chromium/third_party/libaom/source/libaom/aom_dsp/x86/sum_squares_sse2.c

/*
 * Copyright (c) 2016, Alliance for Open Media. All rights reserved.
 *
 * This source code is subject to the terms of the BSD 2 Clause License and
 * the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
 * was not distributed with this source code in the LICENSE file, you can
 * obtain it at www.aomedia.org/license/software. If the Alliance for Open
 * Media Patent License 1.0 was not distributed with this source code in the
 * PATENTS file, you can obtain it at www.aomedia.org/license/patent.
 */

#include <assert.h>
#include <emmintrin.h>
#include <stdio.h>

#include "aom_dsp/x86/synonyms.h"
#include "aom_dsp/x86/sum_squares_sse2.h"
#include "config/aom_dsp_rtcd.h"

static inline __m128i xx_loadh_64(__m128i a, const void *b) { … }

static inline uint64_t xx_cvtsi128_si64(__m128i a) { … }

static inline __m128i sum_squares_i16_4x4_sse2(const int16_t *src, int stride) { … }

uint64_t aom_sum_squares_2d_i16_4x4_sse2(const int16_t *src, int stride) { … }

uint64_t aom_sum_sse_2d_i16_4x4_sse2(const int16_t *src, int stride, int *sum) { … }

uint64_t aom_sum_squares_2d_i16_4xn_sse2(const int16_t *src, int stride,
                                         int height) { … }

uint64_t aom_sum_sse_2d_i16_4xn_sse2(const int16_t *src, int stride, int height,
                                     int *sum) { … }

#ifdef __GNUC__
// This prevents GCC/Clang from inlining this function into
// aom_sum_squares_2d_i16_sse2, which in turn saves some stack
// maintenance instructions in the common case of 4x4.
__attribute__((noinline))
#endif
uint64_t
aom_sum_squares_2d_i16_nxn_sse2(const int16_t *src, int stride, int width,
                                int height) { … }

#ifdef __GNUC__
// This prevents GCC/Clang from inlining this function into
// aom_sum_sse_2d_i16_nxn_sse2, which in turn saves some stack
// maintenance instructions in the common case of 4x4.
__attribute__((noinline))
#endif
uint64_t
aom_sum_sse_2d_i16_nxn_sse2(const int16_t *src, int stride, int width,
                            int height, int *sum) { … }

uint64_t aom_sum_squares_2d_i16_sse2(const int16_t *src, int stride, int width,
                                     int height) { … }

uint64_t aom_sum_sse_2d_i16_sse2(const int16_t *src, int src_stride, int width,
                                 int height, int *sum) { … }

//////////////////////////////////////////////////////////////////////////////
// 1D version
//////////////////////////////////////////////////////////////////////////////

static uint64_t aom_sum_squares_i16_64n_sse2(const int16_t *src, uint32_t n) { … }

uint64_t aom_sum_squares_i16_sse2(const int16_t *src, uint32_t n) { … }

// Accumulate sum of 16-bit elements in the vector
static inline int32_t mm_accumulate_epi16(__m128i vec_a) { … }

// Accumulate sum of 32-bit elements in the vector
static inline int32_t mm_accumulate_epi32(__m128i vec_a) { … }

uint64_t aom_var_2d_u8_sse2(uint8_t *src, int src_stride, int width,
                            int height) { … }

uint64_t aom_var_2d_u16_sse2(uint8_t *src, int src_stride, int width,
                             int height) { … }