From 2291b7f2e66ec6b002fdd32c06c5245d2abb744b Mon Sep 17 00:00:00 2001
From: wilfChen <chenweifeng720@huawei.com>
Date: Fri, 13 Nov 2020 16:23:44 +0800
Subject: [PATCH] dynamic shape check

---
 .../backend/kernel_compiler/gpu/gpu_kernel.h  |  1 -
 .../gpu/batch_norm_add_relu_grad_fusion.cc    | 39 ++++++++++++-------
 .../ccsrc/backend/session/session_basic.cc    |  2 +-
 3 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/mindspore/ccsrc/backend/kernel_compiler/gpu/gpu_kernel.h b/mindspore/ccsrc/backend/kernel_compiler/gpu/gpu_kernel.h
index 54ad8b99ea..863829e55b 100644
--- a/mindspore/ccsrc/backend/kernel_compiler/gpu/gpu_kernel.h
+++ b/mindspore/ccsrc/backend/kernel_compiler/gpu/gpu_kernel.h
@@ -251,7 +251,6 @@ class GpuKernel : public KernelMod {
 
   device::DynamicKernelPtr dynamic_kernel_;
 };
-
 }  // namespace kernel
 }  // namespace mindspore
 
diff --git a/mindspore/ccsrc/backend/optimizer/gpu/batch_norm_add_relu_grad_fusion.cc b/mindspore/ccsrc/backend/optimizer/gpu/batch_norm_add_relu_grad_fusion.cc
index ae9251c543..d80f865260 100644
--- a/mindspore/ccsrc/backend/optimizer/gpu/batch_norm_add_relu_grad_fusion.cc
+++ b/mindspore/ccsrc/backend/optimizer/gpu/batch_norm_add_relu_grad_fusion.cc
@@ -110,44 +110,57 @@ void ReplaceOutput(const FuncGraphPtr &graph, const AnfNodePtr &bn_grad, const A
   manager->Replace(relu_grad, bn_add_relu_grad_output[kBNAddReluGradOutputNum - 1]);
   return;
 }
-}  // namespace
 
-const BaseRef BatchNormAddReluGradFusion::DefinePattern() const {
-  VectorRef relu_grad = VectorRef({prim::kPrimReluGrad, dy_, y_});
-  VectorRef batch_norm_grad =
-    VectorRef({prim::kPrimFusedBatchNormGradEx, relu_grad, x_, scale_, save_mean_, save_var_, reserve_});
-  return batch_norm_grad;
-}
-
-const AnfNodePtr BatchNormAddReluGradFusion::Process(const FuncGraphPtr &graph, const AnfNodePtr &node,
-                                                     const EquivPtr &) const {
+bool PatternCheck(const FuncGraphPtr &graph, const AnfNodePtr &node) {
   MS_EXCEPTION_IF_NULL(graph);
   MS_EXCEPTION_IF_NULL(node);
   auto format_attr = AnfAlgo::GetCNodePrimitive(node)->GetAttr("data_format");
   MS_EXCEPTION_IF_NULL(format_attr);
   auto format = GetValue<std::string>(format_attr);
   if (AnfAlgo::GetInputFormat(node, 0) != kOpFormat_NHWC && format != "NHWC") {
-    return nullptr;
+    return false;
   }
 
   auto relu_grad = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(node), 0);
   MS_EXCEPTION_IF_NULL(relu_grad);
   auto relu_users = GetRealNodeUsedList(graph, relu_grad);
   if (relu_users->size() != 2) {
-    return nullptr;
+    return false;
   }
 
   // process pattern as Relu(TensorAdd(BN#0, BN#1))
   auto tuple_getitem = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(node), 5);
   MS_EXCEPTION_IF_NULL(tuple_getitem);
   if (!utils::isa<CNodePtr>(tuple_getitem) || AnfAlgo::GetCNodeName(tuple_getitem) != prim::kPrimTupleGetItem->name()) {
-    return nullptr;
+    return false;
   }
   auto forward_node = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(tuple_getitem), 0);
   if (AnfAlgo::GetCNodeName(forward_node) != kFusedBatchNormExWithAddAndActivation) {
+    return false;
+  }
+
+  return true;
+}
+}  // namespace
+
+const BaseRef BatchNormAddReluGradFusion::DefinePattern() const {
+  VectorRef relu_grad = VectorRef({prim::kPrimReluGrad, dy_, y_});
+  VectorRef batch_norm_grad =
+    VectorRef({prim::kPrimFusedBatchNormGradEx, relu_grad, x_, scale_, save_mean_, save_var_, reserve_});
+  return batch_norm_grad;
+}
+
+const AnfNodePtr BatchNormAddReluGradFusion::Process(const FuncGraphPtr &graph, const AnfNodePtr &node,
+                                                     const EquivPtr &) const {
+  MS_EXCEPTION_IF_NULL(graph);
+  MS_EXCEPTION_IF_NULL(node);
+
+  if (!PatternCheck(graph, node)) {
     return nullptr;
   }
 
+  auto relu_grad = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(node), 0);
+  MS_EXCEPTION_IF_NULL(relu_grad);
   auto dy = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(relu_grad), 0);
   MS_EXCEPTION_IF_NULL(dy);
   auto y = AnfAlgo::GetInputNode(utils::cast<CNodePtr>(relu_grad), 1);
diff --git a/mindspore/ccsrc/backend/session/session_basic.cc b/mindspore/ccsrc/backend/session/session_basic.cc
index d507bf56e2..a3f2880355 100644
--- a/mindspore/ccsrc/backend/session/session_basic.cc
+++ b/mindspore/ccsrc/backend/session/session_basic.cc
@@ -1432,7 +1432,7 @@ void SessionBasic::RunGraphAsync(const GraphId &graph_id, const std::vector<tens
 }
 
 bool IsDynamicShape(const NotNull<abstract::ShapePtr> &shape) {
-  return std::any_of(shape->shape().begin(), shape->shape().end(), [](int s) { return s < 0; });
+  return std::any_of(shape->shape().begin(), shape->shape().end(), [](int64_t s) { return s < 0; });
 }
 
 bool IsNodeOutputDynamicShape(const CNodePtr &anf_node_ptr) {