Merge pull request #8680 from sylvestre/wc-perf

sylvestre · web-flow · commit cb594452d9b4 · 2025-09-20T19:00:15.000+02:00
evaluate codspeed for perfs
diff --git a/.github/workflows/codspeed.yml b/.github/workflows/codspeed.yml
@@ -0,0 +1,53 @@
+name: CodSpeed Benchmarks
+
+# spell-checker: disable
+
+on:
+  push:
+    branches:
+      - "main"
+  pull_request:
+
+permissions:
+  contents: read
+
+jobs:
+  benchmarks:
+    name: Run benchmarks
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v5
+        with:
+          persist-credentials: false
+
+      - name: Install system dependencies
+        shell: bash
+        run: |
+          sudo apt-get -y update
+          sudo apt-get -y install libselinux1-dev
+
+      - uses: dtolnay/rust-toolchain@stable
+
+      - uses: Swatinem/rust-cache@v2
+
+      - name: Run sccache-cache
+        uses: mozilla-actions/sccache-action@v0.0.9
+
+      - name: Install cargo-codspeed
+        shell: bash
+        run: cargo install cargo-codspeed --locked
+
+      - name: Run benchmarks
+        uses: CodSpeedHQ/action@v4
+        with:
+          mode: instrumentation
+          run: |
+            # Find all utilities with benchmarks and run them
+            find src/uu/*/benches/ -name "*.rs" 2>/dev/null | while read bench_file; do
+              crate_dir=$(dirname $(dirname "$bench_file"))
+              echo "Building benchmarks in $crate_dir"
+              (cd "$crate_dir" && cargo codspeed build)
+              echo "Running benchmarks in $crate_dir"
+              (cd "$crate_dir" && cargo codspeed run)
+            done
+          token: ${{ secrets.CODSPEED_TOKEN }}
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/Cargo.toml b/Cargo.toml
@@ -316,6 +316,7 @@ compare = "0.1.0"
 crossterm = "0.29.0"
 ctor = "0.5.0"
 ctrlc = { version = "3.4.7", features = ["termination"] }
+divan = "0.1"
 dns-lookup = { version = "3.0.0" }
 exacl = "0.12.0"
 file_diff = "1.0.0"
diff --git a/src/uu/wc/Cargo.toml b/src/uu/wc/Cargo.toml
@@ -29,6 +29,14 @@ fluent = { workspace = true }
 nix = { workspace = true }
 libc = { workspace = true }
 
+[dev-dependencies]
+divan = { workspace = true }
+tempfile = { workspace = true }
+
 [[bin]]
 name = "wc"
 path = "src/main.rs"
+
+[[bench]]
+name = "wc_bench"
+harness = false
diff --git a/src/uu/wc/benches/wc_bench.rs b/src/uu/wc/benches/wc_bench.rs
@@ -0,0 +1,246 @@
+// This file is part of the uutils coreutils package.
+//
+// For the full copyright and license information, please view the LICENSE
+// file that was distributed with this source code.
+
+use divan::{Bencher, black_box};
+use std::fs::File;
+use std::io::{BufWriter, Write};
+use tempfile::TempDir;
+
+/// Generate test data with different characteristics
+fn generate_test_data(size_mb: usize, avg_line_length: usize) -> Vec<u8> {
+    let total_size = size_mb * 1024 * 1024;
+    let mut data = Vec::with_capacity(total_size);
+
+    let mut current_size = 0;
+    let mut line_chars = 0;
+
+    while current_size < total_size {
+        if line_chars >= avg_line_length {
+            data.push(b'\n');
+            line_chars = 0;
+        } else {
+            // Use various ASCII characters to make it realistic
+            data.push(b'a' + (current_size % 26) as u8);
+            line_chars += 1;
+        }
+        current_size += 1;
+    }
+
+    // Ensure we end with a newline
+    if data.last() != Some(&b'\n') {
+        data.push(b'\n');
+    }
+
+    data
+}
+
+/// Generate test data by line count instead of size
+fn generate_test_data_by_lines(num_lines: usize, avg_line_length: usize) -> Vec<u8> {
+    let mut data = Vec::new();
+
+    for line_num in 0..num_lines {
+        // Vary line length slightly for realism
+        let line_length = avg_line_length + (line_num % 40).saturating_sub(20);
+
+        for char_pos in 0..line_length {
+            // Create more realistic text with spaces
+            if char_pos > 0 && char_pos % 8 == 0 {
+                data.push(b' '); // Add spaces every 8 characters
+            } else {
+                // Cycle through letters with some variation
+                let char_offset = (line_num + char_pos) % 26;
+                data.push(b'a' + char_offset as u8);
+            }
+        }
+        data.push(b'\n');
+    }
+
+    data
+}
+
+/// Create a temporary file with test data
+fn create_test_file(data: &[u8], temp_dir: &TempDir) -> std::path::PathBuf {
+    let file_path = temp_dir.path().join("test_data.txt");
+    let file = File::create(&file_path).unwrap();
+    let mut writer = BufWriter::new(file);
+    writer.write_all(data).unwrap();
+    writer.flush().unwrap();
+    file_path
+}
+
+/// Run uutils wc with given arguments
+fn run_uutils_wc(args: &[&str]) -> i32 {
+    use std::process::{Command, Stdio};
+
+    // Use the binary instead of calling uumain directly to avoid stdout issues
+    let output = Command::new("../../../target/release/coreutils")
+        .args(["wc"].iter().chain(args.iter()))
+        .stdout(Stdio::null())
+        .stderr(Stdio::null())
+        .status()
+        .expect("Failed to execute wc command");
+
+    i32::from(!output.success())
+}
+
+/// Benchmark different file sizes for line counting
+#[divan::bench(args = [1, 5, 10, 25, 50])]
+fn wc_lines_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-l", file_path_str]));
+    });
+}
+
+/// Benchmark different file sizes for character counting
+#[divan::bench(args = [1, 5, 10, 25])]
+fn wc_chars_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-m", file_path_str]));
+    });
+}
+
+/// Benchmark different file sizes for byte counting
+#[divan::bench(args = [1, 5, 10, 50, 100])]
+fn wc_bytes_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-c", file_path_str]));
+    });
+}
+
+/// Benchmark word counting (should use traditional read path)
+#[divan::bench(args = [1, 5, 10, 25])]
+fn wc_words_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-w", file_path_str]));
+    });
+}
+
+/// Benchmark combined byte+line counting
+#[divan::bench(args = [1, 5, 10, 50])]
+fn wc_bytes_lines_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-cl", file_path_str]));
+    });
+}
+
+/// Benchmark default wc behavior (bytes, lines, words)
+#[divan::bench(args = [1, 5, 10])]
+fn wc_default_synthetic(bencher: Bencher, size_mb: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&[file_path_str]));
+    });
+}
+
+/// Test different line lengths impact on performance
+#[divan::bench(args = [(5, 50), (5, 100), (5, 200), (5, 500)])]
+fn wc_lines_variable_length(bencher: Bencher, (size_mb, avg_line_len): (usize, usize)) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data(size_mb, avg_line_len);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-l", file_path_str]));
+    });
+}
+
+/// Benchmark large files by line count - up to 500K lines!
+#[divan::bench(args = [10_000, 50_000, 100_000, 500_000])]
+fn wc_lines_large_line_count(bencher: Bencher, num_lines: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data_by_lines(num_lines, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-l", file_path_str]));
+    });
+}
+
+/// Benchmark character counting on large line counts
+#[divan::bench(args = [10_000, 50_000, 100_000])]
+fn wc_chars_large_line_count(bencher: Bencher, num_lines: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data_by_lines(num_lines, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-m", file_path_str]));
+    });
+}
+
+/// Benchmark word counting on large line counts
+#[divan::bench(args = [10_000, 50_000, 100_000])]
+fn wc_words_large_line_count(bencher: Bencher, num_lines: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data_by_lines(num_lines, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-w", file_path_str]));
+    });
+}
+
+/// Benchmark default wc (lines, words, bytes) on large line counts
+#[divan::bench(args = [10_000, 50_000, 100_000])]
+fn wc_default_large_line_count(bencher: Bencher, num_lines: usize) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data_by_lines(num_lines, 80);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&[file_path_str]));
+    });
+}
+
+/// Benchmark very short vs very long lines with 100K lines
+#[divan::bench(args = [(100_000, 10), (100_000, 200), (100_000, 1000)])]
+fn wc_lines_extreme_line_lengths(bencher: Bencher, (num_lines, line_len): (usize, usize)) {
+    let temp_dir = tempfile::tempdir().unwrap();
+    let data = generate_test_data_by_lines(num_lines, line_len);
+    let file_path = create_test_file(&data, &temp_dir);
+    let file_path_str = file_path.to_str().unwrap();
+
+    bencher.bench(|| {
+        black_box(run_uutils_wc(&["-l", file_path_str]));
+    });
+}
+
+fn main() {
+    divan::main();
+}
