Automatic update of fbcode/onnx to 83dd62659fc07d5b7fa93b5d1c1879f93509c7db by houseroad · Pull Request #19454 · pytorch/pytorch

houseroad · 2019-04-19T02:21:31Z

Summary:
Previous import was ad7313470a9119d7e1afda7edf1d654497ee80ab

Included changes:

83dd6265: Add NonMaxSuppression operator (Fix compilation with cuDNN 5 #1703)
31ca5d6f: add node tests for quantized ops (Fix broadcasting issues in binary_cross_entropy_with_logits #1944)
e6076c1d: Fix test stat coverage script (Fix exceptions not being caught #1948)
ad036405: Add IsInf to detect infinity values (Broadcast masked_select and implement double backwards for masked_scatter, index_copy, scatter_add, gather #1884)

Differential Revision: D15010015

…09c7db (pytorch#19454) Summary: Pull Request resolved: pytorch#19454 Previous import was ad7313470a9119d7e1afda7edf1d654497ee80ab Included changes: - **[83dd6265](onnx/onnx@83dd6265)**: Add NonMaxSuppression operator (pytorch#1703) <Hector Li> - **[31ca5d6f](onnx/onnx@31ca5d6f)**: add node tests for quantized ops (pytorch#1944) <Ashwini Khade> - **[e6076c1d](onnx/onnx@e6076c1d)**: Fix test stat coverage script (pytorch#1948) <Raymond Yang> - **[ad036405](onnx/onnx@ad036405)**: Add IsInf to detect infinity values (pytorch#1884) <Wei-Sheng Chin> Differential Revision: D15010015 fbshipit-source-id: 9778757752785fe3169ad2ac606b37299aa69da6

ezyang · 2019-04-22T19:06:31Z

After this landed on master, this test failed:

Apr 22 18:44:50 ======================================================================
Apr 22 18:44:50 ERROR: test_dcgan_models_cuda (__main__.TestEndToEndHybridFrontendModels)
Apr 22 18:44:50 ----------------------------------------------------------------------
Apr 22 18:44:50 Traceback (most recent call last):
Apr 22 18:44:50   File "/var/lib/jenkins/workspace/test/common_utils.py", line 338, in wrapper
Apr 22 18:44:50     method(*args, **kwargs)
Apr 22 18:44:50   File "test_jit.py", line 11764, in test_dcgan_models_cuda
Apr 22 18:44:50     self._test_dcgan_models(self, device='cuda', check_export_import=False)
Apr 22 18:44:50   File "test_jit.py", line 11756, in _test_dcgan_models
Apr 22 18:44:50     export_import=check_export_import)
Apr 22 18:44:50   File "test_jit.py", line 583, in checkTrace
Apr 22 18:44:50     allow_unused=allow_unused)
Apr 22 18:44:50   File "/opt/conda/lib/python2.7/site-packages/torch/autograd/__init__.py", line 149, in grad
Apr 22 18:44:50     inputs, allow_unused)
Apr 22 18:44:50 RuntimeError: error in dlopen or dlsym: /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1x;
Apr 22 18:44:50 
Apr 22 18:44:50 /libthnvrtc.so: cannot open shared object file: No such file or directory (checkDL at /var/lib/jenkins/workspace/torch/csrc/jit/fuser/cpu/dynamic_library_unix.cpp:15)
Apr 22 18:44:50 frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0x6a (0x7f8dcbf177ba in /opt/conda/lib/python2.7/site-packages/torch/lib/libc10.so)
Apr 22 18:44:50 frame #1: <unknown function> + 0xc07bbb (0x7f8dcb753bbb in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #2: torch::jit::fuser::cpu::DynamicLibrary::DynamicLibrary(char const*) + 0x20 (0x7f8dcb753c70 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #3: torch::jit::fuser::cuda::loadNVRTC() + 0x9d (0x7f8dcb756c1d in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #4: torch::jit::fuser::cuda::nvrtc() + 0x5c (0x7f8dcb756d4c in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #5: torch::jit::fuser::cuda::FusedKernelCUDA::FusedKernelCUDA(short, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, bool) + 0x3da (0x7f8dcb75756a in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #6: <unknown function> + 0xc0c21d (0x7f8dcb75821d in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #7: std::_Function_handler<std::shared_ptr<torch::jit::fuser::FusedKernel> (short, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, bool), std::shared_ptr<torch::jit::fuser::FusedKernel> (*)(short, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >, bool)>::_M_invoke(std::_Any_data const&, short&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >&&, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >&&, std::vector<torch::jit::fuser::TensorDesc, std::allocator<torch::jit::fuser::TensorDesc> >&&, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >&&, std::vector<torch::jit::fuser::PartitionDesc, std::allocator<torch::jit::fuser::PartitionDesc> >&&, bool&&) + 0x1e5 (0x7f8dcb755fe5 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #8: torch::jit::fuser::compileKernel(torch::jit::fuser::KernelSpec const&, torch::jit::fuser::ArgSpec const&, std::vector<long, std::allocator<long> > const&, c10::Device) + 0x15b6 (0x7f8dcb6d3176 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #9: torch::jit::fuser::runFusion(long, std::vector<c10::IValue, std::allocator<c10::IValue> >&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >*) + 0x858 (0x7f8dcb6dc408 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #10: torch::jit::runFusion(long, std::vector<c10::IValue, std::allocator<c10::IValue> >&) + 0x13 (0x7f8dcb5f6813 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #11: <unknown function> + 0xab0860 (0x7f8dcb5fc860 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #12: <unknown function> + 0x9eb58f (0x7f8dcb53758f in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #13: torch::jit::InterpreterState::run(std::vector<c10::IValue, std::allocator<c10::IValue> >&) + 0x31 (0x7f8dcb531f51 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #14: <unknown function> + 0x9cfdc6 (0x7f8dcb51bdc6 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #15: <unknown function> + 0x9ca15e (0x7f8dcb51615e in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #16: <unknown function> + 0x2a607a (0x7f8dcadf207a in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #17: torch::autograd::Engine::evaluate_function(torch::autograd::FunctionTask&) + 0x1b92 (0x7f8dcadeb312 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #18: torch::autograd::Engine::thread_main(torch::autograd::GraphTask*) + 0xea (0x7f8dcadebdaa in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #19: torch::autograd::Engine::thread_init(int) + 0x2e6 (0x7f8dcade8656 in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch.so.1)
Apr 22 18:44:50 frame #20: torch::autograd::python::PythonEngine::thread_init(int) + 0x2a (0x7f8dd7e8937a in /opt/conda/lib/python2.7/site-packages/torch/lib/libtorch_python.so)
Apr 22 18:44:50 frame #21: <unknown function> + 0xb8678 (0x7f8dcbe67678 in /opt/conda/lib/libstdc++.so.6)
Apr 22 18:44:50 frame #22: <unknown function> + 0x76ba (0x7f8de6ff56ba in /lib/x86_64-linux-gnu/libpthread.so.0)
Apr 22 18:44:50 frame #23: clone + 0x6d (0x7f8de661b41d in /lib/x86_64-linux-gnu/libc.so.6)
Apr 22 18:44:50 
Apr 22 18:44:50 
Apr 22 18:44:50 ----------------------------------------------------------------------
Apr 22 18:44:50 Ran 1867 tests in 175.057s
Apr 22 18:44:50 
Apr 22 18:44:50 FAILED (errors=1, skipped=22, expected failures=1)
Apr 22 18:44:51 Traceback (most recent call last):
Apr 22 18:44:51   File "test/run_test.py", line 438, in <module>
Apr 22 18:44:51     main()
Apr 22 18:44:51   File "test/run_test.py", line 430, in main
Apr 22 18:44:51     raise RuntimeError(message)
Apr 22 18:44:51 RuntimeError: test_jit failed!

I'm not sure if it's related, but it looks pretty serious. https://circleci.com/gh/pytorch/pytorch/1413659?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link/console

facebook-github-bot · 2019-04-22T19:07:56Z

This pull request has been merged in e714429.

…09c7db (pytorch#19454) Summary: Pull Request resolved: pytorch#19454 Previous import was ad7313470a9119d7e1afda7edf1d654497ee80ab Included changes: - **[83dd6265](onnx/onnx@83dd6265)**: Add NonMaxSuppression operator (pytorch#1703) <Hector Li> - **[31ca5d6f](onnx/onnx@31ca5d6f)**: add node tests for quantized ops (pytorch#1944) <Ashwini Khade> - **[e6076c1d](onnx/onnx@e6076c1d)**: Fix test stat coverage script (pytorch#1948) <Raymond Yang> - **[ad036405](onnx/onnx@ad036405)**: Add IsInf to detect infinity values (pytorch#1884) <Wei-Sheng Chin> Reviewed By: benoitsteiner Differential Revision: D15010015 fbshipit-source-id: 4b29de21de60f8e6a2db75309809a4e619c92532

houseroad force-pushed the export-D15010015 branch from e2847fd to e425868 Compare April 19, 2019 03:50

houseroad force-pushed the export-D15010015 branch from e425868 to faf4fc3 Compare April 19, 2019 05:39

houseroad force-pushed the export-D15010015 branch from faf4fc3 to 9c8e2d6 Compare April 19, 2019 05:40

bddppq self-requested a review April 19, 2019 06:05

bddppq approved these changes Apr 19, 2019

View reviewed changes

facebook-github-bot closed this in e714429 Apr 22, 2019

facebook-github-bot added the merged label Apr 22, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Automatic update of fbcode/onnx to 83dd62659fc07d5b7fa93b5d1c1879f93509c7db#19454

Automatic update of fbcode/onnx to 83dd62659fc07d5b7fa93b5d1c1879f93509c7db#19454
houseroad wants to merge 1 commit intopytorch:masterfrom
houseroad:export-D15010015

houseroad commented Apr 19, 2019

Uh oh!

ezyang commented Apr 22, 2019

Uh oh!

facebook-github-bot commented Apr 22, 2019

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

houseroad commented Apr 19, 2019

Uh oh!

ezyang commented Apr 22, 2019

Uh oh!

facebook-github-bot commented Apr 22, 2019

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants