canonicalize includes of form <aten/src/ATen/...>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/78033

This was never intended to be supported.

@override-unit-failures
(Note: this ignores all push blocking failures!)

Differential Revision: [D36567054](https://our.internmc.facebook.com/intern/diff/D36567054/)

Approved by: https://github.com/kit1980

torch/csrc/distributed/autograd/utils.cpp

@@ -1,4 +1,4 @@
-#include <aten/src/ATen/ThreadLocalState.h>
+#include <ATen/ThreadLocalState.h>
 #include <c10/util/ThreadLocalDebugInfo.h>
 #include <torch/csrc/autograd/functions/utils.h>
 #include <torch/csrc/autograd/profiler.h>

torch/csrc/jit/codegen/cuda/type_inference.cpp

@@ -1,6 +1,6 @@
 #include <torch/csrc/jit/codegen/cuda/type_inference.h>
 
-#include <aten/src/ATen/AccumulateType.h>
+#include <ATen/AccumulateType.h>
 #include <c10/core/ScalarType.h>
 #include <torch/csrc/jit/codegen/cuda/instrumentation.h>
 #include <torch/csrc/jit/ir/constants.h>

torch/csrc/jit/cuda/cuda.h

@@ -1,4 +1,4 @@
-#include <aten/src/ATen/cuda/CUDAEvent.h>
+#include <ATen/cuda/CUDAEvent.h>
 #include <c10/core/Device.h>
 #include <c10/cuda/CUDAStream.h>
 #include <torch/custom_class.h>

torch/csrc/jit/passes/onnx/fixup_onnx_controlflow.cpp

@@ -1,6 +1,6 @@
 #include <torch/csrc/jit/passes/onnx/fixup_onnx_controlflow.h>
 
-#include <aten/src/ATen/InitialTensorOptions.h>
+#include <ATen/InitialTensorOptions.h>
 #include <c10/util/irange.h>
 #include <torch/csrc/jit/jit_log.h>
 #include <torch/csrc/jit/passes/dead_code_elimination.h>

torch/csrc/jit/python/python_ir.cpp

@@ -1,6 +1,6 @@
 #include <torch/csrc/jit/python/python_ir.h>
 
-#include <aten/src/ATen/core/jit_type.h>
+#include <ATen/core/jit_type.h>
 #include <pybind11/pybind11.h>
 #include <torch/csrc/Device.h>
 #include <torch/csrc/Dtype.h>

torch/csrc/jit/runtime/register_special_ops.cpp

@@ -1,8 +1,8 @@
-#include <aten/src/ATen/Context.h>
+#include <ATen/Context.h>
 #include <torch/library.h>
 
+#include <ATen/ExpandUtils.h>
 #include <ATen/core/jit_type.h>
-#include <aten/src/ATen/ExpandUtils.h>
 #include <c10/core/DefaultDtype.h>
 #include <c10/util/irange.h>
 #include <torch/csrc/api/include/torch/utils.h>
@@ -11,7 +11,7 @@
 #include <torch/csrc/jit/runtime/operator.h>
 #include <torch/csrc/jit/runtime/vararg_functions.h>
 
-#include <aten/src/ATen/InitialTensorOptions.h>
+#include <ATen/InitialTensorOptions.h>
 #include <c10/core/ScalarType.h>
 #include <torch/csrc/jit/frontend/error_report.h>
 

torch/csrc/jit/serialization/pickler.cpp

@@ -3,7 +3,7 @@
 #ifdef USE_RPC
 #include <torch/csrc/distributed/rpc/rref_context.h>
 #endif
-#include <aten/src/ATen/quantized/Quantizer.h>
+#include <ATen/quantized/Quantizer.h>
 #include <c10/util/irange.h>
 #include <torch/csrc/jit/api/function_impl.h>
 #include <torch/csrc/jit/serialization/pickler.h>

torch/csrc/jit/tensorexpr/external_functions.cpp

@@ -3,6 +3,7 @@
 #include <ATen/ATen.h>
 #include <ATen/Functions.h>
 #include <ATen/NativeFunctions.h>
+#include <ATen/Parallel.h>
 #include <ATen/core/Tensor.h>
 #include <ATen/native/quantized/PackedParams.h>
 #include <ATen/native/quantized/cpu/BinaryOps.h>
@@ -11,7 +12,6 @@
 #include <ATen/native/quantized/cpu/conv_serialization.h>
 #include <ATen/native/xnnpack/OpContext.h>
 #include <ATen/quantized/QTensorImpl.h>
-#include <aten/src/ATen/Parallel.h>
 #include <c10/core/TensorImpl.h>
 #include <c10/core/TensorOptions.h>
 #include <c10/util/ArrayRef.h>

torch/csrc/jit/tensorexpr/llvm_codegen.cpp

@@ -2,7 +2,7 @@
 
 #include <torch/csrc/jit/tensorexpr/llvm_codegen.h>
 
-#include <aten/src/ATen/Parallel.h>
+#include <ATen/Parallel.h>
 #include <c10/util/Exception.h>
 #include <c10/util/irange.h>
 #include <torch/csrc/jit/tensorexpr/analysis.h>

torch/csrc/lazy/core/shape_inference.cpp

@@ -51,13 +51,13 @@
 
 #include <ATen/AccumulateType.h>
 #include <ATen/Dispatch.h>
+#include <ATen/ExpandUtils.h>
 #include <ATen/Functions.h>
 #include <ATen/InferSize.h>
 #include <ATen/WrapDimUtils.h>
 #include <ATen/native/ConvUtils.h>
+#include <ATen/native/ReduceOpsUtils.h>
 #include <ATen/native/TensorConversions.h>
-#include <aten/src/ATen/ExpandUtils.h>
-#include <aten/src/ATen/native/ReduceOpsUtils.h>
 #include <c10/core/ScalarType.h>
 #include <torch/csrc/api/include/torch/enum.h>
 #include <torch/csrc/lazy/core/ops/utils.h>

torch/csrc/utils/throughput_benchmark-inl.h

@@ -7,7 +7,7 @@
 #include <torch/csrc/jit/python/pybind_utils.h>
 #include <torch/csrc/utils/pybind.h>
 
-#include <aten/src/ATen/Parallel.h>
+#include <ATen/Parallel.h>
 #include <c10/util/irange.h>
 
 namespace torch {