Port NEGEMMLowp Part 1

Details:
Port NEGEMMLowpQuantizeDownInt32ScaleKernel to CpuGemmLowpQuantizeDownInt32ScaleKernel
Port NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel to CpuGemmLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel
Port NEGEMMLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel to CpuGemmLowpQuantizeDownInt32ToInt8ScaleByFixedPointKernel
Port NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel to CpuGemmLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel
Port NEGEMMLowpOutputStage functions to CpuGemmLowpOutputStage operators

Partially Resolves: COMPMID-4403

Change-Id: I6d5f45e43f35d731d564ed3b5c0e804d2a318fb1
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5833
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 25 insertions, 25 deletions

diff --git a/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h b/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h
index 2f0c753691..bcb89d997d 100644
--- a/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h
+++ b/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h
@@ -50,7 +50,7 @@ class ITensor;
  * This function calls the following functions/kernels:
  *
  * -# @ref NEGEMMLowpMatrixMultiplyCore                          Quantized matrix multiplication core. Accumulators are 32-bit integers
- * -# @ref NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint   Convert 32-bit integers into QSYMM16
+ * -# @ref NEGEMMLowpOutputStage                                 Convert 32-bit integers into QSYMM16
  * -# @ref NETranspose                                           Matrix transpose
  * -# @ref NEConcatenateLayer                                    Tensor concatenation
  * -# @ref NEActivationLayer                                     Activation functions (tanh and logistic)
@@ -147,30 +147,30 @@ private:
     MemoryGroup _memory_group;
 
     // Functions used
-    NEGEMMLowpMatrixMultiplyCore                        _gemmlowp;
-    NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint _output_stage;
-    NETranspose                                         _transpose_weights;
-    NEConcatenateLayer                                  _concat_input_weights;
-    NEConcatenateLayer                                  _concat_recurrent_weights;
-    NEConcatenateLayer                                  _concat_weights;
-    NEConcatenateLayer                                  _concat_inputs;
-    NEConcatenateLayer                                  _concat_bias;
-    NEActivationLayer                                   _sigmoid_forget_gate;
-    NEActivationLayer                                   _sigmoid_input_gate;
-    NEActivationLayer                                   _sigmoid_output_gate;
-    NEActivationLayer                                   _tanh_modulation_gate;
-    NEActivationLayer                                   _tanh_output_state;
-    NEArithmeticAddition                                _add1;
-    NEArithmeticAddition                                _add2;
-    NEPixelWiseMultiplication                           _mul1;
-    NEPixelWiseMultiplication                           _mul2;
-    NEPixelWiseMultiplication                           _mul3;
-    NESlice                                             _slice_input_tensor;
-    NESlice                                             _slice_forget_tensor;
-    NESlice                                             _slice_cell_tensor;
-    NESlice                                             _slice_output_tensor;
-    NEDequantizationLayer                               _dequantize;
-    NEQuantizationLayer                                 _quantize;
+    NEGEMMLowpMatrixMultiplyCore _gemmlowp;
+    NEGEMMLowpOutputStage        _output_stage;
+    NETranspose                  _transpose_weights;
+    NEConcatenateLayer           _concat_input_weights;
+    NEConcatenateLayer           _concat_recurrent_weights;
+    NEConcatenateLayer           _concat_weights;
+    NEConcatenateLayer           _concat_inputs;
+    NEConcatenateLayer           _concat_bias;
+    NEActivationLayer            _sigmoid_forget_gate;
+    NEActivationLayer            _sigmoid_input_gate;
+    NEActivationLayer            _sigmoid_output_gate;
+    NEActivationLayer            _tanh_modulation_gate;
+    NEActivationLayer            _tanh_output_state;
+    NEArithmeticAddition         _add1;
+    NEArithmeticAddition         _add2;
+    NEPixelWiseMultiplication    _mul1;
+    NEPixelWiseMultiplication    _mul2;
+    NEPixelWiseMultiplication    _mul3;
+    NESlice                      _slice_input_tensor;
+    NESlice                      _slice_forget_tensor;
+    NESlice                      _slice_cell_tensor;
+    NESlice                      _slice_output_tensor;
+    NEDequantizationLayer        _dequantize;
+    NEQuantizationLayer          _quantize;
 
     // Tensor pointers
     const ITensor *_input_to_input_weights;