【Hackathon 6th Fundable Projects 3 No.242】 [fluid_ops] mp_allreduce_s…

…um (#67694)

paddle/fluid/operators/collective/mp_allreduce_sum_op.cc

@@ -82,13 +82,3 @@ REGISTER_OPERATOR(mp_allreduce_sum,
                   ops::MpAllReduceSumOpGradMaker<paddle::imperative::OpBase>,
                   ops::MpAllReduceSumOpMaker,
                   ops::MpAllReduceSumInplaceInferer);
-
-PD_REGISTER_STRUCT_KERNEL(mp_allreduce_sum,
-                          CPU,
-                          ALL_LAYOUT,
-                          ops::MpAllReduceSumCPUKernel,
-                          float,
-                          double,
-                          int,
-                          int64_t,
-                          phi::dtype::float16) {}

paddle/fluid/pir/dialect/op_generator/ops_api_gen.py

@@ -202,6 +202,8 @@
     'nce',
     'lrn',
     'max_pool2d_v2',
+    'mp_allreduce_sum',
+    'mp_allreduce_sum_',
     'partial_sum',
     'pull_gpups_sparse',
     'pull_gpups_sparse_',

paddle/phi/kernels/cpu/mp_allreduce_sum_kernel.cc

@@ -0,0 +1,37 @@
+// Copyright (c) 2024 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/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/all_reduce_kernel.h"
+
+namespace phi {
+template <typename T, typename Context>
+void MpAllReduceSumKernel(const Context& dev_ctx,
+                          const DenseTensor& x,
+                          int ring_id UNUSED,
+                          bool use_calc_stream UNUSED,
+                          DenseTensor* out) {
+  AllReduceKernel<T, Context>(
+      dev_ctx, x, static_cast<int>(ReduceType::kRedSum), out);
+}
+}  // namespace phi
+PD_REGISTER_KERNEL(mp_allreduce_sum,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::MpAllReduceSumKernel,
+                   float,
+                   double,
+                   int,
+                   int64_t,
+                   phi::dtype::float16) {}

paddle/phi/kernels/gpu/mp_allreduce_sum_kernel.cu

@@ -0,0 +1,52 @@
+// Copyright (c) 2024 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/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/all_reduce_kernel.h"
+
+namespace phi {
+template <typename T, typename Context>
+void MpAllReduceSumKernel(const Context& dev_ctx,
+                          const DenseTensor& x,
+                          int ring_id UNUSED,
+                          bool use_calc_stream UNUSED,
+                          DenseTensor* out) {
+  AllReduceKernel<T, Context>(
+      dev_ctx, x, static_cast<int>(ReduceType::kRedSum), out);
+}
+}  // namespace phi
+
+#if (NCCL_VERSION_CODE >= 21000 && CUDA_VERSION >= 11000) || \
+    defined(PADDLE_WITH_HIP)
+PD_REGISTER_KERNEL(mp_allreduce_sum,
+                   GPU,
+                   ALL_LAYOUT,
+                   phi::MpAllReduceSumKernel,
+                   float,
+                   double,
+                   int,
+                   int64_t,
+                   phi::dtype::bfloat16,
+                   phi::dtype::float16) {}
+#else
+PD_REGISTER_KERNEL(mp_allreduce_sum,
+                   GPU,
+                   ALL_LAYOUT,
+                   phi::MpAllReduceSumKernel,
+                   float,
+                   double,
+                   int,
+                   int64_t,
+                   phi::dtype::float16) {}
+#endif

paddle/phi/kernels/kps/mp_allreduce_sum_kernel.kps

@@ -0,0 +1,36 @@
+#ifdef PADDLE_WITH_XPU_KP
+
+// Please do not modify the following code
+#if defined(__CUDA_ARCH__)
+#undef __CUDA_ARCH__
+#endif
+
+#if defined(__CUDACC__)
+#undef __CUDACC__
+#endif
+
+#if defined(__CUDA__)
+#undef __CUDA__
+#endif
+
+#if defined(__NVCC__)
+#undef __NVCC__
+#endif
+
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/all_reduce_kernel.h"
+
+namespace phi {
+template <typename T, typename Context>
+void MpAllReduceSumKernel(const Context& dev_ctx,
+                          const DenseTensor& x,
+                          int ring_id UNUSED,
+                          bool use_calc_stream UNUSED,
+                          DenseTensor* out) {
+  AllReduceKernel<T, Context>(
+      dev_ctx, x, static_cast<int>(ReduceType::kRedSum), out);
+}
+}  // namespace phi
+PD_REGISTER_KERNEL(
+    mp_allreduce_sum, KPS, ALL_LAYOUT, phi::MpAllReduceSumKernel, float) {}
+#endif

paddle/phi/kernels/xpu/mp_allreduce_sum_kernel.cc

@@ -0,0 +1,36 @@
+// Copyright (c) 2024 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/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/all_reduce_kernel.h"
+
+namespace phi {
+template <typename T, typename Context>
+void MpAllReduceSumKernel(const Context& dev_ctx,
+                          const DenseTensor& x,
+                          int ring_id UNUSED,
+                          bool use_calc_stream UNUSED,
+                          DenseTensor* out) {
+  AllReduceKernel<T, Context>(
+      dev_ctx, x, static_cast<int>(ReduceType::kRedSum), out);
+}
+}  // namespace phi
+
+PD_REGISTER_KERNEL(mp_allreduce_sum,
+                   XPU,
+                   ALL_LAYOUT,
+                   phi::MpAllReduceSumKernel,
+                   float,
+                   int,
+                   phi::dtype::float16) {}

paddle/phi/ops/yaml/inconsistent/static_backward.yaml

@@ -313,6 +313,12 @@
     func : minimum_grad
   composite : minimum_grad(x, y, out_grad, x_grad, y_grad)
 
+- backward_op : mp_allreduce_sum_grad
+  forward : mp_allreduce_sum(Tensor x, int ring_id = 0, bool use_calc_stream = false) -> Tensor(out)
+  args : (Tensor out_grad, int ring_id = 0, bool use_calc_stream = false)
+  output : Tensor(x_grad)
+  invoke : c_identity(out_grad, ring_id, false, false)
+
 - backward_op : multiply_double_grad
   forward : multiply_grad (Tensor x, Tensor y, Tensor grad_out, int axis = -1) -> Tensor(grad_x), Tensor(grad_y)
   args : (Tensor x, Tensor y, Tensor grad_out, Tensor grad_x_grad, Tensor grad_y_grad, int axis = -1)

paddle/phi/ops/yaml/inconsistent/static_ops.yaml

@@ -576,6 +576,18 @@
   backward : minimum_grad
   interfaces : paddle::dialect::InferSymbolicShapeInterface
 
+- op : mp_allreduce_sum
+  args : (Tensor x, int ring_id = 0, bool use_calc_stream = false)
+  output : Tensor(out)
+  infer_meta :
+    func : AllReduceInferMeta
+    param: [x]
+  kernel :
+    func : mp_allreduce_sum
+    param: [x, ring_id, use_calc_stream]
+  backward: mp_allreduce_sum_grad
+  inplace: (x -> out)
+
 - op : multiply
   args : (Tensor x, Tensor y)
   output : Tensor

paddle/phi/ops/yaml/op_compat.yaml

@@ -4431,6 +4431,12 @@
   outputs:
     out: Out
 
+- op: mp_allreduce_sum
+  inputs :
+    x : X
+  outputs :
+    out: Out
+
 - op: nce
   backward: nce_grad
   inputs: