Use bit vector allocator for free registers #932

Workflow file for this run

.github/workflows/benchmark.yml at 198f985

	name: Core Benchmarks

	on:
	push:
	branches:
	- master
	pull_request:
	types: [ synchronize, labeled ]
	branches:
	- master
	workflow_dispatch:

	jobs:
	synthesis:
	strategy:
	matrix:
	config: [basic, full]
	name: Run synthesis benchmarks
	if: >
	github.ref == 'refs/heads/master' \|\|
	(
	github.event_name == 'pull_request' &&
	(
	(github.event.action == 'synchronize' && contains(github.event.pull_request.labels.*.name, 'benchmark')) \|\|
	(github.event.action == 'labeled' && github.event.label.name == 'benchmark')
	)
	)
	runs-on: ubuntu-latest
	timeout-minutes: 40
	container: ghcr.io/kuznia-rdzeni/amaranth-synth:ecp5-2023.11.19_v
	steps:
	- uses: actions/checkout@v4

	- name: Set ownership (Github Actions workaround)
	run: \|
	# https://github.com/actions/runner/issues/2033
	chown -R $(id -u):$(id -g) $PWD

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: '3.11'

	- name: Install dependencies
	run: \|
	python3 -m venv venv
	. venv/bin/activate
	python3 -m pip install --upgrade pip
	python3 -m pip install -r requirements-dev.txt

	- name: Synthesize
	run: \|
	. venv/bin/activate
	PYTHONHASHSEED=0 ./scripts/synthesize.py --verbose --strip-debug --config ${{ matrix.config }}

	- name: Print synthesis information
	run: cat ./build/top.tim

	- name: Collect Benchmark information
	run: \|
	. venv/bin/activate
	./scripts/parse_benchmark_info.py -o synth-benchmark-${{ matrix.config }}.json
	cat ./synth-benchmark-${{ matrix.config }}.json

	- uses: actions/upload-artifact@v4
	with:
	name: synth_benchmark_results-${{ matrix.config }}
	path: synth-benchmark-${{ matrix.config }}.json


	build-perf-benchmarks:
	name: Build performance benchmarks
	runs-on: ubuntu-latest
	if: >
	github.ref == 'refs/heads/master' \|\|
	(
	github.event_name == 'pull_request' &&
	(
	(github.event.action == 'synchronize' && contains(github.event.pull_request.labels.*.name, 'benchmark')) \|\|
	(github.event.action == 'labeled' && github.event.label.name == 'benchmark')
	)
	)
	container: ghcr.io/kuznia-rdzeni/riscv-toolchain:2024.03.12
	steps:
	- name: Checkout
	uses: actions/checkout@v4
	with:
	submodules: recursive

	- name: Build embench
	run: cd test/external/embench && make

	- uses: actions/upload-artifact@v4
	with:
	name: "embench"
	path: \|
	test/external/embench/build

	run-perf-benchmarks:
	name: Run performance benchmarks
	runs-on: ubuntu-latest
	timeout-minutes: 30
	container: ghcr.io/kuznia-rdzeni/verilator:v5.008-2023.11.19_v
	needs: build-perf-benchmarks
	steps:
	- name: Checkout
	uses: actions/checkout@v4

	- name: Set ownership (Github Actions workaround)
	run: \|
	# https://github.com/actions/runner/issues/2033
	chown -R $(id -u):$(id -g) $PWD

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: '3.11'

	- name: Install dependencies
	run: \|
	python3 -m venv venv
	. venv/bin/activate
	python3 -m pip install --upgrade pip
	python3 -m pip install -r requirements-dev.txt

	- name: Generate Verilog
	run: \|
	. venv/bin/activate
	PYTHONHASHSEED=0 TRANSACTRON_VERBOSE=1 ./scripts/gen_verilog.py --verbose --config full

	- uses: actions/download-artifact@v4
	with:
	name: "embench"
	path: test/external/embench/build

	- name: Run benchmarks
	run: \|
	. venv/bin/activate
	scripts/run_benchmarks.py -o perf_benchmark.json --summary perf.md
	cat perf.md >> $GITHUB_STEP_SUMMARY

	- uses: actions/upload-artifact@v4
	with:
	name: perf_benchmark_results
	path: perf_benchmark.json

	pr-summary:
	name: Create benchmark summary for a pull request
	if: ${{ github.event_name == 'pull_request' }}
	runs-on: ubuntu-latest
	timeout-minutes: 5
	needs: [run-perf-benchmarks, synthesis]
	steps:
	- name: Checkout
	uses: actions/checkout@v4

	- name: Set ownership (Github Actions workaround)
	run: \|
	# https://github.com/actions/runner/issues/2033
	chown -R $(id -u):$(id -g) $PWD

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: '3.11'

	- name: Install dependencies
	run: \|
	python3 -m venv venv
	. venv/bin/activate
	python3 -m pip install --upgrade pip
	python3 -m pip install -r requirements-dev.txt

	- name: Get baseline benchmarks results
	uses: actions/cache@v4
	env:
	cache-name: benchmark-results
	with:
	key: ${{ env.cache-name }}-${{ runner.os }}-${{ github.event.pull_request.base.sha }}
	path: \|
	perf_benchmark.json
	synth-benchmark-basic.json
	synth-benchmark-full.json

	- uses: actions/download-artifact@v4
	with:
	name: perf_benchmark_results
	path: artifacts
	- uses: actions/download-artifact@v4
	with:
	name: synth_benchmark_results-basic
	path: artifacts
	- uses: actions/download-artifact@v4
	with:
	name: synth_benchmark_results-full
	path: artifacts

	- name: Save PR number
	run: \|
	mkdir -p ./summary
	echo ${{ github.event.number }} > ./summary/pr_number

	- name: Create comment
	id: createComment
	run: \|
	. venv/bin/activate
	{
	echo "## Benchmarks summary"
	echo "### Performance benchmarks"
	ci/print_benchmark_summary.py --precision 3 artifacts/perf_benchmark.json perf_benchmark.json
	echo
	echo "You can view all the metrics [here](${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }})."
	echo "### Synthesis benchmarks (basic)"
	ci/print_benchmark_summary.py --precision 0 artifacts/synth-benchmark-basic.json synth-benchmark-basic.json
	echo
	echo "### Synthesis benchmarks (full)"
	ci/print_benchmark_summary.py --precision 0 artifacts/synth-benchmark-full.json synth-benchmark-full.json
	echo
	} >> summary/comment.md

	- uses: actions/upload-artifact@v4
	with:
	name: benchmark-summary
	path: summary

	publish-benchmarks:
	name: Publish benchmarks
	if: github.ref == 'refs/heads/master'
	runs-on: ubuntu-latest
	timeout-minutes: 5
	needs: [run-perf-benchmarks, synthesis]
	steps:
	- uses: actions/checkout@v4

	- uses: actions/download-artifact@v4
	with:
	name: perf_benchmark_results
	path: .
	- uses: actions/download-artifact@v4
	with:
	name: synth_benchmark_results-basic
	path: .
	- uses: actions/download-artifact@v4
	with:
	name: synth_benchmark_results-full
	path: .

	- name: Put results in a cache
	uses: actions/cache@v4
	env:
	cache-name: benchmark-results
	with:
	key: ${{ env.cache-name }}-${{ runner.os }}-${{ github.sha }}
	path: \|
	perf_benchmark.json
	synth-benchmark-basic.json
	synth-benchmark-full.json

	- name: Store benchmark result (IPC)
	uses: benchmark-action/github-action-benchmark@v1
	with:
	name: Performance (IPC)
	tool: 'customBiggerIsBetter'
	output-file-path: './perf_benchmark.json'
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: true
	benchmark-data-dir-path: "dev/benchmark"

	- name: Store synthesis benchmark result (basic)
	uses: benchmark-action/github-action-benchmark@v1
	with:
	name: Fmax and LCs (basic)
	tool: 'customBiggerIsBetter'
	output-file-path: './synth-benchmark-basic.json'
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: true
	benchmark-data-dir-path: "dev/benchmark"

	- name: Store synthesis benchmark result (full)
	uses: benchmark-action/github-action-benchmark@v1
	with:
	name: Fmax and LCs (full)
	tool: 'customBiggerIsBetter'
	output-file-path: './synth-benchmark-full.json'
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: true
	benchmark-data-dir-path: "dev/benchmark"

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Use bit vector allocator for free registers #932

Workflow file

Use bit vector allocator for free registers #932

Jobs

Run details

Workflow file for this run