.0, 1.0, 0.0, 0.0],
[0.0, 1.0, 2.0, 1.0, 0.0],
[0.0, 0.0, 1.0, 2.0, 1.0],
[0.0, 0.0, 0.0, 1.0, 2.0],
]);
let det = a.ldlt(DEFAULT_SINGULAR_TOL).unwrap().det();
assert!((det - 1.0).abs() <= 1e-12);β οΈ LDLT precondition: The input matrix must be symmetric. Symmetry is verified by a
debug_assert!in debug builds only; release builds silently accept asymmetric inputs and produce a meaningless factorization. Pre-validate withMatrix::is_symmetric(or locate the offending pair withMatrix::first_asymmetry) when you cannot statically guarantee symmetry, or fall back tolu()if your matrices may not be symmetric at all. SeeMatrix::ldltfor details.
Β§β‘ Compile-time determinants (D β€ 4)
det_direct() is a const fn providing closed-form determinants for D=0β4,
using fused multiply-add where applicable. Matrix::<0>::zero().det_direct()
returns Some(1.0) (the empty-product convention). For D=1β4, cofactor
expansion bypasses LU factorization entirely. This enables compile-time
evaluation when inputs are known:
use la_stack::prelude::*;
// Evaluated entirely at compile time β no runtime cost.
const DET: Option<f64> = {
let m = Matrix::<3>::from_rows([
[2.0, 0.0, 0.0],
[0.0, 3.0, 0.0],
[0.0, 0.0, 5.0],
]);
m.det_direct()
};
assert_eq!(DET, Some(30.0));The public det() method automatically dispatches through the closed-form path
for D β€ 4 and falls back to LU for D β₯ 5 β no API change needed.
Β§π¬ Exact arithmetic ("exact" feature)
The default build has zero runtime dependencies. Enable the optional
exact Cargo feature to add exact arithmetic methods using arbitrary-precision
rationals (this pulls in num-bigint, num-rational, and num-traits for
BigRational):
[dependencies]
la-stack = { version = "0.4.1", features = ["exact"] }Determinants:
det_exact()β returns the exact determinant as aBigRationaldet_exact_f64()β returns the exact determinant converted to the nearestf64det_sign_exact()β returns the provably correct sign (β1, 0, or +1)
Linear system solve:
solve_exact(b)β solvesAx = bexactly, returning[BigRational; D]solve_exact_f64(b)β solvesAx = bexactly, converting the result toVector<D>(f64)
use la_stack::prelude::*;
// Exact determinant
let m = Matrix::<3>::from_rows([
[1.0, 2.0, 3.0],
[4.0, 5.0, 6.0],
[7.0, 8.0, 9.0],
]);
assert_eq!(m.det_sign_exact().unwrap(), 0); // exactly singular
let det = m.det_exact().unwrap();
assert_eq!(det, BigRational::from_integer(0.into())); // exact zero
// Exact linear system solve
let a = Matrix::<2>::from_rows([[1.0, 2.0], [3.0, 4.0]]);
let b = Vector::<2>::new([5.0, 11.0]);
let x = a.solve_exact_f64(b).unwrap().into_array();
assert!((x[0] - 1.0).abs() <= f64::EPSILON);
assert!((x[1] - 2.0).abs() <= f64::EPSILON);With the exact feature enabled, BigInt and BigRational are re-exported
from the crate root and prelude, alongside the most commonly needed
num-traits items (FromPrimitive, ToPrimitive, Signed). This lets
consumers construct exact values (BigRational::from_f64, from_i64), query
sign (is_positive / is_negative), and convert back to f64 (to_f64)
with a single use la_stack::prelude::*; β no need to add num-bigint,
num-rational, or num-traits to their own Cargo.toml.
For det_sign_exact(), D β€ 4 matrices use a fast f64 filter (error-bounded
det_direct()) that resolves the sign without allocating. Only near-degenerate
or large (D β₯ 5) matrices fall through to the exact Bareiss algorithm.
Β§Adaptive precision with det_errbound()
det_errbound() returns the conservative absolute error bound used by the fast
filter. This method does NOT require the exact feature β it uses pure f64
arithmetic and is available by default. This enables building custom
adaptive-precision logic for geometric predicates:
use la_stack::prelude::*;
let m = Matrix::<3>::identity();
if let Some(bound) = m.det_errbound() {
let det = m.det_direct().unwrap();
if det.abs() > bound {
// f64 sign is guaranteed correct
let sign = det.signum() as i8;
} else {
// Fall back to exact arithmetic (requires `exact` feature)
let sign = m.det_sign_exact().unwrap();
}
} else {
// D β₯ 5: no fast filter, use exact directly (requires `exact` feature)
let sign = m.det_sign_exact().unwrap();
}The error coefficients (ERR_COEFF_2, ERR_COEFF_3, ERR_COEFF_4) are also
exposed for advanced use cases.
Β§π§© API at a glance
| Type | Storage | Purpose | Key methods |
|---|---|---|---|
Vector<D> | [f64; D] | Fixed-length vector | new, zero, dot, norm2_sq |
Matrix<D> | [[f64; D]; D] | Square matrix | See below |
Lu<D> | Matrix<D> + pivot array | Factorization for solves/det | solve_vec, det |
Ldlt<D> | Matrix<D> | Factorization for symmetric SPD/PSD solves/det | solve_vec, det |
Storage shown above reflects the f64 implementation.
Matrix<D> key methods: lu, ldlt, det, det_direct, det_errbound,
det_exactΒΉ, det_exact_f64ΒΉ, det_sign_exactΒΉ, solve_exactΒΉ, solve_exact_f64ΒΉ.
ΒΉ Requires features = ["exact"].
Β§π Benchmarks (vs nalgebra/faer)
Raw data: docs/assets/bench/vs_linalg_lu_solve_median.csv
Summary (median time; lower is better). The βla-stack vs nalgebra/faerβ columns show the % time reduction relative to each baseline (positive = la-stack faster):
| D | la-stack median (ns) | nalgebra median (ns) | faer median (ns) | la-stack vs nalgebra | la-stack vs faer |
|---|---|---|---|---|---|
| 2 | 2.173 | 4.448 | 139.923 | +51.2% | +98.4% |
| 3 | 13.989 | 34.607 | 180.026 | +59.6% | +92.2% |
| 4 | 27.580 | 48.435 | 203.163 | +43.1% | +86.4% |
| 5 | 53.517 | 75.935 | 274.375 | +29.5% | +80.5% |
| 8 | 134.859 | 162.859 | 371.463 | +17.2% | +63.7% |
| 16 | 635.775 | 576.171 | 846.189 | -10.3% | +24.9% |
| 32 | 2,704.570 | 2,731.740 | 2,589.494 | +1.0% | -4.4% |
| 64 | 17,381.460 | 13,744.505 | 11,276.642 | -26.5% | -54.1% |
Β§π Examples
The examples/ directory contains small, runnable programs:
solve_5x5β solve a 5Γ5 system via LU with partial pivotingdet_5x5β determinant of a 5Γ5 matrix via LUldlt_solve_3x3β solve a 3Γ3 symmetric positive definite system via LDLTconst_det_4x4β compile-time 4Γ4 determinant viadet_direct()exact_det_3x3β exact determinant value of a near-singular 3Γ3 matrix (requiresexactfeature)exact_sign_3x3β exact determinant sign of a near-singular 3Γ3 matrix (requiresexactfeature)exact_solve_3x3β exact solve of a near-singular 3Γ3 system vs f64 LU (requiresexactfeature)
just examples
# or individually:
cargo run --example solve_5x5
cargo run --example det_5x5
cargo run --example ldlt_solve_3x3
cargo run --example const_det_4x4
cargo run --features exact --example exact_det_3x3
cargo run --features exact --example exact_sign_3x3
cargo run --features exact --example exact_solve_3x3Β§π€ Contributing
A short contributor workflow:
cargo install just
just setup # install/verify dev tools + sync Python deps
just check # lint/validate (non-mutating)
just fix # apply auto-fixes (mutating)
just ci # lint + tests + examples + bench compileFor the full set of developer commands, see just --list and AGENTS.md.
Β§π Citation
If you use this library in academic work, please cite it using CITATION.cff (or GitHubβs βCite this repositoryβ feature). A Zenodo DOI will be added for tagged releases.
Β§π References
For canonical references to the algorithms used by this crate, see REFERENCES.md.
Β§π€ AI Agents
This repository contains an AGENTS.md file, which defines the canonical rules and invariants
for all AI coding assistants and autonomous agents working on this codebase.
AI tools (including ChatGPT, Claude, CodeRabbit, KiloCode, and WARP) are expected to read
and follow AGENTS.md when proposing or applying changes.
Portions of this library were developed with the assistance of these AI tools:
All code was written and/or reviewed and validated by the author.
For full tool citation metadata, see the AI-Assisted Development Tools
section of REFERENCES.md.
Β§π License
BSD 3-Clause License. See LICENSE.
ModulesΒ§
- prelude
- Common imports for ergonomic usage.
StructsΒ§
- BigInt
- A big signed integer type.
- Ldlt
- LDLT factorization (
A = L D Lα΅) for symmetric positive (semi)definite matrices. - Lu
- LU decomposition (PA = LU) with partial pivoting.
- Matrix
- Fixed-size square matrix
DΓD, stored inline. - Vector
- Fixed-size vector of length
D, stored inline.
EnumsΒ§
- LaError
- Linear algebra errors.
ConstantsΒ§
- DEFAULT_
PIVOT_ TOL - Default absolute pivot magnitude threshold used for LU pivot selection / singularity detection.
- DEFAULT_
SINGULAR_ TOL - Default absolute threshold used for singularity/degeneracy detection.
- ERR_
COEFF_ 2 - Absolute error coefficient for
Matrix::<2>::det_direct. - ERR_
COEFF_ 3 - Absolute error coefficient for
Matrix::<3>::det_direct. - ERR_
COEFF_ 4 - Absolute error coefficient for
Matrix::<4>::det_direct.
TraitsΒ§
- From
Primitive - A generic trait for converting a number to a value.
- Signed
- Useful functions for signed numbers (i.e. numbers that can be negative).
- ToPrimitive
- A generic trait for converting a value to a number.
Type AliasesΒ§
- BigRational
- Alias for arbitrary precision rationals.