Move c10 dispatcher back to ATen/core (#16050)
authorSebastian Messmer <messmer@fb.com>
Thu, 17 Jan 2019 23:47:16 +0000 (15:47 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Thu, 17 Jan 2019 23:56:52 +0000 (15:56 -0800)
commit3e85a2bcbfb1c0275ed61122e6a70a50ba2deece
tree5412d1025171a3c496505a619e7708c0c78fa577
parenta9438ba62f02af5032171c3773bfc54c348de298
Move c10 dispatcher back to ATen/core (#16050)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/16050

The c10 dispatcher will (soon) depend on IValue and IValue can't be moved to c10 yet because it depends on at::Tensor, which depends on legacy Type dispatch and we don't want the legacy dispatch in c10.

So instead, we move the c10 dispatcher back to ATen/core until we can actually move at::Tensor to c10.

Reviewed By: ezyang

Differential Revision: D13684517

fbshipit-source-id: 1125f4254223907c52f96ff73034f6d4ae9fd0a7
70 files changed:
aten/src/ATen/core/CMakeLists.txt
aten/src/ATen/core/dispatch/DeviceId.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/DeviceId.h [moved from c10/core/dispatch/DeviceId.h with 100% similarity]
aten/src/ATen/core/dispatch/DispatchKey.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/DispatchKey.h [moved from c10/core/dispatch/DispatchKey.h with 97% similarity]
aten/src/ATen/core/dispatch/DispatchTable.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/DispatchTable.h [moved from c10/core/dispatch/DispatchTable.h with 99% similarity]
aten/src/ATen/core/dispatch/Dispatcher.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/Dispatcher.h [moved from c10/core/dispatch/Dispatcher.h with 97% similarity]
aten/src/ATen/core/dispatch/KernelRegistration.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/KernelRegistration.h [moved from c10/core/dispatch/KernelRegistration.h with 98% similarity]
aten/src/ATen/core/dispatch/LayoutId.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/LayoutId.h [moved from c10/core/dispatch/LayoutId.h with 100% similarity]
aten/src/ATen/core/dispatch/OpSchema.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/OpSchema.h [moved from c10/core/dispatch/OpSchema.h with 99% similarity]
aten/src/ATen/core/dispatch/OpSchemaRegistration.cpp [new file with mode: 0644]
aten/src/ATen/core/dispatch/OpSchemaRegistration.h [moved from c10/core/dispatch/OpSchemaRegistration.h with 94% similarity]
aten/src/ATen/core/dispatch/OpSchema_test.cpp [moved from c10/test/core/dispatch/OpSchema_test.cpp with 96% similarity]
aten/src/ATen/core/opschema/layer_norm.cpp [new file with mode: 0644]
aten/src/ATen/core/opschema/layer_norm.h [moved from c10/core/opschema/layer_norm.h with 100% similarity]
c10/core/dispatch/DeviceId.cpp [deleted file]
c10/core/dispatch/DispatchKey.cpp [deleted file]
c10/core/dispatch/DispatchTable.cpp [deleted file]
c10/core/dispatch/Dispatcher.cpp [deleted file]
c10/core/dispatch/KernelRegistration.cpp [deleted file]
c10/core/dispatch/LayoutId.cpp [deleted file]
c10/core/dispatch/OpSchema.cpp [deleted file]
c10/core/dispatch/OpSchemaRegistration.cpp [deleted file]
c10/core/opschema/layer_norm.cpp [deleted file]
caffe2/core/operator_c10wrapper.h
caffe2/operators/experimental/c10/cpu/add_cpu.cc
caffe2/operators/experimental/c10/cpu/averaged_loss_cpu.cc
caffe2/operators/experimental/c10/cpu/batch_gather_cpu.cc
caffe2/operators/experimental/c10/cpu/batch_matmul_cpu.cc
caffe2/operators/experimental/c10/cpu/cast_cpu.cc
caffe2/operators/experimental/c10/cpu/concat_cpu.cc
caffe2/operators/experimental/c10/cpu/enforce_finite_cpu.cc
caffe2/operators/experimental/c10/cpu/expand_dims_cpu.cc
caffe2/operators/experimental/c10/cpu/fc_cpu.cc
caffe2/operators/experimental/c10/cpu/filler_cpu.cc
caffe2/operators/experimental/c10/cpu/flatten_cpu.cc
caffe2/operators/experimental/c10/cpu/mul_cpu.cc
caffe2/operators/experimental/c10/cpu/relu_cpu.cc
caffe2/operators/experimental/c10/cpu/sigmoid_cpu.cc
caffe2/operators/experimental/c10/cpu/sigmoid_cross_entropy_with_logits_cpu.cc
caffe2/operators/experimental/c10/cpu/sparse_lengths_sum_cpu.cc
caffe2/operators/experimental/c10/cpu/stop_gradient_cpu.cc
caffe2/operators/experimental/c10/schemas/add.cc
caffe2/operators/experimental/c10/schemas/averaged_loss.cc
caffe2/operators/experimental/c10/schemas/batch_gather.cc
caffe2/operators/experimental/c10/schemas/batch_matmul.cc
caffe2/operators/experimental/c10/schemas/cast.cc
caffe2/operators/experimental/c10/schemas/concat.cc
caffe2/operators/experimental/c10/schemas/concat.h
caffe2/operators/experimental/c10/schemas/enforce_finite.cc
caffe2/operators/experimental/c10/schemas/expand_dims.cc
caffe2/operators/experimental/c10/schemas/fc.cc
caffe2/operators/experimental/c10/schemas/filler.cc
caffe2/operators/experimental/c10/schemas/filler.h
caffe2/operators/experimental/c10/schemas/flatten.cc
caffe2/operators/experimental/c10/schemas/layer_norm.cc
caffe2/operators/experimental/c10/schemas/mul.cc
caffe2/operators/experimental/c10/schemas/relu.cc
caffe2/operators/experimental/c10/schemas/sigmoid.cc
caffe2/operators/experimental/c10/schemas/sigmoid_cross_entropy_with_logits.cc
caffe2/operators/experimental/c10/schemas/sparse_lengths_sum.cc
caffe2/operators/experimental/c10/schemas/stop_gradient.cc
caffe2/operators/layer_norm_op.cc
setup.py
torch/csrc/jit/c10_ops/layer_norm.cpp