v8/tools/unittests/run_tests_test.py
Michael Achenbach 4a2d9b1b20 [test] Abort testing immediately on interrupt and sigterm
This is a partial revert of:
https://crrev.com/c/890938 and https://crrev.com/c/893982

Before this CL, the test runner blocked on ongoing tests in order to
process their results after an internal timeout. However, the logic
required for this feature was overly complicated and prevented an
acceptable implementation for fast aborts. Furthermore, also the fuzzers
suffered from timeouts on swarming due to hanging tests.

Instead, we now abort immediately on internal timeout (used on
fuzzers), SIGINT (Ctrl-C) and SIGTERM. Ongoing tests are immediately
terminated and their results are disregarded. On SIGTERM and SIGINT,
we return with non-zero exit codes, and zero on internal timeout.

This will also properly return json output, when the external hard
timeout is reached on swarming (causes SIGTERM).

TBR=sergiyb@chromium.org

Bug: v8:7423, chromium:813065
Change-Id: Ib20f835f58a0970693bdd3b21dc5d766d8e115d8
Reviewed-on: https://chromium-review.googlesource.com/924852
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#51399}
2018-02-20 18:02:25 +00:00

689 lines
24 KiB
Python
Executable File

#!/usr/bin/env python
# Copyright 2017 the V8 project authors. All rights reserved.
# Use of this source code is governed by a BSD-style license that can be
# found in the LICENSE file.
"""
Global system tests for V8 test runners and fuzzers.
This hooks up the framework under tools/testrunner testing high-level scenarios
with different test suite extensions and build configurations.
"""
# TODO(machenbach): Mock out util.GuessOS to make these tests really platform
# independent.
# TODO(machenbach): Move coverage recording to a global test entry point to
# include other unittest suites in the coverage report.
# TODO(machenbach): Coverage data from multiprocessing doesn't work.
# TODO(majeski): Add some tests for the fuzzers.
import collections
import contextlib
import json
import os
import shutil
import subprocess
import sys
import tempfile
import unittest
from cStringIO import StringIO
TOOLS_ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
TEST_DATA_ROOT = os.path.join(TOOLS_ROOT, 'unittests', 'testdata')
RUN_TESTS_PY = os.path.join(TOOLS_ROOT, 'run-tests.py')
Result = collections.namedtuple(
'Result', ['stdout', 'stderr', 'returncode'])
Result.__str__ = lambda self: (
'\nReturncode: %s\nStdout:\n%s\nStderr:\n%s\n' %
(self.returncode, self.stdout, self.stderr))
@contextlib.contextmanager
def temp_dir():
"""Wrapper making a temporary directory available."""
path = None
try:
path = tempfile.mkdtemp('v8_test_')
yield path
finally:
if path:
shutil.rmtree(path)
@contextlib.contextmanager
def temp_base(baseroot='testroot1'):
"""Wrapper that sets up a temporary V8 test root.
Args:
baseroot: The folder with the test root blueprint. Relevant files will be
copied to the temporary test root, to guarantee a fresh setup with no
dirty state.
"""
basedir = os.path.join(TEST_DATA_ROOT, baseroot)
with temp_dir() as tempbase:
builddir = os.path.join(tempbase, 'out', 'Release')
testroot = os.path.join(tempbase, 'test')
os.makedirs(builddir)
shutil.copy(os.path.join(basedir, 'v8_build_config.json'), builddir)
shutil.copy(os.path.join(basedir, 'd8_mocked.py'), builddir)
for suite in os.listdir(os.path.join(basedir, 'test')):
os.makedirs(os.path.join(testroot, suite))
for entry in os.listdir(os.path.join(basedir, 'test', suite)):
shutil.copy(
os.path.join(basedir, 'test', suite, entry),
os.path.join(testroot, suite))
yield tempbase
@contextlib.contextmanager
def capture():
"""Wrapper that replaces system stdout/stderr an provides the streams."""
oldout = sys.stdout
olderr = sys.stderr
try:
stdout=StringIO()
stderr=StringIO()
sys.stdout = stdout
sys.stderr = stderr
yield stdout, stderr
finally:
sys.stdout = oldout
sys.stderr = olderr
def run_tests(basedir, *args, **kwargs):
"""Executes the test runner with captured output."""
with capture() as (stdout, stderr):
sys_args = ['--command-prefix', sys.executable] + list(args)
if kwargs.get('infra_staging', False):
sys_args.append('--infra-staging')
else:
sys_args.append('--no-infra-staging')
code = standard_runner.StandardTestRunner(
basedir=basedir).execute(sys_args)
return Result(stdout.getvalue(), stderr.getvalue(), code)
def override_build_config(basedir, **kwargs):
"""Override the build config with new values provided as kwargs."""
path = os.path.join(basedir, 'out', 'Release', 'v8_build_config.json')
with open(path) as f:
config = json.load(f)
config.update(kwargs)
with open(path, 'w') as f:
json.dump(config, f)
class SystemTest(unittest.TestCase):
@classmethod
def setUpClass(cls):
# Try to set up python coverage and run without it if not available.
cls._cov = None
try:
import coverage
if int(coverage.__version__.split('.')[0]) < 4:
cls._cov = None
print 'Python coverage version >= 4 required.'
raise ImportError()
cls._cov = coverage.Coverage(
source=([os.path.join(TOOLS_ROOT, 'testrunner')]),
omit=['*unittest*', '*__init__.py'],
)
cls._cov.exclude('raise NotImplementedError')
cls._cov.exclude('if __name__ == .__main__.:')
cls._cov.exclude('except TestRunnerError:')
cls._cov.exclude('except KeyboardInterrupt:')
cls._cov.exclude('if options.verbose:')
cls._cov.exclude('if verbose:')
cls._cov.exclude('pass')
cls._cov.exclude('assert False')
cls._cov.start()
except ImportError:
print 'Running without python coverage.'
sys.path.append(TOOLS_ROOT)
global standard_runner
from testrunner import standard_runner
from testrunner.local import command
from testrunner.local import pool
command.setup_testing()
pool.setup_testing()
@classmethod
def tearDownClass(cls):
if cls._cov:
cls._cov.stop()
print ''
print cls._cov.report(show_missing=True)
def testPass(self):
"""Test running only passing tests in two variants.
Also test printing durations.
"""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default,stress',
'--time',
'sweet/bananas',
'sweet/raspberries',
)
self.assertIn('Running 2 base tests', result.stdout, result)
self.assertIn('Done running sweet/bananas: pass', result.stdout, result)
# TODO(majeski): Implement for test processors
# self.assertIn('Total time:', result.stderr, result)
# self.assertIn('sweet/bananas', result.stderr, result)
self.assertEqual(0, result.returncode, result)
def testShardedProc(self):
with temp_base() as basedir:
for shard in [1, 2]:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default,stress',
'--shard-count=2',
'--shard-run=%d' % shard,
'sweet/bananas',
'sweet/raspberries',
infra_staging=True,
)
# One of the shards gets one variant of each test.
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
if shard == 1:
self.assertIn('Done running sweet/bananas', result.stdout, result)
else:
self.assertIn('Done running sweet/raspberries', result.stdout, result)
self.assertEqual(0, result.returncode, result)
@unittest.skip("incompatible with test processors")
def testSharded(self):
"""Test running a particular shard."""
with temp_base() as basedir:
for shard in [1, 2]:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default,stress',
'--shard-count=2',
'--shard-run=%d' % shard,
'sweet/bananas',
'sweet/raspberries',
)
# One of the shards gets one variant of each test.
self.assertIn('Running 2 tests', result.stdout, result)
self.assertIn('Done running sweet/bananas', result.stdout, result)
self.assertIn('Done running sweet/raspberries', result.stdout, result)
self.assertEqual(0, result.returncode, result)
def testFailProc(self):
self.testFail(infra_staging=True)
def testFail(self, infra_staging=True):
"""Test running only failing tests in two variants."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default,stress',
'sweet/strawberries',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 2 tests', result.stdout, result)
else:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
self.assertIn('Done running sweet/strawberries: FAIL', result.stdout, result)
self.assertEqual(1, result.returncode, result)
def check_cleaned_json_output(self, expected_results_name, actual_json):
# Check relevant properties of the json output.
with open(actual_json) as f:
json_output = json.load(f)[0]
pretty_json = json.dumps(json_output, indent=2, sort_keys=True)
# Replace duration in actual output as it's non-deterministic. Also
# replace the python executable prefix as it has a different absolute
# path dependent on where this runs.
def replace_variable_data(data):
data['duration'] = 1
data['command'] = ' '.join(
['/usr/bin/python'] + data['command'].split()[1:])
for data in json_output['slowest_tests']:
replace_variable_data(data)
for data in json_output['results']:
replace_variable_data(data)
json_output['duration_mean'] = 1
with open(os.path.join(TEST_DATA_ROOT, expected_results_name)) as f:
expected_test_results = json.load(f)
msg = None # Set to pretty_json for bootstrapping.
self.assertDictEqual(json_output, expected_test_results, msg)
def testFailWithRerunAndJSONProc(self):
self.testFailWithRerunAndJSON(infra_staging=True)
def testFailWithRerunAndJSON(self, infra_staging=True):
"""Test re-running a failing test and output to json."""
with temp_base() as basedir:
json_path = os.path.join(basedir, 'out.json')
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'--rerun-failures-count=2',
'--random-seed=123',
'--json-test-results', json_path,
'sweet/strawberries',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 1 tests', result.stdout, result)
else:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('1 tests ran', result.stdout, result)
self.assertIn('Done running sweet/strawberries: FAIL', result.stdout, result)
if not infra_staging:
# We run one test, which fails and gets re-run twice.
self.assertIn('3 tests failed', result.stdout, result)
else:
# With test processors we don't count reruns as separated failures.
# TODO(majeski): fix it?
self.assertIn('1 tests failed', result.stdout, result)
self.assertEqual(0, result.returncode, result)
# TODO(majeski): Previously we only reported the variant flags in the
# flags field of the test result.
# After recent changes we report all flags, including the file names.
# This is redundant to the command. Needs investigation.
self.maxDiff = None
self.check_cleaned_json_output('expected_test_results1.json', json_path)
def testFlakeWithRerunAndJSONProc(self):
self.testFlakeWithRerunAndJSON(infra_staging=True)
def testFlakeWithRerunAndJSON(self, infra_staging=True):
"""Test re-running a failing test and output to json."""
with temp_base(baseroot='testroot2') as basedir:
json_path = os.path.join(basedir, 'out.json')
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'--rerun-failures-count=2',
'--random-seed=123',
'--json-test-results', json_path,
'sweet',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 1 tests', result.stdout, result)
self.assertIn(
'Done running sweet/bananaflakes: FAIL', result.stdout, result)
self.assertIn('1 tests failed', result.stdout, result)
else:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn(
'Done running sweet/bananaflakes: pass', result.stdout, result)
self.assertIn('All tests succeeded', result.stdout, result)
self.assertEqual(0, result.returncode, result)
self.maxDiff = None
self.check_cleaned_json_output('expected_test_results2.json', json_path)
def testAutoDetect(self):
"""Fake a build with several auto-detected options.
Using all those options at once doesn't really make much sense. This is
merely for getting coverage.
"""
with temp_base() as basedir:
override_build_config(
basedir, dcheck_always_on=True, is_asan=True, is_cfi=True,
is_msan=True, is_tsan=True, is_ubsan_vptr=True, target_cpu='x86',
v8_enable_i18n_support=False, v8_target_cpu='x86',
v8_use_snapshot=False)
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'sweet/bananas',
)
expect_text = (
'>>> Autodetected:\n'
'asan\n'
'cfi_vptr\n'
'dcheck_always_on\n'
'msan\n'
'no_i18n\n'
'no_snap\n'
'tsan\n'
'ubsan_vptr\n'
'>>> Running tests for ia32.release')
self.assertIn(expect_text, result.stdout, result)
self.assertEqual(0, result.returncode, result)
# TODO(machenbach): Test some more implications of the auto-detected
# options, e.g. that the right env variables are set.
def testSkipsProc(self):
self.testSkips(infra_staging=True)
def testSkips(self, infra_staging=True):
"""Test skipping tests in status file for a specific variant."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=nooptimization',
'sweet/strawberries',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 0 tests', result.stdout, result)
else:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('0 tests ran', result.stdout, result)
self.assertEqual(2, result.returncode, result)
def testDefaultProc(self):
self.testDefault(infra_staging=True)
def testDefault(self, infra_staging=True):
"""Test using default test suites, though no tests are run since they don't
exist in a test setting.
"""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Warning: no tests were run!', result.stdout, result)
else:
self.assertIn('Running 0 base tests', result.stdout, result)
self.assertIn('0 tests ran', result.stdout, result)
self.assertEqual(2, result.returncode, result)
def testNoBuildConfig(self):
"""Test failing run when build config is not found."""
with temp_base() as basedir:
result = run_tests(basedir)
self.assertIn('Failed to load build config', result.stdout, result)
self.assertEqual(5, result.returncode, result)
def testGNOption(self):
"""Test using gn option, but no gn build folder is found."""
with temp_base() as basedir:
# TODO(machenbach): This should fail gracefully.
with self.assertRaises(OSError):
run_tests(basedir, '--gn')
def testInconsistentMode(self):
"""Test failing run when attempting to wrongly override the mode."""
with temp_base() as basedir:
override_build_config(basedir, is_debug=True)
result = run_tests(basedir, '--mode=Release')
self.assertIn('execution mode (release) for release is inconsistent '
'with build config (debug)', result.stdout, result)
self.assertEqual(5, result.returncode, result)
def testInconsistentArch(self):
"""Test failing run when attempting to wrongly override the arch."""
with temp_base() as basedir:
result = run_tests(basedir, '--mode=Release', '--arch=ia32')
self.assertIn(
'--arch value (ia32) inconsistent with build config (x64).',
result.stdout, result)
self.assertEqual(5, result.returncode, result)
def testWrongVariant(self):
"""Test using a bogus variant."""
with temp_base() as basedir:
result = run_tests(basedir, '--mode=Release', '--variants=meh')
self.assertEqual(5, result.returncode, result)
def testModeFromBuildConfig(self):
"""Test auto-detection of mode from build config."""
with temp_base() as basedir:
result = run_tests(basedir, '--outdir=out/Release', 'sweet/bananas')
self.assertIn('Running tests for x64.release', result.stdout, result)
self.assertEqual(0, result.returncode, result)
@unittest.skip("not available with test processors")
def testReport(self):
"""Test the report feature.
This also exercises various paths in statusfile logic.
"""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--variants=default',
'sweet',
'--report',
)
self.assertIn(
'3 tests are expected to fail that we should fix',
result.stdout, result)
self.assertEqual(1, result.returncode, result)
@unittest.skip("not available with test processors")
def testWarnUnusedRules(self):
"""Test the unused-rules feature."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--variants=default,nooptimization',
'sweet',
'--warn-unused',
)
self.assertIn( 'Unused rule: carrots', result.stdout, result)
self.assertIn( 'Unused rule: regress/', result.stdout, result)
self.assertEqual(1, result.returncode, result)
@unittest.skip("not available with test processors")
def testCatNoSources(self):
"""Test printing sources, but the suite's tests have none available."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--variants=default',
'sweet/bananas',
'--cat',
)
self.assertIn('begin source: sweet/bananas', result.stdout, result)
self.assertIn('(no source available)', result.stdout, result)
self.assertEqual(0, result.returncode, result)
def testPredictableProc(self):
self.testPredictable(infra_staging=True)
def testPredictable(self, infra_staging=True):
"""Test running a test in verify-predictable mode.
The test will fail because of missing allocation output. We verify that and
that the predictable flags are passed and printed after failure.
"""
with temp_base() as basedir:
override_build_config(basedir, v8_enable_verify_predictable=True)
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'sweet/bananas',
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 1 tests', result.stdout, result)
else:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('1 tests ran', result.stdout, result)
self.assertIn('Done running sweet/bananas: FAIL', result.stdout, result)
self.assertIn('Test had no allocation output', result.stdout, result)
self.assertIn('--predictable --verify_predictable', result.stdout, result)
self.assertEqual(1, result.returncode, result)
def testSlowArch(self):
"""Test timeout factor manipulation on slow architecture."""
with temp_base() as basedir:
override_build_config(basedir, v8_target_cpu='arm64')
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'sweet/bananas',
)
# TODO(machenbach): We don't have a way for testing if the correct
# timeout was used.
self.assertEqual(0, result.returncode, result)
def testRandomSeedStressWithDefaultProc(self):
self.testRandomSeedStressWithDefault(infra_staging=True)
def testRandomSeedStressWithDefault(self, infra_staging=True):
"""Test using random-seed-stress feature has the right number of tests."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'--random-seed-stress-count=2',
'sweet/bananas',
infra_staging=infra_staging,
)
if infra_staging:
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
else:
self.assertIn('Running 2 tests', result.stdout, result)
self.assertEqual(0, result.returncode, result)
def testRandomSeedStressWithSeed(self):
"""Test using random-seed-stress feature passing a random seed."""
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default',
'--random-seed-stress-count=2',
'--random-seed=123',
'sweet/strawberries',
)
self.assertIn('Running 1 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
# We use a failing test so that the command is printed and we can verify
# that the right random seed was passed.
self.assertIn('--random-seed=123', result.stdout, result)
self.assertEqual(1, result.returncode, result)
def testSpecificVariants(self):
"""Test using NO_VARIANTS modifiers in status files skips the desire tests.
The test runner cmd line configures 4 tests to run (2 tests * 2 variants).
But the status file applies a modifier to each skipping one of the
variants.
"""
with temp_base() as basedir:
override_build_config(basedir, v8_use_snapshot=False)
result = run_tests(
basedir,
'--mode=Release',
'--progress=verbose',
'--variants=default,stress',
'sweet/bananas',
'sweet/raspberries',
)
# Both tests are either marked as running in only default or only
# slow variant.
self.assertIn('Running 2 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
self.assertEqual(0, result.returncode, result)
def testStatusFilePresubmit(self):
"""Test that the fake status file is well-formed."""
with temp_base() as basedir:
from testrunner.local import statusfile
self.assertTrue(statusfile.PresubmitCheck(
os.path.join(basedir, 'test', 'sweet', 'sweet.status')))
def testDotsProgressProc(self):
self.testDotsProgress(infra_staging=True)
def testDotsProgress(self, infra_staging=True):
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=dots',
'sweet/cherries',
'sweet/bananas',
'--no-sorting', '-j1', # make results order deterministic
infra_staging=infra_staging,
)
if not infra_staging:
self.assertIn('Running 2 tests', result.stdout, result)
else:
self.assertIn('Running 2 base tests', result.stdout, result)
self.assertIn('2 tests ran', result.stdout, result)
self.assertIn('F.', result.stdout, result)
self.assertEqual(1, result.returncode, result)
def testMonoProgressProc(self):
self._testCompactProgress('mono', True)
def testMonoProgress(self):
self._testCompactProgress('mono', False)
def testColorProgressProc(self):
self._testCompactProgress('color', True)
def testColorProgress(self):
self._testCompactProgress('color', False)
def _testCompactProgress(self, name, infra_staging):
with temp_base() as basedir:
result = run_tests(
basedir,
'--mode=Release',
'--progress=%s' % name,
'sweet/cherries',
'sweet/bananas',
infra_staging=infra_staging,
)
if name == 'color':
expected = ('\033[34m% 100\033[0m|'
'\033[32m+ 1\033[0m|'
'\033[31m- 1\033[0m]: Done')
else:
expected = '% 100|+ 1|- 1]: Done'
self.assertIn(expected, result.stdout)
self.assertIn('sweet/cherries', result.stdout)
self.assertIn('sweet/bananas', result.stdout)
self.assertEqual(1, result.returncode, result)
if __name__ == '__main__':
unittest.main()