does not support summary when task sink and iter num > 1

mindspore/ccsrc/backend/session/ascend_session.cc

@@ -327,6 +327,8 @@ void AscendSession::CompileChildGraph(const KernelGraphPtr &child_graph) {
   runtime_instance->AssignStaticMemoryValueNode(child_graph.get());
 }
 
+bool AscendSession::IsSupportSummary() { return !device::KernelAdjust::NeedInsertSwitch(); }
+
 void AscendSession::RunGraphImpl(const GraphId &graph_id, const std::vector<tensor::TensorPtr> &inputs,
                                  VectorRef *const outputs) {
   MS_LOG(INFO) << "Start";

mindspore/ccsrc/backend/session/ascend_session.h

@@ -71,6 +71,7 @@ class AscendSession : public SessionBasic {
   GraphId CompileGraphImpl(const AnfNodePtrList &lst, const AnfNodePtrList &outputs) override;
   GraphId CompileGraphImpl(NotNull<FuncGraphPtr> func_graph) override;
   GraphId CompileGraphImpl(NotNull<FuncGraphPtr> func_graph, const std::vector<tensor::TensorPtr> &inputs) override;
+  bool IsSupportSummary() override;
   void RunGraphImpl(const GraphId &graph_id, const std::vector<tensor::TensorPtr> &inputs, VectorRef *outputs) override;
   void BuildGraphImpl(GraphId) override;
   void BuildOpImpl(const OpRunInfo &op_run_info, const GraphInfo &graph_info,

mindspore/ccsrc/backend/session/session_basic.cc

@@ -1153,6 +1153,11 @@ void SessionBasic::Summary(KernelGraph *graph) {
   if (!exist_summary) {
     return;
   }
+
+  if (!IsSupportSummary()) {
+    MS_LOG(ERROR) << "The Summary operator can not collect data correctly. Detail: the data sink mode is used and the"
+                     " sink size(in model.train() python api) is not equal to 1.";
+  }
   SetSummaryNodes(graph);
   auto summary_outputs = graph->summary_nodes();
   std::map<std::string, tensor::TensorPtr> params_list;

mindspore/ccsrc/backend/session/session_basic.h

@@ -132,6 +132,7 @@ class SessionBasic : public std::enable_shared_from_this<SessionBasic> {
   friend class RunGraphTask;
   friend class BuildOpTask;
   friend class RunOpTask;
+  virtual bool IsSupportSummary() { return true; }
   virtual void CreateOutputTensors(const GraphId &graph_id, const std::vector<tensor::TensorPtr> &input_tensors,
                                    VectorRef *outputs,
                                    std::map<tensor::TensorPtr, session::KernelWithIndex> *tensor_to_node);