COMPMID-706 - Add GEMMLowp output stage for scaling by a fixed point number

DoD:
- Implement NEON kernel for quantizing down the gemmlowp result. The
  result should be scaled by a fixedpoint number
- Implement OpenCL kernel for quantizing down the gemmlowp result. The
  result should be scaled by a fixedpoint number
- Add test for validating the result

Required for:
- Integration of GEMMLowp in Android NN
- Convolution quantized
- Fully connected quantized

Change-Id: Ia963d25d695471e963961fb49a5600e78374ac4f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110981
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>

6 files changed, 200 insertions, 3 deletions

diff --git a/arm_compute/core/NEON/NEAsymm.h b/arm_compute/core/NEON/NEAsymm.h
new file mode 100644
index 0000000000..d227d3ccbe
--- /dev/null
+++ b/arm_compute/core/NEON/NEAsymm.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright (c) 2017 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_NEASYMM_H__
+#define __ARM_COMPUTE_NEASYMM_H__
+
+#include <arm_neon.h>
+
+namespace arm_compute
+{
+/** Round to the nearest division by a power-of-two using exponent
+ *
+ * @note This function calculates the following expression: (x + 2^n -1 ) / 2^n where n = exponent
+ *
+ * @param[in] x        Vector of 4 elements
+ * @param[in] exponent Integer value used to round to nearest division by a power-of-two
+ *
+ * @return the nearest division by a power-of-two using exponent
+ */
+int32x4_t rounding_divide_by_pow2(int32x4_t x, int exponent);
+} // namespace arm_compute
+#include "arm_compute/core/NEON/NEAsymm.inl"
+#endif // __ARM_COMPUTE_NEASYMM_H__
\ No newline at end of file
diff --git a/arm_compute/core/NEON/NEAsymm.inl b/arm_compute/core/NEON/NEAsymm.inl
new file mode 100644
index 0000000000..bbce308b35
--- /dev/null
+++ b/arm_compute/core/NEON/NEAsymm.inl
@@ -0,0 +1,33 @@
+/*
+ * Copyright (c) 2017 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.
+ */
+namespace arm_compute
+{
+inline int32x4_t rounding_divide_by_pow2(int32x4_t x, int exponent)
+{
+    const int32x4_t shift_vec  = vdupq_n_s32(-exponent);
+    const int32x4_t fixup      = vshrq_n_s32(vandq_s32(x, shift_vec), 31);
+    const int32x4_t fixed_up_x = vqaddq_s32(x, fixup);
+    return vrshlq_s32(fixed_up_x, shift_vec);
+}
+} // namespace arm_compute
\ No newline at end of file
diff --git a/arm_compute/core/NEON/NEKernels.h b/arm_compute/core/NEON/NEKernels.h
index 281f06305f..b23e2ac5a3 100644
--- a/arm_compute/core/NEON/NEKernels.h
+++ b/arm_compute/core/NEON/NEKernels.h
@@ -66,6 +66,7 @@
 #include "arm_compute/core/NEON/kernels/NEGEMMInterleaveBlockedKernel.h"
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h"
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h"
+#include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h"
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel.h"
 #include "arm_compute/core/NEON/kernels/NEGEMMLowpReductionKernel.h"
 #include "arm_compute/core/NEON/kernels/NEGEMMMatrixAccumulateBiasesKernel.h"
diff --git a/arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h b/arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h
index 989260de11..ac0af7cff3 100644
--- a/arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h
+++ b/arm_compute/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h
@@ -77,6 +77,8 @@ public:
      *                           Note: vector_sum_row can be a nullptr in case b_offset = 0. Data type supported: same as @p mm_result
      * @param[in] a_offset       Offset to be added to each element of the matrix A.
      * @param[in] b_offset       Offset to be added to each element of the matrix B.
+     *
+     * @return an error status
      */
     static Error validate(const ITensorInfo *mm_result, const ITensorInfo *vector_sum_col, const ITensorInfo *vector_sum_row, int32_t a_offset, int32_t b_offset);
 
diff --git a/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h b/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
new file mode 100644
index 0000000000..24ba54ebdf
--- /dev/null
+++ b/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel.h
@@ -0,0 +1,116 @@
+/*
+ * Copyright (c) 2017 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_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEBYFIXEDPOINTKERNEL_H__
+#define __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEBYFIXEDPOINTKERNEL_H__
+
+#include "arm_compute/core/NEON/INEKernel.h"
+
+namespace arm_compute
+{
+class ITensor;
+
+/** NEON kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8
+ *
+ * This kernel takes a final int32 accumulator value (the output of @ref NEGEMMLowpMatrixMultiplyKernel), and processes it to obtain the final QASYMM8 value.
+ * The following computations will be performed by the kernel:
+ *
+ *  -# Compute fixed point multiplication between each entry of input by result_fixedpoint_multiplier
+ *  -# Add bias to final result if bias tensor is not a nullptr
+ *  -# Round to nearest division by a power-of-two using result_shift
+ *  -# Add offset to each result
+ *  -# Clamp the value between the specified min and max bounds
+ *  -# Clamp the resulting int32 values to the [0..255] range and cast to QASYMM8.
+ *
+ */
+class NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel : public INEKernel
+{
+public:
+    /** Constructor */
+    NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel();
+    /** Prevent instances of this class from being copied (As this class contains pointers)*/
+    NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel(const NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &) = delete;
+    /** Prevent instances of this class from being copied (As this class contains pointers)*/
+    NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &operator=(const NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &) = delete;
+    /** Allow instances of this class to be moved */
+    NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel(NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &&) = default;
+    /** Allow instances of this class to be moved */
+    NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &operator=(NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel &&) = default;
+    /** Initialise the kernel's input and output.
+     *
+     * @param[in]  input                        Input tensor. Data type supported: S32
+     * @param[in]  bias                         Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required.
+     *                                          Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input.
+     * @param[out] output                       Output tensor. Data type supported: Data type supported: QASYMM8
+     * @param[in]  result_fixedpoint_multiplier Fixed point value to be multiplied to each element of the input matrix when once the result_offset has been add
+     * @param[in]  result_shift                 Integer value used to round to nearest division by a power-of-two the result after the fixed point multiplication
+     * @param[in]  result_offset_after_shift    Offset to be applied to result before converting it back to QASYMM8
+     * @param[in]  min                          (Optional) Min value used to saturate down the output result before converting back to QASYMM8
+     * @param[in]  max                          (Optional) Max value used to saturate up the output result before converting back to QASYMM8,
+     *                                          Along with @p min, this value can be used to implement "rectified linear unit" activation functions
+     */
+    void configure(const ITensor *input, const ITensor *bias, ITensor *output, int result_fixedpoint_multiplier, int result_shift, int result_offset_after_shift, int min = 0, int max = 0);
+    /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel
+     *
+     * @param[in] input  Input tensor. Data type supported: S32
+     * @param[in] bias   Biases tensor. Only shared biases supported and it can be a nullptr if the biases addition is not required.
+     *                   Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input.
+     * @param[in] output Output tensor. Data type supported: Data type supported: QASYMM8
+     * @param[in] min    (Optional) Min value used to saturate down the output result before converting back to QASYMM8
+     * @param[in] max    (Optional) Max value used to saturate up the output result before converting back to QASYMM8,
+     *                   Along with @p min, this value can be used to implement "rectified linear unit" activation functions
+     *
+     * @return an error status
+     */
+    static Error validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0);
+
+    // Inherited methods overridden:
+    void run(const Window &window, const ThreadInfo &info) override;
+
+private:
+    /** Template function to run the NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel
+     *
+     * @param[in] window Region on which to execute the kernel. (Must be a valid region of the window returned by window()).
+     */
+    template <bool is_bounded_relu>
+    void run(const Window &window);
+
+    /** Common signature for all the specialised NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel functions
+     *
+     * @param[in] window Region on which to execute the kernel.
+     */
+    using QuantizeDownFunctionPtr = void (NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel::*)(const Window &window);
+
+    QuantizeDownFunctionPtr _func;
+    const ITensor          *_input;
+    const ITensor          *_bias;
+    ITensor                *_output;
+    int                     _result_fixedpoint_multiplier;
+    int                     _result_shift;
+    int                     _result_offset_after_shift;
+    int                     _min;
+    int                     _max;
+};
+} // namespace arm_compute
+
+#endif /* __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEBYFIXEDPOINTKERNEL_H__ */
diff --git a/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel.h b/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel.h
index a522069330..d873a889d2 100644
--- a/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel.h
+++ b/arm_compute/core/NEON/kernels/NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel.h
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALE_H__
-#define __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALE_H__
+#ifndef __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEKERNEL_H__
+#define __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEKERNEL_H__
 
 #include "arm_compute/core/NEON/INEKernel.h"
 
@@ -79,6 +79,8 @@ public:
      * @param[in] min    (Optional) Min value used to saturate down the output result before converting back to QASYMM8
      * @param[in] max    (Optional) Max value used to saturate up the output result before converting back to QASYMM8,
      *                   Along with @p min, this value can be used to implement "rectified linear unit" activation functions
+     *
+     * @return an error status
      */
     static Error validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, int min = 0, int max = 0);
 
@@ -111,4 +113,4 @@ private:
 };
 } // namespace arm_compute
 
-#endif /* __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALE_H__ */
+#endif /* __ARM_COMPUTE_NEGEMMLOWPQUANTIZEDOWNINT32TOUINT8SCALEKERNEL_H__ */