Skip to content
Latchkey LogoLatchkey home

Benchmark workflow (swc-project/swc)

The Benchmark workflow from swc-project/swc, explained and optimized by Latchkey.

Source: swc-project/swc.github/workflows/bench.ymlLicense Apache-2.0View source
A
CI health grade
85/100
Health score
2
Jobs per trigger
2
Fixes in the optimized version

In short

  1. CI health: A (excellent)
  2. The optimized version below points runs-on at Latchkey managed runners and adds job timeouts.

What it does

This is the Benchmark workflow from the swc-project/swc repository, a real project running GitHub Actions. It is shown here with attribution under its Apache-2.0 license.

Below, Latchkey shows a faster, safer version produced by its optimization engine.

The workflow

workflow (.yml)
name: Benchmark

on:
  push:
    branches:
      - main
  pull_request:
    types: ["opened", "reopened", "synchronize"]

concurrency:
  # Do not block on main branch
  group: ${{ github.event_name == 'pull_request' && format('{0}-{1}', github.workflow, github.event.pull_request.number) || format('{0}-{1}-{2}', github.workflow, github.ref_name, github.run_id) }}
  cancel-in-progress: ${{ github.event_name == 'pull_request' }}

env:
  CI: 1
  CARGO_INCREMENTAL: 0
  CARGO_TERM_COLOR: "always"
  DIFF: 0
  # For faster CI
  RUST_LOG: "off"
  # https://github.com/swc-project/swc/pull/3742
  RUST_MIN_STACK: 4194304
  CARGO_PROFILE_RELEASE_LTO: false
  CARGO_PROFILE_BENCH_LTO: false

permissions: {}

jobs:
  list-crates:
    if: >-
      ${{ !contains(github.event.head_commit.message, 'chore: ') }}
    name: List crates to benchmark
    runs-on: ubuntu-22.04
    permissions:
      contents: read
    outputs:
      crates: ${{ steps.list-crates.outputs.crates }}
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
        with:
          persist-credentials: false

      - name: List crates
        id: list-crates
        run: echo "crates=$(./scripts/bench/list-crates-with-bench.sh)" >> $GITHUB_OUTPUT

  benchmark-crate:
    name: Benchmark ${{ matrix.crate }}
    runs-on: ubuntu-22.04
    permissions:
      contents: read
    needs: list-crates
    strategy:
      matrix:
        crate: ${{fromJson(needs.list-crates.outputs.crates)}}
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
        with:
          persist-credentials: false

      - name: Install Rust
        uses: dtolnay/rust-toolchain@29eef336d9b2848a0b548edc03f92a220660cdb8 # stable

      - uses: Swatinem/rust-cache@c19371144df3bb44fab255c43d04cbc2ab54d1c4 # v2.9.1
        # Some crates are too slow to build
        if: matrix.crate == 'swc'
        with:
          shared-key: "bench-${{ matrix.crate }}"
          key: "bench-${{ matrix.crate }}"
          cache-all-crates: true
          save-if: ${{ github.ref == 'refs/heads/main' }}

      - name: Install cargo-codspeed
        uses: taiki-e/install-action@3235f8901fd37ffed0052b276cec25a362fb82e9 # v2.77.7
        with:
          tool: cargo-codspeed@3.0.2

      - name: Build the benchmark target(s)
        run: ./scripts/bench/build-crate.sh ${{ matrix.crate }}

      - name: Run the benchmarks
        uses: CodSpeedHQ/action@76578c2a7ddd928664caa737f0e962e3085d4e7c # v3.8.1
        with:
          run: cargo codspeed run
          token: ${{ secrets.CODSPEED_TOKEN }}

The same workflow, on Latchkey

Removes redundant runs and caps runaway jobs. Added and changed lines are highlighted.

workflow (.yml), optimized
name: Benchmark

on:
  push:
    branches:
      - main
  pull_request:
    types: ["opened", "reopened", "synchronize"]

concurrency:
  # Do not block on main branch
  group: ${{ github.event_name == 'pull_request' && format('{0}-{1}', github.workflow, github.event.pull_request.number) || format('{0}-{1}-{2}', github.workflow, github.ref_name, github.run_id) }}
  cancel-in-progress: ${{ github.event_name == 'pull_request' }}

env:
  CI: 1
  CARGO_INCREMENTAL: 0
  CARGO_TERM_COLOR: "always"
  DIFF: 0
  # For faster CI
  RUST_LOG: "off"
  # https://github.com/swc-project/swc/pull/3742
  RUST_MIN_STACK: 4194304
  CARGO_PROFILE_RELEASE_LTO: false
  CARGO_PROFILE_BENCH_LTO: false

permissions: {}

jobs:
  list-crates:
    timeout-minutes: 30
    if: >-
      ${{ !contains(github.event.head_commit.message, 'chore: ') }}
    name: List crates to benchmark
    runs-on: ubuntu-22.04
    runs-on: latchkey-small
    permissions:
      contents: read
    outputs:
      crates: ${{ steps.list-crates.outputs.crates }}
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
        with:
          persist-credentials: false

      - name: List crates
        id: list-crates
        run: echo "crates=$(./scripts/bench/list-crates-with-bench.sh)" >> $GITHUB_OUTPUT

  benchmark-crate:
    timeout-minutes: 30
    name: Benchmark ${{ matrix.crate }}
    runs-on: ubuntu-22.04
    runs-on: latchkey-small
    permissions:
      contents: read
    needs: list-crates
    strategy:
      matrix:
        crate: ${{fromJson(needs.list-crates.outputs.crates)}}
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4.3.1
        with:
          persist-credentials: false

      - name: Install Rust
        uses: dtolnay/rust-toolchain@29eef336d9b2848a0b548edc03f92a220660cdb8 # stable

      - uses: Swatinem/rust-cache@c19371144df3bb44fab255c43d04cbc2ab54d1c4 # v2.9.1
        # Some crates are too slow to build
        if: matrix.crate == 'swc'
        with:
          shared-key: "bench-${{ matrix.crate }}"
          key: "bench-${{ matrix.crate }}"
          cache-all-crates: true
          save-if: ${{ github.ref == 'refs/heads/main' }}

      - name: Install cargo-codspeed
        uses: taiki-e/install-action@3235f8901fd37ffed0052b276cec25a362fb82e9 # v2.77.7
        with:
          tool: cargo-codspeed@3.0.2

      - name: Build the benchmark target(s)
        run: ./scripts/bench/build-crate.sh ${{ matrix.crate }}

      - name: Run the benchmarks
        uses: CodSpeedHQ/action@76578c2a7ddd928664caa737f0e962e3085d4e7c # v3.8.1
        with:
          run: cargo codspeed run
          token: ${{ secrets.CODSPEED_TOKEN }}

What changed

  1. Run on Latchkey managed runners with one line (runs-on), which apply the fixes below automatically and self-heal transient failures. Each Linux x64 job moves to the Latchkey size matching its vCPU count, latchkey-small for a standard 2 vCPU runner; macOS, Windows and arm64 jobs stay as written.
  2. Add a job timeout so a hung step cannot burn hours of runner time.

This workflow runs 2 jobs per trigger. All of them run on a standard GitHub-hosted Linux x64 runner, which in a private repository bills $0.006 a minute against $0.0025 on latchkey-small, 58% lower; in a public repository like this one they are free. Estimate the monthly difference with your own run counts.

Actions used in this workflow

  • actions/checkout
  • dtolnay/rust-toolchain
  • Swatinem/rust-cache
  • taiki-e/install-action
  • CodSpeedHQ/action

Frequently asked questions

What does the Benchmark workflow (swc-project/swc) workflow do?
This is the Benchmark workflow from the swc-project/swc repository, a real project running GitHub Actions. It is shown here with attribution under its Apache-2.0 license.
What CI health grade does this workflow get?
This Automation and other workflow grades A. Paste your own workflow into the Latchkey grader to see its grade and the exact fixes.
How can I improve this Automation and other workflow?
Add job timeouts. Latchkey applies these automatically on managed runners when you point runs-on at Latchkey.

References