linkcell

Periodic linked-cell k-nearest neighbour search for molecular simulations.

vesin builds cutoff pair lists. nanoflann builds Euclidean KD-trees without a minimum-image convention. This crate is the piece those two leave open: the linked-cell walk of Allen and Tildesley (Computer Simulation of Liquids), a k-heap per source, shells expanded until the k-th neighbour cannot sit outside the visited cube. The optional gpulite path runs that walk on a CUDA device (linkcell::gpu::Workspace); pair lists stay on the device.

It is a LODE library. The Rust crate is the implementation. The C ABI (lc_*) is the hourglass waist, the same shape as readcon-core. C++ is a RAII header over that ABI.

Released versions live in CHANGELOG.md (Keep a Changelog). Unreleased notes are files under changelog.d/. The cut is towncrier build --version X.Y.Z.

Install

Rust:

cargo add linkcell

C and C++ consumers include linkcell.h or linkcell.hpp and link the installed library. Meson, CMake, and pkg-config install both shared and static CPU libraries. The device API is a separate static library with linkcell_gpu.h / linkcell_gpu.hpp.

Python takes the same search through DLPack (dlpk). Any __dlpack__() object (numpy, torch, jax, cupy) is a valid xyz / cell, on any device. A CUDA xyz stays on device; a CUDA cell is inverted on device. torch.from_dlpack consumes (indices, dist2). Wheels: one CPython 3.12 limited-ABI (abi3) artifact per platform (GIL, 3.12+), and one abi3t artifact (CPython 3.15+ GIL and free-threaded).

pip install linkcell

import numpy as np
import linkcell

xyz = np.array([[0.2, 0.0, 0.0], [9.4, 0.0, 0.0]], dtype=np.float64)
cell = np.array([10.0, 10.0, 10.0], dtype=np.float64)
nn, d2 = linkcell.knearest(xyz, cell, 1)
nn = np.from_dlpack(nn)

Meson

meson setup build
meson compile -C build
meson install -C build

As a wrap, Meson exposes linkcell_dep:

[wrap-git]
url = https://github.com/d-SEAMS/linkcell.git
revision = v0.3.4
depth = 1

[provide]
linkcell = linkcell_dep
linkcell-gpu = linkcell_gpu_dep

linkcell_dep = dependency('linkcell', fallback: ['linkcell', 'linkcell_dep'])
linkcell_gpu_dep = dependency(
  'linkcell-gpu',
  fallback: ['linkcell', 'linkcell_gpu_dep'],
)

with_gpulite controls the device target and defaults to auto.

CMake

cmake -B build -DCMAKE_INSTALL_PREFIX=$PREFIX
cmake --build build
cmake --install build

find_package(linkcell 0.3 REQUIRED)
target_link_libraries(c_app PRIVATE linkcell::linkcell)
target_link_libraries(cpp_app PRIVATE linkcell::cpp)
target_link_libraries(device_app PRIVATE linkcell::gpu)

LINKCELL_WITH_GPULITE=ON builds linkcell::gpu and is the default. linkcell::cpp carries the C++17 requirement; linkcell::linkcell stays valid in a C-only project. The same target names work in the build tree and from an installed prefix.

pkg-config

pkg-config --cflags --libs linkcell
pkg-config --cflags --libs linkcell-gpu

Both Meson and CMake write linkcell.pc and, when the device target is built, linkcell-gpu.pc. Use pkg-config --static when selecting the CPU static archive and its private system libraries.

Rust

use linkcell::{knearest, Cell};

let sim = Cell::ortho(10.0, 10.0, 10.0)?;
let sheared = Cell::from_vectors(
    [10.0, 0.0, 0.0],
    [5.0, 8.66, 0.0],
    [0.0, 0.0, 10.0],
    [0.0, 0.0, 0.0],
)?;
let xyz = [[0.0, 0.0, 0.0], [1.0, 0.0, 0.0]];
let rows = knearest(&xyz, &sim, 1, None, None)?;
assert_eq!(rows[0].indices, vec![1]);

mask[i] == false removes a point as both a source and a candidate. cell_hint is the target cell edge; None uses 3.0 in the box units.

C

#include "linkcell.h"

double xyz[] = {0.0, 0.0, 0.0, 1.0, 0.0, 0.0};
lc_cell box = lc_cell_ortho(10.0, 10.0, 10.0);
int out[2];
if (lc_knearest(xyz, 2, &box, 1, NULL, 0.0, out) != 0) {
  return 1;
}

n and k are size_t. out has length n * k. Unused slots are -1. Neighbours of source i are out[i*k + 0 ..], nearest first.

C++

#include "linkcell.hpp"

const linkcell::Cell box = linkcell::Cell::ortho(10.0, 10.0, 10.0);
const double xyz[] = {0.0, 0.0, 0.0, 1.0, 0.0, 0.0};
const linkcell::Neighbours nn = linkcell::knearest(xyz, 2, box, 1);

nn owns the packed n * k buffer. Unused slots are -1. nn.neighbour(i, j) is the j-th neighbour of i. Failure throws linkcell::Error.

The optional device walk reads occupancy from the environment: LINKCELL_TPP (threads per particle) and LINKCELL_BLOCK (CUDA block size). Unset, the library picks a pair that maximises particles per block under the device thread and 48 KiB shared-memory limits. d-SEAMS writes the same keys from SEAMS_CONFIG / seams --tpp.

Docs

Rust API: docs.rs/linkcell. Map: docs/index.md.

Design

• The cell is a general parallelepiped: three lattice vectors plus an origin. Orthorhombic boxes are Cell::ortho / lc_cell_ortho. Binning is in fractional space, so a sheared dump is not treated as orthogonal.
• Points fold into the primary cell once. Each source then walks Chebyshev shells of neighbour cells. Pair distances are a Cartesian subtract plus that cell's lattice translation (dist2_shifted and lattice_shift), the vesin / LAMMPS ghost construction. Orthorhombic boxes use three independent wraps and skip the two Hinv matvecs.
• One lattice shift per unique cell is wrong unless every wrap of that cell is visited. The walk visits integer cell offsets, so each wrap of a bin is a separate visit.
• The search does not take a cutoff. A cell-size hint only sets the bin width. Shells grow until the k-heap is exact.
• vesin remains the right library for a cutoff pair list.

License

MIT