kunlun add support for c_concat and c_split (#49757)

* kunlun add support for c_concat and c_split

* replace mutable_data() and ShareDataWith()

paddle/fluid/operators/collective/c_concat_op_xpu.cc

+/* Copyright (c) 2021 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/c_concat_op.h"
+
+#include <vector>
+
+#include "paddle/phi/api/include/tensor.h"
+#include "paddle/phi/kernels/funcs//concat_and_split_functor.h"
+
+#if defined(PADDLE_WITH_XPU_BKCL)
+#include "paddle/fluid/distributed/collective/process_group.h"
+#include "paddle/fluid/platform/collective_helper.h"
+#include "paddle/fluid/platform/device/xpu/bkcl_helper.h"
+#endif
+
+namespace paddle {
+namespace operators {
+
+template <typename T>
+class CConcatOpXPUKernel : public framework::OpKernel<T> {
+ public:
+  void Compute(const framework::ExecutionContext& ctx) const override {
+    auto x = ctx.Input<phi::DenseTensor>("X");
+    auto out = ctx.Output<phi::DenseTensor>("Out");
+    BKCLDataType dtype =
+        platform::ToBKCLDataType(framework::TransToProtoVarType(x->dtype()));
+
+    int nranks = ctx.Attr<int>("nranks");
+    int rank = ctx.Attr<int>("rank");
+    int rid = ctx.Attr<int>("ring_id");
+    PADDLE_ENFORCE_GE(rank,
+                      0,
+                      platform::errors::PreconditionNotMet(
+                          "The value of rank (%d) for c_concat must be "
+                          "greater than or equal to 0.",
+                          rank));
+    PADDLE_ENFORCE_GE(nranks,
+                      2,
+                      platform::errors::PreconditionNotMet(
+                          "The value of nranks (%d) for c_concat must be "
+                          "greater than or equal to 2.",
+                          nranks));
+    PADDLE_ENFORCE_LT(rank,
+                      nranks,
+                      platform::errors::PreconditionNotMet(
+                          "The value of rank (%d) for c_concat must be "
+                          "less than that of nranks (%d).",
+                          rank,
+                          nranks));
+
+#if defined(PADDLE_WITH_XPU_BKCL)
+    auto& dev_ctx = ctx.template device_context<phi::XPUContext>();
+    phi::DenseTensor temp_out;
+    framework::DDim temp_out_dims = x->dims();
+    temp_out_dims[0] *= nranks;
+    temp_out.Resize(temp_out_dims);
+    dev_ctx.template Alloc(&temp_out, x->dtype());
+
+    auto map = distributed::ProcessGroupMapFromGid::getInstance();
+    if (map->has(rid)) {
+      // Use ProcessGroup
+      distributed::ProcessGroup* pg = map->get(rid);
+      std::vector<phi::DenseTensor> in_tensor;
+      std::vector<phi::DenseTensor> out_tensor;
+      in_tensor.push_back(*x);
+      out_tensor.push_back(temp_out);
+      auto task = pg->AllGather(in_tensor, out_tensor);
+      task->Wait();
+    } else {
+      auto comm = dev_ctx.bkcl_context();
+
+      int64_t send_numel = x->numel();
+      const T* send_buff = x->data<T>();
+      T* recv_buff = temp_out.data<T>();
+      auto stream = dev_ctx.x_context()->xpu_stream;
+
+      PADDLE_ENFORCE_XPU_SUCCESS(bkcl_all_gather(
+          comm, send_buff, send_numel, recv_buff, dtype, stream));
+    }
+
+    std::vector<phi::DenseTensor> inputs;
+    int axis = x->dims().size() - 1;
+    auto out_dims = x->dims();
+    out_dims[out_dims.size() - 1] *= nranks;
+    int rows_per_tensor = x->dims()[0];
+    int offset = 0;
+    for (int i = 0; i < nranks; i++) {
+      phi::DenseTensor temp = temp_out.Slice(offset, offset + rows_per_tensor);
+      inputs.emplace_back(temp);
+      offset += rows_per_tensor;
+    }
+
+    phi::funcs::ConcatFunctor<phi::XPUContext, T> functor;
+    out->Resize(out_dims);
+    dev_ctx.template Alloc(out, x->dtype());
+    functor(dev_ctx, inputs, axis, out);
+#else
+    PADDLE_THROW(platform::errors::PreconditionNotMet(
+        "PaddlePaddle should compile with XPU."));
+#endif
+  }
+};
+}  // namespace operators
+}  // namespace paddle
+
+namespace ops = paddle::operators;
+namespace plat = paddle::platform;
+
+REGISTER_OP_XPU_KERNEL(c_concat,
+                       ops::CConcatOpXPUKernel<float>,
+                       ops::CConcatOpXPUKernel<plat::float16>);

paddle/fluid/operators/collective/c_split_op_xpu.cc

+/* Copyright (c) 2021 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 <vector>
+
+#include "paddle/fluid/operators/collective/c_split_op.h"
+#if defined(PADDLE_WITH_XPU)
+#include "paddle/phi/backends/xpu/enforce_xpu.h"
+#endif
+
+namespace paddle {
+namespace operators {
+
+template <typename T>
+class CSplitOpXPUKernel : public framework::OpKernel<T> {
+ public:
+  void Compute(const framework::ExecutionContext& ctx) const override {
+    using XPUType = typename XPUTypeTrait<T>::Type;
+    auto x = ctx.Input<phi::DenseTensor>("X");
+    auto out = ctx.Output<phi::DenseTensor>("Out");
+
+    int nranks = ctx.Attr<int>("nranks");
+    int rank = ctx.Attr<int>("rank");
+
+    PADDLE_ENFORCE_GE(rank,
+                      0,
+                      platform::errors::PreconditionNotMet(
+                          "The value of rank (%d) for c_split must be "
+                          "greater than or equal to 0.",
+                          rank));
+    PADDLE_ENFORCE_GE(nranks,
+                      2,
+                      platform::errors::PreconditionNotMet(
+                          "The value of nranks (%d) for c_split must be "
+                          "greater than or equal to 2.",
+                          nranks));
+    PADDLE_ENFORCE_LT(rank,
+                      nranks,
+                      platform::errors::PreconditionNotMet(
+                          "The value of rank (%d) for c_split must be "
+                          "less than that of nranks (%d).",
+                          rank,
+                          nranks));
+
+    auto& dev_ctx = ctx.template device_context<phi::XPUContext>();
+    auto dims = x->dims();
+    auto dims_size = dims.size();
+    // final dim
+    int64_t end_size = dims[dims_size - 1];
+
+    // remain dim
+    auto remain_ddim = phi::slice_ddim(dims, 0, dims_size - 1);
+    int64_t remain_numel = phi::product(remain_ddim);
+
+    dims[dims_size - 1] /= nranks;
+    out->Resize(dims);
+    dev_ctx.template Alloc(out, x->dtype());
+
+    std::vector<XPUType*> output_list(nranks, nullptr);
+    output_list.at(rank) = reinterpret_cast<XPUType*>(out->data<T>());
+    std::vector<int64_t> split_list(nranks, dims[dims_size - 1]);
+    int axis = 1;
+
+    auto ret = xpu::split(dev_ctx.x_context(),
+                          reinterpret_cast<const XPUType*>(x->data<T>()),
+                          output_list,
+                          {remain_numel, end_size},
+                          split_list,
+                          axis);
+    PADDLE_ENFORCE_XDNN_SUCCESS(ret, "split");
+  }
+};
+}  // namespace operators
+}  // namespace paddle
+
+namespace ops = paddle::operators;
+namespace plat = paddle::platform;
+
+REGISTER_OP_XPU_KERNEL(c_split,
+                       ops::CSplitOpXPUKernel<float>,
+                       ops::CSplitOpXPUKernel<int>,
+                       ops::CSplitOpXPUKernel<plat::float16>);

paddle/phi/backends/xpu/xpu2_op_list.cc

@@ -82,6 +82,8 @@ XPUOpMap& get_kl2_ops() {
        XPUKernelSet({phi::DataType::FLOAT16,
                      phi::DataType::FLOAT32,
                      phi::DataType::INT32})},
+      {"c_concat",
+       XPUKernelSet({phi::DataType::FLOAT16, phi::DataType::FLOAT32})},
       {"c_embedding", XPUKernelSet({phi::DataType::FLOAT32})},
       {"c_identity",
        XPUKernelSet({phi::DataType::FLOAT16,
@@ -89,6 +91,10 @@ XPUOpMap& get_kl2_ops() {
                      phi::DataType::FLOAT64,
                      phi::DataType::INT32,
                      phi::DataType::INT64})},
+      {"c_split",
+       XPUKernelSet({phi::DataType::FLOAT16,
+                     phi::DataType::FLOAT32,
+                     phi::DataType::INT32})},
       {"c_sync_calc_stream", XPUKernelSet({phi::DataType::FLOAT32})},
       {"c_sync_comm_stream", XPUKernelSet({phi::DataType::FLOAT32})},
       {"cast",

python/paddle/fluid/tests/unittests/xpu/collective_concat_op.py

+# Copyright (c) 2018 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.
+
+from test_collective_base_xpu import TestCollectiveRunnerBase, runtime_main
+
+import paddle
+import paddle.fluid as fluid
+import paddle.fluid.layers as layers
+from paddle.fluid import core
+
+paddle.enable_static()
+
+
+class TestCollectiveConcat(TestCollectiveRunnerBase):
+    def __init__(self):
+        self.global_ring_id = 0
+
+    def get_model(self, main_prog, startup_program):
+        ring_id = 0
+        nranks = 2
+        with fluid.program_guard(main_prog, startup_program):
+            tindata = layers.data(
+                name="tindata", shape=[10, 1000], dtype='float32'
+            )
+            toutdata = main_prog.current_block().create_var(
+                name="outofconcat",
+                dtype='float32',
+                type=core.VarDesc.VarType.LOD_TENSOR,
+                persistable=False,
+                stop_gradient=False,
+            )
+            main_prog.global_block().append_op(
+                type="c_concat",
+                inputs={'X': tindata},
+                attrs={'ring_id': ring_id, 'rank': self.rank, 'nranks': nranks},
+                outputs={'Out': toutdata},
+            )
+            return toutdata
+
+
+if __name__ == "__main__":
+    runtime_main(TestCollectiveConcat, "c_concat", 0)

python/paddle/fluid/tests/unittests/xpu/collective_split_op.py

+# Copyright (c) 2018 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.
+
+from test_collective_base_xpu import TestCollectiveRunnerBase, runtime_main
+
+import paddle
+import paddle.fluid as fluid
+import paddle.fluid.layers as layers
+from paddle.fluid import core
+
+paddle.enable_static()
+
+
+class TestCollectiveAllGather(TestCollectiveRunnerBase):
+    def __init__(self):
+        self.global_ring_id = 0
+
+    def get_model(self, main_prog, startup_program):
+        ring_id = 0
+        nranks = 2
+        with fluid.program_guard(main_prog, startup_program):
+            tindata = layers.data(
+                name="tindata", shape=[10, 1000], dtype='float32'
+            )
+            toutdata = main_prog.current_block().create_var(
+                name="outofsplit",
+                dtype='float32',
+                type=core.VarDesc.VarType.LOD_TENSOR,
+                persistable=False,
+                stop_gradient=False,
+            )
+            main_prog.global_block().append_op(
+                type="c_split",
+                inputs={'X': tindata},
+                attrs={'ring_id': ring_id, 'rank': self.rank, 'nranks': nranks},
+                outputs={'Out': toutdata},
+            )
+            return toutdata
+
+
+if __name__ == "__main__":
+    runtime_main(TestCollectiveAllGather, "c_split", 0)

python/paddle/fluid/tests/unittests/xpu/test_c_concat.py

+#   Copyright (c) 2018 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.
+
+import sys
+import unittest
+
+from test_collective_base_xpu import TestDistBase
+
+import paddle
+from paddle.fluid import core
+
+sys.path.append("..")
+
+from xpu.get_test_cover_info import (
+    XPUOpTestWrapper,
+    create_test_class,
+    get_xpu_op_support_types,
+)
+
+paddle.enable_static()
+
+
+class XPUTestCConcatOp(XPUOpTestWrapper):
+    def __init__(self):
+        self.op_name = 'c_concat'
+        self.use_dynamic_create_class = False
+
+    class TestConcatOp(TestDistBase):
+        def _setup_config(self):
+            pass
+
+        def test_concat(self, col_type="c_concat"):
+            self.check_with_place(
+                "collective_concat_op.py", col_type, self.in_type_str
+            )
+
+
+support_types = get_xpu_op_support_types('c_concat')
+for stype in support_types:
+    create_test_class(
+        globals(),
+        XPUTestCConcatOp,
+        stype,
+        ignore_device_version=[core.XPUVersion.XPU1],
+    )
+
+if __name__ == '__main__':
+    unittest.main()

python/paddle/fluid/tests/unittests/xpu/test_c_split.py

+#   Copyright (c) 2018 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.
+
+import sys
+import unittest
+
+from test_collective_base_xpu import TestDistBase
+
+import paddle
+from paddle.fluid import core
+
+sys.path.append("..")
+
+from xpu.get_test_cover_info import (
+    XPUOpTestWrapper,
+    create_test_class,
+    get_xpu_op_support_types,
+)
+
+paddle.enable_static()
+
+
+class XPUTestCSplitOp(XPUOpTestWrapper):
+    def __init__(self):
+        self.op_name = 'c_split'
+        self.use_dynamic_create_class = False
+
+    class TestSplitOp(TestDistBase):
+        def _setup_config(self):
+            pass
+
+        def test_split(self, col_type="c_split"):
+            self.check_with_place(
+                "collective_split_op.py", col_type, self.in_type_str
+            )
+
+
+support_types = get_xpu_op_support_types('c_split')
+for stype in support_types:
+    create_test_class(
+        globals(),
+        XPUTestCSplitOp,
+        stype,
+        ignore_device_version=[core.XPUVersion.XPU1],
+    )
+
+if __name__ == '__main__':
+    unittest.main()