[Reshard] Support r to s on cross mesh

paddle/fluid/pybind/auto_parallel_py.cc

@@ -182,6 +182,10 @@ void BindAutoParallel(py::module *m) {
       *m, "RToSReshardFunction", ReshardFunction)
       .def(py::init<>());
 
+  py::class_<phi::distributed::RToSReshardFunctionCrossMesh>(
+      *m, "RToSReshardFunctionCrossMesh", ReshardFunction)
+      .def(py::init<>());
+
   py::class_<phi::distributed::SToRReshardFunction>(
       *m, "SToRReshardFunction", ReshardFunction)
       .def(py::init<>());

paddle/phi/core/distributed/auto_parallel/r_to_s_reshard_function.cc

@@ -18,6 +18,7 @@
 #include "paddle/phi/core/distributed/auto_parallel/dist_attr.h"
 #include "paddle/phi/core/distributed/auto_parallel/dist_tensor.h"
 #include "paddle/phi/core/distributed/auto_parallel/reshard_utils.h"
+#include "paddle/phi/core/distributed/auto_parallel/same_status_reshard_function.h"
 #include "paddle/phi/kernels/split_kernel.h"
 
 namespace phi {
@@ -85,7 +86,57 @@ void RToSReshardFunction::Eval(phi::DeviceContext* dev_ctx,
   SetDistProps(out, in.dims(), out_dist_attr);
 }
 
+bool RToSReshardFunctionCrossMesh::IsSuitable(
+    const DistTensor& in, const TensorDistAttr& out_dist_attr) {
+  const auto& in_dist_attr = in.dist_attr();
+
+  RESHARD_SHORTCUT_IF_FALSE(in_dist_attr.is_replicated());
+  RESHARD_SHORTCUT_IF_FALSE(out_dist_attr.is_shard());
+
+  const auto& in_process_mesh = in_dist_attr.process_mesh();
+  const auto& out_process_mesh = out_dist_attr.process_mesh();
+
+  RESHARD_SHORTCUT_IF_FALSE(in_process_mesh.ndim() == 1);
+  RESHARD_SHORTCUT_IF_FALSE(out_process_mesh.ndim() == 1);
+  RESHARD_SHORTCUT_IF_FALSE(in_process_mesh.shape() ==
+                            out_process_mesh.shape());
+  RESHARD_SHORTCUT_IF_FALSE(in_process_mesh != out_process_mesh);
+
+  return true;
+}
+
+void RToSReshardFunctionCrossMesh::Eval(phi::DeviceContext* dev_ctx,
+                                        const DistTensor& in,
+                                        const TensorDistAttr& out_dist_attr,
+                                        DistTensor* out) {
+  VLOG(3) << "Call RToSReshardFunctionCrossMesh Eval";
+  const auto& in_dist_attr = in.dist_attr();
+
+  DistTensor tmp_result;
+  TensorDistAttr in_dist_attr_shard = in_dist_attr;
+  in_dist_attr_shard.set_dims_mapping(out_dist_attr.dims_mapping());
+  RToSReshardFunction r_to_s_func;
+  PADDLE_ENFORCE(
+      r_to_s_func.IsSuitable(in, in_dist_attr_shard),
+      phi::errors::InvalidArgument(
+          "Invoke the r to s reshard function is not valid from %s to %s.",
+          tmp_result.dist_attr(),
+          out_dist_attr));
+  r_to_s_func.Eval(dev_ctx, in, in_dist_attr_shard, &tmp_result);
+
+  // Step 2: Same status from the input mesh to output mesh
+  SameStatusReshardFunction same_status_func;
+  PADDLE_ENFORCE(
+      same_status_func.IsSuitable(tmp_result, out_dist_attr),
+      phi::errors::InvalidArgument("Invoke the same status reshard function "
+                                   "is not valid from %s to %s.",
+                                   tmp_result.dist_attr(),
+                                   out_dist_attr));
+  same_status_func.Eval(dev_ctx, tmp_result, out_dist_attr, out);
+}
+
 REGISTER_RESHARD_FUNC(RToSReshardFunction);
+REGISTER_RESHARD_FUNC(RToSReshardFunctionCrossMesh);
 
 }  // namespace distributed
 }  // namespace phi

paddle/phi/core/distributed/auto_parallel/r_to_s_reshard_function.h

@@ -21,9 +21,17 @@ namespace distributed {
 
 class RToSReshardFunction final : public ReshardFunction {
  public:
-  RToSReshardFunction() = default;
-  ~RToSReshardFunction() = default;
+  bool IsSuitable(const DistTensor& in,
+                  const TensorDistAttr& out_dist_attr) override;
 
+  void Eval(DeviceContext* dev_ctx,
+            const DistTensor& in,
+            const TensorDistAttr& out_dist_attr,
+            DistTensor* out) override;
+};
+
+class RToSReshardFunctionCrossMesh final : public ReshardFunction {
+ public:
   bool IsSuitable(const DistTensor& in,
                   const TensorDistAttr& out_dist_attr) override;
 

test/auto_parallel/reshard_r_to_s_cross_mesh.py

@@ -0,0 +1,93 @@
+# Copyright (c) 2023 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 os
+
+import numpy as np
+
+import paddle
+import paddle.distributed as dist
+from paddle.base import core
+
+
+class TestReshardRToSCrossMesh:
+    def __init__(self):
+        self._shape = eval(os.getenv("shape"))
+        self._dtype = os.getenv("dtype")
+        self._seeds = eval(os.getenv("seeds"))
+        self._shard = eval(os.getenv("shard"))
+        self._backend = os.getenv("backend")
+        self._in_mesh = dist.ProcessMesh([0, 1], dim_names=["x"])
+        self._out_mesh = dist.ProcessMesh([1, 0], dim_names=["x"])
+
+    def run_test_case(self):
+        # cpu does not support send/recv
+        if self._backend == "cpu":
+            return
+        elif self._backend == "gpu":
+            place = paddle.CUDAPlace(dist.get_rank())
+
+        dev_ctx = core.DeviceContext.create(place)
+
+        paddle.seed(self._seeds)
+        value = paddle.uniform(self._shape, self._dtype)
+
+        in_shard_specs = [None for i in range(len(self._shape))]
+        out_shard_specs = [None for i in range(len(self._shape))]
+        out_shard_specs[self._shard] = "x"
+
+        dist_attr = dist.DistAttr(
+            mesh=self._in_mesh, sharding_specs=in_shard_specs
+        )
+        out_dist_attr = dist.DistAttr(
+            mesh=self._out_mesh, sharding_specs=out_shard_specs
+        )
+
+        input_tensor = dist.shard_tensor(value, dist_attr=dist_attr)
+
+        reshard_func = core.RToSReshardFunctionCrossMesh()
+        assert reshard_func.is_suitable(input_tensor, out_dist_attr)
+
+        out = reshard_func.eval(dev_ctx, input_tensor, out_dist_attr)
+        out_shape = list(self._shape)
+
+        if out_shape[self._shard] % 2 == 0:
+            out_shape[self._shard] = out_shape[self._shard] // 2
+            split_shape = self._in_mesh.shape[0]
+        else:
+            split_shape = [
+                out_shape[self._shard] // 2 + 1,
+                out_shape[self._shard] // 2,
+            ]
+            out_shape[self._shard] = (
+                split_shape[0] if dist.get_rank() == 1 else split_shape[1]
+            )
+
+        out_expected_local_tensor_list = paddle.split(
+            value, num_or_sections=split_shape, axis=self._shard
+        )
+
+        np.testing.assert_equal(
+            out._local_value().numpy(),
+            out_expected_local_tensor_list[0].numpy()
+            if dist.get_rank() == 1
+            else out_expected_local_tensor_list[1].numpy(),
+        )
+
+        assert np.equal(out.shape, input_tensor.shape).all()
+        assert np.equal(out._local_shape, out_shape).all()
+
+
+if __name__ == '__main__':
+    TestReshardRToSCrossMesh().run_test_case()

test/auto_parallel/test_reshard_r_to_s.py

@@ -22,7 +22,7 @@ def setUp(self):
         super().setUp(num_of_devices=2, timeout=120)
         self._default_envs = {
             "dtype": "float32",
-            "seeds": str(self._seeds),
+            "seeds": "2023",
         }
         self._changeable_envs = {
             "shape": ["(10, 20)", "(5, 7)"],
@@ -40,6 +40,17 @@ def test_reshard_r_to_s(self):
                 user_defined_envs=envs,
             )
 
+    def test_reshard_r_to_s_cross_mesh(self):
+        envs_list = test_base.gen_product_envs_list(
+            self._default_envs, self._changeable_envs
+        )
+        for envs in envs_list:
+            if envs["backend"] != "cpu":
+                self.run_test_case(
+                    "reshard_r_to_s_cross_mesh.py",
+                    user_defined_envs=envs,
+                )
+
 
 if __name__ == "__main__":
     unittest.main()