Created by: juncaipeng

disable test_kldiv_loss_op for now, and the log as follows:

40/82 Test #343: test_kldiv_loss_op .........................***Failed   16.76 sec
[02:57:07]	[Step 1/1] W1214 02:56:55.244370 13619 init.cc:159] AVX is available, Please re-compile on local machine
[02:57:07]	[Step 1/1] W1214 02:56:57.910257 13619 device_context.cc:236] Please NOTE: device: 0, CUDA Capability: 61, Driver API Version: 10.1, Runtime API Version: 9.0
[02:57:07]	[Step 1/1] W1214 02:56:57.914772 13619 device_context.cc:244] device: 0, cuDNN Version: 7.6.
[02:57:07]	[Step 1/1] WARNING: Do not have avx core. You may not build with AVX, but AVX is supported on local machine.
[02:57:07]	[Step 1/1]  You could build paddle WITH_AVX=ON to get better performance.
[02:57:07]	[Step 1/1] The original error is: No module named 'paddle.fluid.core_avx'
[02:57:07]	[Step 1/1] /paddle/build/python/paddle/fluid/tests/unittests/test_kldiv_loss_op.py:22: RuntimeWarning: invalid value encountered in log
[02:57:07]	[Step 1/1]   output = target * (np.log(target) - x)
[02:57:07]	[Step 1/1] test_kldiv_loss_op failed
[02:57:07]	[Step 1/1]  F.......
[02:57:07]	[Step 1/1] ======================================================================
[02:57:07]	[Step 1/1] FAIL: test_check_grad (test_kldiv_loss_op.TestKLDivLossOp)
[02:57:07]	[Step 1/1] ----------------------------------------------------------------------
[02:57:07]	[Step 1/1] Traceback (most recent call last):
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/test_kldiv_loss_op.py", line 56, in test_check_grad
[02:57:07]	[Step 1/1]     ['X'], 'Loss', no_grad_set=set(["Target"]), max_relative_error=0.06)
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/op_test.py", line 1417, in check_grad
[02:57:07]	[Step 1/1]     user_defined_grads, check_dygraph)
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/op_test.py", line 1085, in check_grad
[02:57:07]	[Step 1/1]     user_defined_grads, check_dygraph)
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/op_test.py", line 1433, in check_grad_with_place
[02:57:07]	[Step 1/1]     user_defined_grads, check_dygraph)
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/op_test.py", line 1134, in check_grad_with_place
[02:57:07]	[Step 1/1]     "Gradient Check On %s" % str(place))
[02:57:07]	[Step 1/1]   File "/paddle/build/python/paddle/fluid/tests/unittests/op_test.py", line 1069, in _assert_is_close
[02:57:07]	[Step 1/1]     self.assertLessEqual(max_diff, max_relative_error, err_msg())
[02:57:07]	[Step 1/1] AssertionError: 1.0 not less than or equal to 0.06 : Gradient Check On CUDAPlace(0) Variable X max gradient diff 1.000000 over limit 0.060000, the first error element is 2, expected -0.001526, but got 0.000000
[02:57:07]	[Step 1/1] 
[02:57:07]	[Step 1/1] ----------------------------------------------------------------------
[02:57:07]	[Step 1/1] Ran 8 tests in 9.022s
[02:57:07]	[Step 1/1] 
[02:57:07]	[Step 1/1] FAILED (failures=1)