|
|
|
#!/usr/bin/env python3
|
|
|
|
# Copyright 2015 gRPC authors.
|
|
|
|
#
|
|
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
# you may not use this file except in compliance with the License.
|
|
|
|
# You may obtain a copy of the License at
|
|
|
|
#
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
#
|
|
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
# See the License for the specific language governing permissions and
|
|
|
|
# limitations under the License.
|
|
|
|
"""Run tests in parallel."""
|
|
|
|
|
|
|
|
from __future__ import print_function
|
|
|
|
|
|
|
|
import argparse
|
|
|
|
import ast
|
|
|
|
import collections
|
|
|
|
import glob
|
|
|
|
import itertools
|
|
|
|
import json
|
|
|
|
import logging
|
|
|
|
import multiprocessing
|
|
|
|
import os
|
|
|
|
import os.path
|
|
|
|
import platform
|
|
|
|
import random
|
|
|
|
import re
|
|
|
|
import shlex
|
|
|
|
import socket
|
|
|
|
import subprocess
|
|
|
|
import sys
|
|
|
|
import tempfile
|
|
|
|
import time
|
|
|
|
import traceback
|
|
|
|
import uuid
|
|
|
|
|
|
|
|
import six
|
|
|
|
from six.moves import urllib
|
|
|
|
|
|
|
|
import python_utils.jobset as jobset
|
|
|
|
import python_utils.report_utils as report_utils
|
|
|
|
import python_utils.start_port_server as start_port_server
|
|
|
|
import python_utils.watch_dirs as watch_dirs
|
|
|
|
|
|
|
|
try:
|
|
|
|
from python_utils.upload_test_results import upload_results_to_bq
|
|
|
|
except ImportError:
|
|
|
|
pass # It's ok to not import because this is only necessary to upload results to BQ.
|
|
|
|
|
|
|
|
gcp_utils_dir = os.path.abspath(
|
|
|
|
os.path.join(os.path.dirname(__file__), "../gcp/utils")
|
|
|
|
)
|
|
|
|
sys.path.append(gcp_utils_dir)
|
|
|
|
|
|
|
|
_ROOT = os.path.abspath(os.path.join(os.path.dirname(sys.argv[0]), "../.."))
|
|
|
|
os.chdir(_ROOT)
|
|
|
|
|
|
|
|
_FORCE_ENVIRON_FOR_WRAPPERS = {
|
|
|
|
"GRPC_VERBOSITY": "DEBUG",
|
|
|
|
}
|
|
|
|
|
|
|
|
_POLLING_STRATEGIES = {
|
|
|
|
"linux": ["epoll1", "poll"],
|
|
|
|
"mac": ["poll"],
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
def platform_string():
|
|
|
|
return jobset.platform_string()
|
|
|
|
|
|
|
|
|
|
|
|
_DEFAULT_TIMEOUT_SECONDS = 5 * 60
|
|
|
|
_PRE_BUILD_STEP_TIMEOUT_SECONDS = 10 * 60
|
|
|
|
|
|
|
|
|
|
|
|
def run_shell_command(cmd, env=None, cwd=None):
|
|
|
|
try:
|
|
|
|
subprocess.check_output(cmd, shell=True, env=env, cwd=cwd)
|
|
|
|
except subprocess.CalledProcessError as e:
|
|
|
|
logging.exception(
|
|
|
|
"Error while running command '%s'. Exit status %d. Output:\n%s",
|
|
|
|
e.cmd,
|
|
|
|
e.returncode,
|
|
|
|
e.output,
|
|
|
|
)
|
|
|
|
raise
|
|
|
|
|
|
|
|
|
|
|
|
def max_parallel_tests_for_current_platform():
|
|
|
|
# Too much test parallelization has only been seen to be a problem
|
|
|
|
# so far on windows.
|
|
|
|
if jobset.platform_string() == "windows":
|
|
|
|
return 64
|
|
|
|
return 1024
|
|
|
|
|
|
|
|
|
|
|
|
def _print_debug_info_epilogue(dockerfile_dir=None):
|
|
|
|
"""Use to print useful info for debug/repro just before exiting."""
|
|
|
|
print("")
|
|
|
|
print("=== run_tests.py DEBUG INFO ===")
|
|
|
|
print('command: "%s"' % " ".join(sys.argv))
|
|
|
|
if dockerfile_dir:
|
|
|
|
print("dockerfile: %s" % dockerfile_dir)
|
|
|
|
kokoro_job_name = os.getenv("KOKORO_JOB_NAME")
|
|
|
|
if kokoro_job_name:
|
|
|
|
print("kokoro job name: %s" % kokoro_job_name)
|
|
|
|
print("===============================")
|
|
|
|
|
|
|
|
|
|
|
|
# SimpleConfig: just compile with CONFIG=config, and run the binary to test
|
|
|
|
class Config(object):
|
|
|
|
def __init__(
|
|
|
|
self,
|
|
|
|
config,
|
|
|
|
environ=None,
|
|
|
|
timeout_multiplier=1,
|
|
|
|
tool_prefix=[],
|
|
|
|
iomgr_platform="native",
|
|
|
|
):
|
|
|
|
if environ is None:
|
|
|
|
environ = {}
|
|
|
|
self.build_config = config
|
|
|
|
self.environ = environ
|
|
|
|
self.environ["CONFIG"] = config
|
|
|
|
self.tool_prefix = tool_prefix
|
|
|
|
self.timeout_multiplier = timeout_multiplier
|
|
|
|
self.iomgr_platform = iomgr_platform
|
|
|
|
|
|
|
|
def job_spec(
|
|
|
|
self,
|
|
|
|
cmdline,
|
|
|
|
timeout_seconds=_DEFAULT_TIMEOUT_SECONDS,
|
|
|
|
shortname=None,
|
|
|
|
environ={},
|
|
|
|
cpu_cost=1.0,
|
|
|
|
flaky=False,
|
|
|
|
):
|
|
|
|
"""Construct a jobset.JobSpec for a test under this config
|
|
|
|
|
|
|
|
Args:
|
|
|
|
cmdline: a list of strings specifying the command line the test
|
|
|
|
would like to run
|
|
|
|
"""
|
|
|
|
actual_environ = self.environ.copy()
|
|
|
|
for k, v in environ.items():
|
|
|
|
actual_environ[k] = v
|
|
|
|
if not flaky and shortname and shortname in flaky_tests:
|
|
|
|
flaky = True
|
|
|
|
if shortname in shortname_to_cpu:
|
|
|
|
cpu_cost = shortname_to_cpu[shortname]
|
|
|
|
return jobset.JobSpec(
|
|
|
|
cmdline=self.tool_prefix + cmdline,
|
|
|
|
shortname=shortname,
|
|
|
|
environ=actual_environ,
|
|
|
|
cpu_cost=cpu_cost,
|
|
|
|
timeout_seconds=(
|
|
|
|
self.timeout_multiplier * timeout_seconds
|
|
|
|
if timeout_seconds
|
|
|
|
else None
|
|
|
|
),
|
|
|
|
flake_retries=4 if flaky or args.allow_flakes else 0,
|
|
|
|
timeout_retries=1 if flaky or args.allow_flakes else 0,
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def get_c_tests(travis, test_lang):
|
|
|
|
out = []
|
|
|
|
platforms_str = "ci_platforms" if travis else "platforms"
|
|
|
|
with open("tools/run_tests/generated/tests.json") as f:
|
|
|
|
js = json.load(f)
|
|
|
|
return [
|
|
|
|
tgt
|
|
|
|
for tgt in js
|
|
|
|
if tgt["language"] == test_lang
|
|
|
|
and platform_string() in tgt[platforms_str]
|
|
|
|
and not (travis and tgt["flaky"])
|
|
|
|
]
|
|
|
|
|
|
|
|
|
|
|
|
def _check_compiler(compiler, supported_compilers):
|
|
|
|
if compiler not in supported_compilers:
|
|
|
|
raise Exception(
|
|
|
|
"Compiler %s not supported (on this platform)." % compiler
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def _check_arch(arch, supported_archs):
|
|
|
|
if arch not in supported_archs:
|
|
|
|
raise Exception("Architecture %s not supported." % arch)
|
|
|
|
|
|
|
|
|
|
|
|
def _is_use_docker_child():
|
|
|
|
"""Returns True if running running as a --use_docker child."""
|
|
|
|
return True if os.getenv("DOCKER_RUN_SCRIPT_COMMAND") else False
|
|
|
|
|
|
|
|
|
|
|
|
_PythonConfigVars = collections.namedtuple(
|
|
|
|
"_ConfigVars",
|
|
|
|
[
|
|
|
|
"shell",
|
|
|
|
"builder",
|
|
|
|
"builder_prefix_arguments",
|
|
|
|
"venv_relative_python",
|
|
|
|
"toolchain",
|
|
|
|
"runner",
|
|
|
|
],
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def _python_config_generator(name, major, minor, bits, config_vars):
|
|
|
|
build = (
|
|
|
|
config_vars.shell
|
|
|
|
+ config_vars.builder
|
|
|
|
+ config_vars.builder_prefix_arguments
|
|
|
|
+ [_python_pattern_function(major=major, minor=minor, bits=bits)]
|
|
|
|
+ [name]
|
|
|
|
+ config_vars.venv_relative_python
|
|
|
|
+ config_vars.toolchain
|
|
|
|
)
|
|
|
|
# run: [tools/run_tests/helper_scripts/run_python.sh py37/bin/python]
|
|
|
|
python_path = os.path.join(name, config_vars.venv_relative_python[0])
|
|
|
|
run = config_vars.shell + config_vars.runner + [python_path]
|
|
|
|
return PythonConfig(name, build, run, python_path)
|
|
|
|
|
|
|
|
|
|
|
|
def _pypy_config_generator(name, major, config_vars):
|
|
|
|
# Something like "py37/bin/python"
|
|
|
|
python_path = os.path.join(name, config_vars.venv_relative_python[0])
|
|
|
|
return PythonConfig(
|
|
|
|
name,
|
|
|
|
config_vars.shell
|
|
|
|
+ config_vars.builder
|
|
|
|
+ config_vars.builder_prefix_arguments
|
|
|
|
+ [_pypy_pattern_function(major=major)]
|
|
|
|
+ [name]
|
|
|
|
+ config_vars.venv_relative_python
|
|
|
|
+ config_vars.toolchain,
|
|
|
|
config_vars.shell + config_vars.runner + [python_path],
|
|
|
|
python_path,
|
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def _python_pattern_function(major, minor, bits):
|
|
|
|
# Bit-ness is handled by the test machine's environment
|
|
|
|
if os.name == "nt":
|
|
|
|
if bits == "64":
|
|
|
|
return "/c/Python{major}{minor}/python.exe".format(
|
|
|
|
major=major, minor=minor, bits=bits
|
|
|
|
)
|
|
|
|
else:
|
|
|
|
return "/c/Python{major}{minor}_{bits}bits/python.exe".format(
|
|
|
|
major=major, minor=minor, bits=bits
|
|
|
|
)
|
|
|
|
else:
|
|
|
|
return "python{major}.{minor}".format(major=major, minor=minor)
|
|
|
|
|
|
|
|
|
|
|
|
def _pypy_pattern_function(major):
|
|
|
|
if major == "2":
|
|
|
|
return "pypy"
|
|
|
|
elif major == "3":
|
|
|
|
return "pypy3"
|
|
|
|
else:
|
|
|
|
raise ValueError("Unknown PyPy major version")
|
|
|
|
|
|
|
|
|
|
|
|
class CLanguage(object):
|
|
|
|
def __init__(self, lang_suffix, test_lang):
|
|
|
|
self.lang_suffix = lang_suffix
|
|
|
|
self.platform = platform_string()
|
|
|
|
self.test_lang = test_lang
|
|
|
|
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
if self.platform == "windows":
|
|
|
|
_check_compiler(
|
|
|
|
self.args.compiler,
|
|
|
|
[
|
|
|
|
"default",
|
|
|
|
"cmake",
|
|
|
|
"cmake_ninja_vs2019",
|
|
|
|
"cmake_vs2019",
|
|
|
|
"cmake_vs2022",
|
|
|
|
],
|
|
|
|
)
|
|
|
|
_check_arch(self.args.arch, ["default", "x64", "x86"])
|
|
|
|
|
|
|
|
activate_vs_tools = ""
|
|
|
|
if (
|
|
|
|
self.args.compiler == "cmake_ninja_vs2019"
|
|
|
|
or self.args.compiler == "cmake"
|
|
|
|
or self.args.compiler == "default"
|
|
|
|
):
|
|
|
|
# cmake + ninja build is the default because it is faster and supports boringssl assembly optimizations
|
|
|
|
# the compiler used is exactly the same as for cmake_vs2017
|
|
|
|
cmake_generator = "Ninja"
|
|
|
|
activate_vs_tools = "2019"
|
|
|
|
elif self.args.compiler == "cmake_vs2019":
|
|
|
|
cmake_generator = "Visual Studio 16 2019"
|
|
|
|
elif self.args.compiler == "cmake_vs2022":
|
|
|
|
cmake_generator = "Visual Studio 17 2022"
|
|
|
|
else:
|
|
|
|
print("should never reach here.")
|
|
|
|
sys.exit(1)
|
|
|
|
|
|
|
|
self._cmake_configure_extra_args = list(
|
|
|
|
self.args.cmake_configure_extra_args
|
|
|
|
)
|
|
|
|
self._cmake_generator_windows = cmake_generator
|
|
|
|
# required to pass as cmake "-A" configuration for VS builds (but not for Ninja)
|
|
|
|
self._cmake_architecture_windows = (
|
|
|
|
"x64" if self.args.arch == "x64" else "Win32"
|
|
|
|
)
|
|
|
|
# when builing with Ninja, the VS common tools need to be activated first
|
|
|
|
self._activate_vs_tools_windows = activate_vs_tools
|
|
|
|
# "x64_x86" means create 32bit binaries, but use 64bit toolkit to secure more memory for the build
|
|
|
|
self._vs_tools_architecture_windows = (
|
|
|
|
"x64" if self.args.arch == "x64" else "x64_x86"
|
|
|
|
)
|
|
|
|
|
|
|
|
else:
|
|
|
|
if self.platform == "linux":
|
|
|
|
# Allow all the known architectures. _check_arch_option has already checked that we're not doing
|
|
|
|
# something illegal when not running under docker.
|
|
|
|
_check_arch(self.args.arch, ["default", "x64", "x86", "arm64"])
|
|
|
|
else:
|
|
|
|
_check_arch(self.args.arch, ["default"])
|
|
|
|
|
|
|
|
(
|
|
|
|
self._docker_distro,
|
|
|
|
self._cmake_configure_extra_args,
|
|
|
|
) = self._compiler_options(
|
|
|
|
self.args.use_docker,
|
|
|
|
self.args.compiler,
|
|
|
|
self.args.cmake_configure_extra_args,
|
|
|
|
)
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
out = []
|
|
|
|
binaries = get_c_tests(self.args.travis, self.test_lang)
|
|
|
|
for target in binaries:
|
|
|
|
if target.get("boringssl", False):
|
|
|
|
# cmake doesn't build boringssl tests
|
|
|
|
continue
|
|
|
|
auto_timeout_scaling = target.get("auto_timeout_scaling", True)
|
|
|
|
polling_strategies = (
|
|
|
|
_POLLING_STRATEGIES.get(self.platform, ["all"])
|
|
|
|
if target.get("uses_polling", True)
|
|
|
|
else ["none"]
|
|
|
|
)
|
|
|
|
for polling_strategy in polling_strategies:
|
|
|
|
env = {
|
|
|
|
"GRPC_DEFAULT_SSL_ROOTS_FILE_PATH": _ROOT
|
|
|
|
+ "/src/core/tsi/test_creds/ca.pem",
|
|
|
|
"GRPC_POLL_STRATEGY": polling_strategy,
|
|
|
|
"GRPC_VERBOSITY": "DEBUG",
|
|
|
|
}
|
|
|
|
resolver = os.environ.get("GRPC_DNS_RESOLVER", None)
|
|
|
|
if resolver:
|
|
|
|
env["GRPC_DNS_RESOLVER"] = resolver
|
|
|
|
shortname_ext = (
|
|
|
|
""
|
|
|
|
if polling_strategy == "all"
|
|
|
|
else " GRPC_POLL_STRATEGY=%s" % polling_strategy
|
|
|
|
)
|
|
|
|
if polling_strategy in target.get("excluded_poll_engines", []):
|
|
|
|
continue
|
|
|
|
|
|
|
|
timeout_scaling = 1
|
|
|
|
if auto_timeout_scaling:
|
|
|
|
config = self.args.config
|
|
|
|
if (
|
|
|
|
"asan" in config
|
|
|
|
or config == "msan"
|
|
|
|
or config == "tsan"
|
|
|
|
or config == "ubsan"
|
|
|
|
or config == "helgrind"
|
|
|
|
or config == "memcheck"
|
|
|
|
):
|
|
|
|
# Scale overall test timeout if running under various sanitizers.
|
|
|
|
# scaling value is based on historical data analysis
|
|
|
|
timeout_scaling *= 3
|
|
|
|
|
|
|
|
if self.config.build_config in target["exclude_configs"]:
|
|
|
|
continue
|
|
|
|
if self.args.iomgr_platform in target.get("exclude_iomgrs", []):
|
|
|
|
continue
|
|
|
|
|
|
|
|
if self.platform == "windows":
|
|
|
|
if self._cmake_generator_windows == "Ninja":
|
|
|
|
binary = "cmake/build/%s.exe" % target["name"]
|
[test-infra] Re-enable skipped windows tests (#32658)
Followup for https://github.com/grpc/grpc/pull/32649 (which disabled the
tests mentioned below).
Also sets correct path for tests build by ninja on windows, so that they
don't get skipped.
Once merged, I'll backport to 1.54.x and 1.53.x
The original issue with tests being skipped.
```
+ python3 workspace_c_windows_dbg_native/tools/run_tests/run_tests.py -t -j 8 -x run_tests/c_windows_dbg_native/sponge_log.xml --report_suite_name c_windows_dbg_native -l c -c dbg --iomgr_platform native --bq_result_table aggregate_results --measure_cpu_costs
2023-03-20 07:56:53,523 START: tools\run_tests\helper_scripts\build_cxx.bat
2023-03-20 08:04:51,388 PASSED: tools\run_tests\helper_scripts\build_cxx.bat [time=477.9sec, retries=0:0; cpu_cost=0.0; estimated=1.0]
2023-03-20 08:04:52,434 detected port server running version 21
2023-03-20 08:04:52,672 my port server is version 21
2023-03-20 08:04:52,703 SUCCESS: All tests passed
WARNING: binary not found, skipping cmake/build/Debug/bad_server_response_test.exe
WARNING: binary not found, skipping cmake/build/Debug/connection_refused_test.exe
WARNING: binary not found, skipping cmake/build/Debug/goaway_server_test.exe
WARNING: binary not found, skipping cmake/build/Debug/invalid_call_argument_test.exe
WARNING: binary not found, skipping cmake/build/Debug/multiple_server_queues_test.exe
WARNING: binary not found, skipping cmake/build/Debug/no_server_test.exe
WARNING: binary not found, skipping cmake/build/Debug/pollset_windows_starvation_test.exe
WARNING: binary not found, skipping cmake/build/Debug/public_headers_must_be_c89.exe
```
2 years ago
|
|
|
else:
|
|
|
|
binary = "cmake/build/%s/%s.exe" % (
|
|
|
|
_MSBUILD_CONFIG[self.config.build_config],
|
|
|
|
target["name"],
|
|
|
|
)
|
|
|
|
else:
|
|
|
|
binary = "cmake/build/%s" % target["name"]
|
|
|
|
|
|
|
|
cpu_cost = target["cpu_cost"]
|
|
|
|
if cpu_cost == "capacity":
|
|
|
|
cpu_cost = multiprocessing.cpu_count()
|
|
|
|
if os.path.isfile(binary):
|
|
|
|
list_test_command = None
|
|
|
|
filter_test_command = None
|
|
|
|
|
|
|
|
# these are the flag defined by gtest and benchmark framework to list
|
|
|
|
# and filter test runs. We use them to split each individual test
|
|
|
|
# into its own JobSpec, and thus into its own process.
|
|
|
|
if "benchmark" in target and target["benchmark"]:
|
|
|
|
with open(os.devnull, "w") as fnull:
|
|
|
|
tests = subprocess.check_output(
|
|
|
|
[binary, "--benchmark_list_tests"], stderr=fnull
|
|
|
|
)
|
|
|
|
for line in tests.decode().split("\n"):
|
|
|
|
test = line.strip()
|
|
|
|
if not test:
|
|
|
|
continue
|
|
|
|
cmdline = [
|
|
|
|
binary,
|
|
|
|
"--benchmark_filter=%s$" % test,
|
|
|
|
] + target["args"]
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
cmdline,
|
|
|
|
shortname="%s %s"
|
|
|
|
% (" ".join(cmdline), shortname_ext),
|
|
|
|
cpu_cost=cpu_cost,
|
|
|
|
timeout_seconds=target.get(
|
|
|
|
"timeout_seconds",
|
|
|
|
_DEFAULT_TIMEOUT_SECONDS,
|
|
|
|
)
|
|
|
|
* timeout_scaling,
|
|
|
|
environ=env,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
elif "gtest" in target and target["gtest"]:
|
|
|
|
# here we parse the output of --gtest_list_tests to build up a complete
|
|
|
|
# list of the tests contained in a binary for each test, we then
|
|
|
|
# add a job to run, filtering for just that test.
|
|
|
|
with open(os.devnull, "w") as fnull:
|
|
|
|
tests = subprocess.check_output(
|
|
|
|
[binary, "--gtest_list_tests"], stderr=fnull
|
|
|
|
)
|
|
|
|
base = None
|
|
|
|
for line in tests.decode().split("\n"):
|
|
|
|
i = line.find("#")
|
|
|
|
if i >= 0:
|
|
|
|
line = line[:i]
|
|
|
|
if not line:
|
|
|
|
continue
|
|
|
|
if line[0] != " ":
|
|
|
|
base = line.strip()
|
|
|
|
else:
|
|
|
|
assert base is not None
|
|
|
|
assert line[1] == " "
|
|
|
|
test = base + line.strip()
|
|
|
|
cmdline = [
|
|
|
|
binary,
|
|
|
|
"--gtest_filter=%s" % test,
|
|
|
|
] + target["args"]
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
cmdline,
|
|
|
|
shortname="%s %s"
|
|
|
|
% (" ".join(cmdline), shortname_ext),
|
|
|
|
cpu_cost=cpu_cost,
|
|
|
|
timeout_seconds=target.get(
|
|
|
|
"timeout_seconds",
|
|
|
|
_DEFAULT_TIMEOUT_SECONDS,
|
|
|
|
)
|
|
|
|
* timeout_scaling,
|
|
|
|
environ=env,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
else:
|
|
|
|
cmdline = [binary] + target["args"]
|
|
|
|
shortname = target.get(
|
|
|
|
"shortname",
|
|
|
|
" ".join(shlex.quote(arg) for arg in cmdline),
|
|
|
|
)
|
|
|
|
shortname += shortname_ext
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
cmdline,
|
|
|
|
shortname=shortname,
|
|
|
|
cpu_cost=cpu_cost,
|
|
|
|
flaky=target.get("flaky", False),
|
|
|
|
timeout_seconds=target.get(
|
|
|
|
"timeout_seconds", _DEFAULT_TIMEOUT_SECONDS
|
|
|
|
)
|
|
|
|
* timeout_scaling,
|
|
|
|
environ=env,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
elif self.args.regex == ".*" or self.platform == "windows":
|
|
|
|
print("\nWARNING: binary not found, skipping", binary)
|
|
|
|
return sorted(out)
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
if self.platform == "windows":
|
|
|
|
return [
|
|
|
|
[
|
|
|
|
"tools\\run_tests\\helper_scripts\\build_cxx.bat",
|
|
|
|
"-DgRPC_BUILD_MSVC_MP_COUNT=%d" % self.args.jobs,
|
|
|
|
]
|
|
|
|
+ self._cmake_configure_extra_args
|
|
|
|
]
|
|
|
|
else:
|
|
|
|
return [
|
|
|
|
["tools/run_tests/helper_scripts/build_cxx.sh"]
|
|
|
|
+ self._cmake_configure_extra_args
|
|
|
|
]
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
environ = {"GRPC_RUN_TESTS_CXX_LANGUAGE_SUFFIX": self.lang_suffix}
|
|
|
|
if self.platform == "windows":
|
|
|
|
environ["GRPC_CMAKE_GENERATOR"] = self._cmake_generator_windows
|
|
|
|
environ[
|
|
|
|
"GRPC_CMAKE_ARCHITECTURE"
|
|
|
|
] = self._cmake_architecture_windows
|
|
|
|
environ[
|
|
|
|
"GRPC_BUILD_ACTIVATE_VS_TOOLS"
|
|
|
|
] = self._activate_vs_tools_windows
|
|
|
|
environ[
|
|
|
|
"GRPC_BUILD_VS_TOOLS_ARCHITECTURE"
|
|
|
|
] = self._vs_tools_architecture_windows
|
|
|
|
return environ
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
if self.platform == "windows":
|
|
|
|
return []
|
|
|
|
else:
|
|
|
|
return [["tools/run_tests/helper_scripts/post_tests_c.sh"]]
|
|
|
|
|
|
|
|
def _clang_cmake_configure_extra_args(self, version_suffix=""):
|
|
|
|
return [
|
|
|
|
"-DCMAKE_C_COMPILER=clang%s" % version_suffix,
|
|
|
|
"-DCMAKE_CXX_COMPILER=clang++%s" % version_suffix,
|
|
|
|
]
|
|
|
|
|
|
|
|
def _compiler_options(
|
|
|
|
self, use_docker, compiler, cmake_configure_extra_args
|
|
|
|
):
|
|
|
|
"""Returns docker distro and cmake configure args to use for given compiler."""
|
|
|
|
if cmake_configure_extra_args:
|
|
|
|
# only allow specifying extra cmake args for "vanilla" compiler
|
|
|
|
_check_compiler(compiler, ["default", "cmake"])
|
|
|
|
return ("nonexistent_docker_distro", cmake_configure_extra_args)
|
|
|
|
if not use_docker and not _is_use_docker_child():
|
|
|
|
# if not running under docker, we cannot ensure the right compiler version will be used,
|
|
|
|
# so we only allow the non-specific choices.
|
|
|
|
_check_compiler(compiler, ["default", "cmake"])
|
|
|
|
|
|
|
|
if compiler == "default" or compiler == "cmake":
|
|
|
|
return ("debian11", [])
|
|
|
|
elif compiler == "gcc8":
|
|
|
|
return ("gcc_8", [])
|
|
|
|
elif compiler == "gcc10.2":
|
|
|
|
return ("debian11", [])
|
|
|
|
elif compiler == "gcc10.2_openssl102":
|
|
|
|
return (
|
|
|
|
"debian11_openssl102",
|
|
|
|
[
|
|
|
|
"-DgRPC_SSL_PROVIDER=package",
|
|
|
|
],
|
|
|
|
)
|
|
|
|
elif compiler == "gcc10.2_openssl111":
|
|
|
|
return (
|
|
|
|
"debian11_openssl111",
|
|
|
|
[
|
|
|
|
"-DgRPC_SSL_PROVIDER=package",
|
|
|
|
],
|
|
|
|
)
|
|
|
|
elif compiler == "gcc12":
|
|
|
|
return ("gcc_12", ["-DCMAKE_CXX_STANDARD=20"])
|
|
|
|
elif compiler == "gcc12_openssl309":
|
|
|
|
return (
|
|
|
|
"debian12_openssl309",
|
|
|
|
[
|
|
|
|
"-DgRPC_SSL_PROVIDER=package",
|
|
|
|
],
|
|
|
|
)
|
|
|
|
elif compiler == "gcc_musl":
|
|
|
|
return ("alpine", [])
|
|
|
|
elif compiler == "clang6":
|
|
|
|
return ("clang_6", self._clang_cmake_configure_extra_args())
|
|
|
|
elif compiler == "clang17":
|
|
|
|
return ("clang_17", self._clang_cmake_configure_extra_args())
|
|
|
|
else:
|
|
|
|
raise Exception("Compiler %s not supported." % compiler)
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/cxx_%s_%s" % (
|
|
|
|
self._docker_distro,
|
|
|
|
_docker_arch_suffix(self.args.arch),
|
|
|
|
)
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return self.lang_suffix
|
|
|
|
|
|
|
|
|
|
|
|
class Php7Language(object):
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
_check_compiler(self.args.compiler, ["default"])
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
return [
|
|
|
|
self.config.job_spec(
|
|
|
|
["src/php/bin/run_tests.sh"],
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
)
|
|
|
|
]
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
return [["tools/run_tests/helper_scripts/build_php.sh"]]
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
return [["tools/run_tests/helper_scripts/post_tests_php.sh"]]
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/php7_debian11_%s" % _docker_arch_suffix(
|
|
|
|
self.args.arch
|
|
|
|
)
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "php7"
|
|
|
|
|
|
|
|
|
|
|
|
class PythonConfig(
|
|
|
|
collections.namedtuple(
|
|
|
|
"PythonConfig", ["name", "build", "run", "python_path"]
|
|
|
|
)
|
|
|
|
):
|
|
|
|
"""Tuple of commands (named s.t. 'what it says on the tin' applies)"""
|
|
|
|
|
|
|
|
|
|
|
|
class PythonLanguage(object):
|
|
|
|
_TEST_SPECS_FILE = {
|
|
|
|
"native": ["src/python/grpcio_tests/tests/tests.json"],
|
|
|
|
"asyncio": ["src/python/grpcio_tests/tests_aio/tests.json"],
|
|
|
|
}
|
|
|
|
|
|
|
|
_TEST_COMMAND = {
|
|
|
|
"native": "test_lite",
|
|
|
|
"asyncio": "test_aio",
|
|
|
|
}
|
|
|
|
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
self.pythons = self._get_pythons(self.args)
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
# load list of known test suites
|
|
|
|
jobs = []
|
|
|
|
|
|
|
|
# Run tests across all supported interpreters.
|
|
|
|
for python_config in self.pythons:
|
|
|
|
# Run non-io-manager-specific tests.
|
|
|
|
if os.name != "nt":
|
|
|
|
jobs.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
[
|
|
|
|
python_config.python_path,
|
|
|
|
"tools/distrib/python/xds_protos/generated_file_import_test.py",
|
|
|
|
],
|
|
|
|
timeout_seconds=60,
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
shortname=f"{python_config.name}.xds_protos",
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
# Run main test suite across all support IO managers.
|
|
|
|
for io_platform in self._TEST_SPECS_FILE:
|
|
|
|
test_cases = []
|
|
|
|
for tests_json_file_name in self._TEST_SPECS_FILE[io_platform]:
|
|
|
|
with open(tests_json_file_name) as tests_json_file:
|
|
|
|
test_cases.extend(json.load(tests_json_file))
|
|
|
|
|
|
|
|
environment = dict(_FORCE_ENVIRON_FOR_WRAPPERS)
|
|
|
|
# TODO(https://github.com/grpc/grpc/issues/21401) Fork handlers is not
|
|
|
|
# designed for non-native IO manager. It has a side-effect that
|
|
|
|
# overrides threading settings in C-Core.
|
|
|
|
if io_platform != "native":
|
|
|
|
environment["GRPC_ENABLE_FORK_SUPPORT"] = "0"
|
|
|
|
jobs.extend(
|
|
|
|
[
|
|
|
|
self.config.job_spec(
|
|
|
|
python_config.run
|
|
|
|
+ [self._TEST_COMMAND[io_platform]],
|
|
|
|
timeout_seconds=8 * 60,
|
|
|
|
environ=dict(
|
|
|
|
GRPC_PYTHON_TESTRUNNER_FILTER=str(test_case),
|
|
|
|
**environment,
|
|
|
|
),
|
|
|
|
shortname=f"{python_config.name}.{io_platform}.{test_case}",
|
|
|
|
)
|
|
|
|
for test_case in test_cases
|
|
|
|
]
|
|
|
|
)
|
|
|
|
return jobs
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
return [config.build for config in self.pythons]
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
if self.config.build_config != "gcov":
|
|
|
|
return []
|
|
|
|
else:
|
|
|
|
return [["tools/run_tests/helper_scripts/post_tests_python.sh"]]
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/python_%s_%s" % (
|
|
|
|
self._python_docker_distro_name(),
|
|
|
|
_docker_arch_suffix(self.args.arch),
|
|
|
|
)
|
|
|
|
|
|
|
|
def _python_docker_distro_name(self):
|
|
|
|
"""Choose the docker image to use based on python version."""
|
|
|
|
if self.args.compiler == "python_alpine":
|
|
|
|
return "alpine"
|
|
|
|
else:
|
|
|
|
return "debian11_default"
|
|
|
|
|
|
|
|
def _get_pythons(self, args):
|
|
|
|
"""Get python runtimes to test with, based on current platform, architecture, compiler etc."""
|
|
|
|
if args.iomgr_platform != "native":
|
|
|
|
raise ValueError(
|
|
|
|
"Python builds no longer differentiate IO Manager platforms,"
|
|
|
|
' please use "native"'
|
|
|
|
)
|
|
|
|
|
|
|
|
if args.arch == "x86":
|
|
|
|
bits = "32"
|
|
|
|
else:
|
|
|
|
bits = "64"
|
|
|
|
|
|
|
|
if os.name == "nt":
|
|
|
|
shell = ["bash"]
|
|
|
|
builder = [
|
|
|
|
os.path.abspath(
|
|
|
|
"tools/run_tests/helper_scripts/build_python_msys2.sh"
|
|
|
|
)
|
|
|
|
]
|
|
|
|
builder_prefix_arguments = ["MINGW{}".format(bits)]
|
|
|
|
venv_relative_python = ["Scripts/python.exe"]
|
|
|
|
toolchain = ["mingw32"]
|
|
|
|
else:
|
|
|
|
shell = []
|
|
|
|
builder = [
|
|
|
|
os.path.abspath(
|
|
|
|
"tools/run_tests/helper_scripts/build_python.sh"
|
|
|
|
)
|
|
|
|
]
|
|
|
|
builder_prefix_arguments = []
|
|
|
|
venv_relative_python = ["bin/python"]
|
|
|
|
toolchain = ["unix"]
|
|
|
|
|
|
|
|
runner = [
|
|
|
|
os.path.abspath("tools/run_tests/helper_scripts/run_python.sh")
|
|
|
|
]
|
|
|
|
|
|
|
|
config_vars = _PythonConfigVars(
|
|
|
|
shell,
|
|
|
|
builder,
|
|
|
|
builder_prefix_arguments,
|
|
|
|
venv_relative_python,
|
|
|
|
toolchain,
|
|
|
|
runner,
|
|
|
|
)
|
|
|
|
|
|
|
|
# TODO: Supported version range should be defined by a single
|
|
|
|
# source of truth.
|
|
|
|
python37_config = _python_config_generator(
|
|
|
|
name="py37",
|
|
|
|
major="3",
|
|
|
|
minor="7",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
python38_config = _python_config_generator(
|
|
|
|
name="py38",
|
|
|
|
major="3",
|
|
|
|
minor="8",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
python39_config = _python_config_generator(
|
|
|
|
name="py39",
|
|
|
|
major="3",
|
|
|
|
minor="9",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
python310_config = _python_config_generator(
|
|
|
|
name="py310",
|
|
|
|
major="3",
|
|
|
|
minor="10",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
python311_config = _python_config_generator(
|
|
|
|
name="py311",
|
|
|
|
major="3",
|
|
|
|
minor="11",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
python312_config = _python_config_generator(
|
|
|
|
name="py312",
|
|
|
|
major="3",
|
|
|
|
minor="12",
|
|
|
|
bits=bits,
|
|
|
|
config_vars=config_vars,
|
|
|
|
)
|
|
|
|
pypy27_config = _pypy_config_generator(
|
|
|
|
name="pypy", major="2", config_vars=config_vars
|
|
|
|
)
|
|
|
|
pypy32_config = _pypy_config_generator(
|
|
|
|
name="pypy3", major="3", config_vars=config_vars
|
|
|
|
)
|
|
|
|
|
|
|
|
if args.compiler == "default":
|
|
|
|
if os.name == "nt":
|
|
|
|
return (python38_config,)
|
|
|
|
elif os.uname()[0] == "Darwin":
|
|
|
|
# NOTE(rbellevi): Testing takes significantly longer on
|
|
|
|
# MacOS, so we restrict the number of interpreter versions
|
|
|
|
# tested.
|
|
|
|
return (python38_config,)
|
|
|
|
elif platform.machine() == "aarch64":
|
|
|
|
# Currently the python_debian11_default_arm64 docker image
|
|
|
|
# only has python3.9 installed (and that seems sufficient
|
|
|
|
# for arm64 testing)
|
|
|
|
return (python39_config,)
|
|
|
|
else:
|
|
|
|
# Default set tested on master. Test oldest and newest.
|
|
|
|
return (
|
|
|
|
python37_config,
|
|
|
|
python312_config,
|
|
|
|
)
|
|
|
|
elif args.compiler == "python3.7":
|
|
|
|
return (python37_config,)
|
|
|
|
elif args.compiler == "python3.8":
|
|
|
|
return (python38_config,)
|
|
|
|
elif args.compiler == "python3.9":
|
|
|
|
return (python39_config,)
|
|
|
|
elif args.compiler == "python3.10":
|
|
|
|
return (python310_config,)
|
|
|
|
elif args.compiler == "python3.11":
|
|
|
|
return (python311_config,)
|
|
|
|
elif args.compiler == "python3.12":
|
|
|
|
return (python312_config,)
|
|
|
|
elif args.compiler == "pypy":
|
|
|
|
return (pypy27_config,)
|
|
|
|
elif args.compiler == "pypy3":
|
|
|
|
return (pypy32_config,)
|
|
|
|
elif args.compiler == "python_alpine":
|
|
|
|
return (python39_config,)
|
|
|
|
elif args.compiler == "all_the_cpythons":
|
|
|
|
return (
|
|
|
|
python37_config,
|
|
|
|
python38_config,
|
|
|
|
python39_config,
|
|
|
|
python310_config,
|
|
|
|
python311_config,
|
|
|
|
python312_config,
|
|
|
|
)
|
|
|
|
else:
|
|
|
|
raise Exception("Compiler %s not supported." % args.compiler)
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "python"
|
|
|
|
|
|
|
|
|
|
|
|
class RubyLanguage(object):
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
_check_compiler(self.args.compiler, ["default"])
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
tests = [
|
|
|
|
self.config.job_spec(
|
|
|
|
["tools/run_tests/helper_scripts/run_ruby.sh"],
|
|
|
|
timeout_seconds=10 * 60,
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
)
|
|
|
|
]
|
|
|
|
# TODO(apolcyn): re-enable the following tests after
|
|
|
|
# https://bugs.ruby-lang.org/issues/15499 is fixed:
|
|
|
|
# They previously worked on ruby 2.5 but needed to be disabled
|
|
|
|
# after dropping support for ruby 2.5:
|
|
|
|
# - src/ruby/end2end/channel_state_test.rb
|
|
|
|
# - src/ruby/end2end/sig_int_during_channel_watch_test.rb
|
|
|
|
# TODO(apolcyn): the following test is skipped because it sometimes
|
|
|
|
# hits "Bus Error" crashes while requiring the grpc/ruby C-extension.
|
|
|
|
# This crashes have been unreproducible outside of CI. Also see
|
|
|
|
# b/266212253.
|
|
|
|
# - src/ruby/end2end/grpc_class_init_test.rb
|
|
|
|
for test in [
|
|
|
|
"src/ruby/end2end/fork_test.rb",
|
|
|
|
"src/ruby/end2end/simple_fork_test.rb",
|
|
|
|
"src/ruby/end2end/prefork_without_using_grpc_test.rb",
|
|
|
|
"src/ruby/end2end/prefork_postfork_loop_test.rb",
|
|
|
|
"src/ruby/end2end/secure_fork_test.rb",
|
|
|
|
"src/ruby/end2end/bad_usage_fork_test.rb",
|
|
|
|
"src/ruby/end2end/sig_handling_test.rb",
|
|
|
|
"src/ruby/end2end/channel_closing_test.rb",
|
|
|
|
"src/ruby/end2end/killed_client_thread_test.rb",
|
|
|
|
"src/ruby/end2end/forking_client_test.rb",
|
|
|
|
"src/ruby/end2end/multiple_killed_watching_threads_test.rb",
|
|
|
|
"src/ruby/end2end/load_grpc_with_gc_stress_test.rb",
|
|
|
|
"src/ruby/end2end/client_memory_usage_test.rb",
|
|
|
|
"src/ruby/end2end/package_with_underscore_test.rb",
|
|
|
|
"src/ruby/end2end/graceful_sig_handling_test.rb",
|
|
|
|
"src/ruby/end2end/graceful_sig_stop_test.rb",
|
|
|
|
"src/ruby/end2end/errors_load_before_grpc_lib_test.rb",
|
|
|
|
"src/ruby/end2end/logger_load_before_grpc_lib_test.rb",
|
|
|
|
"src/ruby/end2end/status_codes_load_before_grpc_lib_test.rb",
|
|
|
|
"src/ruby/end2end/call_credentials_timeout_test.rb",
|
|
|
|
"src/ruby/end2end/call_credentials_returning_bad_metadata_doesnt_kill_background_thread_test.rb",
|
|
|
|
]:
|
|
|
|
if test in [
|
|
|
|
"src/ruby/end2end/fork_test.rb",
|
|
|
|
"src/ruby/end2end/simple_fork_test.rb",
|
|
|
|
"src/ruby/end2end/secure_fork_test.rb",
|
|
|
|
"src/ruby/end2end/bad_usage_fork_test.rb",
|
|
|
|
"src/ruby/end2end/prefork_without_using_grpc_test.rb",
|
|
|
|
"src/ruby/end2end/prefork_postfork_loop_test.rb",
|
|
|
|
"src/ruby/end2end/fork_test_repro_35489.rb",
|
|
|
|
]:
|
|
|
|
# Skip fork tests in general until https://github.com/grpc/grpc/issues/34442
|
|
|
|
# is fixed. Otherwise we see too many flakes.
|
|
|
|
# After that's fixed, we should continue to skip on mac
|
|
|
|
# indefinitely, and on "dbg" builds until the Event Engine
|
|
|
|
# migration completes.
|
|
|
|
continue
|
|
|
|
tests.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
["ruby", test],
|
|
|
|
shortname=test,
|
|
|
|
timeout_seconds=20 * 60,
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
return tests
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return [["tools/run_tests/helper_scripts/pre_build_ruby.sh"]]
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
return [["tools/run_tests/helper_scripts/build_ruby.sh"]]
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
return [["tools/run_tests/helper_scripts/post_tests_ruby.sh"]]
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/ruby_debian11_%s" % _docker_arch_suffix(
|
|
|
|
self.args.arch
|
|
|
|
)
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "ruby"
|
|
|
|
|
|
|
|
|
|
|
|
class CSharpLanguage(object):
|
|
|
|
def __init__(self):
|
|
|
|
self.platform = platform_string()
|
|
|
|
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
_check_compiler(self.args.compiler, ["default", "coreclr", "mono"])
|
|
|
|
if self.args.compiler == "default":
|
|
|
|
# test both runtimes by default
|
|
|
|
self.test_runtimes = ["coreclr", "mono"]
|
|
|
|
else:
|
|
|
|
# only test the specified runtime
|
|
|
|
self.test_runtimes = [self.args.compiler]
|
|
|
|
|
|
|
|
if self.platform == "windows":
|
|
|
|
_check_arch(self.args.arch, ["default"])
|
|
|
|
self._cmake_arch_option = "x64"
|
|
|
|
else:
|
|
|
|
self._docker_distro = "debian11"
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
with open("src/csharp/tests.json") as f:
|
|
|
|
tests_by_assembly = json.load(f)
|
|
|
|
|
|
|
|
msbuild_config = _MSBUILD_CONFIG[self.config.build_config]
|
|
|
|
nunit_args = ["--labels=All", "--noresult", "--workers=1"]
|
|
|
|
|
|
|
|
specs = []
|
|
|
|
for test_runtime in self.test_runtimes:
|
|
|
|
if test_runtime == "coreclr":
|
|
|
|
assembly_extension = ".dll"
|
|
|
|
assembly_subdir = "bin/%s/netcoreapp3.1" % msbuild_config
|
|
|
|
runtime_cmd = ["dotnet", "exec"]
|
|
|
|
elif test_runtime == "mono":
|
|
|
|
assembly_extension = ".exe"
|
|
|
|
assembly_subdir = "bin/%s/net45" % msbuild_config
|
|
|
|
if self.platform == "windows":
|
|
|
|
runtime_cmd = []
|
|
|
|
elif self.platform == "mac":
|
|
|
|
# mono before version 5.2 on MacOS defaults to 32bit runtime
|
|
|
|
runtime_cmd = ["mono", "--arch=64"]
|
|
|
|
else:
|
|
|
|
runtime_cmd = ["mono"]
|
|
|
|
else:
|
|
|
|
raise Exception('Illegal runtime "%s" was specified.')
|
|
|
|
|
|
|
|
for assembly in six.iterkeys(tests_by_assembly):
|
|
|
|
assembly_file = "src/csharp/%s/%s/%s%s" % (
|
|
|
|
assembly,
|
|
|
|
assembly_subdir,
|
|
|
|
assembly,
|
|
|
|
assembly_extension,
|
|
|
|
)
|
|
|
|
|
|
|
|
# normally, run each test as a separate process
|
|
|
|
for test in tests_by_assembly[assembly]:
|
|
|
|
cmdline = (
|
|
|
|
runtime_cmd
|
|
|
|
+ [assembly_file, "--test=%s" % test]
|
|
|
|
+ nunit_args
|
|
|
|
)
|
|
|
|
specs.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
cmdline,
|
|
|
|
shortname="csharp.%s.%s" % (test_runtime, test),
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
return specs
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
if self.platform == "windows":
|
|
|
|
return [["tools\\run_tests\\helper_scripts\\pre_build_csharp.bat"]]
|
|
|
|
else:
|
|
|
|
return [["tools/run_tests/helper_scripts/pre_build_csharp.sh"]]
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
if self.platform == "windows":
|
|
|
|
return [["tools\\run_tests\\helper_scripts\\build_csharp.bat"]]
|
|
|
|
else:
|
|
|
|
return [["tools/run_tests/helper_scripts/build_csharp.sh"]]
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
if self.platform == "windows":
|
|
|
|
return {"ARCHITECTURE": self._cmake_arch_option}
|
|
|
|
else:
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
if self.platform == "windows":
|
|
|
|
return [["tools\\run_tests\\helper_scripts\\post_tests_csharp.bat"]]
|
|
|
|
else:
|
|
|
|
return [["tools/run_tests/helper_scripts/post_tests_csharp.sh"]]
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/csharp_%s_%s" % (
|
|
|
|
self._docker_distro,
|
|
|
|
_docker_arch_suffix(self.args.arch),
|
|
|
|
)
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "csharp"
|
|
|
|
|
|
|
|
|
|
|
|
class ObjCLanguage(object):
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
_check_compiler(self.args.compiler, ["default"])
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
out = []
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
["src/objective-c/tests/build_one_example.sh"],
|
|
|
|
timeout_seconds=20 * 60,
|
|
|
|
shortname="ios-buildtest-example-sample",
|
|
|
|
cpu_cost=1e6,
|
|
|
|
environ={
|
|
|
|
"SCHEME": "Sample",
|
|
|
|
"EXAMPLE_PATH": "src/objective-c/examples/Sample",
|
|
|
|
},
|
|
|
|
)
|
|
|
|
)
|
|
|
|
# TODO(jtattermusch): Create bazel target for the sample and remove the test task from here.
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
["src/objective-c/tests/build_one_example.sh"],
|
|
|
|
timeout_seconds=20 * 60,
|
|
|
|
shortname="ios-buildtest-example-switftsample",
|
|
|
|
cpu_cost=1e6,
|
|
|
|
environ={
|
|
|
|
"SCHEME": "SwiftSample",
|
|
|
|
"EXAMPLE_PATH": "src/objective-c/examples/SwiftSample",
|
|
|
|
},
|
|
|
|
)
|
|
|
|
)
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
["src/objective-c/tests/build_one_example.sh"],
|
|
|
|
timeout_seconds=20 * 60,
|
|
|
|
shortname="ios-buildtest-example-switft-use-frameworks",
|
|
|
|
cpu_cost=1e6,
|
|
|
|
environ={
|
|
|
|
"SCHEME": "SwiftUseFrameworks",
|
|
|
|
"EXAMPLE_PATH": "src/objective-c/examples/SwiftUseFrameworks",
|
|
|
|
},
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
# Disabled due to #20258
|
|
|
|
# TODO (mxyan): Reenable this test when #20258 is resolved.
|
|
|
|
# out.append(
|
|
|
|
# self.config.job_spec(
|
|
|
|
# ['src/objective-c/tests/build_one_example_bazel.sh'],
|
|
|
|
# timeout_seconds=20 * 60,
|
|
|
|
# shortname='ios-buildtest-example-watchOS-sample',
|
|
|
|
# cpu_cost=1e6,
|
|
|
|
# environ={
|
|
|
|
# 'SCHEME': 'watchOS-sample-WatchKit-App',
|
|
|
|
# 'EXAMPLE_PATH': 'src/objective-c/examples/watchOS-sample',
|
|
|
|
# 'FRAMEWORKS': 'NO'
|
|
|
|
# }))
|
|
|
|
|
|
|
|
# TODO(jtattermusch): move the test out of the test/core/iomgr/CFStreamTests directory?
|
|
|
|
# How does one add the cfstream dependency in bazel?
|
|
|
|
out.append(
|
|
|
|
self.config.job_spec(
|
|
|
|
["test/core/iomgr/ios/CFStreamTests/build_and_run_tests.sh"],
|
|
|
|
timeout_seconds=60 * 60,
|
|
|
|
shortname="ios-test-cfstream-tests",
|
|
|
|
cpu_cost=1e6,
|
|
|
|
environ=_FORCE_ENVIRON_FOR_WRAPPERS,
|
|
|
|
)
|
|
|
|
)
|
|
|
|
return sorted(out)
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return None
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "objc"
|
|
|
|
|
|
|
|
|
|
|
|
class Sanity(object):
|
|
|
|
def __init__(self, config_file):
|
|
|
|
self.config_file = config_file
|
|
|
|
|
|
|
|
def configure(self, config, args):
|
|
|
|
self.config = config
|
|
|
|
self.args = args
|
|
|
|
_check_compiler(self.args.compiler, ["default"])
|
|
|
|
|
|
|
|
def test_specs(self):
|
|
|
|
import yaml
|
|
|
|
|
|
|
|
with open("tools/run_tests/sanity/%s" % self.config_file, "r") as f:
|
|
|
|
environ = {"TEST": "true"}
|
|
|
|
if _is_use_docker_child():
|
|
|
|
environ["CLANG_FORMAT_SKIP_DOCKER"] = "true"
|
|
|
|
environ["CLANG_TIDY_SKIP_DOCKER"] = "true"
|
|
|
|
environ["IWYU_SKIP_DOCKER"] = "true"
|
|
|
|
# sanity tests run tools/bazel wrapper concurrently
|
|
|
|
# and that can result in a download/run race in the wrapper.
|
|
|
|
# under docker we already have the right version of bazel
|
|
|
|
# so we can just disable the wrapper.
|
|
|
|
environ["DISABLE_BAZEL_WRAPPER"] = "true"
|
|
|
|
return [
|
|
|
|
self.config.job_spec(
|
|
|
|
cmd["script"].split(),
|
|
|
|
timeout_seconds=45 * 60,
|
|
|
|
environ=environ,
|
|
|
|
cpu_cost=cmd.get("cpu_cost", 1),
|
|
|
|
)
|
|
|
|
for cmd in yaml.safe_load(f)
|
|
|
|
]
|
|
|
|
|
|
|
|
def pre_build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def build_steps_environ(self):
|
|
|
|
"""Extra environment variables set for pre_build_steps and build_steps jobs."""
|
|
|
|
return {}
|
|
|
|
|
|
|
|
def post_tests_steps(self):
|
|
|
|
return []
|
|
|
|
|
|
|
|
def dockerfile_dir(self):
|
|
|
|
return "tools/dockerfile/test/sanity"
|
|
|
|
|
|
|
|
def __str__(self):
|
|
|
|
return "sanity"
|
|
|
|
|
|
|
|
|
|
|
|
# different configurations we can run under
|
|
|
|
with open("tools/run_tests/generated/configs.json") as f:
|
|
|
|
_CONFIGS = dict(
|
|
|
|
(cfg["config"], Config(**cfg)) for cfg in ast.literal_eval(f.read())
|
|
|
|
)
|
|
|
|
|
|
|
|
_LANGUAGES = {
|
|
|
|
"c++": CLanguage("cxx", "c++"),
|
|
|
|
"c": CLanguage("c", "c"),
|
|
|
|
"php7": Php7Language(),
|
|
|
|
"python": PythonLanguage(),
|
|
|
|
"ruby": RubyLanguage(),
|
|
|
|
"csharp": CSharpLanguage(),
|
|
|
|
"objc": ObjCLanguage(),
|
|
|
|
"sanity": Sanity("sanity_tests.yaml"),
|
|
|
|
"clang-tidy": Sanity("clang_tidy_tests.yaml"),
|
|
|
|
}
|
|
|
|
|
|
|
|
_MSBUILD_CONFIG = {
|
|
|
|
"dbg": "Debug",
|
|
|
|
"opt": "Release",
|
|
|
|
"gcov": "Debug",
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
def _build_step_environ(cfg, extra_env={}):
|
|
|
|
"""Environment variables set for each build step."""
|
|
|
|
environ = {"CONFIG": cfg, "GRPC_RUN_TESTS_JOBS": str(args.jobs)}
|
|
|
|
msbuild_cfg = _MSBUILD_CONFIG.get(cfg)
|
|
|
|
if msbuild_cfg:
|
|
|
|
environ["MSBUILD_CONFIG"] = msbuild_cfg
|
|
|
|
environ.update(extra_env)
|
|
|
|
return environ
|
|
|
|
|
|
|
|
|
|
|
|
def _windows_arch_option(arch):
|
|
|
|
"""Returns msbuild cmdline option for selected architecture."""
|
|
|
|
if arch == "default" or arch == "x86":
|
|
|
|
return "/p:Platform=Win32"
|
|
|
|
elif arch == "x64":
|
|
|
|
return "/p:Platform=x64"
|
|
|
|
else:
|
|
|
|
print("Architecture %s not supported." % arch)
|
|
|
|
sys.exit(1)
|
|
|
|
|
|
|
|
|
|
|
|
def _check_arch_option(arch):
|
|
|
|
"""Checks that architecture option is valid."""
|
|
|
|
if platform_string() == "windows":
|
|
|
|
_windows_arch_option(arch)
|
|
|
|
elif platform_string() == "linux":
|
|
|
|
# On linux, we need to be running under docker with the right architecture.
|
|
|
|
runtime_machine = platform.machine()
|
|
|
|
runtime_arch = platform.architecture()[0]
|
|
|
|
if arch == "default":
|
|
|
|
return
|
|
|
|
elif (
|
|
|
|
runtime_machine == "x86_64"
|
|
|
|
and runtime_arch == "64bit"
|
|
|
|
and arch == "x64"
|
|
|
|
):
|
|
|
|
return
|
|
|
|
elif (
|
|
|
|
runtime_machine == "x86_64"
|
|
|
|
and runtime_arch == "32bit"
|
|
|
|
and arch == "x86"
|
|
|
|
):
|
|
|
|
return
|
|
|
|
elif (
|
|
|
|
runtime_machine == "aarch64"
|
|
|
|
and runtime_arch == "64bit"
|
|
|
|
and arch == "arm64"
|
|
|
|
):
|
|
|
|
return
|
|
|
|
else:
|
|
|
|
print(
|
|
|
|
"Architecture %s does not match current runtime architecture."
|
|
|
|
% arch
|
|
|
|
)
|
|
|
|
sys.exit(1)
|
|
|
|
else:
|
|
|
|
if args.arch != "default":
|
|
|
|
print(
|
|
|
|
"Architecture %s not supported on current platform." % args.arch
|
|
|
|
)
|
|
|
|
sys.exit(1)
|
|
|
|
|
|
|
|
|
|
|
|
def _docker_arch_suffix(arch):
|
|
|
|
"""Returns suffix to dockerfile dir to use."""
|
|
|
|
if arch == "default" or arch == "x64":
|
|
|
|
return "x64"
|
|
|
|
elif arch == "x86":
|
|
|
|
return "x86"
|
|
|
|
elif arch == "arm64":
|
|
|
|
return "arm64"
|
|
|
|
else:
|
|
|
|
print("Architecture %s not supported with current settings." % arch)
|
|
|
|
sys.exit(1)
|
|
|
|
|
|
|
|
|
|
|
|
def runs_per_test_type(arg_str):
|
|
|
|
"""Auxiliary function to parse the "runs_per_test" flag.
|
|
|
|
|
|
|
|
Returns:
|
|
|
|
A positive integer or 0, the latter indicating an infinite number of
|
|
|
|
runs.
|
|
|
|
|
|
|
|
Raises:
|
|
|
|
argparse.ArgumentTypeError: Upon invalid input.
|
|
|
|
"""
|
|
|
|
if arg_str == "inf":
|
|
|
|
return 0
|
|
|
|
try:
|
|
|
|
n = int(arg_str)
|
|
|
|
if n <= 0:
|
|
|
|
raise ValueError
|
|
|
|
return n
|
|
|
|
except:
|
|
|
|
msg = "'{}' is not a positive integer or 'inf'".format(arg_str)
|
|
|
|
raise argparse.ArgumentTypeError(msg)
|
|
|
|
|
|
|
|
|
|
|
|
def percent_type(arg_str):
|
|
|
|
pct = float(arg_str)
|
|
|
|
if pct > 100 or pct < 0:
|
|
|
|
raise argparse.ArgumentTypeError(
|
|
|
|
"'%f' is not a valid percentage in the [0, 100] range" % pct
|
|
|
|
)
|
|
|
|
return pct
|
|
|
|
|
|
|
|
|
|
|
|
# This is math.isclose in python >= 3.5
|
|
|
|
def isclose(a, b, rel_tol=1e-09, abs_tol=0.0):
|
|
|
|
return abs(a - b) <= max(rel_tol * max(abs(a), abs(b)), abs_tol)
|
|
|
|
|
|
|
|
|
|
|
|
def _shut_down_legacy_server(legacy_server_port):
|
|
|
|
"""Shut down legacy version of port server."""
|
|
|
|
try:
|
|
|
|
version = int(
|
|
|
|
urllib.request.urlopen(
|
|
|
|
"http://localhost:%d/version_number" % legacy_server_port,
|
|
|
|
timeout=10,
|
|
|
|
).read()
|
|
|
|
)
|
|
|
|
except:
|
|
|
|
pass
|
|
|
|
else:
|
|
|
|
urllib.request.urlopen(
|
|
|
|
"http://localhost:%d/quitquitquit" % legacy_server_port
|
|
|
|
).read()
|
|
|
|
|
|
|
|
|
|
|
|
def _calculate_num_runs_failures(list_of_results):
|
|
|
|
"""Calculate number of runs and failures for a particular test.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
list_of_results: (List) of JobResult object.
|
|
|
|
Returns:
|
|
|
|
A tuple of total number of runs and failures.
|
|
|
|
"""
|
|
|
|
num_runs = len(list_of_results) # By default, there is 1 run per JobResult.
|
|
|
|
num_failures = 0
|
|
|
|
for jobresult in list_of_results:
|
|
|
|
if jobresult.retries > 0:
|
|
|
|
num_runs += jobresult.retries
|
|
|
|
if jobresult.num_failures > 0:
|
|
|
|
num_failures += jobresult.num_failures
|
|
|
|
return num_runs, num_failures
|
|
|
|
|
|
|
|
|
|
|
|
class BuildAndRunError(object):
|
|
|
|
"""Represents error type in _build_and_run."""
|
|
|
|
|
|
|
|
BUILD = object()
|
|
|
|
TEST = object()
|
|
|
|
POST_TEST = object()
|
|
|
|
|
|
|
|
|
|
|
|
# returns a list of things that failed (or an empty list on success)
|
|
|
|
def _build_and_run(
|
|
|
|
check_cancelled, newline_on_success, xml_report=None, build_only=False
|
|
|
|
):
|
|
|
|
"""Do one pass of building & running tests."""
|
|
|
|
# build latest sequentially
|
|
|
|
num_failures, resultset = jobset.run(
|
|
|
|
build_steps,
|
|
|
|
maxjobs=1,
|
|
|
|
stop_on_failure=True,
|
|
|
|
newline_on_success=newline_on_success,
|
|
|
|
travis=args.travis,
|
|
|
|
)
|
|
|
|
if num_failures:
|
|
|
|
return [BuildAndRunError.BUILD]
|
|
|
|
|
|
|
|
if build_only:
|
|
|
|
if xml_report:
|
|
|
|
report_utils.render_junit_xml_report(
|
|
|
|
resultset, xml_report, suite_name=args.report_suite_name
|
|
|
|
)
|
|
|
|
return []
|
|
|
|
|
|
|
|
# start antagonists
|
|
|
|
antagonists = [
|
|
|
|
subprocess.Popen(["tools/run_tests/python_utils/antagonist.py"])
|
|
|
|
for _ in range(0, args.antagonists)
|
|
|
|
]
|
|
|
|
start_port_server.start_port_server()
|
|
|
|
resultset = None
|
|
|
|
num_test_failures = 0
|
|
|
|
try:
|
|
|
|
infinite_runs = runs_per_test == 0
|
|
|
|
one_run = set(
|
|
|
|
spec
|
|
|
|
for language in languages
|
|
|
|
for spec in language.test_specs()
|
|
|
|
if (
|
|
|
|
re.search(args.regex, spec.shortname)
|
|
|
|
and (
|
|
|
|
args.regex_exclude == ""
|
|
|
|
or not re.search(args.regex_exclude, spec.shortname)
|
|
|
|
)
|
|
|
|
)
|
|
|
|
)
|
|
|
|
# When running on travis, we want out test runs to be as similar as possible
|
|
|
|
# for reproducibility purposes.
|
|
|
|
if args.travis and args.max_time <= 0:
|
|
|
|
massaged_one_run = sorted(one_run, key=lambda x: x.cpu_cost)
|
|
|
|
else:
|
|
|
|
# whereas otherwise, we want to shuffle things up to give all tests a
|
|
|
|
# chance to run.
|
|
|
|
massaged_one_run = list(
|
|
|
|
one_run
|
|
|
|
) # random.sample needs an indexable seq.
|
|
|
|
num_jobs = len(massaged_one_run)
|
|
|
|
# for a random sample, get as many as indicated by the 'sample_percent'
|
|
|
|
# argument. By default this arg is 100, resulting in a shuffle of all
|
|
|
|
# jobs.
|
|
|
|
sample_size = int(num_jobs * args.sample_percent / 100.0)
|
|
|
|
massaged_one_run = random.sample(massaged_one_run, sample_size)
|
|
|
|
if not isclose(args.sample_percent, 100.0):
|
|
|
|
assert (
|
|
|
|
args.runs_per_test == 1
|
|
|
|
), "Can't do sampling (-p) over multiple runs (-n)."
|
|
|
|
print(
|
|
|
|
"Running %d tests out of %d (~%d%%)"
|
|
|
|
% (sample_size, num_jobs, args.sample_percent)
|
|
|
|
)
|
|
|
|
if infinite_runs:
|
|
|
|
assert (
|
|
|
|
len(massaged_one_run) > 0
|
|
|
|
), "Must have at least one test for a -n inf run"
|
|
|
|
runs_sequence = (
|
|
|
|
itertools.repeat(massaged_one_run)
|
|
|
|
if infinite_runs
|
|
|
|
else itertools.repeat(massaged_one_run, runs_per_test)
|
|
|
|
)
|
|
|
|
all_runs = itertools.chain.from_iterable(runs_sequence)
|
|
|
|
|
|
|
|
if args.quiet_success:
|
|
|
|
jobset.message(
|
|
|
|
"START",
|
|
|
|
"Running tests quietly, only failing tests will be reported",
|
|
|
|
do_newline=True,
|
|
|
|
)
|
|
|
|
num_test_failures, resultset = jobset.run(
|
|
|
|
all_runs,
|
|
|
|
check_cancelled,
|
|
|
|
newline_on_success=newline_on_success,
|
|
|
|
travis=args.travis,
|
|
|
|
maxjobs=args.jobs,
|
|
|
|
maxjobs_cpu_agnostic=max_parallel_tests_for_current_platform(),
|
|
|
|
stop_on_failure=args.stop_on_failure,
|
|
|
|
quiet_success=args.quiet_success,
|
|
|
|
max_time=args.max_time,
|
|
|
|
)
|
|
|
|
if resultset:
|
|
|
|
for k, v in sorted(resultset.items()):
|
|
|
|
num_runs, num_failures = _calculate_num_runs_failures(v)
|
|
|
|
if num_failures > 0:
|
|
|
|
if num_failures == num_runs: # what about infinite_runs???
|
|
|
|
jobset.message("FAILED", k, do_newline=True)
|
|
|
|
else:
|
|
|
|
jobset.message(
|
|
|
|
"FLAKE",
|
|
|
|
"%s [%d/%d runs flaked]"
|
|
|
|
% (k, num_failures, num_runs),
|
|
|
|
do_newline=True,
|
|
|
|
)
|
|
|
|
finally:
|
|
|
|
for antagonist in antagonists:
|
|
|
|
antagonist.kill()
|
|
|
|
if args.bq_result_table and resultset:
|
|
|
|
upload_extra_fields = {
|
|
|
|
"compiler": args.compiler,
|
|
|
|
"config": args.config,
|
|
|
|
"iomgr_platform": args.iomgr_platform,
|
|
|
|
"language": args.language[
|
|
|
|
0
|
|
|
|
], # args.language is a list but will always have one element when uploading to BQ is enabled.
|
|
|
|
"platform": platform_string(),
|
|
|
|
}
|
|
|
|
try:
|
|
|
|
upload_results_to_bq(
|
|
|
|
resultset, args.bq_result_table, upload_extra_fields
|
|
|
|
)
|
|
|
|
except NameError as e:
|
|
|
|
logging.warning(
|
|
|
|
e
|
|
|
|
) # It's fine to ignore since this is not critical
|
|
|
|
if xml_report and resultset:
|
|
|
|
report_utils.render_junit_xml_report(
|
|
|
|
resultset,
|
|
|
|
xml_report,
|
|
|
|
suite_name=args.report_suite_name,
|
|
|
|
multi_target=args.report_multi_target,
|
|
|
|
)
|
|
|
|
|
|
|
|
number_failures, _ = jobset.run(
|
|
|
|
post_tests_steps,
|
|
|
|
maxjobs=1,
|
|
|
|
stop_on_failure=False,
|
|
|
|
newline_on_success=newline_on_success,
|
|
|
|
travis=args.travis,
|
|
|
|
)
|
|
|
|
|
|
|
|
out = []
|
|
|
|
if number_failures:
|
|
|
|
out.append(BuildAndRunError.POST_TEST)
|
|
|
|
if num_test_failures:
|
|
|
|
out.append(BuildAndRunError.TEST)
|
|
|
|
|
|
|
|
return out
|
|
|
|
|
|
|
|
|
|
|
|
# parse command line
|
|
|
|
argp = argparse.ArgumentParser(description="Run grpc tests.")
|
|
|
|
argp.add_argument(
|
|
|
|
"-c", "--config", choices=sorted(_CONFIGS.keys()), default="opt"
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"-n",
|
|
|
|
"--runs_per_test",
|
|
|
|
default=1,
|
|
|
|
type=runs_per_test_type,
|
|
|
|
help=(
|
|
|
|
'A positive integer or "inf". If "inf", all tests will run in an '
|
|
|
|
'infinite loop. Especially useful in combination with "-f"'
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument("-r", "--regex", default=".*", type=str)
|
|
|
|
argp.add_argument("--regex_exclude", default="", type=str)
|
|
|
|
argp.add_argument("-j", "--jobs", default=multiprocessing.cpu_count(), type=int)
|
|
|
|
argp.add_argument("-s", "--slowdown", default=1.0, type=float)
|
|
|
|
argp.add_argument(
|
|
|
|
"-p",
|
|
|
|
"--sample_percent",
|
|
|
|
default=100.0,
|
|
|
|
type=percent_type,
|
|
|
|
help="Run a random sample with that percentage of tests",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"-t",
|
|
|
|
"--travis",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help=(
|
|
|
|
"When set, indicates that the script is running on CI (= not locally)."
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--newline_on_success", default=False, action="store_const", const=True
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"-l",
|
|
|
|
"--language",
|
|
|
|
choices=sorted(_LANGUAGES.keys()),
|
|
|
|
nargs="+",
|
|
|
|
required=True,
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"-S", "--stop_on_failure", default=False, action="store_const", const=True
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--use_docker",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help="Run all the tests under docker. That provides "
|
|
|
|
+ "additional isolation and prevents the need to install "
|
|
|
|
+ "language specific prerequisites. Only available on Linux.",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--allow_flakes",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help=(
|
|
|
|
"Allow flaky tests to show as passing (re-runs failed tests up to five"
|
|
|
|
" times)"
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--arch",
|
|
|
|
choices=["default", "x86", "x64", "arm64"],
|
|
|
|
default="default",
|
|
|
|
help=(
|
|
|
|
'Selects architecture to target. For some platforms "default" is the'
|
|
|
|
" only supported choice."
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--compiler",
|
|
|
|
choices=[
|
|
|
|
"default",
|
|
|
|
"gcc8",
|
|
|
|
"gcc10.2",
|
|
|
|
"gcc10.2_openssl102",
|
|
|
|
"gcc10.2_openssl111",
|
|
|
|
"gcc12",
|
|
|
|
"gcc12_openssl309",
|
|
|
|
"gcc_musl",
|
|
|
|
"clang6",
|
|
|
|
"clang17",
|
|
|
|
# TODO: Automatically populate from supported version
|
|
|
|
"python3.7",
|
|
|
|
"python3.8",
|
|
|
|
"python3.9",
|
|
|
|
"python3.10",
|
|
|
|
"python3.11",
|
|
|
|
"python3.12",
|
|
|
|
"pypy",
|
|
|
|
"pypy3",
|
|
|
|
"python_alpine",
|
|
|
|
"all_the_cpythons",
|
|
|
|
"coreclr",
|
|
|
|
"cmake",
|
|
|
|
"cmake_ninja_vs2019",
|
|
|
|
"cmake_vs2019",
|
|
|
|
"cmake_vs2022",
|
|
|
|
"mono",
|
|
|
|
],
|
|
|
|
default="default",
|
|
|
|
help=(
|
|
|
|
"Selects compiler to use. Allowed values depend on the platform and"
|
|
|
|
" language."
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--iomgr_platform",
|
|
|
|
choices=["native", "gevent", "asyncio"],
|
|
|
|
default="native",
|
|
|
|
help="Selects iomgr platform to build on",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--build_only",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help="Perform all the build steps but don't run any tests.",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--measure_cpu_costs",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help="Measure the cpu costs of tests",
|
|
|
|
)
|
|
|
|
argp.add_argument("-a", "--antagonists", default=0, type=int)
|
|
|
|
argp.add_argument(
|
|
|
|
"-x",
|
|
|
|
"--xml_report",
|
|
|
|
default=None,
|
|
|
|
type=str,
|
|
|
|
help="Generates a JUnit-compatible XML report",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--report_suite_name",
|
|
|
|
default="tests",
|
|
|
|
type=str,
|
|
|
|
help="Test suite name to use in generated JUnit XML report",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--report_multi_target",
|
|
|
|
default=False,
|
|
|
|
const=True,
|
|
|
|
action="store_const",
|
|
|
|
help=(
|
|
|
|
"Generate separate XML report for each test job (Looks better in UIs)."
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--quiet_success",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help=(
|
|
|
|
"Don't print anything when a test passes. Passing tests also will not"
|
|
|
|
" be reported in XML report. "
|
|
|
|
)
|
|
|
|
+ "Useful when running many iterations of each test (argument -n).",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--force_default_poller",
|
|
|
|
default=False,
|
|
|
|
action="store_const",
|
|
|
|
const=True,
|
|
|
|
help="Don't try to iterate over many polling strategies when they exist",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--force_use_pollers",
|
|
|
|
default=None,
|
|
|
|
type=str,
|
|
|
|
help=(
|
|
|
|
"Only use the specified comma-delimited list of polling engines. "
|
|
|
|
"Example: --force_use_pollers epoll1,poll "
|
|
|
|
" (This flag has no effect if --force_default_poller flag is also used)"
|
|
|
|
),
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--max_time", default=-1, type=int, help="Maximum test runtime in seconds"
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--bq_result_table",
|
|
|
|
default="",
|
|
|
|
type=str,
|
|
|
|
nargs="?",
|
|
|
|
help="Upload test results to a specified BQ table.",
|
|
|
|
)
|
|
|
|
argp.add_argument(
|
|
|
|
"--cmake_configure_extra_args",
|
|
|
|
default=[],
|
|
|
|
nargs="+",
|
|
|
|
help="Extra arguments that will be passed to the cmake configure command. Only works for C/C++.",
|
|
|
|
)
|
|
|
|
args = argp.parse_args()
|
|
|
|
|
|
|
|
flaky_tests = set()
|
|
|
|
shortname_to_cpu = {}
|
|
|
|
|
|
|
|
if args.force_default_poller:
|
|
|
|
_POLLING_STRATEGIES = {}
|
|
|
|
elif args.force_use_pollers:
|
|
|
|
_POLLING_STRATEGIES[platform_string()] = args.force_use_pollers.split(",")
|
|
|
|
|
|
|
|
jobset.measure_cpu_costs = args.measure_cpu_costs
|
|
|
|
|
|
|
|
# grab config
|
|
|
|
run_config = _CONFIGS[args.config]
|
|
|
|
build_config = run_config.build_config
|
|
|
|
|
|
|
|
languages = set(_LANGUAGES[l] for l in args.language)
|
|
|
|
for l in languages:
|
|
|
|
l.configure(run_config, args)
|
|
|
|
|
|
|
|
if len(languages) != 1:
|
|
|
|
print("Building multiple languages simultaneously is not supported!")
|
|
|
|
sys.exit(1)
|
|
|
|
|
|
|
|
# If --use_docker was used, respawn the run_tests.py script under a docker container
|
|
|
|
# instead of continuing.
|
|
|
|
if args.use_docker:
|
|
|
|
if not args.travis:
|
|
|
|
print("Seen --use_docker flag, will run tests under docker.")
|
|
|
|
print("")
|
|
|
|
print(
|
|
|
|
"IMPORTANT: The changes you are testing need to be locally"
|
|
|
|
" committed"
|
|
|
|
)
|
|
|
|
print(
|
|
|
|
"because only the committed changes in the current branch will be"
|
|
|
|
)
|
|
|
|
print("copied to the docker environment.")
|
|
|
|
time.sleep(5)
|
|
|
|
|
|
|
|
dockerfile_dirs = set([l.dockerfile_dir() for l in languages])
|
|
|
|
if len(dockerfile_dirs) > 1:
|
|
|
|
print(
|
|
|
|
"Languages to be tested require running under different docker "
|
|
|
|
"images."
|
|
|
|
)
|
|
|
|
sys.exit(1)
|
|
|
|
else:
|
|
|
|
dockerfile_dir = next(iter(dockerfile_dirs))
|
|
|
|
|
|
|
|
child_argv = [arg for arg in sys.argv if not arg == "--use_docker"]
|
|
|
|
run_tests_cmd = "python3 tools/run_tests/run_tests.py %s" % " ".join(
|
|
|
|
child_argv[1:]
|
|
|
|
)
|
|
|
|
|
|
|
|
env = os.environ.copy()
|
|
|
|
env["DOCKERFILE_DIR"] = dockerfile_dir
|
|
|
|
env["DOCKER_RUN_SCRIPT"] = "tools/run_tests/dockerize/docker_run.sh"
|
|
|
|
env["DOCKER_RUN_SCRIPT_COMMAND"] = run_tests_cmd
|
|
|
|
|
|
|
|
retcode = subprocess.call(
|
|
|
|
"tools/run_tests/dockerize/build_and_run_docker.sh", shell=True, env=env
|
|
|
|
)
|
|
|
|
_print_debug_info_epilogue(dockerfile_dir=dockerfile_dir)
|
|
|
|
sys.exit(retcode)
|
|
|
|
|
|
|
|
_check_arch_option(args.arch)
|
|
|
|
|
|
|
|
# collect pre-build steps (which get retried if they fail, e.g. to avoid
|
|
|
|
# flakes on downloading dependencies etc.)
|
|
|
|
build_steps = list(
|
|
|
|
set(
|
|
|
|
jobset.JobSpec(
|
|
|
|
cmdline,
|
|
|
|
environ=_build_step_environ(
|
|
|
|
build_config, extra_env=l.build_steps_environ()
|
|
|
|
),
|
|
|
|
timeout_seconds=_PRE_BUILD_STEP_TIMEOUT_SECONDS,
|
|
|
|
flake_retries=2,
|
|
|
|
)
|
|
|
|
for l in languages
|
|
|
|
for cmdline in l.pre_build_steps()
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
# collect build steps
|
|
|
|
build_steps.extend(
|
|
|
|
set(
|
|
|
|
jobset.JobSpec(
|
|
|
|
cmdline,
|
|
|
|
environ=_build_step_environ(
|
|
|
|
build_config, extra_env=l.build_steps_environ()
|
|
|
|
),
|
|
|
|
timeout_seconds=None,
|
|
|
|
)
|
|
|
|
for l in languages
|
|
|
|
for cmdline in l.build_steps()
|
|
|
|
)
|
|
|
|
)
|
|
|
|
|
|
|
|
# collect post test steps
|
|
|
|
post_tests_steps = list(
|
|
|
|
set(
|
|
|
|
jobset.JobSpec(
|
|
|
|
cmdline,
|
|
|
|
environ=_build_step_environ(
|
|
|
|
build_config, extra_env=l.build_steps_environ()
|
|
|
|
),
|
|
|
|
)
|
|
|
|
for l in languages
|
|
|
|
for cmdline in l.post_tests_steps()
|
|
|
|
)
|
|
|
|
)
|
|
|
|
runs_per_test = args.runs_per_test
|
|
|
|
|
|
|
|
errors = _build_and_run(
|
|
|
|
check_cancelled=lambda: False,
|
|
|
|
newline_on_success=args.newline_on_success,
|
|
|
|
xml_report=args.xml_report,
|
|
|
|
build_only=args.build_only,
|
|
|
|
)
|
|
|
|
if not errors:
|
|
|
|
jobset.message("SUCCESS", "All tests passed", do_newline=True)
|
|
|
|
else:
|
|
|
|
jobset.message("FAILED", "Some tests failed", do_newline=True)
|
|
|
|
|
|
|
|
if not _is_use_docker_child():
|
|
|
|
# if --use_docker was used, the outer invocation of run_tests.py will
|
|
|
|
# print the debug info instead.
|
|
|
|
_print_debug_info_epilogue()
|
|
|
|
|
|
|
|
exit_code = 0
|
|
|
|
if BuildAndRunError.BUILD in errors:
|
|
|
|
exit_code |= 1
|
|
|
|
if BuildAndRunError.TEST in errors:
|
|
|
|
exit_code |= 2
|
|
|
|
if BuildAndRunError.POST_TEST in errors:
|
|
|
|
exit_code |= 4
|
|
|
|
sys.exit(exit_code)
|