diff --git a/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.cc b/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.cc
index 9e9387dbeb6d6a3b9d2d3bc83ce3d58bc6a891d7..2c807b411b607de5c09bc8d6a7bc0a5458705273 100755
--- a/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.cc
+++ b/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.cc
@@ -336,27 +336,45 @@ void ComputePropagateScalesMkldnnPass::ComputeWeightScales(
   ComputeLstmWeightScales(graph, scope, "WeightX", "WeightH", var_quant_scales);
 }
 
-void ComputePropagateScalesMkldnnPass::UpdateScaleOpInScale(
+void ComputePropagateScalesMkldnnPass::UpdateScaleOpInOutScales(
     Node* op_node,
     const std::string& input_name,
     const std::string& output_name,
     StringPairMap* var_quant_scales) const {
-  auto iter = var_quant_scales->find(output_name);
-  if (iter != var_quant_scales->end()) {
-    auto pair = iter->second;
-    const auto tensor = pair.second;
-
-    const auto scale = PADDLE_GET_CONST(float, op_node->Op()->GetAttr("scale"));
-    Tensor tmp_tensor;
-    tmp_tensor.Resize(tensor.dims());
-    auto* data = tmp_tensor.mutable_data<float>(platform::CPUPlace());
-    for (int i = 0; i < tensor.numel(); i++) {
-      data[i] = data[i] * scale;
-    }
+  auto out_iter = var_quant_scales->find(output_name);
+  auto input_iter = var_quant_scales->find(input_name);
+  // All the input and output have scales
+  if (out_iter != var_quant_scales->end() &&
+      input_iter != var_quant_scales->end()) {
+    return;
+  }
+  const auto scale = PADDLE_GET_CONST(float, op_node->Op()->GetAttr("scale"));
+  if (std::abs(scale) < 1e-6 && out_iter != var_quant_scales->end()) {
+    return;
+  }
+
+  std::string name = input_name;
+  auto iter = out_iter;
+  if (input_iter != var_quant_scales->end()) {
+    iter = input_iter;
+    name = output_name;
+  }
 
-    auto new_pair = std::make_pair(pair.first, tmp_tensor);
-    var_quant_scales->insert(std::make_pair(input_name, new_pair));
+  phi::DenseTensor tmp_tensor;
+  auto pair = iter->second;
+  const auto tensor = pair.second;
+  tmp_tensor.Resize(tensor.dims());
+  auto* data = tmp_tensor.mutable_data<float>(platform::CPUPlace());
+  auto* src_data = tensor.data<float>();
+  for (int i = 0; i < tensor.numel(); i++) {
+    if (out_iter != var_quant_scales->end()) {
+      data[i] = src_data[i] / scale;
+    } else {
+      data[i] = src_data[i] * scale;
+    }
   }
+  auto new_pair = std::make_pair(pair.first, tmp_tensor);
+  var_quant_scales->insert(std::make_pair(name, new_pair));
 }
 
 std::unordered_set<std::string> ComputePropagateScalesMkldnnPass::UpdateScales(
@@ -403,10 +421,12 @@ std::unordered_set<std::string> ComputePropagateScalesMkldnnPass::UpdateScales(
       }
     } else if (op_name == "scale") {
       const std::string output_name = op_node->Op()->Output("Out")[0];
+      const std::string input_name = op_node->Op()->Input("X")[0];
       auto out_iter = var_quant_scales->find(output_name);
-      if (out_iter != var_quant_scales->end()) {
-        const std::string input_name = op_node->Op()->Input("X")[0];
-        UpdateScaleOpInScale(
+      auto input_iter = var_quant_scales->find(input_name);
+      if (out_iter != var_quant_scales->end() ||
+          input_iter != var_quant_scales->end()) {
+        UpdateScaleOpInOutScales(
             op_node, input_name, output_name, var_quant_scales);
       }
     }
diff --git a/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.h b/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.h
index 0b60eb9f36f3ad36e1fb256a1f241fa35ab44fff..001969887da6534c78ed999c8efdf405f5512f60 100755
--- a/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.h
+++ b/paddle/fluid/framework/ir/mkldnn/compute_propagate_scales_mkldnn_pass.h
@@ -79,10 +79,10 @@ class ComputePropagateScalesMkldnnPass : public FusePassBase {
   void UpdateReluOutputScales(ir::Graph* graph,
                               StringPairMap* var_quant_scales) const;
 
-  void UpdateScaleOpInScale(Node* op_node,
-                            const std::string& input_name,
-                            const std::string& output_name,
-                            StringPairMap* var_quant_scales) const;
+  void UpdateScaleOpInOutScales(Node* op_node,
+                                const std::string& input_name,
+                                const std::string& output_name,
+                                StringPairMap* var_quant_scales) const;
 
   std::unordered_set<std::string> UpdateScales(
       ir::Graph* graph,
diff --git a/paddle/fluid/inference/api/paddle_pass_builder.cc b/paddle/fluid/inference/api/paddle_pass_builder.cc
old mode 100644
new mode 100755
index 14e3ae1d4b6b278c28163c099b77115f86c626f1..31e3f453fc6b936c1abdfabcc335a4f9da21f0e4
--- a/paddle/fluid/inference/api/paddle_pass_builder.cc
+++ b/paddle/fluid/inference/api/paddle_pass_builder.cc
@@ -376,6 +376,7 @@ void CpuPassStrategy::EnableMkldnnInt8() {
     passes_.push_back("quant_dequant_mkldnn_pass");
     passes_.push_back("mkldnn_placement_pass");
     passes_.push_back("simplify_with_basic_ops_pass");
+    passes_.push_back("constant_folding_pass");
     passes_.push_back("layer_norm_fuse_pass");
     passes_.push_back("attention_lstm_fuse_pass");
     passes_.push_back("seqconv_eltadd_relu_fuse_pass");