From a149784db0802c302f957d6e8211f2dfc29f2f0a Mon Sep 17 00:00:00 2001
From: zhaoting <zhaoting23@huawei.com>
Date: Wed, 9 Sep 2020 10:08:04 +0800
Subject: [PATCH] fix CPU BatchNorm infer error

---
 .../cpu/mkldnn/fused_batch_norm_cpu_kernel.cc | 25 +++++++++++++------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/mindspore/ccsrc/backend/kernel_compiler/cpu/mkldnn/fused_batch_norm_cpu_kernel.cc b/mindspore/ccsrc/backend/kernel_compiler/cpu/mkldnn/fused_batch_norm_cpu_kernel.cc
index 8654998ec..9cb821409 100644
--- a/mindspore/ccsrc/backend/kernel_compiler/cpu/mkldnn/fused_batch_norm_cpu_kernel.cc
+++ b/mindspore/ccsrc/backend/kernel_compiler/cpu/mkldnn/fused_batch_norm_cpu_kernel.cc
@@ -50,11 +50,13 @@ void FusedBatchNormCPUKernel::InitKernel(const CNodePtr &kernel_node) {
   dnnl::memory::desc scale_bias_desc = GetDefaultMemDesc({2, channel});
   auto epsilon = AnfAlgo::GetNodeAttr<float>(kernel_node, "epsilon");
   auto prop_kind = dnnl::prop_kind::forward_inference;
+  auto normalization_flags = dnnl::normalization_flags::use_scale_shift | dnnl::normalization_flags::use_global_stats;
   if (is_train) {
     prop_kind = dnnl::prop_kind::forward_training;
+    normalization_flags = dnnl::normalization_flags::use_scale_shift;
   }
   dnnl::batch_normalization_forward::desc desc =
-    dnnl::batch_normalization_forward::desc(prop_kind, x_desc, epsilon, dnnl::normalization_flags::use_scale_shift);
+    dnnl::batch_normalization_forward::desc(prop_kind, x_desc, epsilon, normalization_flags);
   auto prim_desc = dnnl::batch_normalization_forward::primitive_desc(desc, MKLKernelEngine::Get().engine());
   primitive_ = std::make_shared<dnnl::batch_normalization_forward>(prim_desc);
   AddArgument(DNNL_ARG_SRC, x_desc);
@@ -74,14 +76,14 @@ bool FusedBatchNormCPUKernel::Launch(const std::vector<kernel::AddressPtr> &inpu
   auto wksp = reinterpret_cast<float *>(workspace[0]->addr);
   memcpy_s(wksp, workspace[0]->size, inputs[1]->addr, inputs[1]->size);
   memcpy_s(wksp + (inputs[1]->size / sizeof(float)), inputs[2]->size, inputs[2]->addr, inputs[2]->size);
-
-  SetArgumentHandle(DNNL_ARG_SRC, inputs[0]->addr);
-  SetArgumentHandle(DNNL_ARG_MEAN, outputs[3]->addr);
-  SetArgumentHandle(DNNL_ARG_VARIANCE, outputs[4]->addr);
-  SetArgumentHandle(DNNL_ARG_SCALE_SHIFT, workspace[0]->addr);
-  SetArgumentHandle(DNNL_ARG_DST, outputs[0]->addr);
-  ExecutePrimitive();
   if (is_train) {
+    SetArgumentHandle(DNNL_ARG_SRC, inputs[0]->addr);
+    SetArgumentHandle(DNNL_ARG_MEAN, outputs[3]->addr);
+    SetArgumentHandle(DNNL_ARG_VARIANCE, outputs[4]->addr);
+    SetArgumentHandle(DNNL_ARG_SCALE_SHIFT, workspace[0]->addr);
+    SetArgumentHandle(DNNL_ARG_DST, outputs[0]->addr);
+    ExecutePrimitive();
+
     auto moving_mean = reinterpret_cast<float *>(inputs[3]->addr);
     auto moving_variance = reinterpret_cast<float *>(inputs[4]->addr);
     auto mean = reinterpret_cast<float *>(outputs[3]->addr);
@@ -90,6 +92,13 @@ bool FusedBatchNormCPUKernel::Launch(const std::vector<kernel::AddressPtr> &inpu
       moving_mean[i] = moving_mean[i] * (1 - momentum) + mean[i] * momentum;
       moving_variance[i] = moving_variance[i] * (1 - momentum) + variance[i] * momentum;
     }
+  } else {
+    SetArgumentHandle(DNNL_ARG_SRC, inputs[0]->addr);
+    SetArgumentHandle(DNNL_ARG_MEAN, inputs[3]->addr);
+    SetArgumentHandle(DNNL_ARG_VARIANCE, inputs[4]->addr);
+    SetArgumentHandle(DNNL_ARG_SCALE_SHIFT, workspace[0]->addr);
+    SetArgumentHandle(DNNL_ARG_DST, outputs[0]->addr);
+    ExecutePrimitive();
   }
   return true;
 }
-- 
GitLab