223 lines
7.5 KiB
C++
223 lines
7.5 KiB
C++
// Copyright (c) 2025 PaddlePaddle 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.
|
||
|
||
#ifdef PADDLE_WITH_XPU_FFT
|
||
#include "paddle/phi/kernels/complex_kernel.h"
|
||
#include "fft/cuComplex.h"
|
||
#include "paddle/phi/backends/xpu/enforce_xpu.h"
|
||
#include "paddle/phi/common/type_traits.h"
|
||
#include "paddle/phi/core/kernel_registry.h"
|
||
#include "paddle/phi/kernels/expand_kernel.h"
|
||
#include "paddle/phi/kernels/funcs/common_infer_shape_functions.h"
|
||
#include "paddle/phi/kernels/xpu/xpu_api_wrapper.h"
|
||
|
||
namespace xfft_internal::xpu {
|
||
// just for declaration here, the real implementation is in libcufft.so
|
||
template <typename T, typename TComplex>
|
||
int combine_as_complex(
|
||
const XPUStream stream, int N, const T* real, const T* imag, TComplex* out);
|
||
template <>
|
||
int combine_as_complex(const XPUStream stream,
|
||
int N,
|
||
const float* real,
|
||
const float* imag,
|
||
float2* out);
|
||
template <>
|
||
int combine_as_complex(const XPUStream stream,
|
||
int N,
|
||
const double* real,
|
||
const double* imag,
|
||
double2* out);
|
||
|
||
template <typename TComplex, typename T>
|
||
int complex_spilt(
|
||
const XPUStream stream, int N, const TComplex* in, T* real, T* imag);
|
||
template <>
|
||
int complex_spilt(
|
||
const XPUStream stream, int N, const float2* in, float* real, float* imag);
|
||
template <>
|
||
int complex_spilt(const XPUStream stream,
|
||
int N,
|
||
const double2* in,
|
||
double* real,
|
||
double* imag);
|
||
|
||
template <typename T> // T supports float2, double2
|
||
int Conj(const XPUStream stream, int N, const T* input, T* output);
|
||
} // namespace xfft_internal::xpu
|
||
|
||
namespace phi {
|
||
template <typename T, typename Context>
|
||
void ConjKernel(const Context& dev_ctx,
|
||
const DenseTensor& x,
|
||
DenseTensor* out) {
|
||
if (out->numel() == 0) {
|
||
dev_ctx.template Alloc<T>(out);
|
||
return;
|
||
}
|
||
dev_ctx.template Alloc<T>(out);
|
||
if (std::is_same_v<T, phi::complex64>) {
|
||
int r = xfft_internal::xpu::Conj(
|
||
dev_ctx.x_context()->xpu_stream,
|
||
x.numel(),
|
||
reinterpret_cast<const cuFloatComplex*>(x.data<T>()),
|
||
reinterpret_cast<cuFloatComplex*>(out->data<T>()));
|
||
PADDLE_ENFORCE_XPU_SUCCESS(r);
|
||
} else if (std::is_same_v<T, phi::complex128>) {
|
||
int r = xfft_internal::xpu::Conj(
|
||
dev_ctx.x_context()->xpu_stream,
|
||
x.numel(),
|
||
reinterpret_cast<const cuDoubleComplex*>(x.data<T>()),
|
||
reinterpret_cast<cuDoubleComplex*>(out->data<T>()));
|
||
PADDLE_ENFORCE_XPU_SUCCESS(r);
|
||
} else {
|
||
using XPUType = typename XPUCopyTypeTrait<T>::Type;
|
||
const auto* input_data = x.data<T>();
|
||
int r = xpu::copy<XPUType>(dev_ctx.x_context(),
|
||
reinterpret_cast<const XPUType*>(input_data),
|
||
reinterpret_cast<XPUType*>(out->data<T>()),
|
||
x.numel());
|
||
PADDLE_ENFORCE_XDNN_SUCCESS(r, "copy");
|
||
}
|
||
}
|
||
|
||
template <typename T, typename Context>
|
||
void RealKernel(const Context& dev_ctx,
|
||
const DenseTensor& x,
|
||
DenseTensor* out) {
|
||
using XPUComplexType =
|
||
typename XPUComplexTypeTrait<phi::dtype::Real<T>>::Type;
|
||
if (out->numel() == 0) {
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(out);
|
||
return;
|
||
}
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(out);
|
||
// The allocation of imag here is redundant and could be optimized.
|
||
DenseTensor imag;
|
||
imag.Resize(x.dims());
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(&imag);
|
||
int r = xfft_internal::xpu::complex_spilt(
|
||
dev_ctx.x_context()->xpu_stream,
|
||
out->numel(),
|
||
reinterpret_cast<const XPUComplexType*>(x.data<T>()),
|
||
out->data<phi::dtype::Real<T>>(),
|
||
imag.data<phi::dtype::Real<T>>());
|
||
PADDLE_ENFORCE_XPU_SUCCESS(r);
|
||
}
|
||
|
||
template <typename T, typename Context>
|
||
void ImagKernel(const Context& dev_ctx,
|
||
const DenseTensor& x,
|
||
DenseTensor* out) {
|
||
using XPUComplexType =
|
||
typename XPUComplexTypeTrait<phi::dtype::Real<T>>::Type;
|
||
if (out->numel() == 0) {
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(out);
|
||
return;
|
||
}
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(out);
|
||
// The allocation of ‘real’ here is redundant and could be optimized.
|
||
DenseTensor real;
|
||
real.Resize(x.dims());
|
||
dev_ctx.template Alloc<phi::dtype::Real<T>>(&real);
|
||
int r = xfft_internal::xpu::complex_spilt(
|
||
dev_ctx.x_context()->xpu_stream,
|
||
out->numel(),
|
||
reinterpret_cast<const XPUComplexType*>(x.data<T>()),
|
||
real.data<phi::dtype::Real<T>>(),
|
||
out->data<phi::dtype::Real<T>>());
|
||
PADDLE_ENFORCE_XPU_SUCCESS(r);
|
||
}
|
||
|
||
template <typename T, typename Context>
|
||
void ComplexKernel(const Context& dev_ctx,
|
||
const DenseTensor& x,
|
||
const DenseTensor& y,
|
||
DenseTensor* out) {
|
||
using C = phi::dtype::complex<T>;
|
||
using XPUComplexType = typename XPUComplexTypeTrait<T>::Type;
|
||
if (out->numel() == 0) {
|
||
dev_ctx.template Alloc<C>(out);
|
||
return;
|
||
}
|
||
auto x_dims = x.dims();
|
||
auto y_dims = y.dims();
|
||
auto out_dims = funcs::BroadcastTwoDims(x_dims, y_dims);
|
||
std::vector<int64_t> out_dims_vec = vectorize(out_dims);
|
||
|
||
DenseTensor broadcasted_x, broadcasted_y;
|
||
const T* x_data = nullptr;
|
||
const T* y_data = nullptr;
|
||
|
||
if (x_dims == out_dims) {
|
||
x_data = x.data<T>();
|
||
} else {
|
||
broadcasted_x.Resize(out_dims);
|
||
dev_ctx.template Alloc<T>(&broadcasted_x);
|
||
ExpandKernel<T, Context>(
|
||
dev_ctx, x, phi::IntArray(out_dims_vec), &broadcasted_x);
|
||
x_data = broadcasted_x.data<T>();
|
||
}
|
||
|
||
if (y_dims == out_dims) {
|
||
y_data = y.data<T>();
|
||
} else {
|
||
broadcasted_y.Resize(out_dims);
|
||
dev_ctx.template Alloc<T>(&broadcasted_y);
|
||
ExpandKernel<T, Context>(
|
||
dev_ctx, y, phi::IntArray(out_dims_vec), &broadcasted_y);
|
||
y_data = broadcasted_y.data<T>();
|
||
}
|
||
|
||
dev_ctx.template Alloc<C>(out);
|
||
int r = xfft_internal::xpu::combine_as_complex(
|
||
dev_ctx.x_context()->xpu_stream,
|
||
out->numel(),
|
||
x_data,
|
||
y_data,
|
||
reinterpret_cast<XPUComplexType*>(out->data<C>()));
|
||
PADDLE_ENFORCE_XPU_SUCCESS(r);
|
||
}
|
||
} // namespace phi
|
||
|
||
PD_REGISTER_KERNEL(conj,
|
||
XPU,
|
||
ALL_LAYOUT,
|
||
phi::ConjKernel,
|
||
bool,
|
||
int,
|
||
int64_t,
|
||
float,
|
||
double,
|
||
phi::float16,
|
||
phi::bfloat16,
|
||
phi::complex64,
|
||
phi::complex128) {}
|
||
|
||
PD_REGISTER_KERNEL(
|
||
real, XPU, ALL_LAYOUT, phi::RealKernel, phi::complex64, phi::complex128) {
|
||
kernel->OutputAt(0).SetDataType(phi::dtype::ToReal(kernel_key.dtype()));
|
||
}
|
||
|
||
PD_REGISTER_KERNEL(
|
||
imag, XPU, ALL_LAYOUT, phi::ImagKernel, phi::complex64, phi::complex128) {
|
||
kernel->OutputAt(0).SetDataType(phi::dtype::ToReal(kernel_key.dtype()));
|
||
}
|
||
|
||
PD_REGISTER_KERNEL(
|
||
complex, XPU, ALL_LAYOUT, phi::ComplexKernel, float, double) {
|
||
kernel->OutputAt(0).SetDataType(phi::dtype::ToComplex(kernel_key.dtype()));
|
||
}
|
||
#endif // PADDLE_WITH_XPU_FFT
|