クワッドソートで配列を並び替える

set -euo pipefail

WORKDIR="$(mktemp -d)"
trap 'rm -rf "$WORKDIR"' EXIT

cat > "$WORKDIR/Dockerfile" <<'EOF'
FROM rust:1.95.0

WORKDIR /app

RUN mkdir -p src

RUN cat > Cargo.toml <<'CARGO'
[package]
name = "rust-benchmark"
version = "0.1.0"
edition = "2021"

[profile.release]
lto = true
codegen-units = 1
panic = "abort"
CARGO

RUN cat > src/main.rs <<'RUST'
use std::{
    alloc::{GlobalAlloc, Layout, System},
    env,
    process::Command,
    sync::atomic::{AtomicUsize, Ordering as AtomicOrdering},
    time::{Duration, Instant},
};

/// Counts live heap bytes and the high-water mark so auxiliary sort buffers
/// (swap Vecs, etc.) are measured as explicit heap growth during the sort.
struct TrackingAllocator;

static LIVE_BYTES: AtomicUsize = AtomicUsize::new(0);
static PEAK_BYTES: AtomicUsize = AtomicUsize::new(0);

fn record_alloc(size: usize) {
    let live = LIVE_BYTES.fetch_add(size, AtomicOrdering::Relaxed) + size;
    PEAK_BYTES.fetch_max(live, AtomicOrdering::Relaxed);
}

unsafe impl GlobalAlloc for TrackingAllocator {
    unsafe fn alloc(&self, layout: Layout) -> *mut u8 {
        let ptr = System.alloc(layout);
        if !ptr.is_null() {
            record_alloc(layout.size());
        }
        ptr
    }

    unsafe fn dealloc(&self, ptr: *mut u8, layout: Layout) {
        LIVE_BYTES.fetch_sub(layout.size(), AtomicOrdering::Relaxed);
        System.dealloc(ptr, layout);
    }

    unsafe fn alloc_zeroed(&self, layout: Layout) -> *mut u8 {
        let ptr = System.alloc_zeroed(layout);
        if !ptr.is_null() {
            record_alloc(layout.size());
        }
        ptr
    }

    unsafe fn realloc(&self, ptr: *mut u8, layout: Layout, new_size: usize) -> *mut u8 {
        let new_ptr = System.realloc(ptr, layout, new_size);
        if !new_ptr.is_null() {
            LIVE_BYTES.fetch_sub(layout.size(), AtomicOrdering::Relaxed);
            record_alloc(new_size);
        }
        new_ptr
    }
}

#[global_allocator]
static GLOBAL: TrackingAllocator = TrackingAllocator;
const MIN_POWER: u32 = 8;
const MAX_POWER: u32 = 18;
const RUNS: usize = 8192;


/// Sort four elements with a small sorting network (equals keep order via `>`).
fn quad_swap4(a: &mut [usize], i0: usize, i1: usize, i2: usize, i3: usize) {
    if a[i0] > a[i1] {
        a.swap(i0, i1);
    }
    if a[i2] > a[i3] {
        a.swap(i2, i3);
    }
    if a[i0] > a[i2] {
        a.swap(i0, i2);
    }
    if a[i1] > a[i3] {
        a.swap(i1, i3);
    }
    if a[i1] > a[i2] {
        a.swap(i1, i2);
    }
}

fn quad_is_sorted(a: &[usize]) -> bool {
    a.windows(2).all(|w| w[0] <= w[1])
}

fn quad_is_reverse_sorted(a: &[usize]) -> bool {
    a.windows(2).all(|w| w[0] >= w[1])
}

fn quad_reverse(a: &mut [usize]) {
    let mut lo = 0;
    let mut hi = a.len();
    while lo + 1 < hi {
        hi -= 1;
        a.swap(lo, hi);
        lo += 1;
    }
}

/// Stable two-way merge from `src[lo..mid)` and `src[mid..hi)` into `dst[lo..hi)`.
fn quad_merge_two(src: &[usize], dst: &mut [usize], lo: usize, mid: usize, hi: usize) {
    let mut i = lo;
    let mut j = mid;
    let mut k = lo;
    while i < mid && j < hi {
        if src[i] <= src[j] {
            dst[k] = src[i];
            i += 1;
        } else {
            dst[k] = src[j];
            j += 1;
        }
        k += 1;
    }
    while i < mid {
        dst[k] = src[i];
        i += 1;
        k += 1;
    }
    while j < hi {
        dst[k] = src[j];
        j += 1;
        k += 1;
    }
}

/// True when four consecutive sorted blocks of length `block` are already ordered
/// across boundaries (skipping the merge is safe).
fn quad_blocks_ordered(a: &[usize], start: usize, block: usize) -> bool {
    a[start + block - 1] <= a[start + block]
        && a[start + block * 2 - 1] <= a[start + block * 2]
        && a[start + block * 3 - 1] <= a[start + block * 3]
}

/// Ping-pong quad merge: two pairwise merges into swap, then one merge back into `a`.
fn quad_merge_four(a: &mut [usize], swap: &mut [usize], start: usize, block: usize) {
    let mid1 = start + block;
    let mid2 = start + block * 2;
    let mid3 = start + block * 3;
    let end = start + block * 4;
    if quad_blocks_ordered(a, start, block) {
        return;
    }
    quad_merge_two(a, swap, start, mid1, mid2);
    quad_merge_two(a, swap, mid2, mid3, end);
    quad_merge_two(swap, a, start, mid2, end);
}

/// Binary bottom-up merge for a partial span that is not a full group of four blocks.
fn quad_merge_remainder(a: &mut [usize], swap: &mut [usize], start: usize, n: usize, block: usize) {
    let mut width = block;
    while start + width < n {
        let mut lo = start;
        while lo + width < n {
            let mid = lo + width;
            let hi = (lo + width * 2).min(n);
            if a[mid - 1] > a[mid] {
                quad_merge_two(a, swap, lo, mid, hi);
                a[lo..hi].copy_from_slice(&swap[lo..hi]);
            }
            lo = hi;
        }
        width *= 2;
    }
}

fn quad_sort(a: &mut [usize]) {
    let n = a.len();
    if n <= 1 {
        return;
    }
    if quad_is_sorted(a) {
        return;
    }
    if quad_is_reverse_sorted(a) {
        quad_reverse(a);
        return;
    }

    // Analyzer / quad-swap: leave sorted blocks of 4 (educational stand-in for 8).
    let mut i = 0;
    while i + 4 <= n {
        quad_swap4(a, i, i + 1, i + 2, i + 3);
        i += 4;
    }
    if i < n {
        for j in (i + 1)..n {
            let key = a[j];
            let mut k = j;
            while k > i && a[k - 1] > key {
                a[k] = a[k - 1];
                k -= 1;
            }
            a[k] = key;
        }
    }

    let mut swap = vec![0usize; n];
    let mut block = 4usize;
    while block < n {
        let stride = block * 4;
        let mut start = 0usize;
        while start < n {
            let rem = n - start;
            if rem <= block {
                break;
            }
            if rem >= stride {
                quad_merge_four(a, &mut swap, start, block);
                start += stride;
            } else {
                quad_merge_remainder(a, &mut swap, start, n, block);
                break;
            }
        }
        block *= 4;
    }
}


fn benchmark_sort(array: &mut [usize]) {

    quad_sort(array);

}

fn is_non_decreasing(a: &[usize]) -> bool {
    a.windows(2).all(|w| w[0] <= w[1])
}

fn same_multiset(a: &[usize], b: &[usize]) -> bool {
    if a.len() != b.len() {
        return false;
    }

    let mut left = a.to_vec();
    let mut right = b.to_vec();
    left.sort_unstable();
    right.sort_unstable();
    left == right
}

fn check_correctness_case(label: &str, mut input: Vec<usize>) {
    let original = input.clone();

    benchmark_sort(&mut input);

    if !is_non_decreasing(&input) {
        panic!("correctness case {}: output is not sorted", label);
    }

    if !same_multiset(&input, &original) {
        panic!("correctness case {}: elements were lost or added", label);
    }
}

fn few_unique_values(size: usize, unique: usize, seed: u64) -> Vec<usize> {
    let mut state = seed;

    (0..size)
        .map(|_| {
            state ^= state << 13;
            state ^= state >> 7;
            state ^= state << 17;
            (state as usize % unique) + 1
        })
        .collect()
}

fn run_correctness_checks() {
    check_correctness_case("empty", vec![]);
    check_correctness_case("single", vec![42]);
    check_correctness_case("duplicates", vec![3, 1, 3, 2, 1, 2]);
    check_correctness_case("sorted", vec![1, 2, 3, 4, 5]);
    check_correctness_case("reverse", vec![5, 4, 3, 2, 1]);
    check_correctness_case("all_equal", vec![7, 7, 7, 7]);
    check_correctness_case("skewed_range", vec![1_000_000, 2, 1_000_001, 1, 999_999]);
    // Static-buffer Grail skips the in-buffer build when key collection is sparse
    // (ideal_buffer = false). Exercising that path catches regressions in buffer gating.
    check_correctness_case(
        "few_keys_len16",
        vec![2, 2, 2, 2, 2, 2, 2, 2, 4, 3, 1, 2, 3, 4, 1, 4],
    );
    // Seed 0 is a fixed point of the xorshift below, so it would degenerate into
    // yet another all-equal case instead of a 4-value mix. Start at 1.
    for seed in 1..=32 {
        check_correctness_case(
            &format!("few_keys_len32_seed_{seed}"),
            few_unique_values(32, 4, seed),
        );
    }
    // Small-input cutoffs (insertion sort below 32 elements, etc.) hide duplicate-key
    // bugs in the recursive path, so repeat the duplicate cases at the smallest
    // benchmark size, which every algorithm must handle within reasonable time.
    check_correctness_case("all_equal_len256", vec![7; 256]);
    for seed in 1..=4 {
        check_correctness_case(
            &format!("few_keys_len256_seed_{seed}"),
            few_unique_values(256, 4, seed),
        );
    }
}


fn shuffled(size: usize, seed: u64) -> Vec<usize> {
    let mut v: Vec<usize> = (1..=size).collect();

    let mut state = seed;

    for i in (1..size).rev() {
        state ^= state << 13;
        state ^= state >> 7;
        state ^= state << 17;

        let j = (state as usize) % (i + 1);

        v.swap(i, j);
    }

    v
}

fn micros(d: Duration) -> u128 {
    d.as_micros()
}

fn input_array(size: usize, seed: u64) -> Vec<usize> {
    shuffled(size, seed)
}

/// Peak heap growth during `benchmark_sort`, in bytes (explicit buffers such as swap).
/// Kept in bytes so the parent can average before rounding; converting to KiB here
/// would truncate sub-KiB buffers to 0 in every run and hide them from the average.
fn run_once(size: usize, seed: usize) -> (u128, usize) {
    let mut array = input_array(size, seed as u64);

    let base_bytes = LIVE_BYTES.load(AtomicOrdering::Relaxed);
    PEAK_BYTES.store(base_bytes, AtomicOrdering::Relaxed);

    let start = Instant::now();

    benchmark_sort(&mut array);

    let elapsed = start.elapsed();
    let peak_bytes = PEAK_BYTES.load(AtomicOrdering::Relaxed);
    let aux_bytes = peak_bytes.saturating_sub(base_bytes);

    let expected: Vec<usize> = (1..=size).collect();
    if array != expected {
        panic!(
            "sort failed with seed {} for size {}",
            seed,
            size
        );
    }

    (micros(elapsed), aux_bytes)
}

fn run_child(args: &[String]) {
    let size = args[2].parse::<usize>().expect("invalid size");
    let seed = args[3].parse::<usize>().expect("invalid seed");
    let (elapsed_us, mem) = run_once(size, seed);
    println!("{} {}", elapsed_us, mem);
}

fn main() {
    let args: Vec<String> = env::args().collect();
    if args.get(1).is_some_and(|arg| arg == "--run-once") {
        run_child(&args);
        return;
    }

    run_correctness_checks();

    println!(
        "| {:>10} | {:>15} | {:>15} | {:>15} | {:>15} |",
        "Size",
        "Average time",
        "Maximum time",
        "Average memory",
        "Maximum memory"
    );

    println!(
        "|{:-<11}:|{:-<16}:|{:-<16}:|{:-<16}:|{:-<16}:|",
        "",
        "",
        "",
        "",
        ""
    );

    for power in MIN_POWER..=MAX_POWER {
        let size = 1usize << power;

        let mut total_time: u128 = 0;
        let mut max_time: u128 = 0;

        let mut total_mem: usize = 0;
        let mut max_mem: usize = 0;

        for seed in 1..=RUNS {
            let output = Command::new(env::current_exe().expect("failed to find current executable"))
                .arg("--run-once")
                .arg(size.to_string())
                .arg(seed.to_string())
                .output()
                .expect("failed to run benchmark child process");

            if !output.status.success() {
                panic!(
                    "benchmark child process failed: {}",
                    String::from_utf8_lossy(&output.stderr)
                );
            }

            let stdout = String::from_utf8(output.stdout)
                .expect("child process returned non-UTF-8 output");
            let mut fields = stdout.split_whitespace();
            let elapsed_us = fields
                .next()
                .expect("missing elapsed time")
                .parse::<u128>()
                .expect("invalid elapsed time");
            let aux_mem = fields
                .next()
                .expect("missing memory usage")
                .parse::<usize>()
                .expect("invalid memory usage");

            total_time += elapsed_us;

            if elapsed_us > max_time {
                max_time = elapsed_us;
            }

            total_mem += aux_mem;

            if aux_mem > max_mem {
                max_mem = aux_mem;
            }
        }

        let avg_time = total_time / RUNS as u128;
        // Memory is summed in bytes and converted to KiB once, after averaging.
        let avg_mem_kb = total_mem / RUNS / 1024;
        let max_mem_kb = max_mem / 1024;

        println!(
            "| {:>10} | {:>15} | {:>15} | {:>15} | {:>15} |",
            size,
            format!("{}.{:06}", avg_time / 1_000_000, avg_time % 1_000_000),
            format!("{}.{:06}", max_time / 1_000_000, max_time % 1_000_000),
            avg_mem_kb,
            max_mem_kb
        );
    }
}
RUST

RUN cargo build --release

CMD ["./target/release/rust-benchmark"]
EOF

docker build -t rust-benchmark "$WORKDIR"
docker run --rm --init rust-benchmark