1 files changed, 28 insertions, 0 deletions
diff --git a/src/armnn/backends/RefWorkloads/RefMultiplicationFloat32Workload.cpp b/src/armnn/backends/RefWorkloads/RefMultiplicationFloat32Workload.cpp
new file mode 100644
index 0000000000..ed68b1f6db
--- /dev/null
+++ b/src/armnn/backends/RefWorkloads/RefMultiplicationFloat32Workload.cpp
@@ -0,0 +1,28 @@
+//
+// Copyright © 2017 Arm Ltd. All rights reserved.
+// See LICENSE file in the project root for full license information.
+//
+
+#include "RefMultiplicationFloat32Workload.hpp"
+
+#include "Multiplication.hpp"
+#include "RefWorkloadUtils.hpp"
+
+#include "Profiling.hpp"
+
+namespace armnn
+{
+
+void RefMultiplicationFloat32Workload::Execute() const
+{
+    ARMNN_SCOPED_PROFILING_EVENT(Compute::CpuRef, "RefMultiplicationFloat32Workload_Execute");
+
+    const TensorInfo& inputInfo0 = GetTensorInfo(m_Data.m_Inputs[0]);
+
+    float* outputData = GetOutputTensorDataFloat(0, m_Data);
+    const float* inputData0 = GetInputTensorDataFloat(0, m_Data);
+    const float* inputData1 = GetInputTensorDataFloat(1, m_Data);
+    Multiplication(inputData0, inputData1, inputInfo0.GetNumElements(), outputData);
+}
+
+} //namespace armnn