pytorch/.github/workflows/_mac-test.yml
Catherine Lee d6e5898e8d Quieter logs in CI (#110033)
To reduce the amount of logs
* for successes, only print the part that says what tests ran and don't print the rest.  Zip the log into an artifact.  The line listing al the test names is really long, but if you view source of the raw logs, it will not wrap so it will only be one line.  The log classifier can also be configured to ignored this line. Gets rid of lines like `test_ops.py::TestCommonCPU::test_multiple_devices_round_cpu_int64 SKIPPED [0.0010s] (Only runs on cuda) [  9%]`
* for failures/reruns, print logs.  Do not zip.

Also
* change log artifact name

Examples of various logs:
a074db0f7f failures
1b439e24c4 failures

possibly controversial haha
should i include an option for always printing?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110033
Approved by: https://github.com/huydhn
2023-10-05 16:40:37 +00:00

208 lines
7.5 KiB
YAML

name: mac-test
on:
workflow_call:
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
test-matrix:
required: true
type: string
description: JSON description of what test configs to run.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
python-version:
required: false
type: string
default: "3.8"
description: |
The python version to be used. Will be 3.8 by default
arch:
required: true
type: string
description: |
Contains the architecture to run the tests with
timeout-minutes:
required: false
type: number
default: 270
description: |
Set the maximum (in minutes) how long the workflow should take to finish
jobs:
test:
# Don't run on forked repos or empty test matrix
if: github.repository_owner == 'pytorch' && toJSON(fromJSON(inputs.test-matrix).include) != '[]'
# For setup-miniconda, see https://github.com/conda-incubator/setup-miniconda/issues/179
# Also ensure that we always run with the right architecture
defaults:
run:
shell: arch -arch ${{ inputs.arch }} bash -e -l {0}
strategy:
matrix: ${{ fromJSON(inputs.test-matrix) }}
fail-fast: false
runs-on: ${{ matrix.runner }}
timeout-minutes: ${{ inputs.timeout-minutes }}
env:
GIT_DEFAULT_BRANCH: ${{ github.event.repository.default_branch }}
BUILD_ENVIRONMENT: ${{ inputs.build-environment }}
TEST_CONFIG: ${{ matrix.config }}
SHARD_NUMBER: ${{ matrix.shard }}
NUM_TEST_SHARDS: ${{ matrix.num_shards }}
PR_BODY: ${{ github.event.pull_request.body }}
PYTORCH_RETRY_TEST_CASES: 1
PYTORCH_OVERRIDE_FLAKY_SIGNAL: 1
steps:
- name: Clean up leftover processes on MacOS pet runner
continue-on-error: true
run: |
for PROCESS in "python" "conda" "ninja" "clang"; do
echo "Cleaning up all remaining ${PROCESS} process"
pkill "${PROCESS}" || true
done
- name: Clean up disk space before running MacOS workflow
uses: pytorch/test-infra/.github/actions/check-disk-space@main
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
- name: Download build artifacts
uses: ./.github/actions/download-build-artifacts
with:
name: ${{ inputs.build-environment }}
use-gha: true
- name: Setup miniconda
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
with:
python-version: ${{ inputs.python-version }}
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}
pip-requirements-file: .github/requirements/pip-requirements-${{ runner.os }}.txt
- name: Start monitoring script
id: monitor-script
continue-on-error: true
run: |
${CONDA_RUN} python3 -m tools.stats.monitor > usage_log.txt 2>&1 &
echo "monitor-script-pid=${!}" >> "${GITHUB_OUTPUT}"
- name: Parse ref
id: parse-ref
run: .github/scripts/parse_ref.py
- name: Check for keep-going label and re-enabled test issues
# This uses the filter-test-configs action because it conviniently
# checks for labels and re-enabled test issues. It does not actually do
# any filtering. All filtering is done in the build step.
id: keep-going
uses: ./.github/actions/filter-test-configs
with:
github-token: ${{ secrets.GITHUB_TOKEN }}
test-matrix: ${{ inputs.test-matrix }}
- name: Pre-process arm64 wheels
if: inputs.build-environment == 'macos-12-py3-arm64'
run: |
# As wheels are cross-compiled they are reported as x86_64 ones
ORIG_WHLNAME=$(ls -1 dist/*.whl); ARM_WHLNAME=${ORIG_WHLNAME/x86_64/arm64}; mv "${ORIG_WHLNAME}" "${ARM_WHLNAME}"
- name: Set Test step time
id: test-timeout
shell: bash
env:
JOB_TIMEOUT: ${{ inputs.timeout-minutes }}
run: |
echo "timeout=$((JOB_TIMEOUT-30))" >> "${GITHUB_OUTPUT}"
- name: Test
id: test
timeout-minutes: ${{ fromJson(steps.test-timeout.outputs.timeout) }}
env:
PYTORCH_TEST_CUDA_MEM_LEAK_CHECK: ${{ matrix.mem_leak_check && '1' || '0' }}
PYTORCH_TEST_RERUN_DISABLED_TESTS: ${{ matrix.rerun_disabled_tests && '1' || '0' }}
CONTINUE_THROUGH_ERROR: ${{ steps.keep-going.outputs.keep-going }}
PIP_REQUIREMENTS_FILE: .github/requirements/pip-requirements-${{ runner.os }}.txt
GITHUB_REPOSITORY: ${{ github.repository }}
GITHUB_WORKFLOW: ${{ github.workflow }}
GITHUB_JOB: ${{ github.job }}
GITHUB_RUN_ID: ${{ github.run_id }}
GITHUB_RUN_NUMBER: ${{ github.run_number }}
GITHUB_RUN_ATTEMPT: ${{ github.run_attempt }}
REENABLED_ISSUES: ${{ steps.keep-going.outputs.reenabled-issues }}
run: |
# shellcheck disable=SC1090
set -ex
arch
if [[ -n "$CONDA_ENV" ]]; then
# Use binaries under conda environment
export PATH="$CONDA_ENV/bin":$PATH
fi
# Print out some information about the test environment
which conda
conda --version
${CONDA_RUN} which python3
${CONDA_RUN} python3 --version
${CONDA_RUN} which python
${CONDA_RUN} python --version
${CONDA_RUN} python3 -mpip install --no-index --no-deps dist/*.whl
set +e
pushd "${RUNNER_TEMP}"
# Install pip dependencies if they are not found. This is to mitigate a peculiar
# flaky missing dependencies on MacOS
${CONDA_RUN} python3 -c "import torch"
RC=$?
popd
if [ "${RC}" -ne 0 ]; then
${CONDA_RUN} python3 -mpip install --ignore-installed -r "${PIP_REQUIREMENTS_FILE}"
fi
set -e
${CONDA_RUN} .ci/pytorch/macos-test.sh
- name: Print remaining test logs
shell: bash
if: always() && steps.test.conclusion
run: |
cat test/**/*_toprint.log || true
- name: Get workflow job id
id: get-job-id
uses: ./.github/actions/get-workflow-job-id
if: always()
with:
github-token: ${{ secrets.GITHUB_TOKEN }}
- name: Stop monitoring script
if: always() && ${{ steps.monitor-script.outputs.monitor-script-pid }}
continue-on-error: true
env:
MONITOR_SCRIPT_PID: ${{ steps.monitor-script.outputs.monitor-script-pid }}
run: |
kill "$MONITOR_SCRIPT_PID"
- name: Upload test artifacts
uses: ./.github/actions/upload-test-artifacts
if: always() && steps.test.conclusion && steps.test.conclusion != 'skipped'
with:
use-gha: true
file-suffix: ${{ github.job }}-${{ matrix.config }}-${{ matrix.shard }}-${{ matrix.num_shards }}-${{ matrix.runner }}_${{ steps.get-job-id.outputs.job-id }}
- name: Clean up disk space
if: always()
continue-on-error: true
uses: pytorch/test-infra/.github/actions/check-disk-space@main