[SemiAuto] move ut of auto_parallel (#55217)

* move ut of auto_parallel

* fix ut

paddle/fluid/distributed/auto_parallel/CMakeLists.txt

@@ -6,5 +6,3 @@ cc_library(
   DEPS phi auto_parallel_proto proto_desc)
 
 cc_library(auto_parallel DEPS op_dist_attr spmd_rules)
-
-add_subdirectory(test)

paddle/fluid/distributed/auto_parallel/spmd_rules/layer_norm_spmd_rule.cc

@@ -96,8 +96,8 @@ LayerNormSPMDRule::InferForward(const std::vector<DistTensorSpec>& input_specs,
   std::string mean_axes;
   std::string variance_axes;
   if (begin_norm_axis > 1) {
-    mean_axes = "x";
-    variance_axes = "x";
+    mean_axes = "z";
+    variance_axes = "z";
   } else {
     mean_axes = "j";
     variance_axes = "j";
@@ -109,9 +109,8 @@ LayerNormSPMDRule::InferForward(const std::vector<DistTensorSpec>& input_specs,
           << x_axes << "," << scale_axes << "," << bias_axes << " --> "
           << out_axes << "," << mean_axes << "," << variance_axes
           << "](begin_norm_axis:" << begin_norm_axis
-          << ",x=" << x_axes.substr(0, begin_norm_axis)
           << ",y=" << x_axes.substr(begin_norm_axis, x_ndim - begin_norm_axis)
-          << ").";
+          << ",z=" << x_axes.substr(0, begin_norm_axis) << ").";
 
   // step2: Sharding Propogation
   TensorDistAttr output_dist_attr_dst =

test/cpp/CMakeLists.txt

+add_subdirectory(auto_parallel)
 add_subdirectory(phi)
 add_subdirectory(jit)
 add_subdirectory(new_executor)

test/cpp/auto_parallel/CMakeLists.txt

+cc_test(device_mesh_test SRCS device_mesh_test.cc)
+
+cc_test(process_mesh_test SRCS process_mesh_test.cc)
+
+cc_test(dist_attr_test SRCS dist_attr_test.cc)
+
+cc_test_old(dist_mapper_test SRCS dist_mapper_test.cc DEPS phi)
+
+cc_test_old(spmd_rule_test SRCS spmd_rule_test.cc DEPS spmd_rules)

paddle/fluid/distributed/auto_parallel/test/spmd_rule_test.cc → test/cpp/auto_parallel/spmd_rule_test.cc

@@ -236,7 +236,8 @@ TEST(LayerNormSPMDRule, Ctor) {
 
   SPMDRuleBase* layer_norm_rule = SPMDRuleMap::Instance().Get("layer_norm");
 
-  // ijk[1, -1, -1],k[-1],k[-1] --> ijk[1, -1, -1] partial[1]
+  // ijk[1, -1, -1], k[-1], k[-1] --> ijk[1, -1, -1], z[1], z[1], z=ij,
+  // begin_norm_axis=2
   std::pair<std::vector<TensorDistAttr>, std::vector<TensorDistAttr>>
       infered_dist_attrs = layer_norm_rule->InferForward(
           {x_dist_tensor_spec, scale_dist_tensor_spec, bias_dist_tensor_spec},
@@ -261,7 +262,7 @@ TEST(LayerNormSPMDRule, Ctor) {
             std::vector<int64_t>({1}));
   VLOG(4) << "test1 done.";
 
-  // ijk[1, 0, -1],k[0],k[0] --> ijk[1, 0, -1]
+  // ijk[1, 0, -1],k[0],k[0] --> error, begin_norm_axis=2
   x_dist_tensor_spec.set_dims_mapping({1, 0, -1});
   scale_dist_tensor_spec.set_dims_mapping({0});
   bias_dist_tensor_spec.set_dims_mapping({0});
@@ -271,7 +272,8 @@ TEST(LayerNormSPMDRule, Ctor) {
           attrs););
   VLOG(4) << "test2 done.";
 
-  // ijk[0, -1, -1],z[-1],z[1] --> ijk[0, 1, -1, -1], z=jk
+  // ijk[0, -1, -1],y[-1],y[1] --> ijk[0, 1, -1], i[0], i[0], y=jk,
+  // begin_norm_axis=1
   x_dist_tensor_spec.set_dims_mapping({0, -1, -1});
   scale_dist_tensor_spec.set_dims_mapping({-1});
   bias_dist_tensor_spec.set_dims_mapping({1});