COMPMID-3239: Implement QSYMM16 LayerNormalizationKernel for CL

Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Change-Id: Ib1577c4a9aa29293a903731b2a4083b0d2243e1e
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2994
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
diff --git a/arm_compute/core/CL/CLKernels.h b/arm_compute/core/CL/CLKernels.h
index b265aa2..583cf27 100644
--- a/arm_compute/core/CL/CLKernels.h
+++ b/arm_compute/core/CL/CLKernels.h
@@ -125,6 +125,7 @@
 #include "arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h"
 #include "arm_compute/core/CL/kernels/CLPoolingLayerKernel.h"
 #include "arm_compute/core/CL/kernels/CLPriorBoxLayerKernel.h"
+#include "arm_compute/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h"
 #include "arm_compute/core/CL/kernels/CLQuantizationLayerKernel.h"
 #include "arm_compute/core/CL/kernels/CLROIAlignLayerKernel.h"
 #include "arm_compute/core/CL/kernels/CLROIPoolingLayerKernel.h"
diff --git a/arm_compute/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h b/arm_compute/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h
new file mode 100644
index 0000000..3206eda
--- /dev/null
+++ b/arm_compute/core/CL/kernels/CLQLSTMLayerNormalizationKernel.h
@@ -0,0 +1,88 @@
+/*
+ * Copyright (c) 2020 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef ARM_COMPUTE_CLQLSTMLAYERVNORMALIZATIONKERNEL_H
+#define ARM_COMPUTE_CLQLSTMLAYERVNORMALIZATIONKERNEL_H
+
+#include "arm_compute/core/CL/ICLKernel.h"
+
+namespace arm_compute
+{
+class ICLTensor;
+
+/** Interface for the kernel to do layer normalization. */
+class CLQLSTMLayerNormalizationKernel : public ICLKernel
+{
+public:
+    /** Default constructor */
+    CLQLSTMLayerNormalizationKernel();
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    CLQLSTMLayerNormalizationKernel(const CLQLSTMLayerNormalizationKernel &) = delete;
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    CLQLSTMLayerNormalizationKernel &operator=(const CLQLSTMLayerNormalizationKernel &) = delete;
+    /** Allow instances of this class to be moved */
+    CLQLSTMLayerNormalizationKernel(CLQLSTMLayerNormalizationKernel &&) = default;
+    /** Allow instances of this class to be moved */
+    CLQLSTMLayerNormalizationKernel &operator=(CLQLSTMLayerNormalizationKernel &&) = default;
+    /** Default destructor */
+    ~CLQLSTMLayerNormalizationKernel() = default;
+    /** Initialise the kernel's input and outputs.
+     *
+     * @param[in]  input  Source tensor with 2 dimensions. Data types supported: QSYMM16.
+     * @param[out] output Destination tensor. Data type supported: same as @p input
+     * @param[in]  weight Weight tensor. Data types supported: Same as @p input.
+     * @param[in]  bias   Bias tensor. Data types supported: S32.
+     *
+     */
+    void configure(const ICLTensor *input, ICLTensor *output, const ICLTensor *weight, const ICLTensor *bias);
+    /** Initialise the kernel's input and outputs.
+     *
+     * @param[in]  compile_context The compile context to be used.
+     * @param[in]  input           Source tensor with 2 dimensions. Data types supported: QSYMM16.
+     * @param[out] output          Destination tensor. Data type supported: same as @p input
+     * @param[in]  weight          Weight tensor. Data types supported: Same as @p input.
+     * @param[in]  bias            Bias tensor. Data types supported: S32.
+     */
+    void configure(CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *weight, const ICLTensor *bias);
+    /** Static function to check if given info will lead to a valid configuration of @ref CLQLSTMLayerNormalizationKernel
+     *
+     * @param[in] input  Source tensor info with 2 dimensions. Data types supported: QSYMM16.
+     * @param[in] output Destination info tensor. Data type supported: same as @p input
+     * @param[in] weight Weight info tensor. Data types supported: Same as @p input.
+     * @param[in] bias   Bias tensor info. Data types supported: S32.
+     *
+     * @return a status
+     */
+    static Status validate(const ITensorInfo *input, ITensorInfo *output, const ITensorInfo *weight, const ITensorInfo *bias);
+
+    // Inherited methods overridden:
+    void run(const Window &window, cl::CommandQueue &queue) override;
+
+private:
+    const ICLTensor *_input;
+    const ICLTensor *_weight;
+    const ICLTensor *_bias;
+    ICLTensor       *_output;
+};
+} // namespace arm_compute
+#endif /* ARM_COMPUTE_CLQLSTMLAYERVNORMALIZATIONKERNEL_H */
diff --git a/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h b/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h
index 36a35d3..27c1c51 100644
--- a/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h
+++ b/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019 ARM Limited.
+ * Copyright (c) 2019-2020 ARM Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -66,4 +66,4 @@
     static Status validate(const ITensorInfo *boxes, const ITensorInfo *pred_boxes, const ITensorInfo *deltas, const BoundingBoxTransformInfo &info);
 };
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_CLBOUNDINGBOXTRANSFORM_H */
+#endif /* ARM_COMPUTE_NEBOUNDINGBOXTRANSFORM_H */