From 426a565908e47f791938baa3a876a87de306b436 Mon Sep 17 00:00:00 2001 From: Mashiro <57566630+HAOCHENYE@users.noreply.github.com> Date: Thu, 25 May 2023 16:22:17 +0800 Subject: [PATCH] Fix unit test error due to program abort caused by limited resources (#1162) * Fix unit test error due to program abort cause by limited resources * clean the code --- mmengine/testing/_internal/distributed.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/mmengine/testing/_internal/distributed.py b/mmengine/testing/_internal/distributed.py index 20e3b601..df10c008 100644 --- a/mmengine/testing/_internal/distributed.py +++ b/mmengine/testing/_internal/distributed.py @@ -1,10 +1,10 @@ # Copyright (c) OpenMMLab. All rights reserved. # Copyright (c) https://github.com/pytorch/pytorch # Modified from https://github.com/pytorch/pytorch/blob/master/torch/testing/_internal/common_distributed.py # noqa: E501 - import faulthandler import logging import multiprocessing +import signal import sys import tempfile import threading @@ -339,6 +339,9 @@ class MultiProcessTestCase(TestCase): raise RuntimeError( f'Process {i} terminated or timed out after ' '{elapsed_time} seconds') + if p.exitcode == signal.SIGABRT: + self.skipTest(f'Skip test {self._testMethodName} due to ' + 'the program abort') self.assertEqual( p.exitcode, first_process.exitcode,