AArch64: Hoist zeta values for [inv]NTT in C file and auto-generate #831
Workflow file for this run
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # SPDX-License-Identifier: Apache-2.0 | |
| name: Bench | |
| on: | |
| workflow_dispatch: | |
| push: | |
| branches: ["main"] | |
| pull_request: | |
| branches: ["main"] | |
| types: [ "labeled" ] | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }} | |
| cancel-in-progress: true | |
| jobs: | |
| bench: | |
| permissions: | |
| contents: write | |
| pull-requests: write | |
| name: ${{ matrix.target.name }} | |
| strategy: | |
| fail-fast: true | |
| matrix: | |
| target: | |
| - system: rpi4 | |
| name: Arm Cortex-A72 (Raspberry Pi 4) benchmarks | |
| bench_pmu: PMU | |
| archflags: -mcpu=cortex-a72 -DSYS_AARCH64_SLOW_BARREL_SHIFTER | |
| cflags: "-flto -DFORCE_AARCH64" | |
| bench_extra_args: "" | |
| - system: rpi5 | |
| name: Arm Cortex-A76 (Raspberry Pi 5) benchmarks | |
| bench_pmu: PERF | |
| archflags: "-mcpu=cortex-a76 -march=armv8.2-a" | |
| cflags: "-flto -DFORCE_AARCH64" | |
| bench_extra_args: "" | |
| - system: a55 | |
| name: Arm Cortex-A55 (Snapdragon 888) benchmarks | |
| bench_pmu: PERF | |
| archflags: "-mcpu=cortex-a55 -march=armv8.2-a" | |
| cflags: "-flto -static -DFORCE_AARCH64 -DFIPS202_NATIVE_PROFILE=\"aarch64/profiles/cortex_a55.h\"" | |
| bench_extra_args: -w exec-on-a55 | |
| if: github.repository_owner == 'pq-code-package' && (github.event.label.name == 'benchmark' || github.ref == 'refs/heads/main') | |
| runs-on: self-hosted-${{ matrix.target.system }} | |
| steps: | |
| - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 | |
| - uses: ./.github/actions/bench | |
| with: | |
| name: ${{ matrix.target.name }} | |
| cflags: ${{ matrix.target.cflags }} | |
| archflags: ${{ matrix.target.archflags }} | |
| perf: ${{ matrix.target.bench_pmu }} | |
| store_results: ${{ github.repository_owner == 'pq-code-package' && github.ref == 'refs/heads/main' }} | |
| bench_extra_args: ${{ matrix.target.bench_extra_args }} | |
| gh_token: ${{ secrets.AWS_GITHUB_TOKEN }} | |
| ec2_all: | |
| name: ${{ matrix.target.name }} ${{ matrix.opt.name }} | |
| permissions: | |
| contents: write | |
| pull-requests: write | |
| id-token: write | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| target: | |
| - name: Graviton2 | |
| ec2_instance_type: t4g.small | |
| ec2_ami: ubuntu-latest (aarch64) | |
| archflags: -mcpu=cortex-a76 -march=armv8.2-a | |
| cflags: "-flto -DFORCE_AARCH64" | |
| perf: PERF | |
| - name: Graviton3 | |
| ec2_instance_type: c7g.medium | |
| ec2_ami: ubuntu-latest (aarch64) | |
| archflags: -march=armv8.4-a+sha3 | |
| cflags: "-flto -DFORCE_AARCH64" | |
| perf: PERF | |
| - name: Graviton4 | |
| ec2_instance_type: c8g.medium | |
| ec2_ami: ubuntu-latest (aarch64) | |
| archflags: -march=armv9-a+sha3 | |
| cflags: "-flto -DFORCE_AARCH64" | |
| perf: PERF | |
| - name: AMD EPYC 4th gen (c7a) | |
| ec2_instance_type: c7a.medium | |
| ec2_ami: ubuntu-latest (x86_64) | |
| archflags: -mavx2 -mbmi2 -mpopcnt -maes | |
| cflags: "-flto -DFORCE_X86_64" | |
| perf: PMU | |
| - name: Intel Xeon 4th gen (c7i) | |
| ec2_instance_type: c7i.metal-24xl | |
| ec2_ami: ubuntu-latest (x86_64) | |
| archflags: -mavx2 -mbmi2 -mpopcnt -maes | |
| cflags: "-flto -DFORCE_X86_64" | |
| perf: PMU | |
| - name: AMD EPYC 3rd gen (c6a) | |
| ec2_instance_type: c6a.large | |
| ec2_ami: ubuntu-latest (x86_64) | |
| archflags: -mavx2 -mbmi2 -mpopcnt -maes | |
| cflags: "-flto -DFORCE_X86_64" | |
| perf: PMU | |
| - name: Intel Xeon 3rd gen (c6i) | |
| ec2_instance_type: c6i.large | |
| ec2_ami: ubuntu-latest (x86_64) | |
| archflags: -mavx2 -mbmi2 -mpopcnt -maes | |
| cflags: "-flto -DFORCE_X86_64" | |
| perf: PMU | |
| uses: ./.github/workflows/bench_ec2_reusable.yml | |
| if: github.repository_owner == 'pq-code-package' && (github.event.label.name == 'benchmark' || github.ref == 'refs/heads/main') | |
| with: | |
| ec2_instance_type: ${{ matrix.target.ec2_instance_type }} | |
| ec2_ami: ${{ matrix.target.ec2_ami }} | |
| archflags: ${{ matrix.target.archflags }} | |
| cflags: ${{ matrix.target.cflags }} | |
| opt: "all" | |
| store_results: ${{ github.repository_owner == 'pq-code-package' && github.ref == 'refs/heads/main' }} # Only store optimized results | |
| name: ${{ matrix.target.name }} | |
| perf: ${{ matrix.target.perf }} | |
| secrets: inherit |