Flaky Test Issue of GPU Operator

[zhanghang1989]

Description

kernel_error_check_imperative() and kernel_error_check_symbolic() in test_operator_gpu.py have flaky issues.

test_operator_gpu.test_kernel_error_checking ... Process SpawnProcess-1:

Traceback (most recent call last):

  File "/usr/lib/python3.5/multiprocessing/process.py", line 249, in _bootstrap

    self.run()

  File "/usr/lib/python3.5/multiprocessing/process.py", line 93, in run

    self._target(*self._args, **self._kwargs)

  File "/work/mxnet/tests/python/gpu/test_operator_gpu.py", line 1832, in kernel_error_check_imperative

    c = (a / b).asnumpy()

  File "/work/mxnet/tests/python/unittest/../../../python/mxnet/ndarray/ndarray.py", line 1910, in asnumpy

    ctypes.c_size_t(data.size)))

  File "/work/mxnet/tests/python/unittest/../../../python/mxnet/base.py", line 210, in check_call

    raise MXNetError(py_str(_LIB.MXGetLastError()))

mxnet.base.MXNetError: [19:32:53] src/operator/tensor/././../mxnet_op.h:586: Check failed: err == cudaSuccess (9 vs. 0) Name: mxnet_generic_kernel_ex ErrStr:invalid configuration argument

[marcoabreu]

Hi, would you mind looking a CI run where this occurred?

[zhanghang1989]

This fails at "/work/mxnet/tests/python/gpu/test_operator_gpu.py", line 1832, in kernel_error_check_imperative http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-11502/25/pipeline/799

[haojin2]

I cannot reproduce this error on my local machine, are you sure that this failure is not related to your PR? Seems like this is happening to other environments for the same build as well.

[zhanghang1989]

I am closing this issue, since I don't have this issue any more. My PR just passed CI test.

[marcoabreu]

Agree with @haojin2 , the error seems to be related to the PR. Thanks everybody!

[szha]

I just had the exact same error in another PR. http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-11482/17/pipeline/859#step-1530-log-1018

@zhanghang1989 what did you do to resolve the problem?

[ChaiBapchya]

I had the exact same error too. Any pointers to the same?

http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-12749/28/pipeline/1003/#step-1612-log-1562

[ChaiBapchya]

@szha Can we reopen this issue? Your latest PR on Notice #14043 has the same CI issue.