Skip to content

Conversation

@hameerabbasi
Copy link
Collaborator

No description provided.

@hameerabbasi hameerabbasi requested a review from apaszke as a code owner August 12, 2020 10:05
@facebook-github-bot facebook-github-bot added the oncall: jit Add this issue/PR to JIT oncall triage queue label Aug 12, 2020
@hameerabbasi hameerabbasi mentioned this pull request Aug 12, 2020
8 tasks
@dr-ci
Copy link

dr-ci bot commented Aug 12, 2020

💊 CI failures summary and remediations

As of commit d02863a (more details on the Dr. CI page):



🕵️ 14 new failures recognized by patterns

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

See CircleCI build pytorch_linux_xenial_py3_6_gcc5_4_test (1/14)

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

Aug 18 10:10:59 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n ^\n" }
Aug 18 10:10:59 Traceback (most recent call last): 
Aug 18 10:10:59   File "test/run_test.py", line 719, in <module> 
Aug 18 10:10:59     main() 
Aug 18 10:10:59   File "test/run_test.py", line 708, in main 
Aug 18 10:10:59     raise RuntimeError(err) 
Aug 18 10:10:59 RuntimeError: test_jit failed! 
Aug 18 10:10:59 + cleanup 
Aug 18 10:10:59 + retcode=1 
Aug 18 10:10:59 + set +x 
Aug 18 10:10:59 =================== sccache compilation log =================== 
Aug 18 10:10:59 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n                       ^\n" } 
Aug 18 10:10:59  
Aug 18 10:10:59 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Aug 18 10:10:59 Compile requests                 65 
Aug 18 10:10:59 Compile requests executed        35 
Aug 18 10:10:59 Cache hits                        2 
Aug 18 10:10:59 Cache misses                     32 
Aug 18 10:10:59 Cache timeouts                    0 
Aug 18 10:10:59 Cache read errors                 0 
Aug 18 10:10:59 Forced recaches                   0 
Aug 18 10:10:59 Cache write errors                0 

See CircleCI build pytorch_macos_10_13_py3_test (2/14)

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

Aug 18 03:31:18 RuntimeError: test_jit failed!
Aug 18 03:31:18 Generated XML report: test-reports/dist-gloo/TEST-jit.test_type_sharing.TestTypeSharing-20200818032847.xml 
Aug 18 03:31:18 Generated XML report: test-reports/dist-gloo/TEST-jit.test_unsupported_ops.TestUnsupportedOps-20200818032847.xml 
Aug 18 03:31:18 Generated XML report: test-reports/dist-gloo/TEST-jit.test_with.TestWith-20200818032847.xml 
Aug 18 03:31:18 Generated XML report: test-reports/dist-gloo/TEST-jit.test_autodiff_subgraph_slicing.TestAutodiffSubgraphSlicing-20200818032847.xml 
Aug 18 03:31:18 Generated XML report: test-reports/dist-gloo/TEST-jit.test_data_parallel.TestDataParallel-20200818032847.xml 
Aug 18 03:31:18 Traceback (most recent call last): 
Aug 18 03:31:18   File "test/run_test.py", line 719, in <module> 
Aug 18 03:31:18     main() 
Aug 18 03:31:18   File "test/run_test.py", line 708, in main 
Aug 18 03:31:18     raise RuntimeError(err) 
Aug 18 03:31:18 RuntimeError: test_jit failed! 
Aug 18 03:31:18 + cleanup 
Aug 18 03:31:18 + retcode=1 
Aug 18 03:31:18 + set +x 

See CircleCI build pytorch_linux_bionic_py3_8_gcc9_test (3/14)

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

Aug 18 10:27:50 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function ‘int main()’:\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:22: error: expected ‘;’ before ‘}’ token\n 2 | int main() { return 0 }\n | ^~\n | ;\n" }
Aug 18 10:27:50     raise RuntimeError(err) 
Aug 18 10:27:50 RuntimeError: test_jit failed! 
Aug 18 10:27:50  
Aug 18 10:27:50 real	16m49.588s 
Aug 18 10:27:50 user	9m3.244s 
Aug 18 10:27:50 sys	0m43.319s 
Aug 18 10:27:50 + cleanup 
Aug 18 10:27:50 + retcode=1 
Aug 18 10:27:50 + set +x 
Aug 18 10:27:50 =================== sccache compilation log =================== 
Aug 18 10:27:50 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function ‘int main()’:\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:22: error: expected ‘;’ before ‘}’ token\n    2 | int main() { return 0 }\n      |                      ^~\n      |                      ;\n" } 
Aug 18 10:27:50  
Aug 18 10:27:50 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Aug 18 10:27:50 Compile requests                 65 
Aug 18 10:27:50 Compile requests executed        35 
Aug 18 10:27:50 Cache hits                        2 
Aug 18 10:27:50 Cache misses                     32 
Aug 18 10:27:50 Cache timeouts                    0 
Aug 18 10:27:50 Cache read errors                 0 
Aug 18 10:27:50 Forced recaches                   0 
Aug 18 10:27:50 Cache write errors                0 

See CircleCI build binary_windows_libtorch_3_7_cpu_release_build (4/14)

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

FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj
[264/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\extension_set.cc.obj 
[265/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\generated_enum_util.cc.obj 
[266/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\io\coded_stream.cc.obj 
[267/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\stubs\int128.cc.obj 
[268/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\implicit_weak_message.cc.obj 
[269/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\generated_message_table_driven_lite.cc.obj 
[270/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\generated_message_util.cc.obj 
[271/2275] Building RC object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\version.rc.res 
[272/2275] Building CXX object third_party\protobuf\cmake\CMakeFiles\libprotobuf-lite.dir\__\src\google\protobuf\io\zero_copy_stream_impl_lite.cc.obj 
[273/2275] Building CXX object third_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj 
FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj  
arty\pybind11\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG   -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
C:\w\b\windows\pytorch\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp(401) : fatal error C1001: Internal compiler error.
(compiler file 'd:\agent\_work\7\s\src\vctools\Compiler\Utc\src\p2\main.c', line 195)
 To work around this problem, try simplifying or changing the program near the locations listed above.
If possible please provide a repro here: https://developercommunity.visualstudio.com 
Please choose the Technical Support command on the Visual C++ 
 Help menu, or open the Technical Support help file for more information
  cl!RaiseException()+0x69
  cl!RaiseException()+0x69
  cl!CloseTypeServerPDB()+0x22e6b

See CircleCI build pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_profiling_test (5/14)

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

Aug 18 11:59:49 ERROR [0.107s]: test_math_ops (test_jit.TestScript)
Aug 18 11:59:49   test_tracing_gives_different_types (jit.test_type_sharing.TestTypeSharing) ... ok (0.014s) 
Aug 18 11:59:49   test_factory_ops_requires_grad_fail (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.004s) 
Aug 18 11:59:49   test_init_ops (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.017s) 
Aug 18 11:59:49   test_ops_bound_in_functional (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.003s) 
Aug 18 11:59:49   test_with_as (jit.test_with.TestWith) ... ok (0.110s) 
Aug 18 11:59:49   test_with_errors (jit.test_with.TestWith) ... ok (0.016s) 
Aug 18 11:59:49   test_with_exceptions (jit.test_with.TestWith) ... ok (0.006s) 
Aug 18 11:59:49   test_with_no_as (jit.test_with.TestWith) ... ok (0.107s) 
Aug 18 11:59:49  
Aug 18 11:59:49 ====================================================================== 
Aug 18 11:59:49 ERROR [0.107s]: test_math_ops (test_jit.TestScript) 
Aug 18 11:59:49 ---------------------------------------------------------------------- 
Aug 18 11:59:49 Traceback (most recent call last): 
Aug 18 11:59:49   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops 
Aug 18 11:59:49     checkMathWrap(op, 1) 
Aug 18 11:59:49   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap 
Aug 18 11:59:49     checkMath(func_name, num_args, True, **args) 
Aug 18 11:59:49   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath 
Aug 18 11:59:49     cu = torch.jit.CompilationUnit(funcs_str) 
Aug 18 11:59:49   File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__ 
Aug 18 11:59:49     self.define(lang, _frames_up=_frames_up + 1) 

See CircleCI build pytorch_windows_vs2019_py36_cuda10.1_build (6/14)

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

FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj

UDA\v10.1\include" /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\matmul\ref_matmul.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\matmul\ref_matmul.cpp 
Microsoft (R) C/C++ Optimizing Compiler Version 19.27.29111 for x64
Copyright (C) Microsoft Corporation.  All rights reserved.

clude" /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\matmul\gemm_bf16_matmul.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\matmul\gemm_bf16_matmul.cpp 
Microsoft (R) C/C++ Optimizing Compiler Version 19.27.29111 for x64
Copyright (C) Microsoft Corporation.  All rights reserved.

include" /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj  
include" /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
C:\Users\circleci\project\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp(401) : fatal error C1001: Internal compiler error.
(compiler file 'd:\agent\_work\7\s\src\vctools\Compiler\Utc\src\p2\main.c', line 195)
 To work around this problem, try simplifying or changing the program near the locations listed above.
If possible please provide a repro here: https://developercommunity.visualstudio.com 
Please choose the Technical Support command on the Visual C++ 
 Help menu, or open the Technical Support help file for more information
  cl!RaiseException()+0x69
  cl!RaiseException()+0x69
  cl!CloseTypeServerPDB()+0x22e6b

See CircleCI build pytorch_linux_xenial_py3_clang5_asan_test2 (7/14)

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

Aug 18 10:16:21 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in
Aug 18 10:16:21     #7 0x562b9eef17eb in PyEval_EvalCode /tmp/build/80754af9/python_1588903631989/work/Python/ceval.c:731 
Aug 18 10:16:21     #8 0x562b9ef71e73 in run_mod /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:1025 
Aug 18 10:16:21     #9 0x562b9ef71f0c in PyRun_StringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:949 
Aug 18 10:16:21     #10 0x562b9ef71f6e in PyRun_SimpleStringFlags /tmp/build/80754af9/python_1588903631989/work/Python/pythonrun.c:445 
Aug 18 10:16:21     #11 0x562b9ef75d72 in run_command /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:301 
Aug 18 10:16:21     #12 0x562b9ef75d72 in Py_Main /tmp/build/80754af9/python_1588903631989/work/Modules/main.c:749 
Aug 18 10:16:21     #13 0x562b9ee3ff2d in main /tmp/build/80754af9/python_1588903631989/work/Programs/python.c:69 
Aug 18 10:16:21     #14 0x7f2cf421883f in __libc_start_main /build/glibc-e6zv40/glibc-2.23/csu/../csu/libc-start.c:291 
Aug 18 10:16:21     #15 0x562b9ef1f27e 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 
Aug 18 10:16:21  
Aug 18 10:16:21 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/aten/src/ATen/Utils.cpp:11:3 in  
Aug 18 10:16:21 + retcode=1 
Aug 18 10:16:21 + set -e 
Aug 18 10:16:21 + return 1 
Aug 18 10:16:21 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *-NO_AVX-* ]] 
Aug 18 10:16:21 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *-NO_AVX2-* ]] 
Aug 18 10:16:21 + '[' -n '' ']' 
Aug 18 10:16:21 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *tbb* ]] 
Aug 18 10:16:21 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *libtorch* ]] 
Aug 18 10:16:21 + [[ pytorch-linux-xenial-py3-clang5-asan-test2 == *-bazel-* ]] 
Aug 18 10:16:21 + cd test 

See CircleCI build pytorch_linux_xenial_py3_6_gcc5_4_ge_config_profiling_test (8/14)

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

Aug 18 09:43:44 ERROR [0.097s]: test_math_ops (test_jit.TestScript)
Aug 18 09:43:44   test_tracing_gives_different_types (jit.test_type_sharing.TestTypeSharing) ... ok (0.014s) 
Aug 18 09:43:44   test_factory_ops_requires_grad_fail (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.004s) 
Aug 18 09:43:44   test_init_ops (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.017s) 
Aug 18 09:43:44   test_ops_bound_in_functional (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.003s) 
Aug 18 09:43:44   test_with_as (jit.test_with.TestWith) ... ok (0.109s) 
Aug 18 09:43:44   test_with_errors (jit.test_with.TestWith) ... ok (0.018s) 
Aug 18 09:43:44   test_with_exceptions (jit.test_with.TestWith) ... ok (0.008s) 
Aug 18 09:43:44   test_with_no_as (jit.test_with.TestWith) ... ok (0.107s) 
Aug 18 09:43:44  
Aug 18 09:43:44 ====================================================================== 
Aug 18 09:43:44 ERROR [0.097s]: test_math_ops (test_jit.TestScript) 
Aug 18 09:43:44 ---------------------------------------------------------------------- 
Aug 18 09:43:44 Traceback (most recent call last): 
Aug 18 09:43:44   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops 
Aug 18 09:43:44     checkMathWrap(op, 1) 
Aug 18 09:43:44   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap 
Aug 18 09:43:44     checkMath(func_name, num_args, True, **args) 
Aug 18 09:43:44   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath 
Aug 18 09:43:44     cu = torch.jit.CompilationUnit(funcs_str) 
Aug 18 09:43:44   File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__ 
Aug 18 09:43:44     self.define(lang, _frames_up=_frames_up + 1) 

See CircleCI build pytorch_linux_xenial_py3_6_gcc5_4_ge_config_simple_test (9/14)

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

Aug 18 10:11:05 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n ^\n" }
Aug 18 10:11:05 Traceback (most recent call last): 
Aug 18 10:11:05   File "test/run_test.py", line 719, in <module> 
Aug 18 10:11:05     main() 
Aug 18 10:11:05   File "test/run_test.py", line 708, in main 
Aug 18 10:11:05     raise RuntimeError(err) 
Aug 18 10:11:05 RuntimeError: test_jit failed! 
Aug 18 10:11:05 =================== sccache compilation log =================== 
Aug 18 10:11:05 + cleanup 
Aug 18 10:11:05 + retcode=1 
Aug 18 10:11:05 + set +x 
Aug 18 10:11:05 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n                       ^\n" } 
Aug 18 10:11:05  
Aug 18 10:11:05 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Aug 18 10:11:05 Compile requests                 65 
Aug 18 10:11:05 Compile requests executed        35 
Aug 18 10:11:05 Cache hits                        4 
Aug 18 10:11:05 Cache misses                     30 
Aug 18 10:11:05 Cache timeouts                    0 
Aug 18 10:11:05 Cache read errors                 0 
Aug 18 10:11:05 Forced recaches                   0 
Aug 18 10:11:05 Cache write errors                0 

See CircleCI build pytorch_windows_vs2019_py36_cpu_build (10/14)

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

FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj
Microsoft (R) C/C++ Optimizing Compiler Version 19.27.29111 for x64
Copyright (C) Microsoft Corporation.  All rights reserved.

cl : Command line warning D9025 : overriding '/O2' with '/Od'
 -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /Od /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\x64\CMakeFiles\dnnl_cpu_x64.dir\gemm\f32\jit_sse41_f32_copy_at_kern_autogen.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\x64\CMakeFiles\dnnl_cpu_x64.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\x64\gemm\f32\jit_sse41_f32_copy_at_kern_autogen.cpp 
Microsoft (R) C/C++ Optimizing Compiler Version 19.27.29111 for x64
Copyright (C) Microsoft Corporation.  All rights reserved.

cl : Command line warning D9025 : overriding '/O2' with '/Od'
ird_party\pybind11\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj  
ird_party\pybind11\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
C:\Users\circleci\project\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp(401) : fatal error C1001: Internal compiler error.
(compiler file 'd:\agent\_work\7\s\src\vctools\Compiler\Utc\src\p2\main.c', line 195)
 To work around this problem, try simplifying or changing the program near the locations listed above.
If possible please provide a repro here: https://developercommunity.visualstudio.com 
Please choose the Technical Support command on the Visual C++ 
 Help menu, or open the Technical Support help file for more information
  cl!RaiseException()+0x69
  cl!RaiseException()+0x69
  cl!CloseTypeServerPDB()+0x22e6b

See CircleCI build pytorch_windows_vs2019_py36_cuda11.0_build (11/14)

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

FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj

_tmp\magma\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\simple_resampling.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\simple_resampling.cpp 
Microsoft (R) C/C++ Optimizing Compiler Version 19.27.29111 for x64
Copyright (C) Microsoft Corporation.  All rights reserved.

\CMakeFiles\onnx.dir\onnx\optimizer\optimize.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\optimizer\pass.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\optimizer\pass_manager.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\optimizer\pass_registry.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\shape_inference\implementation.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\version_converter\convert.cc.obj third_party\onnx\CMakeFiles\onnx.dir\onnx\version_converter\helper.cc.obj  && cd ." 
Microsoft (R) Library Manager Version 14.27.29111.0 
Copyright (C) Microsoft Corporation.  All rights reserved. 
 
\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
FAILED: third_party/ideep/mkl-dnn/src/cpu/CMakeFiles/dnnl_cpu.dir/gemm_convolution_utils.cpp.obj  
\include /DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -openmp:experimental -DNDEBUG -openmp:experimental  /MP    /wd4800 /wd4068 /wd4305 /wd4551 /wd4244  /MD /O2 /Ob2 /DNDEBUG /w /bigobj -DNDEBUG -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -std:c++14 /showIncludes /Fothird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\gemm_convolution_utils.cpp.obj /Fdthird_party\ideep\mkl-dnn\src\cpu\CMakeFiles\dnnl_cpu.dir\ /FS -c ..\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp 
C:\Users\circleci\project\third_party\ideep\mkl-dnn\src\cpu\gemm_convolution_utils.cpp(401) : fatal error C1001: Internal compiler error.
(compiler file 'd:\agent\_work\7\s\src\vctools\Compiler\Utc\src\p2\main.c', line 195)
 To work around this problem, try simplifying or changing the program near the locations listed above.
If possible please provide a repro here: https://developercommunity.visualstudio.com 
Please choose the Technical Support command on the Visual C++ 
 Help menu, or open the Technical Support help file for more information
  cl!RaiseException()+0x69
  cl!RaiseException()+0x69
  cl!CloseTypeServerPDB()+0x22e6b

See CircleCI build pytorch_linux_xenial_py3_6_gcc5_4_ge_config_legacy_test (12/14)

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

Aug 18 09:44:55 ERROR [0.107s]: test_math_ops (test_jit.TestScript)
Aug 18 09:44:55   test_tracing_gives_different_types (jit.test_type_sharing.TestTypeSharing) ... ok (0.020s) 
Aug 18 09:44:55   test_factory_ops_requires_grad_fail (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.005s) 
Aug 18 09:44:55   test_init_ops (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.019s) 
Aug 18 09:44:55   test_ops_bound_in_functional (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.003s) 
Aug 18 09:44:55   test_with_as (jit.test_with.TestWith) ... ok (0.103s) 
Aug 18 09:44:55   test_with_errors (jit.test_with.TestWith) ... ok (0.018s) 
Aug 18 09:44:55   test_with_exceptions (jit.test_with.TestWith) ... ok (0.008s) 
Aug 18 09:44:55   test_with_no_as (jit.test_with.TestWith) ... ok (0.094s) 
Aug 18 09:44:55  
Aug 18 09:44:55 ====================================================================== 
Aug 18 09:44:55 ERROR [0.107s]: test_math_ops (test_jit.TestScript) 
Aug 18 09:44:55 ---------------------------------------------------------------------- 
Aug 18 09:44:55 Traceback (most recent call last): 
Aug 18 09:44:55   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops 
Aug 18 09:44:55     checkMathWrap(op, 1) 
Aug 18 09:44:55   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap 
Aug 18 09:44:55     checkMath(func_name, num_args, True, **args) 
Aug 18 09:44:55   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath 
Aug 18 09:44:55     cu = torch.jit.CompilationUnit(funcs_str) 
Aug 18 09:44:55   File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__ 
Aug 18 09:44:55     self.define(lang, _frames_up=_frames_up + 1) 

See CircleCI build pytorch_linux_bionic_py3_6_clang9_test (13/14)

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

Aug 18 10:08:30 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n ^\n" }
Aug 18 10:08:30     raise RuntimeError(err) 
Aug 18 10:08:30 RuntimeError: test_jit failed! 
Aug 18 10:08:30  
Aug 18 10:08:30 real	19m15.362s 
Aug 18 10:08:30 user	10m24.515s 
Aug 18 10:08:30 sys	2m15.250s 
Aug 18 10:08:30 + cleanup 
Aug 18 10:08:30 + retcode=1 
Aug 18 10:08:30 + set +x 
Aug 18 10:08:30 =================== sccache compilation log =================== 
Aug 18 10:08:30 ERROR:sccache::server: Compilation failed: Output { status: ExitStatus(ExitStatus(256)), stdout: "", stderr: "/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp: In function \'int main()\':\n/var/lib/jenkins/.cache/torch_extensions/test_compilation_error_formatting/main.cpp:2:23: error: expected \';\' before \'}\' token\n int main() { return 0 }\n                       ^\n" } 
Aug 18 10:08:30  
Aug 18 10:08:30 =========== If your build fails, please take a look at the log above for possible reasons =========== 
Aug 18 10:08:30 Compile requests                 65 
Aug 18 10:08:30 Compile requests executed        35 
Aug 18 10:08:30 Cache hits                        2 
Aug 18 10:08:30 Cache misses                     32 
Aug 18 10:08:30 Cache timeouts                    0 
Aug 18 10:08:30 Cache read errors                 0 
Aug 18 10:08:30 Forced recaches                   0 
Aug 18 10:08:30 Cache write errors                0 

See CircleCI build pytorch_linux_xenial_cuda10_2_cudnn7_py3_ge_config_legacy_test (14/14)

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

Aug 18 12:00:12 ERROR [0.108s]: test_math_ops (test_jit.TestScript)
Aug 18 12:00:12   test_tracing_gives_different_types (jit.test_type_sharing.TestTypeSharing) ... ok (0.015s) 
Aug 18 12:00:12   test_factory_ops_requires_grad_fail (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.004s) 
Aug 18 12:00:12   test_init_ops (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.018s) 
Aug 18 12:00:12   test_ops_bound_in_functional (jit.test_unsupported_ops.TestUnsupportedOps) ... ok (0.003s) 
Aug 18 12:00:12   test_with_as (jit.test_with.TestWith) ... ok (0.092s) 
Aug 18 12:00:12   test_with_errors (jit.test_with.TestWith) ... ok (0.016s) 
Aug 18 12:00:12   test_with_exceptions (jit.test_with.TestWith) ... ok (0.006s) 
Aug 18 12:00:12   test_with_no_as (jit.test_with.TestWith) ... ok (0.087s) 
Aug 18 12:00:12  
Aug 18 12:00:12 ====================================================================== 
Aug 18 12:00:12 ERROR [0.108s]: test_math_ops (test_jit.TestScript) 
Aug 18 12:00:12 ---------------------------------------------------------------------- 
Aug 18 12:00:12 Traceback (most recent call last): 
Aug 18 12:00:12   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops 
Aug 18 12:00:12     checkMathWrap(op, 1) 
Aug 18 12:00:12   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap 
Aug 18 12:00:12     checkMath(func_name, num_args, True, **args) 
Aug 18 12:00:12   File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath 
Aug 18 12:00:12     cu = torch.jit.CompilationUnit(funcs_str) 
Aug 18 12:00:12   File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__ 
Aug 18 12:00:12     self.define(lang, _frames_up=_frames_up + 1) 

❄️ 1 failure tentatively classified as flaky

but reruns have not yet been triggered to confirm:

See CircleCI build pytorch_linux_xenial_cuda10_2_cudnn7_py3_gcc7_test (1/1)

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

Aug 18 12:45:52 ConnectionResetError: [Errno 104] Connection reset by peer
Aug 18 12:45:52   File "/opt/conda/lib/python3.6/multiprocessing/connection.py", line 456, in accept 
Aug 18 12:45:52     answer_challenge(c, self._authkey) 
Aug 18 12:45:52   File "/opt/conda/lib/python3.6/multiprocessing/connection.py", line 737, in answer_challenge 
Aug 18 12:45:52     response = connection.recv_bytes(256)        # reject large message 
Aug 18 12:45:52   File "/opt/conda/lib/python3.6/multiprocessing/connection.py", line 216, in recv_bytes 
Aug 18 12:45:52     buf = self._recv_bytes(maxlength) 
Aug 18 12:45:52   File "/opt/conda/lib/python3.6/multiprocessing/connection.py", line 407, in _recv_bytes 
Aug 18 12:45:52     buf = self._recv(4) 
Aug 18 12:45:52   File "/opt/conda/lib/python3.6/multiprocessing/connection.py", line 379, in _recv 
Aug 18 12:45:52     chunk = read(handle, remaining) 
Aug 18 12:45:52 ConnectionResetError: [Errno 104] Connection reset by peer 
Aug 18 12:45:52 /opt/conda/lib/python3.6/multiprocessing/semaphore_tracker.py:143: UserWarning: semaphore_tracker: There appear to be 14 leaked semaphores to clean up at shutdown 
Aug 18 12:45:52   len(cache)) 
Aug 18 12:45:55 Process ErrorTrackingProcess-156: 
Aug 18 12:45:55 Traceback (most recent call last): 
Aug 18 12:45:55   File "/opt/conda/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap 
Aug 18 12:45:55     self.run() 
Aug 18 12:45:55   File "/var/lib/jenkins/workspace/test/test_dataloader.py", line 361, in run 
Aug 18 12:45:55     super(ErrorTrackingProcess, self).run() 
Aug 18 12:45:55   File "/opt/conda/lib/python3.6/multiprocessing/process.py", line 93, in run 
Aug 18 12:45:55     self._target(*self._args, **self._kwargs) 

ci.pytorch.org: 1 failed


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 70 times.

@hameerabbasi hameerabbasi changed the title Add degree alias to deg2rad. Add degree alias to rad2deg. Aug 12, 2020
@hameerabbasi hameerabbasi changed the title Add degree alias to rad2deg. Add degrees alias to rad2deg. Aug 12, 2020
@mruberry
Copy link
Collaborator

Unfortunately you'll need to rebase because #42802 introduced a new way of adding aliases. It should be pretty simple, though.

@mruberry mruberry requested review from mruberry and removed request for apaszke August 12, 2020 17:46
@mruberry mruberry added module: numpy Related to numpy support, and also numpy compatibility of our operators triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module labels Aug 12, 2020
@hameerabbasi
Copy link
Collaborator Author

The Windows CI builds have an internal compiler error, possibly related, but this PR isn't the source of the issue I think.

The Linux CI jobs fail with:

Traceback (most recent call last):
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops
    checkMathWrap(op, 1)
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap
    checkMath(func_name, num_args, True, **args)
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath
    cu = torch.jit.CompilationUnit(funcs_str)
  File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__
    self.define(lang, _frames_up=_frames_up + 1)
  File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1083, in define
    self._c.define(lang, rcb)
RuntimeError: replace_node->maybeOperator() INTERNAL ASSERT FAILED at "/var/lib/jenkins/workspace/torch/csrc/jit/passes/normalize_ops.cpp":38, please report a bug to PyTorch. invalid symbol replacement:140136

Which I've tried and failed to fix, even after asking internally.

@mruberry
Copy link
Collaborator

The Windows CI builds have an internal compiler error, possibly related, but this PR isn't the source of the issue I think.

The Linux CI jobs fail with:

Traceback (most recent call last):
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6330, in test_math_ops
    checkMathWrap(op, 1)
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6256, in checkMathWrap
    checkMath(func_name, num_args, True, **args)
  File "/var/lib/jenkins/workspace/test/test_jit.py", line 6286, in checkMath
    cu = torch.jit.CompilationUnit(funcs_str)
  File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1078, in __init__
    self.define(lang, _frames_up=_frames_up + 1)
  File "/opt/conda/lib/python3.6/site-packages/torch/jit/_script.py", line 1083, in define
    self._c.define(lang, rcb)
RuntimeError: replace_node->maybeOperator() INTERNAL ASSERT FAILED at "/var/lib/jenkins/workspace/torch/csrc/jit/passes/normalize_ops.cpp":38, please report a bug to PyTorch. invalid symbol replacement:140136

Which I've tried and failed to fix, even after asking internally.

Unfortunately Windows CI builds were broken for awhile. Sorry about that and the confusion it caused.

The Linux issue is more interesting. It turns out that the JIT already defines degrees and radians analogous to Python:

(math.degrees, "aten::degrees"),

It does this because the JIT supports a subset of Python, so it has implemented many Python builtins.

This means that unconditionally converting from degrees to another name won't work, because the alias is attempting to convert these calls, too!

We could probably reverse the alias and have rad2deg alias to degrees, @eellison, what would you think of that? Although I don't want to send you on a wild goose chase, @hameerabbasi, so maybe a different issue would be more interesting until we resolve this one.

@eellison
Copy link
Contributor

@mruberry I think we can add a check that the node has inputs and the first input is a Tensor before doing the alias replacement, that would fix this issue. Another fix would be to match on the full schema instead of just the kind in checking if there's an alias.

@hameerabbasi hameerabbasi deleted the add-degree-alias branch October 2, 2020 11:11
@hameerabbasi hameerabbasi restored the add-degree-alias branch October 2, 2020 11:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

module: numpy Related to numpy support, and also numpy compatibility of our operators oncall: jit Add this issue/PR to JIT oncall triage queue open source triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants