kernel_factory.cc 8.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
//   Copyright (c) 2021 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.

15
#include "paddle/phi/core/kernel_factory.h"
16 17

// See Note [ Why still include the fluid headers? ]
18
#include "paddle/phi/core/enforce.h"
19

20
namespace phi {
21

22 23
const static Kernel empty_kernel;  // NOLINT

24 25 26 27 28 29 30 31 32
uint32_t KernelKey::Hash::operator()(const KernelKey& key) const {
  uint32_t hash_value = 0;
  // |----31-20------|---19-12---|---11-8----|---7-0---|
  // | For extension | DataType | DataLayout | Backend |
  hash_value |= static_cast<uint8_t>(key.backend());
  hash_value |=
      (static_cast<uint8_t>(key.layout()) << KernelKey::kBackendBitLength);
  hash_value |=
      (static_cast<uint16_t>(key.dtype())
33
       << (KernelKey::kBackendBitLength + KernelKey::kDataLayoutBitLength));
34 35 36 37 38 39 40 41
  return hash_value;
}

KernelFactory& KernelFactory::Instance() {
  static KernelFactory g_op_kernel_factory;
  return g_op_kernel_factory;
}

42 43
const Kernel& KernelFactory::SelectKernel(const std::string& kernel_name,
                                          const KernelKey& kernel_key) const {
44 45
  auto iter = kernels_.find(kernel_name);
  if (iter == kernels_.end()) {
46
    return empty_kernel;
47 48 49
  }
  auto kernel_iter = iter->second.find(kernel_key);
  if (kernel_iter == iter->second.end()) {
50
    return empty_kernel;
51 52 53 54
  }
  return kernel_iter->second;
}

55 56
KernelKeyMap KernelFactory::SelectKernelMap(
    const std::string& kernel_name) const {
57 58
  auto iter = kernels_.find(kernel_name);
  if (iter == kernels_.end()) {
59
    return KernelKeyMap();
60 61 62 63
  }
  return iter->second;
}

64 65
bool KernelFactory::HasKernel(const std::string& kernel_name,
                              const KernelKey& kernel_key) const {
66 67 68 69 70 71 72 73 74 75 76 77 78
  auto iter = kernels_.find(kernel_name);
  PADDLE_ENFORCE_NE(
      iter,
      kernels_.end(),
      phi::errors::NotFound("The kernel `%s` is not registered.", kernel_name));

  auto kernel_iter = iter->second.find(kernel_key);
  if (kernel_iter == iter->second.end()) {
    return false;
  }
  return true;
}

79
const Kernel& KernelFactory::SelectKernelOrThrowError(
Z
zyfncg 已提交
80 81
    const std::string& kernel_name,
    const KernelKey& kernel_key,
82
    bool use_gpudnn) const {
83
  auto iter = kernels_.find(kernel_name);
84 85 86 87
  PADDLE_ENFORCE_NE(
      iter,
      kernels_.end(),
      phi::errors::NotFound("The kernel `%s` is not registered.", kernel_name));
88

Z
zyfncg 已提交
89
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
90
  if (use_gpudnn && kernel_key.backend() == Backend::GPU) {
Z
zyfncg 已提交
91 92 93 94 95 96 97 98 99 100 101 102 103 104
    auto kernel_iter = iter->second.find(
        {Backend::GPUDNN, kernel_key.layout(), kernel_key.dtype()});
    if (kernel_iter == iter->second.end() &&
        kernel_key.layout() != phi::DataLayout::ALL_LAYOUT) {
      kernel_iter = iter->second.find(
          {Backend::GPUDNN, DataLayout::ALL_LAYOUT, kernel_key.dtype()});
    }
    if (kernel_iter != iter->second.end()) {
      return kernel_iter->second;
    }
    LOG(WARNING) << "The cudnn kernel for [" << kernel_name
                 << "] is not registered.";
  }
#endif
105 106
  auto kernel_iter = iter->second.find(kernel_key);
  // TODO(chenweihang): polish refind impl here
107
  if (kernel_iter == iter->second.end() &&
108 109 110
      kernel_key.layout() != phi::DataLayout::ALL_LAYOUT) {
    phi::KernelKey any_layout_kernel_key(
        kernel_key.backend(), phi::DataLayout::ALL_LAYOUT, kernel_key.dtype());
111 112 113 114 115
    kernel_iter = iter->second.find(any_layout_kernel_key);
  }
  PADDLE_ENFORCE_NE(
      kernel_iter,
      iter->second.end(),
116
      phi::errors::NotFound(
117 118 119 120 121 122 123 124
          "The kernel with key %s of kernel `%s` is not registered.",
          kernel_key,
          kernel_name));

  return kernel_iter->second;
}

const Kernel& KernelFactory::SelectKernelOrThrowError(
Y
YuanRisheng 已提交
125
    const std::string& kernel_name,
126 127 128 129 130 131 132
    Backend backend,
    DataLayout layout,
    DataType dtype) const {
  return SelectKernelOrThrowError(kernel_name,
                                  KernelKey(backend, layout, dtype));
}

133 134 135 136 137 138 139 140 141 142
const KernelArgsDef& KernelFactory::GetFirstKernelArgsDef(
    const std::string& kernel_name) const {
  auto iter = kernels_.find(kernel_name);
  PADDLE_ENFORCE_NE(
      iter,
      kernels_.end(),
      phi::errors::NotFound("The kernel `%s` is not registered.", kernel_name));
  return iter->second.cbegin()->second.args_def();
}

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
std::ostream& operator<<(std::ostream& os, AttributeType attr_type) {
  switch (attr_type) {
    case AttributeType::BOOL:
      os << "bool";
      break;
    case AttributeType::INT32:
      os << "int";
      break;
    case AttributeType::INT64:
      os << "int64_t";
      break;
    case AttributeType::FLOAT32:
      os << "float";
      break;
    case AttributeType::FLOAT64:
      os << "double";
      break;
    case AttributeType::STRING:
      os << "string";
      break;
    case AttributeType::BOOLS:
      os << "vector<bool>";
      break;
    case AttributeType::INT32S:
      os << "vector<int>";
      break;
    case AttributeType::INT64S:
      os << "vector<int64_t>";
      break;
    case AttributeType::FLOAT32S:
      os << "vector<float>";
      break;
    case AttributeType::FLOAT64S:
      os << "vector<double>";
      break;
    case AttributeType::STRINGS:
      os << "vector<string>";
      break;
    case AttributeType::SCALAR:
      os << "Scalar";
      break;
    case AttributeType::SCALARS:
      os << "vector<Scalar>";
      break;
    case AttributeType::INT_ARRAY:
      os << "IntArray";
      break;
    case AttributeType::DATA_TYPE:
      os << "DataType";
      break;
    case AttributeType::DATA_LAYOUT:
      os << "DataLayout";
      break;
    case AttributeType::PLACE:
      os << "Place";
      break;
    default:
      os << "Undefined";
  }
  return os;
}

205 206 207 208 209 210 211
// print kernel info with json format:
// {
//   "(CPU, Undefined(AnyLayout), complex64)": {
//   "input": ["CPU, NCHW, complex64", "CPU, NCHW, complex64"],
//   "output": ["CPU, NCHW, complex64"],
//   "attribute": ["i"]
// }
212
std::ostream& operator<<(std::ostream& os, const Kernel& kernel) {
213 214 215
  // input
  os << "{\"input\":[";
  bool need_comma = false;
216
  for (auto& in_def : kernel.args_def().input_defs()) {
217 218 219 220
    if (need_comma) os << ",";
    os << "\"" << in_def.backend << ", " << in_def.layout << ", "
       << in_def.dtype << "\"";
    need_comma = true;
221
  }
222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
  os << "],";

  // output
  os << "\"output\":[";
  need_comma = false;
  for (auto& out_def : kernel.args_def().output_defs()) {
    if (need_comma) os << ",";
    os << "\"" << out_def.backend << ", " << out_def.layout << ", "
       << out_def.dtype << "\"";
    need_comma = true;
  }
  os << "],";

  // attr
  os << "\"attribute\":[";
  need_comma = false;
  for (auto& arg_def : kernel.args_def().attribute_defs()) {
    if (need_comma) os << ",";
240
    os << "\"" << arg_def.type_index << "\"";
241 242 243 244
    need_comma = true;
  }
  os << "]}";

245 246 247
  return os;
}

248 249 250 251 252 253 254 255 256 257 258 259 260 261 262
// print all kernels info with json format:
// {
//  "kernel_name1":
//      [
//        {
//          "(CPU, Undefined(AnyLayout), complex64)": {
//          "input": ["CPU, NCHW, complex64", "CPU, NCHW, complex64"],
//          "output": ["CPU, NCHW, complex64"],
//          "attribute": ["i"]
//        },
//        ...
//      ],
//    "kernel_name2": []
//    ...
// }
263
std::ostream& operator<<(std::ostream& os, KernelFactory& kernel_factory) {
264 265
  os << "{";
  bool need_comma_kernels = false;
266
  for (const auto& op_kernel_pair : kernel_factory.kernels()) {
267 268 269
    if (need_comma_kernels) os << ",";
    os << "\"" << op_kernel_pair.first << "\":[";
    bool need_comma_per_kernel = false;
270
    for (const auto& kernel_pair : op_kernel_pair.second) {
271 272 273
      if (need_comma_per_kernel) os << ",";
      os << "{\"" << kernel_pair.first << "\":" << kernel_pair.second << "}";
      need_comma_per_kernel = true;
274
    }
275 276
    os << "]";
    need_comma_kernels = true;
277
  }
278 279
  os << "}";

280 281 282
  return os;
}

283
}  // namespace phi