COMPMID-2237

Implement SPACE_TO_DEPTH for NEON

Change-Id: I9f427bceca6da52671e0096be08772612f4be152
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1227
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>

2 files changed, 81 insertions, 0 deletions

diff --git a/arm_compute/core/NEON/NEKernels.h b/arm_compute/core/NEON/NEKernels.h
index d19b0c71ba..119f9b8702 100644
--- a/arm_compute/core/NEON/NEKernels.h
+++ b/arm_compute/core/NEON/NEKernels.h
@@ -129,6 +129,7 @@
 #include "arm_compute/core/NEON/kernels/NESobel7x7Kernel.h"
 #include "arm_compute/core/NEON/kernels/NESoftmaxLayerKernel.h"
 #include "arm_compute/core/NEON/kernels/NESpaceToBatchLayerKernel.h"
+#include "arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h"
 #include "arm_compute/core/NEON/kernels/NEStackLayerKernel.h"
 #include "arm_compute/core/NEON/kernels/NEStridedSliceKernel.h"
 #include "arm_compute/core/NEON/kernels/NETableLookupKernel.h"
diff --git a/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h b/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h
new file mode 100644
index 0000000000..c9ecdd26f8
--- /dev/null
+++ b/arm_compute/core/NEON/kernels/NESpaceToDepthLayerKernel.h
@@ -0,0 +1,80 @@
+/*
+ * Copyright (c) 2019 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_NESPACETODEPTHLAYERKERNEL_H__
+#define __ARM_COMPUTE_NESPACETODEPTHLAYERKERNEL_H__
+
+#include "arm_compute/core/NEON/INEKernel.h"
+#include "arm_compute/core/Types.h"
+
+namespace arm_compute
+{
+class ITensor;
+
+/** Interface for the space to depth kernel */
+class NESpaceToDepthLayerKernel : public INEKernel
+{
+public:
+    const char *name() const override
+    {
+        return "NESpaceToDepthLayerKernel";
+    }
+    /** Default constructor */
+    NESpaceToDepthLayerKernel();
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    NESpaceToDepthLayerKernel(const NESpaceToDepthLayerKernel &) = delete;
+    /** Prevent instances of this class from being copied (As this class contains pointers) */
+    NESpaceToDepthLayerKernel &operator=(const NESpaceToDepthLayerKernel &) = delete;
+    /** Allow instances of this class to be moved */
+    NESpaceToDepthLayerKernel(NESpaceToDepthLayerKernel &&) = default;
+    /** Allow instances of this class to be moved */
+    NESpaceToDepthLayerKernel &operator=(NESpaceToDepthLayerKernel &&) = default;
+    /** Default destructor */
+    ~NESpaceToDepthLayerKernel() = default;
+    /** Initialise the kernel's inputs and output.
+     *
+     * @param[in]  input       Tensor input. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[out] output      Tensor output. Data types supported: same as @p input
+     * @param[in]  block_shape Block shape value
+     */
+    void configure(const ITensor *input, ITensor *output, int32_t block_shape);
+    /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToDepthLayerKernel
+     *
+     * @param[in] input       Tensor input info. Supported tensor rank: 4. Data types supported: U8/S8/QASYMM8/U16/S16/F16/U32/S32/F32.
+     * @param[in] output      Tensor output info. Data types supported: same as @p input
+     * @param[in] block_shape Block shape value
+     *
+     * @return a status
+     */
+    static Status validate(const ITensorInfo *input, const ITensorInfo *output, int32_t block_shape);
+
+    // Inherited methods overridden:
+    void run(const Window &window, const ThreadInfo &info) override;
+
+private:
+    const ITensor *_input;       /**< Source tensor */
+    ITensor       *_output;      /**< Destination tensor */
+    int32_t        _block_shape; /**< Block shape */
+};
+} // namespace arm_compute
+#endif /* __ARM_COMPUTE_NESPACETODEPTHLAYERKERNEL_H__ */