Use G++ to compile some cu operators.

paddle/operators/CMakeLists.txt

@@ -9,6 +9,7 @@ function(op_library TARGET)
     set(OP_LIBRARY ${TARGET} ${OP_LIBRARY} PARENT_SCOPE)
     set(cc_srcs)
     set(cu_srcs)
+    set(cu_cc_srcs)
     set(op_common_deps operator op_registry math_function)
     set(options "")
     set(oneValueArgs "")
@@ -22,6 +23,9 @@ function(op_library TARGET)
         if (EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${TARGET}.cc)
             list(APPEND cc_srcs ${TARGET}.cc)
         endif()
+        if (EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${TARGET}.cu.cc)
+            list(APPEND cu_cc_srcs ${TARGET}.cu.cc)
+        endif()
         if (EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${TARGET}.cu)
             list(APPEND cu_srcs ${TARGET}.cu)
         endif()
@@ -29,6 +33,8 @@ function(op_library TARGET)
         foreach(src ${op_library_SRCS})
             if (${src} MATCHES ".*\\.cu$")
                 list(APPEND cu_srcs ${src})
+            elseif(${src} MATCHES ".*\\.cu.cc$")
+                list(APPEND cu_cc_srcs ${src})
             elseif(${src} MATCHES ".*\\.cc$")
                 list(APPEND cc_srcs ${src})
             else()
@@ -43,7 +49,7 @@ function(op_library TARGET)
     endif()
 
     if (WITH_GPU)
-        nv_library(${TARGET} SRCS ${cc_srcs} ${cu_srcs} DEPS ${op_library_DEPS}
+        nv_library(${TARGET} SRCS ${cc_srcs} ${cu_cc_srcs} ${cu_srcs} DEPS ${op_library_DEPS}
                 ${op_common_deps})
     else()
         cc_library(${TARGET} SRCS ${cc_srcs} DEPS ${op_library_DEPS}
@@ -140,7 +146,9 @@ function(op_library TARGET)
 
     # pybind USE_CPU_ONLY_OP
     list(LENGTH cu_srcs cu_srcs_len)
-    if (${pybind_flag} EQUAL 0 AND ${cu_srcs_len} EQUAL 0)
+    list(LENGTH cu_cc_srcs cu_cc_srcs_len)
+
+    if (${pybind_flag} EQUAL 0 AND ${cu_srcs_len} EQUAL 0 AND ${cu_cc_srcs_len} EQUAL 0)
         file(APPEND ${pybind_file} "USE_CPU_ONLY_OP(${TARGET});\n")
         set(pybind_flag 1)
     endif()
@@ -219,6 +227,6 @@ cc_test(dynamic_recurrent_op_test SRCS dynamic_recurrent_op_test.cc
         rnn/recurrent_op_utils.cc
         DEPS dynamic_recurrent_op)
 if(WITH_GPU)
-  nv_test(nccl_op_test SRCS nccl_op_test.cu DEPS nccl_op gpu_info device_context)
+  cc_test(nccl_op_test SRCS nccl_op_test.cu.cc DEPS nccl_op gpu_info device_context)
 endif()
 cc_test(save_load_op_test SRCS save_load_op_test.cc DEPS save_op load_op)

paddle/operators/conv2d_transpose_cudnn_op.cu.cc

@@ -200,9 +200,7 @@ class CudnnConvTransposeGradOpKernel : public framework::OpKernel<T> {
     T alpha = 1.0f, beta = 0.0f;
     if (input_grad) {
       T* input_grad_data = input_grad->mutable_data<T>(ctx.GetPlace());
-      auto t = framework::EigenVector<T>::Flatten(*input_grad);
+      math::set_constant(ctx.device_context(), input_grad, 0);
-      t.device(ctx.GetEigenDevice<platform::GPUPlace>()) =
-          t.constant(static_cast<T>(0));
 
       PADDLE_ENFORCE(platform::dynload::cudnnConvolutionForward(
           handle, &alpha, cudnn_output_desc, output_grad_data,
@@ -214,9 +212,8 @@ class CudnnConvTransposeGradOpKernel : public framework::OpKernel<T> {
     // ------------------- cudnn conv backward filter ---------------------
     if (filter_grad) {
       T* filter_grad_data = filter_grad->mutable_data<T>(ctx.GetPlace());
-      auto t = framework::EigenVector<T>::Flatten(*filter_grad);
+      math::set_constant(ctx.device_context(), filter_grad, 0);
-      t.device(ctx.GetEigenDevice<platform::GPUPlace>()) =
+
-          t.constant(static_cast<T>(0));
       // Gradient with respect to the filter
       PADDLE_ENFORCE(platform::dynload::cudnnConvolutionBackwardFilter(
           handle, &alpha, cudnn_output_desc, output_grad_data, cudnn_input_desc,

paddle/operators/fill_constant_batch_size_like_op.cu.cc

@@ -12,8 +12,8 @@
    See the License for the specific language governing permissions and
    limitations under the License. */
 
-#include "paddle/framework/op_registry.h"
 #include "paddle/operators/fill_constant_batch_size_like_op.h"
+#include "paddle/framework/op_registry.h"
 
 namespace ops = paddle::operators;
 REGISTER_OP_GPU_KERNEL(

paddle/operators/fill_zeros_like_op.cu.cc

@@ -12,8 +12,8 @@
    See the License for the specific language governing permissions and
    limitations under the License. */
 
-#include "paddle/framework/op_registry.h"
 #include "paddle/operators/fill_zeros_like_op.h"
+#include "paddle/framework/op_registry.h"
 
 namespace ops = paddle::operators;
 REGISTER_OP_GPU_KERNEL(

paddle/operators/gru_op.cu.cc

@@ -12,7 +12,6 @@
    See the License for the specific language governing permissions and
    limitations under the License. */
 
-#define EIGEN_USE_GPU
 #include "paddle/operators/gru_op.h"
 
 namespace ops = paddle::operators;

paddle/operators/gru_op.h

@@ -27,10 +27,6 @@ namespace operators {
 using Tensor = framework::Tensor;
 using LoDTensor = framework::LoDTensor;
 
-template <typename T, int MajorType = Eigen::RowMajor,
-          typename IndexType = Eigen::DenseIndex>
-using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;
-
 template <typename Place, typename T>
 class GRUKernel : public framework::OpKernel<T> {
  public:
@@ -57,19 +53,15 @@ class GRUKernel : public framework::OpKernel<T> {
 
     bool is_reverse = context.Attr<bool>("is_reverse");
     math::LoDTensor2BatchFunctor<Place, T> to_batch;
-    to_batch(context.device_context(), *input, *batch_gate, true, is_reverse);
+    auto& dev_ctx = context.device_context();
+    to_batch(dev_ctx, *input, *batch_gate, true, is_reverse);
 
-    int frame_size = hidden_dims[1];
-    int batch_size = hidden_dims[0];
-    auto g = EigenMatrix<T>::From(*batch_gate);
-    auto place = context.GetEigenDevice<Place>();
     if (bias) {
-      auto b = EigenMatrix<T>::From(*bias);
+      math::RowwiseAdd<Place, T> add_bias;
-      g.device(place) = g +
+      add_bias(dev_ctx, *batch_gate, *bias, batch_gate);
-                        b.reshape(Eigen::array<int, 2>({{1, frame_size * 3}}))
-                            .broadcast(Eigen::array<int, 2>({{batch_size, 1}}));
     }
 
+    int frame_size = hidden_dims[1];
     math::hl_gru_value<T> gru_value;
     gru_value.gateWeight = const_cast<T*>(weight_data);
     gru_value.stateWeight =
@@ -89,7 +81,7 @@ class GRUKernel : public framework::OpKernel<T> {
       gru_value.gateValue = gate_t.data<T>();
       gru_value.resetOutputValue = reset_hidden_prev_t.data<T>();
       math::GRUUnitFunctor<Place, T>::compute(
-          context.device_context(), gru_value, frame_size, cur_batch_size,
+          dev_ctx, gru_value, frame_size, cur_batch_size,
           math::ActiveType(context.Attr<std::string>("activation")),
           math::ActiveType(context.Attr<std::string>("gate_activation")));
       gru_value.prevOutValue = gru_value.outputValue;
@@ -97,7 +89,7 @@ class GRUKernel : public framework::OpKernel<T> {
 
     math::Batch2LoDTensorFunctor<Place, T> to_seq;
     batch_hidden->set_lod(batch_gate->lod());
-    to_seq(context.device_context(), *batch_hidden, *hidden);
+    to_seq(dev_ctx, *batch_hidden, *hidden);
   }
 
   void Compute(const framework::ExecutionContext& context) const override {
@@ -138,15 +130,14 @@ class GRUGradKernel : public framework::OpKernel<T> {
     batch_reset_hidden_prev_grad.mutable_data<T>(hidden_dims,
                                                  context.GetPlace());
     math::SetConstant<Place, T> zero;
-    zero(context.device_context(), &batch_hidden_grad, static_cast<T>(0.0));
+    auto& dev_ctx = context.device_context();
-    zero(context.device_context(), &batch_gate_grad, static_cast<T>(0.0));
+    zero(dev_ctx, &batch_hidden_grad, static_cast<T>(0.0));
-    zero(context.device_context(), &batch_reset_hidden_prev_grad,
+    zero(dev_ctx, &batch_gate_grad, static_cast<T>(0.0));
-         static_cast<T>(0.0));
+    zero(dev_ctx, &batch_reset_hidden_prev_grad, static_cast<T>(0.0));
 
     bool is_reverse = context.Attr<bool>("is_reverse");
     batch_hidden_grad.set_lod(batch_hidden->lod());
-    to_batch(context.device_context(), *hidden_grad, batch_hidden_grad, false,
+    to_batch(dev_ctx, *hidden_grad, batch_hidden_grad, false, is_reverse);
-             is_reverse);
 
     math::hl_gru_value<T> gru_value;
     gru_value.gateWeight = const_cast<T*>(weight_data);
@@ -157,7 +148,7 @@ class GRUGradKernel : public framework::OpKernel<T> {
     if (weight_grad) {
       gru_grad.gateWeightGrad =
           weight_grad->mutable_data<T>(context.GetPlace());
-      zero(context.device_context(), weight_grad, static_cast<T>(0.0));
+      zero(dev_ctx, weight_grad, static_cast<T>(0.0));
       gru_grad.stateWeightGrad =
           weight_grad->data<T>() + 2 * frame_size * frame_size;
     } else {
@@ -188,7 +179,7 @@ class GRUGradKernel : public framework::OpKernel<T> {
         gru_value.prevOutValue = const_cast<T*>(h0_data);
         if (h0_grad) {
           T* h0_grad_data = h0_grad->mutable_data<T>(context.GetPlace());
-          zero(context.device_context(), h0_grad, static_cast<T>(0.0));
+          zero(dev_ctx, h0_grad, static_cast<T>(0.0));
           gru_grad.prevOutGrad = h0_grad_data;
         } else {
           gru_grad.prevOutGrad = nullptr;
@@ -202,8 +193,7 @@ class GRUGradKernel : public framework::OpKernel<T> {
       }
 
       math::GRUUnitGradFunctor<Place, T>::compute(
-          context.device_context(), gru_value, gru_grad, frame_size,
+          dev_ctx, gru_value, gru_grad, frame_size, cur_batch_size,
-          cur_batch_size,
           math::ActiveType(context.Attr<std::string>("activation")),
           math::ActiveType(context.Attr<std::string>("gate_activation")));
     }
@@ -211,14 +201,18 @@ class GRUGradKernel : public framework::OpKernel<T> {
       input_grad->mutable_data<T>(context.GetPlace());
       math::Batch2LoDTensorFunctor<Place, T> to_seq;
       batch_gate_grad.set_lod(batch_gate->lod());
-      to_seq(context.device_context(), batch_gate_grad, *input_grad);
+      to_seq(dev_ctx, batch_gate_grad, *input_grad);
     }
     if (bias_grad) {
       bias_grad->mutable_data<T>(context.GetPlace());
-      auto d_b = EigenMatrix<T>::From(*bias_grad);
+      int m = static_cast<int>(batch_gate_grad.dims()[0]);
-      auto d_g = EigenMatrix<T>::From(batch_gate_grad);
+      int n = static_cast<int>(batch_gate_grad.dims()[1]);
-      auto place = context.GetEigenDevice<Place>();
+      Tensor ones;
-      d_b.device(place) = d_g.sum(Eigen::array<int, 1>({{0}}));
+      ones.mutable_data<T>({m}, context.GetPlace());
+      math::SetConstant<Place, T> set;
+      set(dev_ctx, &ones, static_cast<T>(1));
+      math::gemv<Place, T>(dev_ctx, true, m, n, 1., batch_gate_grad.data<T>(),
+                           ones.data<T>(), 0., bias_grad->data<T>());
     }
   }
 

paddle/operators/lstm_op.cu.cc

@@ -12,7 +12,6 @@
    See the License for the specific language governing permissions and
    limitations under the License. */
 
-#define EIGEN_USE_GPU
 #include "paddle/operators/lstm_op.h"
 
 namespace ops = paddle::operators;

paddle/operators/lstm_op.h

@@ -24,10 +24,6 @@ namespace operators {
 using LoDTensor = framework::LoDTensor;
 using Tensor = framework::Tensor;
 
-template <typename T, int MajorType = Eigen::RowMajor,
-          typename IndexType = Eigen::DenseIndex>
-using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;
-
 template <typename Place, typename T>
 inline void ReorderInitState(const platform::DeviceContext& ctx,
                              const framework::Tensor& src, const size_t* index,
@@ -65,16 +61,11 @@ class LSTMKernel : public framework::OpKernel<T> {
     framework::DDim dims({in_dims[0], frame_size});
 
     if (bias) {
-      Eigen::array<int, 2> extents({{1, 4 * frame_size}});
+      Tensor b = *bias;
-      Eigen::array<int, 2> offsets({{0, 0}});
+      b.Resize({bias->numel(), 1});
-      auto b = EigenMatrix<T>::From(*bias);
+      Tensor gate_bias = b.Slice(0, 4 * frame_size);
-      auto gate = EigenMatrix<T>::From(*batch_gate);
+      math::RowwiseAdd<Place, T> add_bias;
-      gate.device(ctx.GetEigenDevice<Place>()) =
+      add_bias(device_ctx, *batch_gate, gate_bias, batch_gate);
-          gate +
-          b.slice(offsets, extents)
-              .reshape(Eigen::array<int, 2>({{1, frame_size * 4}}))
-              .broadcast(
-                  Eigen::array<int, 2>({{static_cast<int>(in_dims[0]), 1}}));
     }
 
     math::LstmMetaValue<T> lstm_value;

paddle/operators/math/context_project.h

@@ -14,9 +14,9 @@ limitations under the License. */
 
 #pragma once
 
-#include "paddle/framework/eigen.h"
 #include "paddle/framework/lod_tensor.h"
 #include "paddle/operators/math/im2col.h"
+#include "paddle/operators/math/math_function.h"
 
 namespace paddle {
 namespace operators {
@@ -24,9 +24,6 @@ namespace math {
 
 using Tensor = framework::Tensor;
 using LoDTensor = framework::LoDTensor;
-template <typename T, int MajorType = Eigen::RowMajor,
-          typename IndexType = Eigen::DenseIndex>
-using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;
 
 /*
  * \brief Context projection concatenates features in adjacent time-steps in
@@ -94,6 +91,9 @@ class ContextProjectFunctor {
     auto lod_level_0 = in.lod()[0];
 
     math::Im2ColFunctor<math::ColFormat::kOCF, Place, float> im2col_ocf;
+    if (platform::is_gpu_place(context.GetPlace())) {
+      LOG(INFO) << "========= gpu ==========";
+    }
 
     int input_row_begin, input_row_end;
     int sequence_height, sequence_width;
@@ -150,9 +150,7 @@ class ContextProjectFunctor {
             Tensor out_t_sub = out_t.Slice(k * context_length,
                                            k * context_length + padding_size);
             Tensor w_sub = padding_data.Slice(k, k + padding_size);
-            auto out_t_sub_e = EigenMatrix<T>::From(out_t_sub);
+            out_t_sub.CopyFrom(w_sub, context.GetPlace(), context);
-            auto w_sub_e = EigenMatrix<T>::From(w_sub);
-            out_t_sub_e.device(*context.GetEigenDevice<Place>()) = w_sub_e;
           }
         }
         if (down_pad > 0) {  // add down pad
@@ -182,9 +180,7 @@ class ContextProjectFunctor {
                 (down_pad_begin_row + t) * context_length);
             Tensor w_sub = padding_data.Slice(
                 up_pad + padding_idx, up_pad + padding_idx + padding_size);
-            auto out_t_sub_e = EigenMatrix<T>::From(out_t_sub);
+            out_t_sub.CopyFrom(w_sub, context.GetPlace(), context);
-            auto w_sub_e = EigenMatrix<T>::From(w_sub);
-            out_t_sub_e.device(*context.GetEigenDevice<Place>()) = w_sub_e;
           }
         }
         out_t.Resize({sequence_height, context_length * sequence_width});
@@ -260,10 +256,8 @@ class ContextProjectGradFunctor {
               Tensor out_t_sub = out_t.Slice(k * context_length,
                                              k * context_length + padding_size);
               Tensor w_sub = padding_data.Slice(k, k + padding_size);
-              auto out_t_sub_e = EigenMatrix<T>::From(out_t_sub);
+              axpy<Place, T>(context, w_sub.numel(), static_cast<T>(1),
-              auto w_sub_e = EigenMatrix<T>::From(w_sub);
+                             out_t_sub.data<T>(), w_sub.data<T>());
-              w_sub_e.device(*context.GetEigenDevice<Place>()) =
-                  w_sub_e + out_t_sub_e;
             }
           }
           if (down_pad > 0) {
@@ -294,10 +288,8 @@ class ContextProjectGradFunctor {
                   (down_pad_begin_row + t) * context_length);
               Tensor w_sub = padding_data.Slice(
                   up_pad + padding_idx, up_pad + padding_idx + padding_size);
-              auto out_t_sub_e = EigenMatrix<T>::From(out_t_sub);
+              axpy<Place, T>(context, w_sub.numel(), static_cast<T>(1),
-              auto w_sub_e = EigenMatrix<T>::From(w_sub);
+                             out_t_sub.data<T>(), w_sub.data<T>());
-              w_sub_e.device(*context.GetEigenDevice<Place>()) =
-                  w_sub_e + out_t_sub_e;
             }
           }
           out_t.Resize({sequence_height, context_length * sequence_width});

paddle/operators/math/math_function.cc

@@ -14,6 +14,7 @@ limitations under the License. */
 
 #include "paddle/operators/math/math_function.h"
 #include "paddle/framework/data_type.h"
+#include "paddle/operators/math/math_function_impl.h"
 
 namespace paddle {
 namespace operators {
@@ -232,7 +233,34 @@ void gemv<platform::CPUPlace, double>(const platform::DeviceContext& context,
   cblas_dgemv(CblasRowMajor, transA, M, N, alpha, A, N, B, 1, beta, C, 1);
 }
 
+template <>
+void axpy<platform::CPUPlace, float>(const platform::DeviceContext& context,
+                                     const int n, const float alpha,
+                                     const float* x, float* y) {
+  cblas_saxpy(n, alpha, x, 1, y, 1);
+}
+
+template <>
+void axpy<platform::CPUPlace, double>(const platform::DeviceContext& context,
+                                      const int n, const double alpha,
+                                      const double* x, double* y) {
+  cblas_daxpy(n, alpha, x, 1, y, 1);
+}
+
 template struct SetConstant<platform::CPUPlace, float>;
+template struct SetConstant<platform::CPUPlace, double>;
+template struct SetConstant<platform::CPUPlace, int>;
+
+#define DEFINE_CPU_TRANS(RANK)                                \
+  template struct Transpose<platform::CPUPlace, float, RANK>; \
+  template struct Transpose<platform::CPUPlace, double, RANK>;
+
+DEFINE_CPU_TRANS(1);
+DEFINE_CPU_TRANS(2);
+DEFINE_CPU_TRANS(3);
+DEFINE_CPU_TRANS(4);
+DEFINE_CPU_TRANS(5);
+DEFINE_CPU_TRANS(6);
 
 struct TensorSetConstant {
   TensorSetConstant(framework::Tensor* tensor, float value)

paddle/operators/math/math_function.cu

@@ -12,8 +12,10 @@ 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. */
 
+#define EIGEN_USE_GPU
 #include "paddle/framework/data_type.h"
 #include "paddle/operators/math/math_function.h"
+#include "paddle/operators/math/math_function_impl.h"
 
 namespace paddle {
 namespace operators {
@@ -231,7 +233,40 @@ void gemv<platform::GPUPlace, double>(const platform::DeviceContext& context,
       cuTransA, N, M, &alpha, A, N, B, 1, &beta, C, 1));
 }
 
+template <>
+void axpy<platform::GPUPlace, float>(const platform::DeviceContext& context,
+                                     const int n, const float alpha,
+                                     const float* x, float* y) {
+  PADDLE_ENFORCE(platform::dynload::cublasSaxpy(
+      reinterpret_cast<const platform::CUDADeviceContext&>(context)
+          .cublas_handle(),
+      n, alpha, x, 1, y, 1));
+}
+
+template <>
+void axpy<platform::GPUPlace, double>(const platform::DeviceContext& context,
+                                      const int n, const double alpha,
+                                      const double* x, double* y) {
+  PADDLE_ENFORCE(platform::dynload::cublasDaxpy(
+      reinterpret_cast<const platform::CUDADeviceContext&>(context)
+          .cublas_handle(),
+      n, alpha, x, 1, y, 1));
+}
+
 template struct SetConstant<platform::GPUPlace, float>;
+template struct SetConstant<platform::GPUPlace, double>;
+template struct SetConstant<platform::GPUPlace, int>;
+
+#define DEFINE_GPU_TRANS(RANK)                                \
+  template struct Transpose<platform::GPUPlace, float, RANK>; \
+  template struct Transpose<platform::GPUPlace, double, RANK>;
+
+DEFINE_GPU_TRANS(1);
+DEFINE_GPU_TRANS(2);
+DEFINE_GPU_TRANS(3);
+DEFINE_GPU_TRANS(4);
+DEFINE_GPU_TRANS(5);
+DEFINE_GPU_TRANS(6);
 
 struct TensorSetConstant {
   TensorSetConstant(const platform::DeviceContext& context,

paddle/operators/math/math_function.h

@@ -93,14 +93,21 @@ void gemv(const platform::DeviceContext& context, const bool trans_a,
           const int M, const int N, const T alpha, const T* A, const T* B,
           const T beta, T* C);
 
+template <typename Place, typename T>
+void axpy(const platform::DeviceContext& context, const int n, const T alpha,
+          const T* x, T* y);
+
+template <typename Place, typename T, int Rank>
+struct Transpose {
+  void operator()(const platform::DeviceContext& context,
+                  const framework::Tensor& in, framework::Tensor* out,
+                  const std::vector<int>& axis);
+};
+
 template <typename Place, typename T>
 struct SetConstant {
   void operator()(const platform::DeviceContext& context,
-                  framework::Tensor* tensor, T num) {
+                  framework::Tensor* tensor, T num);
-    auto t = framework::EigenVector<T>::Flatten(*tensor);
-    t.device(*context.GetEigenDevice<Place>()) =
-        t.constant(static_cast<T>(num));
-  }
 };
 
 template <typename Place>

paddle/operators/math/math_function_impl.h

@@ -0,0 +1,48 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+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/framework/data_type.h"
+#include "paddle/operators/math/math_function.h"
+
+namespace paddle {
+namespace operators {
+namespace math {
+
+template <typename Place, typename T>
+void SetConstant<Place, T>::operator()(const platform::DeviceContext& context,
+                                       framework::Tensor* tensor, T num) {
+  auto t = framework::EigenVector<T>::Flatten(*tensor);
+  t.device(*context.GetEigenDevice<platform::CPUPlace>()) =
+      t.constant(static_cast<T>(num));
+}
+
+template <typename Place, typename T, int Rank>
+void Transpose<Place, T, Rank>::operator()(
+    const platform::DeviceContext& context, const framework::Tensor& in,
+    framework::Tensor* out, const std::vector<int>& axis) {
+  Eigen::array<int, Rank> permute;
+  for (int i = 0; i < Rank; i++) {
+    permute[i] = axis[i];
+  }
+  auto in_dim = in.dims();
+  auto out_dim = out->dims();
+
+  auto eigen_in = framework::EigenTensor<T, Rank>::From(in);
+  auto eigen_out = framework::EigenTensor<T, Rank>::From(*out);
+  auto* dev = context.GetEigenDevice<Place>();
+  eigen_out.device(*dev) = eigen_in.shuffle(permute);
+}
+}
+}
+}

paddle/operators/math/sequence2batch.cc

@@ -56,6 +56,29 @@ template class LoDTensor2BatchFunctor<platform::CPUPlace, double>;
 template class Batch2LoDTensorFunctor<platform::CPUPlace, float>;
 template class Batch2LoDTensorFunctor<platform::CPUPlace, double>;
 
+template <typename T>
+struct RowwiseAdd<platform::CPUPlace, T> {
+  void operator()(const platform::DeviceContext& context,
+                  const framework::Tensor& input, const framework::Tensor& bias,
+                  framework::Tensor* output) {
+    auto in_dims = input.dims();
+    auto size = input.numel() / in_dims[0];
+    PADDLE_ENFORCE_EQ(bias.numel(), size);
+    PADDLE_ENFORCE_EQ(output->dims(), in_dims);
+
+    auto in = EigenMatrix<T>::From(input);
+    auto b = EigenMatrix<T>::From(bias);
+    auto out = EigenMatrix<T>::From(*output);
+    Eigen::array<int, 2> bshape({{1, static_cast<int>(size)}});
+    Eigen::array<int, 2> bcast({{static_cast<int>(in_dims[0]), 1}});
+    out.device(*context.GetEigenDevice<platform::CPUPlace>()) =
+        in + b.reshape(bshape).broadcast(bcast);
+  }
+};
+
+template struct RowwiseAdd<platform::CPUPlace, float>;
+template struct RowwiseAdd<platform::CPUPlace, double>;
+
 }  // namespace math
 }  // namespace operators
 }  // namespace paddle

paddle/operators/math/sequence2batch.cu

@@ -12,6 +12,7 @@ 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. */
 
+#define EIGEN_USE_GPU
 #include "paddle/operators/math/sequence2batch.h"
 
 namespace paddle {
@@ -73,6 +74,37 @@ template class LoDTensor2BatchFunctor<platform::GPUPlace, double>;
 template class Batch2LoDTensorFunctor<platform::GPUPlace, float>;
 template class Batch2LoDTensorFunctor<platform::GPUPlace, double>;
 
+template <typename T>
+__global__ void RowwiseAddKernel(const T* src, const T* b, T* dst,
+                                 int64_t height, int64_t width) {
+  for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < height * width;
+       i += blockDim.x * gridDim.x) {
+    int64_t h = i / width;
+    int64_t w = i % width;
+    dst[h * width + w] = src[h * width + w] + b[w];
+  }
+}
+
+template <typename T>
+struct RowwiseAdd<platform::GPUPlace, T> {
+  void operator()(const platform::DeviceContext& context,
+                  const framework::Tensor& input, const framework::Tensor& bias,
+                  framework::Tensor* output) {
+    auto in_dims = input.dims();
+    auto size = input.numel() / in_dims[0];
+    PADDLE_ENFORCE_EQ(bias.numel(), size);
+    PADDLE_ENFORCE_EQ(output->dims(), in_dims);
+    int block = 512;
+    int grid = (input.numel() + block - 1) / block;
+    auto stream =
+        reinterpret_cast<const platform::CUDADeviceContext&>(context).stream();
+    RowwiseAddKernel<T><<<grid, block, 0, stream>>>(
+        input.data<T>(), bias.data<T>(), output->data<T>(), in_dims[0], size);
+  }
+};
+
+template struct RowwiseAdd<platform::GPUPlace, float>;
+template struct RowwiseAdd<platform::GPUPlace, double>;
 }  // namespace math
 }  // namespace operators
 }  // namespace paddle

paddle/operators/math/sequence2batch.h

@@ -13,6 +13,7 @@ See the License for the specific language governing permissions and
 limitations under the License. */
 
 #pragma once
+#include "paddle/framework/eigen.h"
 #include "paddle/framework/lod_tensor.h"
 #include "paddle/framework/tensor.h"
 #include "paddle/platform/device_context.h"
@@ -21,6 +22,10 @@ namespace paddle {
 namespace operators {
 namespace math {
 
+template <typename T, int MajorType = Eigen::RowMajor,
+          typename IndexType = Eigen::DenseIndex>
+using EigenMatrix = framework::EigenMatrix<T, MajorType, IndexType>;
+
 template <typename Place, typename T>
 class CopyMatrixRowsFunctor {
  public:
@@ -159,6 +164,13 @@ class Batch2LoDTensorFunctor {
   }
 };
 
+template <typename Place, typename T>
+struct RowwiseAdd {
+  void operator()(const platform::DeviceContext& context,
+                  const framework::Tensor& input, const framework::Tensor& bias,
+                  framework::Tensor* output);
+};
+
 }  // namespace math
 }  // namespace operators
 }  // namespace paddle

paddle/operators/matmul_op.h

@@ -15,8 +15,8 @@
 #pragma once
 
 #include "paddle/framework/op_registry.h"
+#include "paddle/operators/math/math_function.h"
 #include "paddle/operators/math/matmul.h"
-#include "paddle/operators/transpose_op.h"
 
 namespace paddle {
 namespace operators {
@@ -74,11 +74,13 @@ Tensor CombineBatchAndN(const framework::ExecutionContext& context,
   Tensor output;
   auto in_dims = input.dims();
   if (in_dims.size() == 3) {
-    output.Resize(in_dims);
+    output.Resize({in_dims[1], in_dims[0], in_dims[2]});
     output.mutable_data<T>(context.GetPlace());
-    EigenTranspose<Place, T, 3>(context, input, output, {1, 0, 2});
+    std::vector<int> axis = {1, 0, 2};
+    math::Transpose<Place, T, 3> trans;
+    trans(context.device_context(), input, &output, axis);
     std::vector<int64_t> out_dims = {in_dims[1], in_dims[0] * in_dims[2]};
-    output.Resize(make_ddim(out_dims));
+    output.Resize({in_dims[1], in_dims[0] * in_dims[2]});
   } else {
     output.ShareDataWith(input);
   }