Add ComplexAbs kernels and tests.
Also add the LowerComplexPass to KernelCreator. This is needed for lowering ComplexAbs. PiperOrigin-RevId: 352388803 Change-Id: I45ddefd239c99f36d62af2f6bcf62f1b062df511
This commit is contained in:
parent
92bc3920b9
commit
e19e3d081a
tensorflow
@ -115,6 +115,7 @@ Status LowerTFtoGPU(mlir::ModuleOp module, llvm::ArrayRef<uint32_t> tile_sizes,
|
||||
/*allow_partial_conversion=*/false, /*legalize_chlo=*/false));
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::createTransformUnrankedHloPass());
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::mhlo::createChloLegalizeToHloPass());
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::mhlo::createLowerComplexPass());
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::createCanonicalizerPass());
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::createCSEPass());
|
||||
pm.addNestedPass<mlir::FuncOp>(mlir::createCanonicalizerPass());
|
||||
|
@ -24,7 +24,10 @@ REGISTER2(UnaryOp, CPU, "ComplexAbs", functor::abs, complex64, complex128);
|
||||
#ifndef MLIR_GENERATED_GPU_KERNELS_ENABLED
|
||||
REGISTER4(UnaryOp, GPU, "Abs", functor::abs, Eigen::half, float, double, int64);
|
||||
#endif
|
||||
#if !defined(MLIR_GENERATED_GPU_KERNELS_ENABLED) || \
|
||||
!defined(MLIR_GENERATED_EXPERIMENTAL_GPU_KERNELS_ENABLED)
|
||||
REGISTER2(UnaryOp, GPU, "ComplexAbs", functor::abs, complex64, complex128);
|
||||
#endif
|
||||
|
||||
// A special GPU kernel for int32.
|
||||
// TODO(b/25387198): Also enable int32 in device memory. This kernel
|
||||
|
@ -52,6 +52,7 @@ filegroup(
|
||||
"gpu_op_atan.cc",
|
||||
"gpu_op_ceil.cc",
|
||||
"gpu_op_complex.cc",
|
||||
"gpu_op_complex_abs.cc",
|
||||
"gpu_op_conj.cc",
|
||||
"gpu_op_cos.cc",
|
||||
"gpu_op_cosh.cc",
|
||||
@ -118,6 +119,7 @@ tf_kernel_library(
|
||||
":asinh_kernels",
|
||||
":atan_kernels",
|
||||
":ceil_kernels",
|
||||
":complex_abs_kernels",
|
||||
":complex_kernels",
|
||||
":conj_kernels",
|
||||
":cos_kernels",
|
||||
@ -465,6 +467,15 @@ gen_kernel_library(
|
||||
unroll_factors = "2",
|
||||
)
|
||||
|
||||
gen_kernel_library(
|
||||
name = "complex_abs",
|
||||
tile_size = "256",
|
||||
types = [
|
||||
"f32",
|
||||
"f64",
|
||||
],
|
||||
)
|
||||
|
||||
gen_kernel_library(
|
||||
name = "div",
|
||||
tile_size = "256,1,1",
|
||||
|
29
tensorflow/core/kernels/mlir_generated/gpu_op_complex_abs.cc
Normal file
29
tensorflow/core/kernels/mlir_generated/gpu_op_complex_abs.cc
Normal file
@ -0,0 +1,29 @@
|
||||
/* Copyright 2021 The TensorFlow Authors. All Rights Reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License");
|
||||
you may not use this file except in compliance with the License.
|
||||
You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software
|
||||
distributed under the License is distributed on an "AS IS" BASIS,
|
||||
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
See the License for the specific language governing permissions and
|
||||
limitations under the License.
|
||||
==============================================================================*/
|
||||
|
||||
#include <complex>
|
||||
|
||||
#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor"
|
||||
#include "tensorflow/core/kernels/mlir_generated/gpu_ops_base.h"
|
||||
|
||||
namespace tensorflow {
|
||||
|
||||
GENERATE_UNARY_KERNEL2(ComplexAbs, f32, DT_FLOAT, float, std::complex<float>);
|
||||
REGISTER_COMPLEX_KERNEL(ComplexAbs, f32, float, std::complex<float>);
|
||||
GENERATE_UNARY_KERNEL2(ComplexAbs, f64, DT_DOUBLE, double,
|
||||
std::complex<double>);
|
||||
REGISTER_COMPLEX_KERNEL(ComplexAbs, f64, double, std::complex<double>);
|
||||
|
||||
} // namespace tensorflow
|
@ -218,6 +218,20 @@ GENERATE_DEFAULT_TEST(Ceil, DT_DOUBLE, DT_DOUBLE, std::ceil,
|
||||
GENERATE_DEFAULT_TEST_2(Ceil, DT_HALF, DT_FLOAT, DT_HALF, DT_FLOAT, std::ceil,
|
||||
test::GpuOpsTestConfig().ExpectStrictlyEqual())
|
||||
|
||||
/// Test `tf.ComplexAbs`.
|
||||
|
||||
template <typename T>
|
||||
typename T::value_type baseline_complex_abs(T x) {
|
||||
return std::abs(x);
|
||||
}
|
||||
|
||||
GENERATE_DEFAULT_TEST(ComplexAbs, DT_COMPLEX64, DT_FLOAT, baseline_complex_abs,
|
||||
test::GpuOpsTestConfig().AddTout().NoBufferReuse())
|
||||
|
||||
GENERATE_DEFAULT_TEST(ComplexAbs, DT_COMPLEX128, DT_DOUBLE,
|
||||
baseline_complex_abs,
|
||||
test::GpuOpsTestConfig().AddTout().NoBufferReuse())
|
||||
|
||||
/// Test `tf.Conj`.
|
||||
|
||||
template <typename T>
|
||||
|
@ -0,0 +1,5 @@
|
||||
func @ComplexAbs_elem_type(%arg0: tensor<*xcomplex<elem_type>>)
|
||||
-> tensor<*xelem_type> attributes {tf_entry, llvm.emit_c_interface} {
|
||||
%0 = "tf.ComplexAbs"(%arg0) : (tensor<*xcomplex<elem_type>>) -> tensor<*xelem_type>
|
||||
return %0 : tensor<*xelem_type>
|
||||
}
|
Loading…
Reference in New Issue
Block a user