OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
533cb9530e
pytorch/torch/csrc/jit/codegen/cuda
History
Jane Xu 533cb9530e Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 
Summary:
Sub-step of my attempt to split up the torch_cuda library, as it is huge. Please look at https://github.com/pytorch/pytorch/issues/49050 for details on the split and which files are in which target.

This PR introduces two new macros for Windows DLL purposes, TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API. Both are defined as TORCH_CUDA_API for the time being.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/50627

Reviewed By: mruberry

Differential Revision: D25955441

Pulled By: janeyx99

fbshipit-source-id: ff226026833b8fb2fb7c77df6f2d6c824f006869
2021-01-21 19:09:11 -08:00
..
docs
runtime [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
tools Replace kernel resource strings with real .cu source files (#48283) 2020-12-02 21:22:29 -08:00
arith.cpp Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
arith.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
codegen.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
codegen.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
compute_at.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
compute_at.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
dispatch.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
dispatch.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
executor_kernel_arg.cpp
executor_kernel_arg.h
executor_launch_params.cpp
executor_launch_params.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
executor_utils.cpp Replace kernel resource strings with real .cu source files (#48283) 2020-12-02 21:22:29 -08:00
executor_utils.h Replace kernel resource strings with real .cu source files (#48283) 2020-12-02 21:22:29 -08:00
executor.cpp Make empty c10-full (#46092) 2020-11-12 17:08:21 -08:00
executor.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
expr_evaluator.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
expr_evaluator.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
fusion.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
fusion.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
graph_fuser.cpp Remove output used only by sizes (#448) (#47665) 2020-12-03 11:14:30 -08:00
index_compute.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
index_compute.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
instrumentation.cpp Reorganize and refine the Windows.h import in C++ files (#48009) 2020-11-20 14:21:09 -08:00
instrumentation.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
interface.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
interface.h adding profile_ivalue (#47666) 2020-12-09 15:29:15 -08:00
ir_all_nodes.h
ir_base_nodes.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
ir_base_nodes.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_cloner.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
ir_cloner.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_graphviz.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
ir_graphviz.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_interface_nodes.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_internal_nodes.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_iostream.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
ir_iostream.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_nodes.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
ir_printer.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
ir_utils.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
iter_visitor.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
iter_visitor.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
kernel_cache.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
kernel_cache.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
kernel_ir_builder.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
kernel_ir_builder.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
kernel_ir_printer.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
kernel_ir_printer.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
kernel_ir.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
kernel_ir.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
kernel.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
kernel.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower_alias_memory.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
lower_alias_memory.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_index.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_index.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower_insert_syncs.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
lower_insert_syncs.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_loops.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
lower_loops.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower_thread_predicate.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_thread_predicate.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower_unroll.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_unroll.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower_utils.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
lower_utils.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower_validation.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
lower_validation.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
lower2device.cpp Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
lower2device.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
manager.cpp [PyTorch][JIT] Skip unnecessary refcounting in TensorType::merge (#47959) 2020-11-18 17:49:16 -08:00
manager.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
mutator.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
mutator.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
parser.cpp Remove output used only by sizes (#448) (#47665) 2020-12-03 11:14:30 -08:00
parser.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
partition.cpp [PyTorch][codemod] Replace immediately-dereferenced expect calls w/expectRef (#50228) 2021-01-13 16:13:55 -08:00
partition.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
predicate_compute.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
predicate_compute.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
register_interface.cpp adding profile_ivalue (#47666) 2020-12-09 15:29:15 -08:00
scheduler.cpp Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
scheduler.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
shape_inference.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
shape_inference.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
tensor_view.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
transform_iter.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
transform_iter.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
transform_replay.cpp [codemod][fbcode/caffe2] Apply clang-format update fixes 2021-01-09 14:37:36 -08:00
transform_replay.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
transform_rfactor.cpp [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00
transform_rfactor.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
type.cpp Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
type.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
utils.h [nvFuser] Switching to CudaFusionGuard from BailOut for nvfuser - update 2 (#46452) 2020-10-19 15:44:31 -07:00