【CUDA Kernel No.92】partial_send算子Kernel修复-part (#75640)

* Add partial_send_kernel.h

* Change name

* chore: retry CI checks

* Fix

Author: Le-soleile

paddle/phi/kernels/cpu/partial_send_kernel.cc

@@ -15,7 +15,6 @@
 #include <algorithm>
 #include <utility>
 #include <vector>
-
 #include "paddle/phi/core/kernel_registry.h"
 
 namespace phi {

paddle/phi/kernels/gpu/partial_send_kernel.cu

@@ -19,7 +19,7 @@
 #include "glog/logging.h"
 #include "paddle/phi/core/distributed/utils.h"
 #include "paddle/phi/core/kernel_registry.h"
-
+#include "paddle/phi/kernels/gpu/partial_send_kernel.h"
 #if defined(PADDLE_WITH_NCCL) || defined(PADDLE_WITH_RCCL)
 #include "paddle/phi/core/distributed/nccl_comm_context.h"
 #endif

paddle/phi/kernels/gpu/partial_send_kernel.h

@@ -0,0 +1,26 @@
+// 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.
+
+#pragma once
+
+#include "paddle/phi/core/dense_tensor.h"
+#include "paddle/phi/core/device_context.h"
+
+namespace phi {
+
+template <typename T, typename Context>
+void PartialSendKernel(
+    const Context& dev_ctx, const DenseTensor& x, int peer, int num, int id);
+
+}  // namespace phi