import faulthandler import logging import multiprocessing import os import queue import subprocess import sys import tempfile import threading import time import traceback import types import unittest from contextlib import contextmanager from dataclasses import dataclass from datetime import timedelta from enum import Enum from functools import partial, reduce, wraps from io import StringIO from typing import Dict, NamedTuple, Optional, Union from unittest.mock import patch import torch import torch._dynamo.test_case import torch.cuda.nccl import torch.distributed as c10d import torch.nn as nn from torch.testing._internal.common_utils import ( FILE_SCHEMA, find_free_port, IS_SANDCASTLE, retry_on_connect_failures, sandcastle_skip, sandcastle_skip_if, TEST_WITH_ROCM, TEST_WITH_TSAN, TestCase, ) from torch.testing._internal.distributed.multi_threaded_pg import ( _install_threaded_pg, _uninstall_threaded_pg, ProcessLocalGroup, ) logging.basicConfig(level=logging.INFO) logger = logging.getLogger(__name__) class TestSkip(NamedTuple): exit_code: int message: str TEST_SKIPS = { "backend_unavailable": TestSkip( 72, "Skipped because distributed backend is not available." ), "small_worldsize": TestSkip(73, "Skipped due to small world size."), "odd_worldsize": TestSkip(87, "Skipped due to odd world size."), "no_cuda": TestSkip(74, "CUDA is not available."), "multi-gpu-1": TestSkip(75, "Need at least 1 CUDA device"), "multi-gpu-2": TestSkip(77, "Need at least 2 CUDA devices"), "multi-gpu-3": TestSkip(80, "Need at least 3 CUDA devices"), "multi-gpu-4": TestSkip(81, "Need at least 4 CUDA devices"), "multi-gpu-5": TestSkip(82, "Need at least 5 CUDA devices"), "multi-gpu-6": TestSkip(83, "Need at least 6 CUDA devices"), "multi-gpu-7": TestSkip(84, "Need at least 7 CUDA devices"), "multi-gpu-8": TestSkip(85, "Need at least 8 CUDA devices"), "nccl": TestSkip(76, "c10d not compiled with NCCL support"), "skipIfRocm": TestSkip(78, "Test skipped for ROCm"), "no_peer_access": TestSkip(79, "Test skipped because no GPU peer access"), "generic": TestSkip( 86, "Test skipped at subprocess level, look at subprocess log for skip reason" ), "importerror": TestSkip(88, "Test skipped due to missing import"), } @dataclass class DistTestCases: # Backends that do not support a specific collective skip_collective = {} skip_collective["allgather_coalesced"] = {"nccl", "mpi", "ucc"} skip_collective["reduce"] = set() skip_collective["sendrecv anysource"] = {"nccl", "ucc"} skip_collective["cpu barrier"] = {"nccl", "ucc"} # Sets showing that something is implemented backend_feature = {} backend_feature["gpu"] = {"nccl", "gloo", "ucc"} backend_feature["cuda"] = {"nccl", "gloo", "ucc"} backend_feature["ddp"] = {"nccl", "gloo", "ucc"} backend_feature["subgroup"] = {"nccl", "gloo", "ucc"} backend_feature["plugin"] = set() def skip_if_no_gpu(func): """Skips if the world size exceeds the number of GPUs, ensuring that if the test is run, each rank has its own GPU via ``torch.cuda.device(rank)``.""" @wraps(func) def wrapper(*args, **kwargs): if not torch.cuda.is_available(): sys.exit(TEST_SKIPS["no_cuda"].exit_code) world_size = int(os.environ["WORLD_SIZE"]) if torch.cuda.device_count() < world_size: sys.exit(TEST_SKIPS[f"multi-gpu-{world_size}"].exit_code) return func(*args, **kwargs) return wrapper def skip_if_small_worldsize(func): @wraps(func) def wrapper(*args, **kwargs): if (os.environ["BACKEND"] != "mpi") and int(os.environ["WORLD_SIZE"]) <= 2: sys.exit(TEST_SKIPS["small_worldsize"].exit_code) return func(*args, **kwargs) return wrapper def skip_if_odd_worldsize(func): @wraps(func) def wrapper(*args, **kwargs): if (os.environ["BACKEND"] != "mpi") and int(os.environ["WORLD_SIZE"]) % 2 == 1: sys.exit(TEST_SKIPS["odd_worldsize"].exit_code) return func(*args, **kwargs) return wrapper def require_n_gpus_for_nccl_backend(n, backend): def decorator(func): @wraps(func) def wrapper(*args, **kwargs): if backend == "nccl" and torch.cuda.device_count() < n: sys.exit(TEST_SKIPS[f"multi-gpu-{n}"].exit_code) else: return func(*args, **kwargs) return wrapper return decorator def import_transformers_or_skip(): def decorator(func): @wraps(func) def wrapper(*args, **kwargs): try: from transformers import ( # noqa: Unused AutoModelForMaskedLM, BertConfig, ) return func(*args, **kwargs) except ImportError: sys.exit(TEST_SKIPS["importerror"].exit_code) return wrapper return decorator def skip_if_lt_x_gpu(x): def decorator(func): @wraps(func) def wrapper(*args, **kwargs): if torch.cuda.is_available() and torch.cuda.device_count() >= x: return func(*args, **kwargs) sys.exit(TEST_SKIPS[f"multi-gpu-{x}"].exit_code) return wrapper return decorator # This decorator helps avoiding initializing cuda while testing other backends def nccl_skip_if_lt_x_gpu(backend, x): def decorator(func): @wraps(func) def wrapper(*args, **kwargs): if backend != "nccl": return func(*args, **kwargs) if torch.cuda.is_available() and torch.cuda.device_count() >= x: return func(*args, **kwargs) sys.exit(TEST_SKIPS[f"multi-gpu-{x}"].exit_code) return wrapper return decorator def verify_ddp_error_logged(model_DDP, err_substr): # Verify error was logged in ddp_logging_data. ddp_logging_data = model_DDP._get_ddp_logging_data() assert "iteration" in ddp_logging_data assert "has_error" in ddp_logging_data assert "error" in ddp_logging_data logging_err = ddp_logging_data["error"] # Remove C++ stacktrace if needed. actual = ( err_substr if err_substr.find("\nException raised from ") == -1 else err_substr.split("\nException raised from ")[0] ) assert ( actual in logging_err ), f"Did not find expected {actual} in ddp logging data error: {logging_err}" def with_nccl_blocking_wait(func): """ Convenience decorator to set/unset NCCL_BLOCKING_WAIT flag. Note that use of this decorator will override the setting of NCCL_ASYNC_ERROR_HANDLING for the particular test. After the test, both NCCL_BLOCKING_WAIT and NCCL_ASYNC_ERROR_HANDLING will be restored to their original values. """ @wraps(func) def wrapper(*args, **kwargs): # Save and unset NCCL_ASYNC_ERROR_HANDLING try: cached_nccl_async_error_handling: Union[str, None] = os.environ[ "NCCL_ASYNC_ERROR_HANDLING" ] del os.environ["NCCL_ASYNC_ERROR_HANDLING"] except KeyError: # NCCL_ASYNC_ERROR_HANDLING was unset cached_nccl_async_error_handling = None # Save val of NCCL_BLOCKING_WAIT and set it. try: cached_nccl_blocking_wait: Union[str, None] = os.environ[ "NCCL_BLOCKING_WAIT" ] except KeyError: cached_nccl_blocking_wait = None finally: os.environ["NCCL_BLOCKING_WAIT"] = "1" try: ret = func(*args, **kwargs) return ret finally: # restore old values. if cached_nccl_async_error_handling is not None: os.environ[ "NCCL_ASYNC_ERROR_HANDLING" ] = cached_nccl_async_error_handling if cached_nccl_blocking_wait is not None: os.environ["NCCL_BLOCKING_WAIT"] = cached_nccl_blocking_wait return wrapper def with_dist_debug_levels(levels): """ Runs a test for each distributed debug level specified in levels. """ def decorator(func): @wraps(func) def wrapper(*args, **kwargs): old_level = os.environ.get("TORCH_DISTRIBUTED_DEBUG", None) for level in levels: os.environ["TORCH_DISTRIBUTED_DEBUG"] = level c10d.set_debug_level_from_env() ret = func(*args, **kwargs) c10d.barrier() if old_level is not None: os.environ["TORCH_DISTRIBUTED_DEBUG"] = old_level # Only returns test return for last test, but since these are # unittests the return value is not really used and earlier tests # would've raised had they failed. return ret return wrapper return decorator def requires_gloo(): return sandcastle_skip_if( not c10d.is_gloo_available(), "c10d was not compiled with the Gloo backend", ) def requires_nccl_version(version, msg): if not c10d.is_nccl_available(): return sandcastle_skip( "c10d was not compiled with the NCCL backend", ) else: return sandcastle_skip_if( torch.cuda.nccl.version() < version, "Requires NCCL version greater than or equal to: {}, found: {}, reason: {}".format( version, torch.cuda.nccl.version(), msg ), ) def requires_nccl(): return sandcastle_skip_if( not c10d.is_nccl_available(), "c10d was not compiled with the NCCL backend", ) def requires_ucc(): return sandcastle_skip_if( not c10d.is_ucc_available(), "c10d was not compiled with the UCC backend", ) def requires_mpi(): return sandcastle_skip_if( not c10d.is_mpi_available(), "c10d was not compiled with the MPI backend", ) def skip_if_rocm(func): """Skips a test for ROCm""" func.skip_if_rocm = True @wraps(func) def wrapper(*args, **kwargs): if not TEST_WITH_ROCM: return func(*args, **kwargs) sys.exit(TEST_SKIPS["skipIfRocm"].exit_code) return wrapper def skip_if_win32(): return sandcastle_skip_if( sys.platform == "win32", "This unit test case is not supported on Windows platform", ) @retry_on_connect_failures def create_tcp_store( addr="localhost", world_size=1, is_master=True, timeout=timedelta(minutes=5), wait_for_workers=True, jit_class=False, ): """ Creates a TCP store. Retries if the chosen port is already in use. """ port = find_free_port() if jit_class: timeout_millisecond = int(timeout / timedelta(milliseconds=1)) return torch.classes.dist_c10d.TCPStore( addr, port, world_size, is_master, timeout_millisecond ) else: return c10d.TCPStore( addr, port, world_size, is_master, wait_for_workers=wait_for_workers ) if TEST_WITH_TSAN: # TSAN runs much slower. TIMEOUT_DEFAULT = 500 else: TIMEOUT_DEFAULT = int(os.getenv('DISTRIBUTED_TESTS_DEFAULT_TIMEOUT', '300')) TIMEOUT_OVERRIDE = {"test_ddp_uneven_inputs": 400} # https://github.com/pytorch/pytorch/issues/75665 if TEST_WITH_ROCM: TIMEOUT_OVERRIDE["test_join_kwargs"] = 200 def create_device(interface=None): if sys.platform == "win32" or interface is None: return c10d.ProcessGroupGloo.create_device(hostname="127.0.0.1") else: return c10d.ProcessGroupGloo.create_device(interface=interface) def get_timeout(test_id) -> int: return TIMEOUT_OVERRIDE.get(test_id.split(".")[-1], TIMEOUT_DEFAULT) @contextmanager def captured_output(): new_out, new_err = StringIO(), StringIO() old_out, old_err = sys.stdout, sys.stderr try: sys.stdout, sys.stderr = new_out, new_err yield sys.stdout, sys.stderr finally: sys.stdout, sys.stderr = old_out, old_err def simple_sparse_reduce_tests(rank: int, world_size: int, num_inputs: int = 1): """ Generate a number of basic test cases for sparse reduction. These cover tensors with a varying number of sparse dimensions and a varying number of dense dimensions. The only reduction operation we support is sum. """ def generate(rank: int, world_size: int, sparse_dims: int = 1, dense_dims: int = 0): # First sparse dimension is [0..rank]. # Subsequent dimensions are always 0, so we know there is # a non-empty intersection between any two sparse tensors. indices = torch.reshape(torch.arange(rank + 1), (1, rank + 1)) shape = [world_size] + [2 for _ in range(dense_dims)] for _ in range(sparse_dims - 1): indices = torch.cat((indices, torch.zeros(1, rank + 1))) shape.append(world_size) values = torch.ones([rank + 1] + [2 for _ in range(dense_dims)]) return torch.sparse_coo_tensor(indices, values, shape) def compute_sum(fn, world_size: int): return reduce( lambda a, b: a + b, [fn(rank, world_size) for rank in range(world_size)] ) return [ ( [ fn(num_inputs * rank + i, num_inputs * world_size) for i in range(num_inputs) ], [compute_sum(fn, num_inputs * world_size) for i in range(num_inputs)], ) for fn in [ partial(generate, sparse_dims=1), partial(generate, sparse_dims=2), partial(generate, sparse_dims=3), partial(generate, dense_dims=1), partial(generate, dense_dims=2), partial(generate, dense_dims=3), ] ] # HELPER FOR MULTIGPU TESTS def init_multigpu_helper(world_size: int, backend: str): """Multigpu tests are designed to simulate the multi nodes with multi GPUs on each node. Nccl backend requires equal #GPUs in each process. On a single node, all visible GPUs are evenly divided to subsets, each process only uses a subset. """ nGPUs = torch.cuda.device_count() visible_devices = range(nGPUs) if backend == "nccl": # This is a hack for a known NCCL issue using multiprocess # in conjunction with multiple threads to manage different GPUs which # may cause ncclCommInitRank to fail. # http://docs.nvidia.com/deeplearning/sdk/nccl-release-notes/rel_2.1.4.html#rel_2.1.4 # It slows down the performance of collective operations. # Without this setting NCCL might throw unhandled error. os.environ["NCCL_MAX_NRINGS"] = "1" # If rank is less than or equal to number of available GPU's # then each rank can be mapped to corresponding GPU. nGPUs_per_process = 1 if world_size > nGPUs: nGPUs_per_process = nGPUs // world_size rank_to_GPU = { i: list(visible_devices[i * nGPUs_per_process : (i + 1) * nGPUs_per_process]) for i in range(world_size) } return rank_to_GPU tmp_dir: Optional[tempfile.TemporaryDirectory] = None def initialize_temp_directories(init_method: Optional[str] = None) -> None: global tmp_dir tmp_dir = tempfile.TemporaryDirectory() os.environ["TEMP_DIR"] = tmp_dir.name os.mkdir(os.path.join(tmp_dir.name, "barrier")) os.mkdir(os.path.join(tmp_dir.name, "test_dir")) init_dir_path = os.path.join(tmp_dir.name, "init_dir") os.mkdir(init_dir_path) # Set init method if specified. if init_method is not None: os.environ["INIT_METHOD"] = init_method else: os.environ["INIT_METHOD"] = FILE_SCHEMA + os.path.join( init_dir_path, "shared_init_file" ) def cleanup_temp_dir() -> None: if tmp_dir is not None: tmp_dir.cleanup() # Most tests operate with this worldsize DEFAULT_WORLD_SIZE = 4 # [How does MultiProcessTestCase work?] # Each MultiProcessTestCase instance uses 1 + `world_size()` processes, by # default `world_size()` returns 4. Let's take `test_rpc_spawn.py` as an # example which inherits from this class. Its `Setup()` methods calls into # `MultiProcessTestCase._spawn_processes()` which spawns `world_size()` # subprocesses. During the spawn, the main process passes the test name to # subprocesses, and the name is acquired from self.id(). The subprocesses # then use the provided test function name to retrieve the function attribute # from the test instance and run it. The main process simply waits for all # subprocesses to join. class MultiProcessTestCase(TestCase): MAIN_PROCESS_RANK = -1 # This exit code is used to indicate that the test code had an error and # exited abnormally. There are certain tests that might use sys.exit() to # simulate failures and in those cases, we can't have an exit code of 0, # but we still want to ensure we didn't run into any other errors. TEST_ERROR_EXIT_CODE = 10 # do not early terminate for distributed tests. def _should_stop_test_suite(self) -> bool: return False @property def world_size(self) -> int: return DEFAULT_WORLD_SIZE def join_or_run(self, fn): @wraps(fn) def wrapper(self): if self.rank == self.MAIN_PROCESS_RANK: self._join_processes(fn) else: fn() return types.MethodType(wrapper, self) # The main process spawns N subprocesses that run the test. # Constructor patches current instance test method to # assume the role of the main process and join its subprocesses, # or run the underlying test function. def __init__(self, method_name: str = "runTest") -> None: super().__init__(method_name) fn = getattr(self, method_name) setattr(self, method_name, self.join_or_run(fn)) def setUp(self) -> None: super().setUp() self.skip_return_code_checks = [] # type: ignore[var-annotated] self.processes = [] # type: ignore[var-annotated] self.rank = self.MAIN_PROCESS_RANK self.file_name = tempfile.NamedTemporaryFile(delete=False).name # pid to pipe consisting of error message from process. self.pid_to_pipe = {} # type: ignore[var-annotated] def tearDown(self) -> None: super().tearDown() for p in self.processes: p.terminate() # Each Process instance holds a few open file descriptors. The unittest # runner creates a new TestCase instance for each test method and keeps # it alive until the end of the entire suite. We must thus reset the # processes to prevent an effective file descriptor leak. self.processes = [] def _current_test_name(self) -> str: # self.id() == e.g. '__main__.TestDistributed.TestAdditive.test_get_rank' return self.id().split(".")[-1] def _start_processes(self, proc) -> None: self.processes = [] for rank in range(int(self.world_size)): parent_conn, child_conn = torch.multiprocessing.Pipe() process = proc( target=self.__class__._run, name="process " + str(rank), args=(rank, self._current_test_name(), self.file_name, child_conn), ) process.start() logger.info(f"Started process {rank} with pid {process.pid}") self.pid_to_pipe[process.pid] = parent_conn self.processes.append(process) def _spawn_processes(self) -> None: proc = torch.multiprocessing.get_context("spawn").Process self._start_processes(proc) class Event(Enum): GET_TRACEBACK = 1 @staticmethod def _event_listener(parent_pipe, signal_pipe, rank: int): logger.info(f"Starting event listener thread for rank {rank}") while True: ready_pipes = multiprocessing.connection.wait([parent_pipe, signal_pipe]) if parent_pipe in ready_pipes: if parent_pipe.closed: logger.info( f"Pipe closed for process {rank}, stopping event listener thread" ) return event = parent_pipe.recv() logger.info(f"Received event {event} on process {rank}") if event == MultiProcessTestCase.Event.GET_TRACEBACK: # Return traceback to the parent process. with tempfile.NamedTemporaryFile(mode="r+") as tmp_file: faulthandler.dump_traceback(tmp_file) # Flush buffers and seek to read from the beginning tmp_file.flush() tmp_file.seek(0) parent_pipe.send(tmp_file.read()) logger.info(f"Process {rank} sent traceback") if signal_pipe in ready_pipes: return @classmethod def _run(cls, rank: int, test_name: str, file_name: str, parent_pipe) -> None: # Enable DDP + ReplicatedTensor from torch.nn.parallel._replicated_tensor_ddp_utils import ( _set_ddp_with_replicated_tensor, ) _set_ddp_with_replicated_tensor(True) self = cls(test_name) self.rank = rank self.file_name = file_name self.run_test(test_name, parent_pipe) def run_test(self, test_name: str, parent_pipe) -> None: # Start event listener thread. signal_recv_pipe, signal_send_pipe = torch.multiprocessing.Pipe(duplex=False) event_listener_thread = threading.Thread( target=MultiProcessTestCase._event_listener, args=(parent_pipe, signal_recv_pipe, self.rank), daemon=True, ) event_listener_thread.start() if sys.platform != "win32" and sys.platform != "darwin": # Register signal handler to dump stack traces on FATALs. # Windows and MacOS do not support the signal handlers. torch._C._set_print_stack_traces_on_fatal_signal(True) # Show full C++ stacktraces when a Python error originating from C++ is raised. os.environ["TORCH_SHOW_CPP_STACKTRACES"] = "1" # self.id() == e.g. '__main__.TestDistributed.test_get_rank' # We're retrieving a corresponding test and executing it. try: getattr(self, test_name)() except unittest.SkipTest as se: logger.info( f"Process {self.rank} skipping test {test_name} for following reason: {str(se)}" ) sys.exit(TEST_SKIPS["generic"].exit_code) except Exception as e: logger.error( f"Caught exception: \n{traceback.format_exc()} exiting " f"process {self.rank} with exit code: {MultiProcessTestCase.TEST_ERROR_EXIT_CODE}" ) # Send error to parent process. parent_pipe.send(traceback.format_exc()) sys.exit(MultiProcessTestCase.TEST_ERROR_EXIT_CODE) finally: if signal_send_pipe is not None: signal_send_pipe.send(None) assert event_listener_thread is not None event_listener_thread.join() # Close pipe after done with test. parent_pipe.close() def _get_timedout_process_traceback(self) -> None: pipes = [] for i, process in enumerate(self.processes): if process.exitcode is None: pipe = self.pid_to_pipe[process.pid] try: pipe.send(MultiProcessTestCase.Event.GET_TRACEBACK) pipes.append((i, pipe)) except ConnectionError as e: logger.error( f"Encountered error while trying to get traceback for process {i}: {e}" ) # Wait for results. for rank, pipe in pipes: try: # Wait for traceback if pipe.poll(5): if pipe.closed: logger.info( f"Pipe closed for process {rank}, cannot retrieve traceback" ) continue traceback = pipe.recv() logger.error( f"Process {rank} timed out with traceback: \n\n{traceback}" ) else: logger.error( f"Could not retrieve traceback for timed out process: {rank}" ) except ConnectionError as e: logger.error( f"Encountered error while trying to get traceback for process {rank}: {e}" ) def _join_processes(self, fn) -> None: timeout = get_timeout(self.id()) start_time = time.time() subprocess_error = False try: while True: # check to see if any subprocess exited with an error early. for (i, p) in enumerate(self.processes): # This is the exit code processes exit with if they # encountered an exception. if p.exitcode == MultiProcessTestCase.TEST_ERROR_EXIT_CODE: print( f"Process {i} terminated with exit code {p.exitcode}, terminating remaining processes." ) active_children = torch.multiprocessing.active_children() for ac in active_children: ac.terminate() subprocess_error = True break if subprocess_error: break # All processes have joined cleanly if they all a valid exitcode if all([p.exitcode is not None for p in self.processes]): break # Check if we should time out the test. If so, we terminate each process. elapsed = time.time() - start_time if elapsed > timeout: self._get_timedout_process_traceback() print( f"Timing out after {timeout} seconds and killing subprocesses." ) for p in self.processes: p.terminate() break # Sleep to avoid excessive busy polling. time.sleep(0.1) elapsed_time = time.time() - start_time if fn in self.skip_return_code_checks: self._check_no_test_errors(elapsed_time) else: self._check_return_codes(elapsed_time) finally: # Close all pipes for pid, pipe in self.pid_to_pipe.items(): pipe.close() def _check_no_test_errors(self, elapsed_time) -> None: """ Checks that we didn't have any errors thrown in the child processes. """ for i, p in enumerate(self.processes): if p.exitcode is None: raise RuntimeError( "Process {} timed out after {} seconds".format(i, elapsed_time) ) self.assertNotEqual(self.TEST_ERROR_EXIT_CODE, p.exitcode) def _check_return_codes(self, elapsed_time) -> None: """ Checks that the return codes of all spawned processes match, and skips tests if they returned a return code indicating a skipping condition. """ first_process = self.processes[0] # first, we check if there are errors in actual processes # (via TEST_ERROR_EXIT CODE), and raise an exception for those. # the reason we do this is to attempt to raise a more helpful error # message than "Process x terminated/timed out" # TODO: we should pipe the exception of the failed subprocess here. # Currently, the actual exception is displayed as a logging output. errored_processes = [ (i, p) for i, p in enumerate(self.processes) if p.exitcode == MultiProcessTestCase.TEST_ERROR_EXIT_CODE ] if errored_processes: error = "" for i, process in errored_processes: # Get error from pipe. error_message = self.pid_to_pipe[process.pid].recv() error += ( "Process {} exited with error code {} and exception:\n{}\n".format( i, MultiProcessTestCase.TEST_ERROR_EXIT_CODE, error_message ) ) raise RuntimeError(error) # If no process exited uncleanly, we check for timeouts, and then ensure # each process exited cleanly. for i, p in enumerate(self.processes): if p.exitcode is None: raise RuntimeError( "Process {} terminated or timed out after {} seconds".format( i, elapsed_time ) ) self.assertEqual( p.exitcode, first_process.exitcode, msg="Expect process {} exit code to match Process 0 exit code of {}, but got {}".format( i, first_process.exitcode, p.exitcode ), ) for skip in TEST_SKIPS.values(): if first_process.exitcode == skip.exit_code: if IS_SANDCASTLE: # Don't use unittest.skip to skip the test on sandcastle # since it creates tasks for skipped tests assuming there # is some follow-up needed. Instead just "pass" the test # with an appropriate message. logger.info( f"Skipping {self.id()} on sandcastle for the following reason: {skip.message}" ) return else: raise unittest.SkipTest(skip.message) self.assertEqual( first_process.exitcode, 0, msg="Expected zero exit code but got {} for pid: {}".format( first_process.exitcode, first_process.pid ), ) @property def is_master(self) -> bool: return self.rank == 0 # Cannot use functools.cache as it requires python 3.9 EFA_PROBE_RESULT = None def has_efa() -> bool: """ If shell command `fi_info -p efa -t FI_EP_RDM` returns exit code 0 then we assume that the machine has Libfabric EFA interfaces and EFA software components installed, see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/efa-start.html. """ global EFA_PROBE_RESULT if EFA_PROBE_RESULT is not None: return EFA_PROBE_RESULT try: EFA_PROBE_RESULT = ( subprocess.run(["fi_info", "-p", "efa", "-t", "FI_EP_RDM"]).returncode == 0 ) except FileNotFoundError: EFA_PROBE_RESULT = False return EFA_PROBE_RESULT def tp_transports(): """ If the machine has Libfabric EFA interfaces and EFA software components installed it may cause 'RuntimeError: In operator() at tensorpipe/common/ibv.h:172 "": Operation not supported' if tensorpipe uses InfiniBand transport, so we exclude it from tensorpipe transports, see https://github.com/pytorch/pytorch/issues/73885 and https://github.com/pytorch/pytorch/issues/65022 """ return ["shm", "uv"] if has_efa() else None def spawn_threads_and_init_comms( func=None, timeout=TIMEOUT_DEFAULT, world_size=DEFAULT_WORLD_SIZE ): """ Wrapper to use with a test method """ if func is None: return partial( spawn_threads_and_init_comms, timeout=timeout, world_size=world_size ) def _run_test_method_with_multi_threads(world_size, callback): world = _install_threaded_pg() global_store = c10d.HashStore() def world_is_valid(): return world == c10d.distributed_c10d._world def worker(rank, world_pg, store): c10d.init_process_group( backend="threaded", rank=rank, world_size=world_size, store=store ) try: callback() except BaseException as ex: # Exceptions are handled in MultiThreadedTestCase MultiThreadedTestCase.exception_queue.put((rank, sys.exc_info())) ProcessLocalGroup.exception_handle(ex) # trigger _terminate event and awaken worker threads finally: if world_is_valid(): c10d.destroy_process_group() threads = [] for rank in range(world_size): t = threading.Thread(target=worker, args=(rank, world, global_store)) t.start() threads.append(t) return threads @wraps(func) def wrapper(self, *args, **kwargs): # TODO: get test name from kwargs threads = _run_test_method_with_multi_threads(world_size, lambda: func(self, *args, **kwargs)) # join and error handling MultiThreadedTestCase._join_threads(threads, func) return wrapper class MultiThreadedTestCase(TestCase): """ Test runner that runs all tests with the in-proc process group using multiple threads with the threaded process group. Each test spawns world_size threads and run the test method in each thread. Difference from regular MultiProcess test runner: Must explicitly defines SetUp and call self._spawn_threads() to run the tests. Cannot use setUp / tearDown (must use perThreadSetup / perThreadShutdown) to set up / tear down each thread when running each test. No global state possible How bad of a limitation is this? """ exception_queue = queue.Queue() MAIN_THREAD_RANK = -1 def join_or_run(self, fn): @wraps(fn) def wrapper(self): if self.rank == self.MAIN_THREAD_RANK: self._join_threads(self.threads, fn) else: fn() return types.MethodType(wrapper, self) def __init__(self, method_name: str = "runTest") -> None: super().__init__(method_name) test_fn = getattr(self, method_name, None) setattr(self, method_name, self.join_or_run(test_fn)) def perThreadSetUp(self): # super().setUp() # TestCase.setUp() calls torch.manual_seed() pass def perThreadTearDown(self): pass def setUp(self) -> None: """ setUp only set up things in the main thread, if you want to configure things in the spawned threads, use perThreadSetUp """ super().setUp() self.rank = self.MAIN_THREAD_RANK self.threads = [] # Show full C++ stacktraces when a Python error originating from C++ is raised. os.environ["TORCH_SHOW_CPP_STACKTRACES"] = "1" def tearDown(self): """ tearDown only set up things in the main thread, if you want to configure things in the spawned threads, use perThreadTearDown """ super().tearDown() self.threads = [] def _spawn_threads(self): """ class method to spawn threads and run test, use this method in the SetUp of your TestCase """ test_name = self._current_test_name # for each test case, we need to create thread local world, and a global store world = _install_threaded_pg() self.__class__.global_store = c10d.HashStore() def world_is_valid(): return world == c10d.distributed_c10d._world if not world_is_valid(): raise RuntimeError("Invalid world") for rank in range(self.world_size): t = threading.Thread(target=self.__class__._run, args=(test_name, rank, self.world_size)) t.start() self.threads.append(t) @classmethod def _run(cls, test_name, rank, world_size): self = cls(test_name) self.rank = rank # precision/rel_tol is a thread-local setting since it may be overridden per test, need to make # every thread have the same value. This would be relevant when we use op db tests, where it # needs those states to be set i.e. using instantiate_device_type_tests() # TODO: figure out a better way to do this if hasattr(self, "_tls"): self._tls = threading.local() self._tls.precision = TestCase._precision self._tls.rel_tol = TestCase._rel_tol self.run_test_with_threaded_pg(test_name, rank, world_size) def run_test_with_threaded_pg(self, test_name, rank, world_size): """ Run the current test associated with `test_name` using the threaded process group. """ c10d.init_process_group( backend="threaded", rank=rank, world_size=world_size, store=self.__class__.global_store ) self.perThreadSetUp() try: getattr(self, test_name)() except BaseException as ex: self.exception_queue.put((rank, sys.exc_info())) ProcessLocalGroup.exception_handle(ex) # trigger _terminate event and awaken worker threads finally: c10d.destroy_process_group() self.perThreadTearDown() @classmethod def _join_threads(cls, threads, fn): timeout = TIMEOUT_DEFAULT try: for idx, thread in enumerate(threads): thread.join(max(0, timeout)) if thread.is_alive(): MultiThreadedTestCase.exception_queue.put( ( idx, ( TimeoutError, TimeoutError( f"Rank failed to join in under {timeout} seconds" ), None, ), ) ) ProcessLocalGroup.reset() failed_ranks = [] while not cls.exception_queue.empty(): failure = cls.exception_queue.get() failed_ranks.append(failure) finally: _uninstall_threaded_pg() cls._check_return_codes(failed_ranks, timeout, fn) @classmethod def _check_return_codes(cls, failed_ranks, timeout, fn): # Print based on exceptions raised from threads # SkipTest: print info for each thread # TimeoutError: raise RuntimeError for any timed out thread # Normal Exception: print error for each thread that raises exception # and raise a RuntimeError error_msg = "" skip_code = -1 for rank, exc_info in failed_ranks: exc = exc_info[1] if isinstance(exc, unittest.SkipTest): logger.info( f"Thread {rank} skipping test {fn} for following reason: {str(exc)}" ) if skip_code < 0: skip_code = TEST_SKIPS["generic"].exit_code elif isinstance(exc, TimeoutError): msg = "Thread {} terminated or timed out after {} seconds\n".format( rank, timeout ) logger.error(msg) raise RuntimeError(msg) elif isinstance(exc, Exception): msg = "".join(traceback.format_exception(*exc_info)) logger.error( f"Caught exception: \n{msg} exiting thread {rank}" ) error_msg += ( "Thread {} exited with exception:\n{}\n".format(rank, msg) ) elif isinstance(exc, SystemExit): if type(exc.code) == int and skip_code < 0: skip_code = exc.code # check exceptions if len(error_msg) > 0: raise RuntimeError(error_msg) # check skip if skip_code > 0: for skip in TEST_SKIPS.values(): if skip_code == skip.exit_code: if IS_SANDCASTLE: # "pass" the test with an appropriate message. logger.info( f"Skipping {fn} on sandcastle for the following reason: {skip.message}" ) return else: raise unittest.SkipTest(skip.message) @property def world_size(self) -> int: return DEFAULT_WORLD_SIZE @property def _current_test_name(self) -> str: # self.id() == e.g. '__main__.TestDistributed.TestAdditive.test_get_rank' return self.id().split(".")[-1] def assertEqualOnRank(self, x, y, msg=None, *, rank=0): """ The reason why we have this util function instead of self.assertEqual is all threads are sharing one CPU RNG so the assertion result is only reliable on rank 0 """ if self.rank == rank: self.assertEqual(x, y, msg) def assertNotEqualOnRank(self, x, y, msg=None, *, rank=0): if self.rank == rank: self.assertNotEqual(x, y) class SaveForwardInputsModule(nn.Module): def __init__( self, forward_inputs: Dict[nn.Module, torch.Tensor], cast_forward_inputs: bool, ) -> None: super().__init__() self.l = nn.Linear(100, 100) self.forward_inputs = forward_inputs self.cast_forward_inputs = cast_forward_inputs def forward(self, x: torch.Tensor) -> torch.Tensor: self.forward_inputs[self] = x return self.l(x.to(self.l.weight.dtype) if self.cast_forward_inputs else x) class SaveForwardInputsModel(nn.Module): def __init__( self, forward_inputs: Dict[nn.Module, torch.Tensor], cast_forward_inputs: bool, ) -> None: super().__init__() self.c1 = SaveForwardInputsModule(forward_inputs, cast_forward_inputs) self.c2 = SaveForwardInputsModule(forward_inputs, cast_forward_inputs) self.forward_inputs = forward_inputs def forward(self, x: torch.Tensor) -> torch.Tensor: self.forward_inputs[self] = x return self.c2(self.c1(x)) @contextmanager def _dynamo_dist_per_rank_init(rank, world_size, init_pg=True): # To avoid multiple inheritance from _dynamo.test_case.TestCase and MultiProcessTestCase, # Just manually implement the most important part of the dynamo behavior to reset/clear. torch.cuda.set_device(rank) os.environ['MASTER_ADDR'] = 'localhost' os.environ['MASTER_PORT'] = '6789' if init_pg: c10d.init_process_group("nccl", rank=rank, world_size=world_size) torch._dynamo.reset() torch._dynamo.utils.counters.clear() yield torch._dynamo.reset() torch._dynamo.utils.counters.clear() if init_pg: c10d.destroy_process_group() class DynamoDistributedSingleProcTestCase(torch._dynamo.test_case.TestCase): """ Test harness for single-process dynamo distributed tests, initializes dist process group. Prefer this for simple tests, as it's easier to debug. """ @classmethod def setUpClass(cls): super().setUpClass() # _exit_stack is set up in TestCase cls._exit_stack.enter_context( patch.dict( os.environ, { "MASTER_ADDR": "localhost", "MASTER_PORT": "12355", }, ) ) cls.rank = 0 cls.device = f"cuda:{cls.rank}" cls.device_ids = None if "cuda" in cls.device else [cls.rank] c10d.init_process_group("nccl", rank=cls.rank, world_size=1) @classmethod def tearDownClass(cls): c10d.destroy_process_group() super().tearDownClass() class DynamoDistributedMultiProcTestCase(MultiProcessTestCase): """ Use this for tests that actually run on multiple GPUs. Decorate tests with @skip_if_lt_x_gpu(ngpu) Note: MultiProcTestCase spawns processes per test and is slow. Prefer MultiThreadedTestCase for most tests. Perhaps use this one sparingly for integration tests. """ def setUp(self): super().setUp() self._spawn_processes() def tearDown(self): super().tearDown() try: os.remove(self.file_name) except OSError: pass @property def world_size(self) -> int: return torch.cuda.device_count() @classmethod def _run(cls, rank: int, test_name: str, file_name: str, parent_pipe) -> None: # Don't enable DDP + ReplicatedTensor, as that breaks Dynamo+DDP # TODO(whc) why is ReplicatedTensor defaulted=True in MultiProcessTestCase, and should we support it? # from torch.nn.parallel._replicated_tensor_ddp_utils import _set_ddp_with_replicated_tensor # _set_ddp_with_replicated_tensor(True) # The rest is copypasta from MultiProcessTestCase._run self = cls(test_name) self.rank = rank self.file_name = file_name self.run_test(test_name, parent_pipe)