chromium/third_party/libaom/source/libaom/av1/common/x86/resize_ssse3.c

/*
 *
 * Copyright (c) 2020, 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 <tmmintrin.h>  // SSSE3
#include "config/av1_rtcd.h"
#include "config/aom_scale_rtcd.h"

#include "aom_dsp/x86/convolve_sse2.h"
#include "aom_dsp/x86/convolve_ssse3.h"
#include "aom_dsp/x86/mem_sse2.h"
#include "aom_dsp/x86/transpose_sse2.h"
#include "av1/common/resize.h"

static inline __m128i scale_plane_2_to_1_phase_0_kernel(
    const uint8_t *const src, const __m128i *const mask) { … }

static inline void shuffle_filter_odd_ssse3(const int16_t *const filter,
                                            __m128i *const f) { … }

static inline __m128i convolve8_8_even_offset_ssse3(const __m128i *const s,
                                                    const __m128i *const f) { … }

static inline __m128i convolve8_8_odd_offset_ssse3(const __m128i *const s,
                                                   const __m128i *const f) { … }

static void scale_plane_2_to_1_phase_0(const uint8_t *src,
                                       const ptrdiff_t src_stride, uint8_t *dst,
                                       const ptrdiff_t dst_stride,
                                       const int dst_w, const int dst_h) { … }

static void scale_plane_4_to_1_phase_0(const uint8_t *src,
                                       const ptrdiff_t src_stride, uint8_t *dst,
                                       const ptrdiff_t dst_stride,
                                       const int dst_w, const int dst_h) { … }

static inline __m128i scale_plane_bilinear_kernel(const __m128i *const s,
                                                  const __m128i c0c1) { … }

static void scale_plane_2_to_1_bilinear(const uint8_t *src,
                                        const ptrdiff_t src_stride,
                                        uint8_t *dst,
                                        const ptrdiff_t dst_stride,
                                        const int dst_w, const int dst_h,
                                        const __m128i c0c1) { … }

static void scale_plane_4_to_1_bilinear(const uint8_t *src,
                                        const ptrdiff_t src_stride,
                                        uint8_t *dst,
                                        const ptrdiff_t dst_stride,
                                        const int dst_w, const int dst_h,
                                        const __m128i c0c1) { … }

static void scale_plane_4_to_1_general(const uint8_t *src, const int src_stride,
                                       uint8_t *dst, const int dst_stride,
                                       const int w, const int h,
                                       const int16_t *const coef,
                                       uint8_t *const temp_buffer) { … }

static void scale_plane_2_to_1_general(const uint8_t *src, const int src_stride,
                                       uint8_t *dst, const int dst_stride,
                                       const int w, const int h,
                                       const int16_t *const coef,
                                       uint8_t *const temp_buffer) { … }

shuffle_filter_funcs;

convolve8_funcs;

static void scale_plane_4_to_3_general(const uint8_t *src, const int src_stride,
                                       uint8_t *dst, const int dst_stride,
                                       const int w, const int h,
                                       const InterpKernel *const coef,
                                       const int phase,
                                       uint8_t *const temp_buffer) { … }

static inline __m128i scale_1_to_2_phase_0_kernel(const __m128i *const s,
                                                  const __m128i *const f) { … }

// Only calculate odd columns since even columns are just src pixels' copies.
static void scale_1_to_2_phase_0_row(const uint8_t *src, uint8_t *dst,
                                     const int w, const __m128i *const f) { … }

static void scale_plane_1_to_2_phase_0(const uint8_t *src,
                                       const ptrdiff_t src_stride, uint8_t *dst,
                                       const ptrdiff_t dst_stride,
                                       const int src_w, const int src_h,
                                       const int16_t *const coef,
                                       uint8_t *const temp_buffer) { … }

// There's SIMD optimizations for 1/4, 1/2 and 3/4 downscaling and 2x upscaling
// in SSSE3.
static inline bool has_normative_scaler_ssse3(const int src_width,
                                              const int src_height,
                                              const int dst_width,
                                              const int dst_height) { … }

void av1_resize_and_extend_frame_ssse3(const YV12_BUFFER_CONFIG *src,
                                       YV12_BUFFER_CONFIG *dst,
                                       const InterpFilter filter,
                                       const int phase, const int num_planes) { … }