From ead54eab366d1b661c968da1c09c26a62de2c814 Mon Sep 17 00:00:00 2001
From: Chenxiao Niu <ncxinhanzhong@gmail.com>
Date: Fri, 23 Sep 2022 12:46:40 +0800
Subject: [PATCH] [MLU] add barrier_op kernel. (#46417)

---
 .../operators/collective/barrier_op_mlu.cc    | 63 +++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 paddle/fluid/operators/collective/barrier_op_mlu.cc

diff --git a/paddle/fluid/operators/collective/barrier_op_mlu.cc b/paddle/fluid/operators/collective/barrier_op_mlu.cc
new file mode 100644
index 00000000000..5a68afe3567
--- /dev/null
+++ b/paddle/fluid/operators/collective/barrier_op_mlu.cc
@@ -0,0 +1,63 @@
+/* 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/operators/collective/barrier_op.h"
+#if defined(PADDLE_WITH_CNCL)
+#include "paddle/fluid/platform/collective_helper.h"
+#include "paddle/fluid/platform/device/mlu/cncl_helper.h"
+#endif
+
+namespace paddle {
+namespace operators {
+
+template <typename T>
+class BarrierOpMLUKernel : public framework::OpKernel<T> {
+ public:
+  void Compute(const framework::ExecutionContext& ctx) const override {
+#if defined(PADDLE_WITH_CNCL)
+    auto in = ctx.Input<framework::Tensor>("X");
+    auto out = ctx.Output<framework::Tensor>("Out");
+
+    auto place = ctx.GetPlace();
+    cnclDataType_t dtype =
+        platform::ToCNCLDataType(framework::TransToProtoVarType(in->dtype()));
+    int64_t numel = in->numel();
+    const void* sendbuff = in->data();
+    void* recvbuff = out->mutable_data<T>(place);
+
+    int rid = ctx.Attr<int>("ring_id");
+    auto cncl_comm = platform::CNCLCommContext::Instance().Get(rid, place);
+    auto* comm = cncl_comm->comm();
+    auto comm_stream = cncl_comm->stream();
+    auto& dev_ctx =
+        ctx.template device_context<paddle::platform::MLUDeviceContext>();
+    cnclReduceOp_t cncl_red_type = cnclSum;
+    dev_ctx.Wait();
+    PADDLE_ENFORCE_MLU_SUCCESS(cnclAllReduce(
+        sendbuff, recvbuff, numel, dtype, cncl_red_type, comm, comm_stream));
+    PADDLE_ENFORCE_MLU_SUCCESS(cnrtQueueSync(comm_stream));
+#else
+    PADDLE_THROW(platform::errors::Unavailable(
+        "PaddlePaddle should compile with CNCL."));
+#endif
+  }
+};
+
+}  // namespace operators
+}  // namespace paddle
+
+namespace ops = paddle::operators;
+namespace plat = paddle::platform;
+
+REGISTER_OP_MLU_KERNEL(barrier, ops::BarrierOpMLUKernel<int>);
-- 
GitLab