diff --git a/src/caffe/layers/batch_norm_layer.cpp b/src/caffe/layers/batch_norm_layer.cpp
index bc19c53b8fa..a69d8f99316 100644
--- a/src/caffe/layers/batch_norm_layer.cpp
+++ b/src/caffe/layers/batch_norm_layer.cpp
@@ -48,9 +48,7 @@ void BatchNormLayer<Dtype>::Reshape(const vector<Blob<Dtype>*>& bottom,
   mean_.Reshape(sz);
   variance_.Reshape(sz);
   temp_.ReshapeLike(*bottom[0]);
-  if (use_global_stats_) {
-    x_norm_.ReshapeLike(*bottom[0]);
-  }
+  x_norm_.ReshapeLike(*bottom[0]);
   sz[0]=bottom[0]->shape(0);
   batch_sum_multiplier_.Reshape(sz);
 
diff --git a/src/caffe/layers/smooth_L1_loss_ohem_layer.cu b/src/caffe/layers/smooth_L1_loss_ohem_layer.cu
index 669a4f8e48e..515c340b98e 100644
--- a/src/caffe/layers/smooth_L1_loss_ohem_layer.cu
+++ b/src/caffe/layers/smooth_L1_loss_ohem_layer.cu
@@ -75,7 +75,7 @@ namespace caffe {
 
     // Output per-instance loss
     if (top.size() >= 2) {
-      kernel_channel_sum<Dtype> << <CAFFE_GET_BLOCKS(top[0]->count()),
+      kernel_channel_sum<Dtype> << <CAFFE_GET_BLOCKS(top[1]->count()),
         CAFFE_CUDA_NUM_THREADS >> > (outer_num_, bottom[0]->channels(),
         inner_num_, errors_.gpu_data(), top[1]->mutable_gpu_data());
     }