infershape_utils.cc

/* 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. */

#include "paddle/fluid/framework/infershape_utils.h"

#include <algorithm>
#include <string>

#include "paddle/fluid/framework/convert_utils.h"
#include "paddle/fluid/framework/framework.pb.h"
#include "paddle/fluid/framework/phi_utils.h"
#include "paddle/fluid/platform/enforce.h"
#include "paddle/phi/common/int_array.h"
#include "paddle/phi/common/scalar.h"
#include "paddle/phi/core/compat/arg_map_context.h"
#include "paddle/phi/core/compat/convert_utils.h"
#include "paddle/phi/core/compat/op_utils.h"
#include "paddle/phi/core/dense_tensor.h"
#include "paddle/phi/core/infermeta_utils.h"
#include "paddle/phi/core/kernel_factory.h"
#include "paddle/phi/core/tensor_utils.h"

namespace paddle {
namespace framework {

class InferShapeArgumentMappingContext : public phi::ArgumentMappingContext {
 public:
  explicit InferShapeArgumentMappingContext(const InferShapeContext& ctx)
      : ctx_(ctx) {}

  bool HasInput(const std::string& name) const override {
    return ctx_.HasInput(name);
  }

  bool HasOutput(const std::string& name) const override {
    return ctx_.HasOutput(name);
  }

  bool HasAttr(const std::string& name) const override {
    return ctx_.HasAttr(name);
  }

  paddle::any Attr(const std::string& name) const override {
    auto* attr = ctx_.Attrs().GetAttr(name);
    PADDLE_ENFORCE_NOT_NULL(
        attr,
        platform::errors::NotFound("Attribute (%s) should be in AttributeMap.",
                                   name));
    return GetAttrValue(*attr);
  }

  size_t InputSize(const std::string& name) const override {
    if (ctx_.HasInputs(name)) {
      return ctx_.Inputs(name).size();
    } else if (ctx_.HasInput(name)) {
      return 1;
    }
    return 0;
  }

  size_t OutputSize(const std::string& name) const override {
    return ctx_.Outputs(name).size();
  }

  bool IsDenseTensorInput(const std::string& name) const override {
    auto var_type = ctx_.GetInputVarType(name);
    return var_type == proto::VarType::LOD_TENSOR;
  }

  bool IsDenseTensorInputs(const std::string& name) const override {
    auto var_types = ctx_.GetInputsVarType(name);
    return std::all_of(var_types.begin(),
                       var_types.end(),
                       [](const proto::VarType::Type& type) {
                         return type == proto::VarType::LOD_TENSOR;
                       });
  }

  bool IsSelectedRowsInput(const std::string& name) const override {
    auto var_type = ctx_.GetInputVarType(name);
    return var_type == proto::VarType::SELECTED_ROWS;
  }

  bool IsDenseTensorVectorInput(const std::string& name) const override {
    auto var_types = ctx_.GetInputsVarType(name);
    return std::all_of(var_types.begin(),
                       var_types.end(),
                       [](const proto::VarType::Type& type) {
                         return type == proto::VarType::LOD_TENSOR_ARRAY;
                       });
  }

  bool IsDenseTensorOutput(const std::string& name) const override {
    auto var_types = ctx_.GetOutputsVarType(name);
    return std::all_of(var_types.begin(),
                       var_types.end(),
                       [](const proto::VarType::Type& type) {
                         return type == proto::VarType::LOD_TENSOR;
                       });
  }

  bool IsSelectedRowsOutput(const std::string& name) const override {
    auto var_types = ctx_.GetOutputsVarType(name);
    return std::all_of(var_types.begin(),
                       var_types.end(),
                       [](const proto::VarType::Type& type) {
                         return type == proto::VarType::SELECTED_ROWS;
                       });
  }

  bool IsForInferShape() const override { return true; }

  bool IsRuntime() const override { return ctx_.IsRuntime(); }

 private:
  const InferShapeContext& ctx_;
};

int64_t CompatMetaTensor::numel() const {
  if (is_runtime_) {
    auto* var = PADDLE_GET_CONST(Variable*, var_);
    return var->Get<Tensor>().numel();
  } else {
    auto* var = PADDLE_GET_CONST(VarDesc*, var_);
    return var->ElementSize();
  }
}

DDim CompatMetaTensor::dims() const {
  if (is_runtime_) {
    auto* var = PADDLE_GET_CONST(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      return var->Get<phi::DenseTensor>().dims();
    } else if (var->IsType<phi::SelectedRows>()) {
      return var->Get<phi::SelectedRows>().dims();
    } else if (var->IsType<framework::LoDTensorArray>()) {
      // use tensor array size as dims
      auto& tensor_array = var->Get<framework::LoDTensorArray>();
      return phi::make_ddim({static_cast<int64_t>(tensor_array.size())});
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can get dims from DenseTensor or SelectedRows or "
          "DenseTensorArray."));
    }
  } else {
    auto* var = PADDLE_GET_CONST(VarDesc*, var_);

    return var->GetShape().empty() ? phi::make_ddim({0UL})
                                   : phi::make_ddim(var->GetShape());
  }
}

phi::DataType CompatMetaTensor::dtype() const {
  if (is_runtime_) {
    auto* var = PADDLE_GET_CONST(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      return var->Get<phi::DenseTensor>().dtype();
    } else if (var->IsType<phi::SelectedRows>()) {
      return var->Get<phi::SelectedRows>().dtype();
    } else if (var->IsType<framework::LoDTensorArray>()) {
      // NOTE(chenweihang): do nothing
      // Unsupported get dtype from LoDTensorArray now
      return phi::DataType::UNDEFINED;
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can get dtype from DenseTensor or SelectedRows."));
    }
  } else {
    auto* var = PADDLE_GET_CONST(VarDesc*, var_);
    return paddle::framework::TransToPhiDataType(var->GetDataType());
  }
}

DataLayout CompatMetaTensor::layout() const {
  if (is_runtime_) {
    auto* var = PADDLE_GET_CONST(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      return var->Get<phi::DenseTensor>().layout();
    } else if (var->IsType<phi::SelectedRows>()) {
      return var->Get<phi::SelectedRows>().layout();
    } else if (var->IsType<framework::LoDTensorArray>()) {
      // NOTE(chenweihang): do nothing
      // Unsupported get layout from LoDTensorArray now
      return phi::DataLayout::UNDEFINED;
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can get layout from DenseTensor or "
          "SelectedRows."));
    }
  } else {
    // NOTE(chenweihang): do nothing
    // Unsupported get layout for VarDesc now
    return DataLayout::UNDEFINED;
  }
}

void CompatMetaTensor::set_dims(const DDim& dims) {
  if (is_runtime_) {
    auto* var = PADDLE_GET(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      auto* tensor = var->GetMutable<phi::DenseTensor>();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->dims = dims;
    } else if (var->IsType<phi::SelectedRows>()) {
      auto* tensor = var->GetMutable<phi::SelectedRows>()->mutable_value();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->dims = dims;
    } else if (var->IsType<framework::LoDTensorArray>()) {
      auto* tensor_array = var->GetMutable<framework::LoDTensorArray>();
      // Note: Here I want enforce `tensor_array->size() == 0UL`, because
      // inplace using on LoDTensorArray is dangerous, but the unittest
      // `test_list` contains this behavior
      PADDLE_ENFORCE_EQ(dims.size(),
                        1UL,
                        platform::errors::InvalidArgument(
                            "LoDTensorArray can only have one dimension."));
      // only set the array size for LoDTensorArray input
      tensor_array->resize(dims[0]);
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can set dims from DenseTensor or SelectedRows."));
    }
  } else {
    auto* var = PADDLE_GET(VarDesc*, var_);
    var->SetShape(vectorize(dims));
  }
}

void CompatMetaTensor::set_dtype(phi::DataType dtype) {
  if (is_runtime_) {
    auto* var = PADDLE_GET(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      auto* tensor = var->GetMutable<phi::DenseTensor>();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->dtype = dtype;
    } else if (var->IsType<phi::SelectedRows>()) {
      auto* tensor = var->GetMutable<phi::SelectedRows>()->mutable_value();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->dtype = dtype;
    } else if (var->IsType<framework::LoDTensorArray>()) {
      // NOTE(chenweihang): do nothing
      // Unsupported set dtype for LoDTensorArray now
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can set dtype from DenseTensor or SelectedRows."));
    }
  } else {
    auto* var = PADDLE_GET(VarDesc*, var_);
    var->SetDataType(paddle::framework::TransToProtoVarType(dtype));
  }
}

void CompatMetaTensor::set_layout(DataLayout layout) {
  if (is_runtime_) {
    auto* var = PADDLE_GET(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      auto* tensor = var->GetMutable<phi::DenseTensor>();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->layout = layout;
    } else if (var->IsType<phi::SelectedRows>()) {
      auto* tensor = var->GetMutable<phi::SelectedRows>()->mutable_value();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->layout = layout;
    } else if (var->IsType<framework::LoDTensorArray>()) {
      // NOTE(chenweihang): do nothing
      // Unsupported set dtype for LoDTensorArray now
    } else {
      PADDLE_THROW(platform::errors::Unimplemented(
          "Currently, only can set layout from DenseTensor or "
          "SelectedRows."));
    }
  } else {
    // NOTE(chenweihang): do nothing
    // Unsupported set layout for VarDesc now
  }
}

void CompatMetaTensor::share_lod(const MetaTensor& meta_tensor) {
  if (is_runtime_) {
    auto* var = PADDLE_GET(Variable*, var_);
    if (var->IsType<phi::DenseTensor>()) {
      auto* tensor = var->GetMutable<phi::DenseTensor>();
      phi::DenseTensorUtils::GetMutableMeta(tensor)->lod =
          static_cast<const CompatMetaTensor&>(meta_tensor).GetRuntimeLoD();
    } else {
      // NOTE(chenweihang): do nothing
      // only LoDTensor need to share lod
    }
  } else {
    auto* var = PADDLE_GET(VarDesc*, var_);
    var->SetLoDLevel(
        static_cast<const CompatMetaTensor&>(meta_tensor).GetCompileTimeLoD());
  }
}

void CompatMetaTensor::share_dims(const MetaTensor& meta_tensor) {
  set_dims(meta_tensor.dims());
  if (is_runtime_) {
    auto* var = PADDLE_GET(Variable*, var_);
    if (var->IsType<phi::SelectedRows>()) {
      auto* selected_rows = var->GetMutable<phi::SelectedRows>();
      auto& input_selected_rows =
          static_cast<const CompatMetaTensor&>(meta_tensor).GetSelectedRows();
      selected_rows->set_rows(input_selected_rows.rows());
      selected_rows->set_height(input_selected_rows.height());
    }
  }
}

void CompatMetaTensor::share_meta(const MetaTensor& meta_tensor) {
  share_dims(meta_tensor);
  set_dtype(meta_tensor.dtype());
  set_layout(meta_tensor.layout());
  // special case: share lod of LoDTensor
  share_lod(meta_tensor);
}

void CompatInferMetaContext::EmplaceBackInput(CompatMetaTensor input) {
  int index = compat_inputs_.size();
  compat_inputs_.emplace_back(std::move(input));
  input_range_.emplace_back(std::pair<int, int>(index, index + 1));
}
void CompatInferMetaContext::EmplaceBackOutput(CompatMetaTensor output) {
  int index = compat_outputs_.size();
  compat_outputs_.emplace_back(std::move(output));
  output_range_.emplace_back(std::pair<int, int>(index, index + 1));
}

void CompatInferMetaContext::EmplaceBackInputs(
    paddle::small_vector<CompatMetaTensor, phi::kInputSmallVectorSize> inputs) {
  int index = compat_inputs_.size();
  input_range_.emplace_back(std::pair<int, int>(index, index + inputs.size()));
  compat_inputs_.insert(compat_inputs_.end(),
                        std::make_move_iterator(inputs.begin()),
                        std::make_move_iterator(inputs.end()));
}

void CompatInferMetaContext::EmplaceBackOutputs(
    paddle::small_vector<CompatMetaTensor, phi::kOutputSmallVectorSize>
        outputs) {
  int index = compat_outputs_.size();
  output_range_.emplace_back(
      std::pair<int, int>(index, index + outputs.size()));
  compat_outputs_.insert(compat_outputs_.end(),
                         std::make_move_iterator(outputs.begin()),
                         std::make_move_iterator(outputs.end()));
}

const phi::MetaTensor& CompatInferMetaContext::InputAt(size_t idx) const {
  return compat_inputs_.at(idx);
}

std::vector<const phi::MetaTensor*> CompatInferMetaContext::InputsBetween(
    size_t start, size_t end) const {
  std::vector<const phi::MetaTensor*> result;
  result.reserve(end - start);

  for (size_t i = start; i < end; ++i) {
    auto& in = compat_inputs_.at(i);
    result.emplace_back(in.initialized() ? &in : nullptr);
  }

  return result;
}

paddle::optional<std::vector<const phi::MetaTensor*>>
CompatInferMetaContext::OptionalInputsBetween(size_t start, size_t end) const {
  const auto& first = compat_inputs_.at(start);

  if (first.initialized()) {
    std::vector<const phi::MetaTensor*> result;
    result.reserve(end - start);

    for (size_t i = start; i < end; ++i) {
      auto& in = compat_inputs_.at(i);
      result.emplace_back(in.initialized() ? &in : nullptr);
    }

    return paddle::optional<std::vector<const phi::MetaTensor*>>(
        std::move(result));
  }
  return paddle::none;
}

phi::MetaTensor* CompatInferMetaContext::MutableOutputAt(size_t idx) {
  auto& out = compat_outputs_.at(idx);
  return out.initialized() ? &out : nullptr;
}

std::vector<phi::MetaTensor*> CompatInferMetaContext::MutableOutputBetween(
    size_t start, size_t end) {
  std::vector<phi::MetaTensor*> result;
  result.reserve(end - start);
  for (size_t i = start; i < end; ++i) {
    auto& out = compat_outputs_.at(i);
    result.emplace_back(out.initialized() ? &out : nullptr);
  }
  return result;
}

CompatInferMetaContext BuildInferMetaContext(InferShapeContext* ctx,
                                             const std::string& op_type) {
  // 1. get kernel args
  auto* arg_map_fn = ctx->GetPhiArgumentMappingFn();
  InferShapeArgumentMappingContext arg_map_context(*ctx);
  phi::KernelSignature signature = arg_map_fn
                                       ? (*arg_map_fn)(arg_map_context)
                                       : *ctx->GetPhiDefaultKernelSignature();
  VLOG(3) << "BuildInferMetaContext: op kernel signature - " << signature;

  // 2. build infermeta context
  CompatInferMetaContext infer_meta_context(
      {ctx->IsRuntime(), ctx->IsRunMKLDNNKernel()});

  const auto& input_names = signature.input_names;
  const auto& attr_names = signature.attr_names;
  const auto& output_names = signature.output_names;

  const auto& args_def =
      phi::KernelFactory::Instance().GetFirstKernelArgsDef(signature.name);
  const auto& attr_defs = args_def.attribute_defs();

  for (auto& in_name : input_names) {
    if (ctx->HasInputs(in_name)) {
      auto input_var = std::move(ctx->GetInputVarPtrs(in_name));
      if (input_var.size() == 1) {
        infer_meta_context.EmplaceBackInput(
            std::move(CompatMetaTensor(input_var[0], ctx->IsRuntime())));
      } else {
        paddle::small_vector<CompatMetaTensor, phi::kInputSmallVectorSize>
            inputs;
        for (const auto& in : input_var) {
          inputs.emplace_back(
              std::move(CompatMetaTensor(in, ctx->IsRuntime())));
        }
        infer_meta_context.EmplaceBackInputs(std::move(inputs));
      }
    } else {
      infer_meta_context.EmplaceBackInput(
          std::move(CompatMetaTensor(ctx->IsRuntime())));
    }
  }

  VLOG(6) << "BuildInferMetaContext: Done inputs";

  auto attr_reader = ctx->Attrs();
  for (size_t i = 0; i < attr_names.size(); ++i) {
    auto& attr_name = attr_names[i];
    VLOG(6) << "BuildInferMetaContext: " << attr_name << ": "
            << attr_defs[i].type_index;
    auto* attr_ptr = attr_reader.GetAttr(attr_name);
    switch (attr_defs[i].type_index) {
      case phi::AttributeType::SCALAR:
        if (attr_ptr) {
          auto& attr = *attr_ptr;
          switch (AttrTypeID(attr)) {
            case framework::proto::AttrType::FLOAT:
              infer_meta_context.EmplaceBackAttr(
                  phi::Scalar(PADDLE_GET_CONST(float, attr)));
              break;
            case framework::proto::AttrType::INT:
              infer_meta_context.EmplaceBackAttr(
                  phi::Scalar(PADDLE_GET_CONST(int, attr)));
              break;
            case framework::proto::AttrType::STRING:
              infer_meta_context.EmplaceBackAttr(
                  phi::Scalar(PADDLE_GET_CONST(std::string, attr)));
              break;
            default:
              PADDLE_THROW(platform::errors::Unimplemented(
                  "Unsupported cast op attribute `%s` to Scalar when construct "
                  "InferMetaContext.",
                  attr_name));
          }
        } else if (ctx->HasInput(attr_name)) {
          auto infershape_input = std::move(ctx->GetInputVarPtrs(attr_name));
          if (infershape_input.size() == 1) {
            if (ctx->IsRuntime()) {
              Variable* var = PADDLE_GET_CONST(Variable*, infershape_input[0]);
              infer_meta_context.EmplaceBackAttr(
                  std::move(experimental::MakePhiScalarFromVar(*var)));
            } else {
              phi::Scalar tensor_scalar(-1);
              tensor_scalar.SetFromTensor(true);
              infer_meta_context.EmplaceBackAttr(std::move(tensor_scalar));
            }
          } else {
            PADDLE_THROW(platform::errors::InvalidArgument(
                "Invalid input.size() when cast op attribute `%s` to Scalar, "
                "expected 1, but actually is %d .",
                attr_name,
                infershape_input.size()));
          }
        } else {
          // do nothing, skip current attr
        }
        break;
      case phi::AttributeType::INT_ARRAY:
        // When attr is a vector_tensor or tensor, transform it to IntArray
        if (attr_ptr) {
          auto& attr = *attr_ptr;
          switch (AttrTypeID(attr)) {
            case framework::proto::AttrType::INTS:
              infer_meta_context.EmplaceBackAttr(std::move(
                  phi::IntArray(PADDLE_GET_CONST(std::vector<int32_t>, attr))));
              break;
            case framework::proto::AttrType::LONGS:
              infer_meta_context.EmplaceBackAttr(std::move(
                  phi::IntArray(PADDLE_GET_CONST(std::vector<int64_t>, attr))));
              break;
            case framework::proto::AttrType::INT:
              infer_meta_context.EmplaceBackAttr(
                  phi::IntArray({PADDLE_GET_CONST(int, attr)}));
              break;
            default:
              PADDLE_THROW(platform::errors::Unimplemented(
                  "Unsupported cast op attribute `%s` to IntArray when "
                  "construct InferMetaContext.",
                  attr_name));
          }
        } else if (ctx->HasInputs(attr_name) || ctx->HasInput(attr_name)) {
          auto infershape_inputs = std::move(ctx->GetInputVarPtrs(attr_name));
          if (ctx->IsRuntime()) {
            // If is in runtime, we will get tensor's value for IntArray
            // and push it into attrs
            std::vector<Variable*> vars;
            vars.reserve(infershape_inputs.size());
            for (size_t i = 0; i < infershape_inputs.size(); i++) {
              vars.push_back(PADDLE_GET_CONST(Variable*, infershape_inputs[i]));
            }
            if (infershape_inputs.size() != 1) {
              infer_meta_context.EmplaceBackAttr(
                  std::move(experimental::MakePhiIntArrayFromVarList(vars)));
            } else {
              infer_meta_context.EmplaceBackAttr(
                  std::move(experimental::MakePhiIntArrayFromVar(*vars[0])));
            }
          } else {
            // If is not in runtime, we will set default value(-1) for IntArray
            std::vector<VarDesc*> vars;
            vars.reserve(infershape_inputs.size());
            for (size_t i = 0; i < infershape_inputs.size(); ++i) {
              vars.push_back(PADDLE_GET_CONST(VarDesc*, infershape_inputs[i]));
            }

            int64_t num_ele = 0;
            if (vars.size() == 1) {
              num_ele = 1;
              const auto& tensor_dims = vars[0]->GetShape();
              for (size_t i = 0; i < tensor_dims.size(); ++i) {
                num_ele *= tensor_dims[i];
              }

              if (num_ele <= 0) {
                num_ele = tensor_dims.size();
              }

            } else {
              num_ele = vars.size();
            }
            phi::IntArray tensor_attr(std::vector<int32_t>(num_ele, -1));
            tensor_attr.SetFromTensor(true);
            infer_meta_context.EmplaceBackAttr(std::move(tensor_attr));
          }
        } else {
          // do nothing, skip current attr
        }
        break;
      case phi::AttributeType::SCALARS:
        if (attr_ptr) {
          auto& attr = *attr_ptr;
          switch (AttrTypeID(attr)) {
            case framework::proto::AttrType::INTS: {
              const auto& vec = PADDLE_GET_CONST(std::vector<int32_t>, attr);
              std::vector<phi::Scalar> scalar_list;
              scalar_list.reserve(vec.size());
              for (const auto& val : vec) {
                scalar_list.emplace_back(val);
              }
              infer_meta_context.EmplaceBackAttr(std::move(scalar_list));
            } break;
            case framework::proto::AttrType::LONGS: {
              const auto& vec = PADDLE_GET_CONST(std::vector<int64_t>, attr);
              std::vector<phi::Scalar> scalar_list;
              scalar_list.reserve(vec.size());
              for (const auto& val : vec) {
                scalar_list.emplace_back(val);
              }
              infer_meta_context.EmplaceBackAttr(std::move(scalar_list));
            } break;
            case framework::proto::AttrType::FLOATS: {
              const auto& vec = PADDLE_GET_CONST(std::vector<float>, attr);
              std::vector<phi::Scalar> scalar_list;
              scalar_list.reserve(vec.size());
              for (const auto& val : vec) {
                scalar_list.emplace_back(val);
              }
              infer_meta_context.EmplaceBackAttr(std::move(scalar_list));
            } break;
            case framework::proto::AttrType::FLOAT64S: {
              const auto& vec = PADDLE_GET_CONST(std::vector<double>, attr);
              std::vector<phi::Scalar> scalar_list;
              scalar_list.reserve(vec.size());
              for (const auto& val : vec) {
                scalar_list.emplace_back(val);
              }
              infer_meta_context.EmplaceBackAttr(std::move(scalar_list));
            } break;
            default:
              PADDLE_THROW(platform::errors::Unimplemented(
                  "Unsupported cast op attribute `%s` to vector<Scalar> when "
                  "construct KernelContext.",
                  attr_names[i]));
          }
        } else {
          // do nothing, skip current attr
        }
        break;
      default:
        if (attr_ptr) {
          auto& attr = *attr_ptr;
          switch (attr_defs[i].type_index) {
            case phi::AttributeType::FLOAT32:
              infer_meta_context.EmplaceBackAttr(PADDLE_GET_CONST(float, attr));
              break;
            case phi::AttributeType::INT32:
              infer_meta_context.EmplaceBackAttr(PADDLE_GET_CONST(int, attr));
              break;
            case phi::AttributeType::BOOL:
              infer_meta_context.EmplaceBackAttr(PADDLE_GET_CONST(bool, attr));
              break;
            case phi::AttributeType::INT64:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(int64_t, attr));
              break;
            case phi::AttributeType::INT32S:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::vector<int>, attr));
              break;
            case phi::AttributeType::DATA_TYPE: {
              auto data_type = paddle::framework::TransToPhiDataType(
                  static_cast<framework::proto::VarType::Type>(
                      PADDLE_GET_CONST(int, attr)));
              infer_meta_context.EmplaceBackAttr(data_type);
            } break;
            case phi::AttributeType::STRING:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::string, attr));
              break;
            case phi::AttributeType::INT64S:
              switch (AttrTypeID(attr)) {
                case framework::proto::AttrType::LONGS:
                  infer_meta_context.EmplaceBackAttr(
                      PADDLE_GET_CONST(std::vector<int64_t>, attr));
                  break;
                case framework::proto::AttrType::INTS: {
                  const auto& vector_int_attr =
                      PADDLE_GET_CONST(std::vector<int>, attr);
                  const std::vector<int64_t> vector_int64_attr(
                      vector_int_attr.begin(), vector_int_attr.end());
                  infer_meta_context.EmplaceBackAttr(vector_int64_attr);
                } break;
                default:
                  PADDLE_THROW(platform::errors::Unimplemented(
                      "Unsupported cast op attribute `%s` to vector<int64_t> "
                      "when "
                      "construct KernelContext.",
                      attr_names[i]));
              }
              break;
            case phi::AttributeType::FLOAT32S:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::vector<float>, attr));
              break;
            case phi::AttributeType::STRINGS:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::vector<std::string>, attr));
              break;
            case phi::AttributeType::BOOLS:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::vector<bool>, attr));
              break;
            case phi::AttributeType::FLOAT64S:
              infer_meta_context.EmplaceBackAttr(
                  PADDLE_GET_CONST(std::vector<double>, attr));
              break;
            default:
              PADDLE_THROW(platform::errors::Unimplemented(
                  "Unsupported cast op attribute `%s` when construct "
                  "KernelContext in dygraph.",
                  attr_names[i]));
          }
        } else {
          // do nothing, skip currnet attr
        }
    }
  }

  VLOG(6) << "BuildInferMetaContext: Done attrs";

  for (auto& out_name : output_names) {
    if (ctx->HasOutputs(out_name, true)) {
      auto output_var = std::move(ctx->GetOutputVarPtrs(out_name));
      if (output_var.size() == 1) {
        infer_meta_context.EmplaceBackOutput(
            std::move(CompatMetaTensor(output_var[0], ctx->IsRuntime())));
      } else {
        paddle::small_vector<CompatMetaTensor, phi::kOutputSmallVectorSize>
            outputs;
        for (const auto& out : output_var) {
          if (ctx->IsRuntime()) {
            if (PADDLE_GET_CONST(Variable*, out)) {
              outputs.emplace_back(
                  std::move(CompatMetaTensor(out, ctx->IsRuntime())));
              continue;
            }
          } else if (PADDLE_GET_CONST(VarDesc*, out)) {
            outputs.emplace_back(
                std::move(CompatMetaTensor(out, ctx->IsRuntime())));
            continue;
          }
          outputs.emplace_back(std::move(CompatMetaTensor(ctx->IsRuntime())));
        }
        infer_meta_context.EmplaceBackOutputs(std::move(outputs));
      }
    } else {
      infer_meta_context.EmplaceBackOutput(
          std::move(CompatMetaTensor(ctx->IsRuntime())));
    }
  }

  VLOG(6) << "BuildInferMetaContext: Done outputs";

  return infer_meta_context;
}

}  // namespace framework
}  // namespace paddle