[PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

cccclai · 2021-08-30T00:28:53Z

Stack from ghstack:

-> [PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

As title, introduce the file TracerRunner shared by internal/external tracer and the main function is

TracerResult trace_run(const std::string& input_module_path);

which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is

the dependency on <yaml-cpp/yaml.h>.
the output yaml file from internal tracer includes model_version and model_asset. These are only needed for internal.

Differential Revision: D30632224

NOTE FOR REVIEWERS: This PR has internal Facebook specific changes or comments, please review them on Phabricator!

Differential Revision: D30632224

Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! [ghstack-poisoned]

facebook-github-bot · 2021-08-30T00:28:56Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/64152
📄 Preview docs built from this PR
📄 Preview C++ docs built from this PR
↩️ [fb-only] Re-run with SSH instructions
🔧 Opt-in to CIFlow to control what jobs run on your PRs

💊 CI failures summary and remediations

As of commit 55705d3 (more details on the Dr. CI page):

1/1 failures introduced in this PR

🕵️ 1 new failure recognized by patterns

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

win-vs2019-cuda11.3-py3 / build (1/1)

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

2021-10-15T04:31:09.0401002Z Catastrophic error: out of memory

2021-10-15T04:31:09.0393234Z C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\include\cub\device/device_segmented_radix_sort.cuh(366): here
2021-10-15T04:31:09.0394786Z             instantiation of "cudaError_t at::cuda::detail::cub::DeviceSegmentedRadixSort::SortPairsDescending(void *, size_t &, const KeyT *, KeyT *, const ValueT *, ValueT *, int, int, OffsetIteratorT, OffsetIteratorT, int, int, cudaStream_t, __nv_bool) [with KeyT=c10::BFloat16, ValueT=int64_t, OffsetIteratorT=at::native::<unnamed>::offset_t]" 
2021-10-15T04:31:09.0396176Z C:/actions-runner/_work/pytorch/pytorch/aten/src\ATen/cuda/cub.cuh(186): here
2021-10-15T04:31:09.0397335Z             instantiation of "void at::cuda::cub::segmented_sort_pairs(const key_t *, key_t *, const value_t *, value_t *, int64_t, int64_t, OffsetIteratorT, OffsetIteratorT, __nv_bool, int64_t, int64_t) [with key_t=c10::BFloat16, value_t=int64_t, OffsetIteratorT=at::native::<unnamed>::offset_t]" 
2021-10-15T04:31:09.0398457Z C:\actions-runner\_work\pytorch\pytorch\aten\src\ATen\native\cuda\Sort.cu(424): here
2021-10-15T04:31:09.0398847Z 
2021-10-15T04:31:09.0399422Z 1 catastrophic error detected in the compilation of "C:/actions-runner/_work/pytorch/pytorch/aten/src/ATen/native/cuda/Sort.cu".
2021-10-15T04:31:09.0400077Z Compilation terminated.
2021-10-15T04:31:09.0400381Z Sort.cu
2021-10-15T04:31:09.0400658Z Retry attempt: 2
2021-10-15T04:31:09.0401002Z Catastrophic error: out of memory
2021-10-15T04:31:09.0401251Z 
2021-10-15T04:31:09.0401838Z 1 catastrophic error detected in the compilation of "C:/actions-runner/_work/pytorch/pytorch/aten/src/ATen/native/cuda/Sort.cu".
2021-10-15T04:31:09.0402473Z Compilation terminated.
2021-10-15T04:31:09.0402791Z Sort.cu
2021-10-15T04:31:09.0403068Z Retry attempt: 3
2021-10-15T04:31:09.0403373Z Sort.cu
2021-10-15T04:31:14.5678226Z [5650/6354] C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\bin\randomtemp.exe C:/actions-runner/_work/pytorch/pytorch/build/win_tmp\bin\sccache.exe C:\PROGRA~1\NVIDIA~2\CUDA\v11.3\bin\nvcc.exe -forward-unknown-to-host-compiler -DBUILD_SPLIT_CUDA -DIDEEP_USE_MKL -DMAGMA_V2 -DMINIZ_DISABLE_ZIP_READER_CRC32_CHECKS -DONNXIFI_ENABLE_EXT=1 -DONNX_ML=1 -DONNX_NAMESPACE=onnx_torch -DTH_BLAS_MKL -DTORCH_CUDA_CU_BUILD_MAIN_LIB -DUSE_C10D_GLOO -DUSE_CUDA -DUSE_DISTRIBUTED -DUSE_EXTERNAL_MZCRC -DWIN32_LEAN_AND_MEAN -D_CRT_SECURE_NO_DEPRECATE=1 -D_OPENMP_NOFORCE_MANIFEST -Dtorch_cuda_cu_EXPORTS -IC:\actions-runner\_work\pytorch\pytorch\build\aten\src -IC:\actions-runner\_work\pytorch\pytorch\aten\src -IC:\actions-runner\_work\pytorch\pytorch\build -IC:\actions-runner\_work\pytorch\pytorch -IC:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\benchmark\include -IC:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\cudnn_frontend\include -IC:\actions-runner\_work\pytorch\pytorch\build\caffe2\contrib\aten -IC:\actions-runner\_work\pytorch\pytorch\third_party\onnx -IC:\actions-runner\_work\pytorch\pytorch\build\third_party\onnx -IC:\actions-runner\_work\pytorch\pytorch\third_party\foxi -IC:\actions-runner\_work\pytorch\pytorch\build\third_party\foxi -IC:\actions-runner\_work\pytorch\pytorch\build\include -IC:\actions-runner\_work\pytorch\pytorch\build\caffe2\aten\src\TH -IC:\actions-runner\_work\pytorch\pytorch\aten\src\TH -IC:\actions-runner\_work\pytorch\pytorch\aten\src\THC -IC:\actions-runner\_work\pytorch\pytorch\aten\src\ATen\cuda -IC:\actions-runner\_work\pytorch\pytorch\build\caffe2\aten\src -IC:\actions-runner\_work\pytorch\pytorch\aten\..\third_party\catch\single_include -IC:\actions-runner\_work\pytorch\pytorch\aten\src\ATen\.. -IC:\actions-runner\_work\pytorch\pytorch\build\caffe2\aten\src\ATen -IC:\actions-runner\_work\pytorch\pytorch\c10\cuda\..\.. -IC:\actions-runner\_work\pytorch\pytorch\c10\.. -IC:\actions-runner\_work\pytorch\pytorch\torch\csrc\api -IC:\actions-runner\_work\pytorch\pytorch\torch\csrc\api\include -IC:\actions-runner\_work\pytorch\pytorch\build\third_party\ideep\mkl-dnn\include -IC:\actions-runner\_work\pytorch\pytorch\third_party\ideep\mkl-dnn\src\..\include -isystem=C:\actions-runner\_work\pytorch\pytorch\build\third_party\gloo -isystem=C:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\gloo -isystem=C:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\googletest\googlemock\include -isystem=C:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\googletest\googletest\include -isystem=C:\actions-runner\_work\pytorch\pytorch\third_party\protobuf\src -isystem=C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\mkl\include -isystem=C:\actions-runner\_work\pytorch\pytorch\third_party\XNNPACK\include -isystem=C:\actions-runner\_work\pytorch\pytorch\third_party -isystem=C:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\eigen -isystem=C:\Jenkins\Miniconda3\include -isystem=C:\Jenkins\Miniconda3\lib\site-packages\numpy\core\include -isystem=C:\actions-runner\_work\pytorch\pytorch\cmake\..\third_party\pybind11\include -isystem="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\include" -isystem=C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\magma\include -isystem=C:\actions-runner\_work\pytorch\pytorch\third_party\ideep\mkl-dnn\include -isystem=C:\actions-runner\_work\pytorch\pytorch\third_party\ideep\include -isystem="C:\Program Files\NVIDIA Corporation\NvToolsExt\include" -Xcompiler /w -w -Xfatbin -compress-all -DONNX_NAMESPACE=onnx_torch --use-local-env -gencode arch=compute_70,code=sm_70 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=integer_sign_change,--diag_suppress=useless_using_declaration,--diag_suppress=set_but_not_used,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=implicit_return_from_non_void_function,--diag_suppress=unsigned_compare_with_zero,--diag_suppress=declared_but_not_referenced,--diag_suppress=bad_friend_decl --Werror cross-execution-space-call --no-host-device-move-forward --expt-relaxed-constexpr --expt-extended-lambda  -Xcompiler=/wd4819,/wd4503,/wd4190,/wd4244,/wd4251,/wd4275,/wd4522 -Wno-deprecated-gpu-targets --expt-extended-lambda -DCUDA_HAS_FP16=1 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -Xcompiler="-MD -O2 -Ob2" -DNDEBUG -Xcompiler /MD -DCAFFE2_USE_GLOO -DCUDA_HAS_FP16=1 -DUSE_GCC_GET_CPUID -DUSE_AVX -DUSE_AVX2 -DTH_HAVE_THREAD -Xcompiler= -DTORCH_CUDA_CU_BUILD_MAIN_LIB -std=c++14 -MD -MT caffe2\CMakeFiles\torch_cuda_cu.dir\__\aten\src\ATen\native\cuda\ZetaKernel.cu.obj -MF caffe2\CMakeFiles\torch_cuda_cu.dir\__\aten\src\ATen\native\cuda\ZetaKernel.cu.obj.d -x cu -c C:\actions-runner\_work\pytorch\
2021-10-15T04:31:14.5693318Z LLVM ERROR: out of memory
2021-10-15T04:31:14.5693663Z ZetaKernel.cu
2021-10-15T04:31:14.5694025Z nvcc error   : 'cicc' died with status 0xC0000409

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 to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! ghstack-source-id: 136951291 Pull Request resolved: #64152

…l and external" Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! [ghstack-poisoned]

Pull Request resolved: #64152 ghstack-source-id: 136954259 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! [ghstack-poisoned]

Pull Request resolved: #64152 ghstack-source-id: 136957213 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! [ghstack-poisoned]

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 137009462 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 137009462 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: pytorch/pytorch#64152 ghstack-source-id: 139900294 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: pytorch/pytorch#64152 ghstack-source-id: 139146459 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140206623 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140269696 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140522807 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140563354 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140662931 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

…l and external" As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)! Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224) [ghstack-poisoned]

As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140692467 Differential Revision: [D30632224](https://our.internmc.facebook.com/intern/diff/D30632224/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D30632224/)!

facebook-github-bot · 2021-10-15T09:21:26Z

This pull request has been merged in 76efbcc.

…nal (#64152) Summary: As title, introduce the file `TracerRunner` shared by internal/external tracer and the main function is ``` TracerResult trace_run(const std::string& input_module_path); ``` which basically takes the path to model file and generate the trace result. The main difference between external tracer and internal tracer is 1. the dependency on `<yaml-cpp/yaml.h>`. 2. the output yaml file from internal tracer includes `model_version` and `model_asset`. These are only needed for internal. Pull Request resolved: #64152 ghstack-source-id: 140692467 Test Plan: ``` ./build/bin/model_tracer --model_input_path "/Users/chenlai/Documents/pytorch/tracing/deeplabv3_scripted_with_bundled_input.ptl" --build_yaml_path "/Users/chenlai/Documents/pytorch/tracing/tmp.yaml" ``` ``` ./fbcode/caffe2/fb/model_tracer/run_model_with_bundled_inputs.sh ~/local/notebooks/prod_models/deeplabv3_scripted_with_bundled_input.ptl ``` have the same operator output selected_operators.yaml (P460296279) selected_mobile_ops.h (P460296258) Reviewed By: dhruvbird Differential Revision: D30632224 fbshipit-source-id: eb0321dbc0f1fcf6d2e05384695eebb59ac04f8c

facebook-github-bot added oncall: jit Add this issue/PR to JIT oncall triage queue cla signed labels Aug 30, 2021

This was referenced Aug 30, 2021

[Pytorch Edge][tracing-based] reorganize model tracer dependency #63421

Closed

[Pytorch Edge][tracing-based] build tracer in OSS #64087

Closed

cccclai requested review from raziel, iseeyuan, JacobSzwejbka, dhruvbird and pavithranrao August 30, 2021 20:18

cccclai added this to the 1.10.0 milestone Aug 30, 2021

cccclai mentioned this pull request Sep 1, 2021

[PyTorch Edge][tracing-based] tracing-based build libtorch_lite #64406

Closed

chenlai added 3 commits October 9, 2021 12:24

facebook-github-bot closed this in 76efbcc Oct 15, 2021

facebook-github-bot added the Merged label Oct 15, 2021

facebook-github-bot deleted the gh/cccclai/108/head branch October 18, 2021 14:43

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

[PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

Uh oh!

cccclai commented Aug 30, 2021 •

edited

Loading

Uh oh!

facebook-github-bot commented Aug 30, 2021 •

edited

Loading

Uh oh!

facebook-github-bot commented Oct 15, 2021

Uh oh!

Uh oh!

[PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

[PyTorch Edge][tracing-based] Unify tracer between internal and external #64152

Uh oh!

Conversation

cccclai commented Aug 30, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Aug 30, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

🕵️ 1 new failure recognized by patterns

win-vs2019-cuda11.3-py3 / build (1/1)

Uh oh!

facebook-github-bot commented Oct 15, 2021

Uh oh!

Uh oh!

cccclai commented Aug 30, 2021 •

edited

Loading

facebook-github-bot commented Aug 30, 2021 •

edited

Loading