Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)" by ezyang · Pull Request #44554 · pytorch/pytorch

ezyang · 2020-09-11T15:33:51Z

Stack from ghstack:

Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)" #44554 Revert "Allow Tensor-likes in torch.autograd.gradcheck (Allow Tensor-likes in torch.autograd.gradcheck #43877)"

This reverts commit f9a0d0c.

This reverts commit f9a0d0c. [ghstack-poisoned]

This reverts commit f9a0d0c. ghstack-source-id: d662f56 Pull Request resolved: #44554

dr-ci · 2020-09-11T16:01:42Z

💊 CI failures summary and remediations

As of commit 7c6878f (more details on the Dr. CI page):

14/16 failures possibly* introduced in this PR
- 1/14 non-CircleCI failure(s)
1/16 tentatively recognized as flaky ❄️
- Click here to rerun these jobs
1/16 broken upstream at merge base 77cc7d1 since Sep 11

🕵️ 13 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

pytorch_windows_vs2019_py36_cuda10.1_test1 (1/13)

Step: "Test" (full log | diagnosis details | 🔁 rerun)

RuntimeError: test_jit_profiling failed!

Generated XML report: test-reports\python-unittest\TEST-jit.test_tracer.TestTracer-20200911160101.xml 
Generated XML report: test-reports\python-unittest\TEST-jit.test_type_sharing.TestTypeSharing-20200911160101.xml 
Generated XML report: test-reports\python-unittest\TEST-jit.test_unsupported_ops.TestUnsupportedOps-20200911160101.xml 
Generated XML report: test-reports\python-unittest\TEST-jit.test_with.TestWith-20200911160101.xml 
Generated XML report: test-reports\python-unittest\TEST-jit.test_data_parallel.TestDataParallel-20200911160101.xml 
Traceback (most recent call last): 
  File "run_test.py", line 740, in <module> 
    main() 
  File "run_test.py", line 723, in main 
    raise RuntimeError(err_message) 
RuntimeError: test_jit_profiling failed! 
+ cleanup
+ retcode=1
+ set +x

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_legacy_test (2/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:11:19 ERROR [0.009s]: test_trace_tensor_factory (jit.test_tracer.TestTracer)

Sep 11 16:11:19     self.checkTrace(stuff, (example, example[0] + 1)) 
Sep 11 16:11:19   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:11:19     allow_unused=allow_unused) 
Sep 11 16:11:19   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:11:19     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs)) 
Sep 11 16:11:19   File "/opt/conda/lib/python3.6/site-packages/torch/tensor.py", line 563, in __len__ 
Sep 11 16:11:19     raise TypeError("len() of a 0-d tensor") 
Sep 11 16:11:19 TypeError: len() of a 0-d tensor 
Sep 11 16:11:19  
Sep 11 16:11:19 ====================================================================== 
Sep 11 16:11:19 ERROR [0.009s]: test_trace_tensor_factory (jit.test_tracer.TestTracer) 
Sep 11 16:11:19 ---------------------------------------------------------------------- 
Sep 11 16:11:19 Traceback (most recent call last): 
Sep 11 16:11:19   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 531, in test_trace_tensor_factory 
Sep 11 16:11:19     run() 
Sep 11 16:11:19   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 527, in run 
Sep 11 16:11:19     self.checkTrace(fn, (input,), inputs_require_grads=inputs_require_grads) 
Sep 11 16:11:19   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:11:19     allow_unused=allow_unused) 
Sep 11 16:11:19   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:11:19     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs))

pytorch_linux_xenial_py3_clang5_asan_test2 (3/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:10:24 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in

Sep 11 16:10:24     #7 0x55c649af07eb in PyEval_EvalCode /tmp/build/80754af9/python_1588903631989/work/Python/ceval.c:731 
Sep 11 16:10:24     #8 0x55c649b70e73 in run_mod /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:1025 
Sep 11 16:10:24     #9 0x55c649b70f0c in PyRun_StringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:949 
Sep 11 16:10:24     #10 0x55c649b70f6e in PyRun_SimpleStringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:445 
Sep 11 16:10:24     #11 0x55c649b74d72 in run_command /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:301 
Sep 11 16:10:24     #12 0x55c649b74d72 in Py_Main /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:749 
Sep 11 16:10:24     #13 0x55c649a3ef2d in main /tmp/build/80754af9/python_1588903631989/work/Programs/python.c:69 
Sep 11 16:10:24     #14 0x7fac6bf2483f in __libc_start_main /build/glibc-e6zv40/glibc-2.23/csu/../csu/libc-start.c:291 
Sep 11 16:10:24     #15 0x55c649b1e27e in _start /home/rdonnelly/mc/conda-bld/compilers_linux-64_1534865402226/work/.build/src/glibc-2.12.2/csu/../sysdeps/x86_64/elf/start.S:103 
Sep 11 16:10:24  
Sep 11 16:10:24 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in  
Sep 11 16:10:24 + retcode=1 
Sep 11 16:10:24 + set -e 
Sep 11 16:10:24 + return 1 
Sep 11 16:10:24 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *-NO_AVX-* ]] 
Sep 11 16:10:24 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *-NO_AVX2-* ]] 
Sep 11 16:10:24 + '[' -n https://github.com/pytorch/pytorch/pull/44554 ']' 
Sep 11 16:10:24 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 != *coverage* ]] 
Sep 11 16:10:24 ++ mktemp 
Sep 11 16:10:24 + DETERMINE_FROM=/tmp/tmp.CgKMmdFlCq 
Sep 11 16:10:24 + file_diff_from_base /tmp/tmp.CgKMmdFlCq

pytorch_macos_10_13_py3_test (4/13)

Step: "Test" (full log | diagnosis details | 🔁 rerun)

Sep 11 09:02:13 RuntimeError: test_autograd failed! Received signal: SIGSEGV

Sep 11 09:02:13   test_nansum_scalar_keepdim_dim_cpu (__main__.TestAutogradDeviceTypeCPU) ... ERROR (0.012s) 
Sep 11 09:02:13   test_nansum_scalar_keepdim_dim_neg0_cpu (__main__.TestAutogradDeviceTypeCPU) ... ERROR (0.012s) 
Sep 11 09:02:13   test_narrow_dim_cpu (__main__.TestAutogradDeviceTypeCPU) ... ERROR (0.012s) 
Sep 11 09:02:13   test_narrow_dim_neg0_cpu (__main__.TestAutogradDeviceTypeCPU) ... ERROR (0.012s) 
Sep 11 09:02:13   test_narrow_empty_dim_cpu (__main__.TestAutogradDeviceTypeCPU) ... ERROR (0.034s) 
Sep 11 09:02:13   test_narrow_empty_dim_neg0_cpu (__main__.TestAutogradDeviceTypeCPU) ... Traceback (most recent call last): 
Sep 11 09:02:13   File "test/run_test.py", line 740, in <module> 
Sep 11 09:02:13     main() 
Sep 11 09:02:13   File "test/run_test.py", line 723, in main 
Sep 11 09:02:13     raise RuntimeError(err_message) 
Sep 11 09:02:13 RuntimeError: test_autograd failed! Received signal: SIGSEGV 
Sep 11 09:02:13 + cleanup 
Sep 11 09:02:13 + retcode=1 
Sep 11 09:02:13 + set +x

pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_profiling_test (5/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:26:50 ERROR [0.007s]: test_trace_tensor_factory (jit.test_tracer.TestTracer)

Sep 11 16:26:50     self.checkTrace(stuff, (example, example[0] + 1)) 
Sep 11 16:26:50   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:26:50     allow_unused=allow_unused) 
Sep 11 16:26:50   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:26:50     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs)) 
Sep 11 16:26:50   File "/opt/conda/lib/python3.6/site-packages/torch/tensor.py", line 563, in __len__ 
Sep 11 16:26:50     raise TypeError("len() of a 0-d tensor") 
Sep 11 16:26:50 TypeError: len() of a 0-d tensor 
Sep 11 16:26:50  
Sep 11 16:26:50 ====================================================================== 
Sep 11 16:26:50 ERROR [0.007s]: test_trace_tensor_factory (jit.test_tracer.TestTracer) 
Sep 11 16:26:50 ---------------------------------------------------------------------- 
Sep 11 16:26:50 Traceback (most recent call last): 
Sep 11 16:26:50   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 531, in test_trace_tensor_factory 
Sep 11 16:26:50     run() 
Sep 11 16:26:50   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 527, in run 
Sep 11 16:26:50     self.checkTrace(fn, (input,), inputs_require_grads=inputs_require_grads) 
Sep 11 16:26:50   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:26:50     allow_unused=allow_unused) 
Sep 11 16:26:50   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:26:50     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs))

pytorch_xla_linux_bionic_py3_6_clang9_test (6/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 17:09:39 ERROR [0.054s]: test_polygamma_xla_float64 (__main__.TestDevicePrecisionXLA)

Sep 11 17:09:37   test_multidevice_serialization_xla (__main__.TestDevicePrecisionXLA) ... skip (0.002s) 
Sep 11 17:09:37   test_polygamma_xla_float64 (__main__.TestDevicePrecisionXLA) ... ERROR (0.054s) 
Sep 11 17:09:37   test_solve_methods_arg_device_xla (__main__.TestDevicePrecisionXLA) ... skip (0.004s) 
Sep 11 17:09:37   test_sum_cpu_device_mismatch_xla (__main__.TestDevicePrecisionXLA) ... skip (0.002s) 
Sep 11 17:09:38   test_sum_noncontig_xla_float64 (__main__.TestDevicePrecisionXLA) ... ok (0.994s) 
Sep 11 17:09:38   test_type_conversions_same_device_xla (__main__.TestDevicePrecisionXLA) ... skip (0.003s) 
Sep 11 17:09:39   test_var_large_input_xla (__main__.TestDevicePrecisionXLA) ... ok (0.424s) 
Sep 11 17:09:39   test_var_xla (__main__.TestDevicePrecisionXLA) ... ok (0.122s) 
Sep 11 17:09:39  
Sep 11 17:09:39 ====================================================================== 
Sep 11 17:09:39 ERROR [0.054s]: test_polygamma_xla_float64 (__main__.TestDevicePrecisionXLA) 
Sep 11 17:09:39 ---------------------------------------------------------------------- 
Sep 11 17:09:39 Traceback (most recent call last): 
Sep 11 17:09:39   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_device_type.py", line 270, in instantiated_test 
Sep 11 17:09:39     result = test_fn(self, *args) 
Sep 11 17:09:39   File "/var/lib/jenkins/workspace/xla/test/../../test/test_torch.py", line 18853, in test_polygamma 
Sep 11 17:09:39     cpu_tensor) 
Sep 11 17:09:39   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/gradcheck.py", line 308, in gradcheck 
Sep 11 17:09:39     nondet_tol=nondet_tol) 
Sep 11 17:09:39   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/gradcheck.py", line 156, in get_analytical_jacobian 
Sep 11 17:09:39     retain_graph=True, allow_unused=True)

pytorch_linux_xenial_py3_clang5_asan_test1 (7/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:11:39 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in

Sep 11 16:11:39     #7 0x559ea6a917eb in PyEval_EvalCode /tmp/build/80754af9/python_1588903631989/work/Python/ceval.c:731 
Sep 11 16:11:39     #8 0x559ea6b11e73 in run_mod /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:1025 
Sep 11 16:11:39     #9 0x559ea6b11f0c in PyRun_StringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:949 
Sep 11 16:11:39     #10 0x559ea6b11f6e in PyRun_SimpleStringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:445 
Sep 11 16:11:39     #11 0x559ea6b15d72 in run_command /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:301 
Sep 11 16:11:39     #12 0x559ea6b15d72 in Py_Main /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:749 
Sep 11 16:11:39     #13 0x559ea69dff2d in main /tmp/build/80754af9/python_1588903631989/work/Programs/python.c:69 
Sep 11 16:11:39     #14 0x7f48144cf83f in __libc_start_main /build/glibc-e6zv40/glibc-2.23/csu/../csu/libc-start.c:291 
Sep 11 16:11:39     #15 0x559ea6abf27e in _start /home/rdonnelly/mc/conda-bld/compilers_linux-64_1534865402226/work/.build/src/glibc-2.12.2/csu/../sysdeps/x86_64/elf/start.S:103 
Sep 11 16:11:39  
Sep 11 16:11:39 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in  
Sep 11 16:11:39 + retcode=1 
Sep 11 16:11:39 + set -e 
Sep 11 16:11:39 + return 1 
Sep 11 16:11:39 + [[ pytorch-linux-xenial-py3-clang5-asan-test1 == *-NO_AVX-* ]] 
Sep 11 16:11:39 + [[ pytorch-linux-xenial-py3-clang5-asan-test1 == *-NO_AVX2-* ]] 
Sep 11 16:11:39 + '[' -n https://github.com/pytorch/pytorch/pull/44554 ']' 
Sep 11 16:11:39 + [[ pytorch-linux-xenial-py3-clang5-asan-test1 != *coverage* ]] 
Sep 11 16:11:39 ++ mktemp 
Sep 11 16:11:39 + DETERMINE_FROM=/tmp/tmp.cA4bsz4RCS 
Sep 11 16:11:39 + file_diff_from_base /tmp/tmp.cA4bsz4RCS

pytorch_linux_xenial_py3_6_gcc5_4_test (8/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:10:04 caused by: Connection refused (os error 111)

Sep 11 16:10:04 ++++ extract_trap_cmd 
Sep 11 16:10:04 ++++ printf '%s\n' '' 
Sep 11 16:10:04 +++ printf '%s\n' cleanup 
Sep 11 16:10:04 ++ trap -- ' 
Sep 11 16:10:04 cleanup' EXIT 
Sep 11 16:10:04 ++ [[ pytorch-linux-xenial-py3.6-gcc5.4-test != *pytorch-win-* ]] 
Sep 11 16:10:04 ++ which sccache 
Sep 11 16:10:04 ++ sccache --stop-server 
Sep 11 16:10:04 Stopping sccache server... 
Sep 11 16:10:04 error: couldn't connect to server 
Sep 11 16:10:04 caused by: Connection refused (os error 111) 
Sep 11 16:10:04 ++ true 
Sep 11 16:10:04 ++ rm /var/lib/jenkins/sccache_error.log 
Sep 11 16:10:04 ++ [[ pytorch-linux-xenial-py3.6-gcc5.4-test == *rocm* ]] 
Sep 11 16:10:04 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Sep 11 16:10:04 ++ SCCACHE_IDLE_TIMEOUT=1200 
Sep 11 16:10:04 ++ RUST_LOG=sccache::server=error 
Sep 11 16:10:04 ++ sccache --start-server 
Sep 11 16:10:04 Starting sccache server... 
Sep 11 16:10:04 ++ sccache --zero-stats 
Sep 11 16:10:04 Compile requests                 0

pytorch_linux_bionic_py3_6_clang9_test (9/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:05:55 caused by: Connection refused (os error 111)

Sep 11 16:05:55 ++++ extract_trap_cmd 
Sep 11 16:05:55 ++++ printf '%s\n' '' 
Sep 11 16:05:55 +++ printf '%s\n' cleanup 
Sep 11 16:05:55 ++ trap -- ' 
Sep 11 16:05:55 cleanup' EXIT 
Sep 11 16:05:55 ++ [[ pytorch-linux-bionic-py3.6-clang9-test != *pytorch-win-* ]] 
Sep 11 16:05:55 ++ which sccache 
Sep 11 16:05:55 ++ sccache --stop-server 
Sep 11 16:05:55 Stopping sccache server... 
Sep 11 16:05:55 error: couldn't connect to server 
Sep 11 16:05:55 caused by: Connection refused (os error 111) 
Sep 11 16:05:55 ++ true 
Sep 11 16:05:55 ++ rm /var/lib/jenkins/sccache_error.log 
Sep 11 16:05:55 ++ [[ pytorch-linux-bionic-py3.6-clang9-test == *rocm* ]] 
Sep 11 16:05:55 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Sep 11 16:05:55 ++ SCCACHE_IDLE_TIMEOUT=1200 
Sep 11 16:05:55 ++ RUST_LOG=sccache::server=error 
Sep 11 16:05:55 ++ sccache --start-server 
Sep 11 16:05:55 Starting sccache server... 
Sep 11 16:05:55 ++ sccache --zero-stats 
Sep 11 16:05:55 Compile requests                 0

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_simple_test (10/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:09:42 caused by: Connection refused (os error 111)

Sep 11 16:09:42 ++++ extract_trap_cmd 
Sep 11 16:09:42 ++++ printf '%s\n' '' 
Sep 11 16:09:42 +++ printf '%s\n' cleanup 
Sep 11 16:09:42 ++ trap -- ' 
Sep 11 16:09:42 cleanup' EXIT 
Sep 11 16:09:42 ++ [[ pytorch-linux-xenial-py3.6-gcc5.4-ge_config_simple-test != *pytorch-win-* ]] 
Sep 11 16:09:42 ++ which sccache 
Sep 11 16:09:42 ++ sccache --stop-server 
Sep 11 16:09:42 Stopping sccache server... 
Sep 11 16:09:42 error: couldn't connect to server 
Sep 11 16:09:42 caused by: Connection refused (os error 111) 
Sep 11 16:09:42 ++ true 
Sep 11 16:09:42 ++ rm /var/lib/jenkins/sccache_error.log 
Sep 11 16:09:42 ++ [[ pytorch-linux-xenial-py3.6-gcc5.4-ge_config_simple-test == *rocm* ]] 
Sep 11 16:09:42 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Sep 11 16:09:42 ++ SCCACHE_IDLE_TIMEOUT=1200 
Sep 11 16:09:42 ++ RUST_LOG=sccache::server=error 
Sep 11 16:09:42 ++ sccache --start-server 
Sep 11 16:09:42 Starting sccache server... 
Sep 11 16:09:42 ++ sccache --zero-stats 
Sep 11 16:09:42 Compile requests                 0

pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_legacy_test (11/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:27:44 ERROR [0.007s]: test_trace_tensor_factory (jit.test_tracer.TestTracer)

Sep 11 16:27:44     self.checkTrace(stuff, (example, example[0] + 1)) 
Sep 11 16:27:44   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:27:44     allow_unused=allow_unused) 
Sep 11 16:27:44   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:27:44     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs)) 
Sep 11 16:27:44   File "/opt/conda/lib/python3.6/site-packages/torch/tensor.py", line 563, in __len__ 
Sep 11 16:27:44     raise TypeError("len() of a 0-d tensor") 
Sep 11 16:27:44 TypeError: len() of a 0-d tensor 
Sep 11 16:27:44  
Sep 11 16:27:44 ====================================================================== 
Sep 11 16:27:44 ERROR [0.007s]: test_trace_tensor_factory (jit.test_tracer.TestTracer) 
Sep 11 16:27:44 ---------------------------------------------------------------------- 
Sep 11 16:27:44 Traceback (most recent call last): 
Sep 11 16:27:44   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 531, in test_trace_tensor_factory 
Sep 11 16:27:44     run() 
Sep 11 16:27:44   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 527, in run 
Sep 11 16:27:44     self.checkTrace(fn, (input,), inputs_require_grads=inputs_require_grads) 
Sep 11 16:27:44   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:27:44     allow_unused=allow_unused) 
Sep 11 16:27:44   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:27:44     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs))

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_profiling_test (12/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:11:22 ERROR [0.012s]: test_trace_tensor_factory (jit.test_tracer.TestTracer)

Sep 11 16:11:22     self.checkTrace(stuff, (example, example[0] + 1)) 
Sep 11 16:11:22   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:11:22     allow_unused=allow_unused) 
Sep 11 16:11:22   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:11:22     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs)) 
Sep 11 16:11:22   File "/opt/conda/lib/python3.6/site-packages/torch/tensor.py", line 563, in __len__ 
Sep 11 16:11:22     raise TypeError("len() of a 0-d tensor") 
Sep 11 16:11:22 TypeError: len() of a 0-d tensor 
Sep 11 16:11:22  
Sep 11 16:11:22 ====================================================================== 
Sep 11 16:11:22 ERROR [0.012s]: test_trace_tensor_factory (jit.test_tracer.TestTracer) 
Sep 11 16:11:22 ---------------------------------------------------------------------- 
Sep 11 16:11:22 Traceback (most recent call last): 
Sep 11 16:11:22   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 531, in test_trace_tensor_factory 
Sep 11 16:11:22     run() 
Sep 11 16:11:22   File "/var/lib/jenkins/workspace/test/jit/test_tracer.py", line 527, in run 
Sep 11 16:11:22     self.checkTrace(fn, (input,), inputs_require_grads=inputs_require_grads) 
Sep 11 16:11:22   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/jit_utils.py", line 518, in checkTrace 
Sep 11 16:11:22     allow_unused=allow_unused) 
Sep 11 16:11:22   File "/opt/conda/lib/python3.6/site-packages/torch/autograd/__init__.py", line 179, in grad 
Sep 11 16:11:22     grad_outputs_ = _tensor_or_tensors_to_tuple(grad_outputs, len(outputs))

pytorch_linux_xenial_cuda10_2_cudnn7_py3_gcc7_test (13/13)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Sep 11 16:20:25 caused by: Connection refused (os error 111)

Sep 11 16:20:25 ++++ extract_trap_cmd 
Sep 11 16:20:25 ++++ printf '%s\n' '' 
Sep 11 16:20:25 +++ printf '%s\n' cleanup 
Sep 11 16:20:25 ++ trap -- ' 
Sep 11 16:20:25 cleanup' EXIT 
Sep 11 16:20:25 ++ [[ pytorch-linux-xenial-cuda10.2-cudnn7-py3-gcc7-test != *pytorch-win-* ]] 
Sep 11 16:20:25 ++ which sccache 
Sep 11 16:20:25 ++ sccache --stop-server 
Sep 11 16:20:25 Stopping sccache server... 
Sep 11 16:20:25 error: couldn't connect to server 
Sep 11 16:20:25 caused by: Connection refused (os error 111) 
Sep 11 16:20:25 ++ true 
Sep 11 16:20:25 ++ rm /var/lib/jenkins/sccache_error.log 
Sep 11 16:20:25 ++ [[ pytorch-linux-xenial-cuda10.2-cudnn7-py3-gcc7-test == *rocm* ]] 
Sep 11 16:20:25 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Sep 11 16:20:25 ++ SCCACHE_IDLE_TIMEOUT=1200 
Sep 11 16:20:25 ++ RUST_LOG=sccache::server=error 
Sep 11 16:20:25 ++ sccache --start-server 
Sep 11 16:20:25 Starting sccache server... 
Sep 11 16:20:25 ++ sccache --zero-stats 
Sep 11 16:20:25 Compile requests                 0

❄️ 1 failure tentatively classified as flaky

but reruns have not yet been triggered to confirm:

pytorch_windows_vs2019_py36_cuda10.1_test2 (1/1)

Step: "Test" (full log | diagnosis details | 🔁 rerun) ❄️

WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', ConnectionResetError(10054, 'An existing connection was forcibly closed by the remote host', None, 10054, None))': /simple/scikit-learn/

Collecting pillow 
  Downloading Pillow-7.2.0-cp36-cp36m-win_amd64.whl (2.0 MB) 
Collecting unittest-xml-reporting 
  Downloading unittest_xml_reporting-3.0.4-py2.py3-none-any.whl (19 kB) 
Collecting attrs>=19.2.0 
  Downloading attrs-20.2.0-py2.py3-none-any.whl (48 kB) 
Collecting audioread>=2.0.0 
  Downloading audioread-2.1.8.tar.gz (21 kB) 
Requirement already satisfied: numpy>=1.15.0 in c:\jenkins\miniconda3\lib\site-packages (from librosa>=0.6.2) (1.19.1) 
Requirement already satisfied: scipy>=1.0.0 in c:\jenkins\miniconda3\lib\site-packages (from librosa>=0.6.2) (1.5.0) 
WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', ConnectionResetError(10054, 'An existing connection was forcibly closed by the remote host', None, 10054, None))': /simple/scikit-learn/ 
Collecting scikit-learn!=0.19.0,>=0.14.0 
  Downloading scikit_learn-0.23.2-cp36-cp36m-win_amd64.whl (6.8 MB) 
Collecting joblib>=0.14 
  Downloading joblib-0.16.0-py3-none-any.whl (300 kB) 
Collecting decorator>=3.0.0 
  Downloading decorator-4.4.2-py2.py3-none-any.whl (9.2 kB) 
Collecting resampy>=0.2.2 
  Downloading resampy-0.2.2.tar.gz (323 kB) 
Requirement already satisfied: numba>=0.43.0 in c:\jenkins\miniconda3\lib\site-packages (from librosa>=0.6.2) (0.44.0) 
Collecting soundfile>=0.9.0

🚧 1 ongoing upstream failure:

These were probably caused by upstream breakages that are not fixed yet:

pytorch_linux_bionic_py3_8_gcc9_coverage_test since Sep 11
- 🔁 rerun

ci.pytorch.org: 1 failed

Failed: pr/pytorch-linux-bionic-rocm3.7-py3.6

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group.

See how this bot performed.

This comment has been revised 7 times.

ezyang · 2020-09-14T15:07:41Z

@anjali411 and I agreed that we would just disable the tensor-like test instead.

Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)"

7c6878f

This reverts commit f9a0d0c. [ghstack-poisoned]

ezyang requested review from albanD and apaszke as code owners September 11, 2020 15:33

ezyang added a commit that referenced this pull request Sep 11, 2020

Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)"

353ef02

This reverts commit f9a0d0c. ghstack-source-id: d662f56 Pull Request resolved: #44554

anjali411 self-requested a review September 11, 2020 15:36

anjali411 approved these changes Sep 11, 2020

View reviewed changes

ezyang closed this Sep 14, 2020

facebook-github-bot deleted the gh/ezyang/836/head branch October 15, 2020 14:21

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)"#44554

Revert "Allow Tensor-likes in torch.autograd.gradcheck (#43877)"#44554
ezyang wants to merge 1 commit intogh/ezyang/836/basefrom
gh/ezyang/836/head

ezyang commented Sep 11, 2020 •

edited

Loading

Uh oh!

dr-ci Bot commented Sep 11, 2020 •

edited

Loading

Uh oh!

ezyang commented Sep 14, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

ezyang commented Sep 11, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dr-ci Bot commented Sep 11, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💊 CI failures summary and remediations

🕵️ 13 new failures recognized by patterns

pytorch_windows_vs2019_py36_cuda10.1_test1 (1/13)

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_legacy_test (2/13)

pytorch_linux_xenial_py3_clang5_asan_test2 (3/13)

pytorch_macos_10_13_py3_test (4/13)

pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_profiling_test (5/13)

pytorch_xla_linux_bionic_py3_6_clang9_test (6/13)

pytorch_linux_xenial_py3_clang5_asan_test1 (7/13)

pytorch_linux_xenial_py3_6_gcc5_4_test (8/13)

pytorch_linux_bionic_py3_6_clang9_test (9/13)

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_simple_test (10/13)

pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_legacy_test (11/13)

pytorch_linux_xenial_py3_6_gcc5_4_ge_config_profiling_test (12/13)

pytorch_linux_xenial_cuda10_2_cudnn7_py3_gcc7_test (13/13)

❄️ 1 failure tentatively classified as flaky

pytorch_windows_vs2019_py36_cuda10.1_test2 (1/1)

🚧 1 ongoing upstream failure:

ci.pytorch.org: 1 failed

Uh oh!

ezyang commented Sep 14, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

ezyang commented Sep 11, 2020 •

edited

Loading

dr-ci Bot commented Sep 11, 2020 •

edited

Loading