elementwise_kernel.cc 4.3 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
//   Copyright (c) 2022 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.

Y
YuanRisheng 已提交
15
#include "paddle/phi/kernels/elementwise_kernel.h"
16

17 18
#include "paddle/phi/backends/all_context.h"
#include "paddle/phi/core/kernel_registry.h"
19

20
namespace phi {
21

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
template <typename T, typename Context>
void MaximumKernel(const Context& dev_ctx,
                   const DenseTensor& x,
                   const DenseTensor& y,
                   DenseTensor* out) {
  int axis = -1;
  MaximumRawKernel<T>(dev_ctx, x, y, axis, out);
}

template <typename T, typename Context>
void MinimumKernel(const Context& dev_ctx,
                   const DenseTensor& x,
                   const DenseTensor& y,
                   DenseTensor* out) {
  int axis = -1;
  MinimumRawKernel<T>(dev_ctx, x, y, axis, out);
}

template <typename T, typename Context>
void ModuloKernel(const Context& dev_ctx,
                  const DenseTensor& x,
                  const DenseTensor& y,
                  DenseTensor* out) {
  int axis = -1;
  ModuloRawKernel<T>(dev_ctx, x, y, axis, out);
}
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

template <typename T, typename Context>
void FloorDivideKernel(const Context& dev_ctx,
                       const DenseTensor& x,
                       const DenseTensor& y,
                       DenseTensor* out) {
  int axis = -1;
  FloorDivideRawKernel<T>(dev_ctx, x, y, axis, out);
}

template <typename T, typename Context>
void ElementwisePowKernel(const Context& dev_ctx,
                          const DenseTensor& x,
                          const DenseTensor& y,
                          DenseTensor* out) {
  int axis = -1;
  ElementwisePowRawKernel<T>(dev_ctx, x, y, axis, out);
}

67
}  // namespace phi
68

69 70
using complex64 = ::phi::dtype::complex<float>;
using complex128 = ::phi::dtype::complex<double>;
71

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91
PD_REGISTER_KERNEL(maximum,
                   CPU,
                   ALL_LAYOUT,
                   phi::MaximumKernel,
                   float,
                   double,
                   int,
                   int64_t,
                   phi::dtype::bfloat16) {}
PD_REGISTER_KERNEL(minimum,
                   CPU,
                   ALL_LAYOUT,
                   phi::MinimumKernel,
                   float,
                   double,
                   int,
                   int64_t,
                   phi::dtype::bfloat16) {}
PD_REGISTER_KERNEL(
    modulo, CPU, ALL_LAYOUT, phi::ModuloKernel, float, double, int, int64_t) {}
92 93 94 95 96 97 98 99 100 101
PD_REGISTER_KERNEL(
    floor_divide, CPU, ALL_LAYOUT, phi::FloorDivideKernel, int, int64_t) {}
PD_REGISTER_KERNEL(elementwise_pow,
                   CPU,
                   ALL_LAYOUT,
                   phi::ElementwisePowKernel,
                   float,
                   double,
                   int,
                   int64_t) {}
102 103

#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
104

105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
PD_REGISTER_KERNEL(maximum,
                   GPU,
                   ALL_LAYOUT,
                   phi::MaximumKernel,
                   float,
                   double,
                   int,
                   int64_t,
                   phi::dtype::float16,
                   phi::dtype::bfloat16) {}
PD_REGISTER_KERNEL(minimum,
                   GPU,
                   ALL_LAYOUT,
                   phi::MinimumKernel,
                   float,
                   double,
                   int,
                   int64_t,
                   phi::dtype::float16,
                   phi::dtype::bfloat16) {}
PD_REGISTER_KERNEL(
    modulo, GPU, ALL_LAYOUT, phi::ModuloKernel, float, double, int, int64_t) {}
127 128 129 130 131 132 133 134 135 136
PD_REGISTER_KERNEL(
    floor_divide, GPU, ALL_LAYOUT, phi::FloorDivideKernel, int, int64_t) {}
PD_REGISTER_KERNEL(elementwise_pow,
                   GPU,
                   ALL_LAYOUT,
                   phi::ElementwisePowKernel,
                   float,
                   double,
                   int,
                   int64_t) {}
137
#endif