Source code
Revision control
Copy as Markdown
Other Tools
/*
* Copyright (c) 2023, 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.
*/
#ifndef AOM_AV1_COMMON_ARM_HIGHBD_WARP_PLANE_NEON_H_
#define AOM_AV1_COMMON_ARM_HIGHBD_WARP_PLANE_NEON_H_
#include <arm_neon.h>
#include <assert.h>
#include <stdbool.h>
#include "aom_dsp/aom_dsp_common.h"
#include "aom_dsp/arm/mem_neon.h"
#include "aom_dsp/arm/sum_neon.h"
#include "aom_ports/mem.h"
#include "av1/common/scale.h"
#include "av1/common/warped_motion.h"
#include "config/av1_rtcd.h"
static AOM_FORCE_INLINE int16x8_t
highbd_horizontal_filter_4x1_f4(int16x8_t rv0, int16x8_t rv1, int16x8_t rv2,
int16x8_t rv3, int bd, int sx, int alpha);
static AOM_FORCE_INLINE int16x8_t highbd_horizontal_filter_8x1_f8(
int16x8_t rv0, int16x8_t rv1, int16x8_t rv2, int16x8_t rv3, int16x8_t rv4,
int16x8_t rv5, int16x8_t rv6, int16x8_t rv7, int bd, int sx, int alpha);
static AOM_FORCE_INLINE int16x8_t highbd_horizontal_filter_4x1_f1(
int16x8_t rv0, int16x8_t rv1, int16x8_t rv2, int16x8_t rv3, int bd, int sx);
static AOM_FORCE_INLINE int16x8_t highbd_horizontal_filter_8x1_f1(
int16x8_t rv0, int16x8_t rv1, int16x8_t rv2, int16x8_t rv3, int16x8_t rv4,
int16x8_t rv5, int16x8_t rv6, int16x8_t rv7, int bd, int sx);
static AOM_FORCE_INLINE int32x4_t vertical_filter_4x1_f1(const int16x8_t *tmp,
int sy);
static AOM_FORCE_INLINE int32x4x2_t vertical_filter_8x1_f1(const int16x8_t *tmp,
int sy);
static AOM_FORCE_INLINE int32x4_t vertical_filter_4x1_f4(const int16x8_t *tmp,
int sy, int gamma);
static AOM_FORCE_INLINE int32x4x2_t vertical_filter_8x1_f8(const int16x8_t *tmp,
int sy, int gamma);
static AOM_FORCE_INLINE int16x8_t load_filters_1(int ofs) {
const int ofs0 = ROUND_POWER_OF_TWO(ofs, WARPEDDIFF_PREC_BITS);
const int16_t *base =
(int16_t *)av1_warped_filter + WARPEDPIXEL_PREC_SHIFTS * 8;
return vld1q_s16(base + ofs0 * 8);
}
static AOM_FORCE_INLINE void load_filters_4(int16x8_t out[], int ofs,
int stride) {
const int ofs0 = ROUND_POWER_OF_TWO(ofs + stride * 0, WARPEDDIFF_PREC_BITS);
const int ofs1 = ROUND_POWER_OF_TWO(ofs + stride * 1, WARPEDDIFF_PREC_BITS);
const int ofs2 = ROUND_POWER_OF_TWO(ofs + stride * 2, WARPEDDIFF_PREC_BITS);
const int ofs3 = ROUND_POWER_OF_TWO(ofs + stride * 3, WARPEDDIFF_PREC_BITS);
const int16_t *base =
(int16_t *)av1_warped_filter + WARPEDPIXEL_PREC_SHIFTS * 8;
out[0] = vld1q_s16(base + ofs0 * 8);
out[1] = vld1q_s16(base + ofs1 * 8);
out[2] = vld1q_s16(base + ofs2 * 8);
out[3] = vld1q_s16(base + ofs3 * 8);
}
static AOM_FORCE_INLINE void load_filters_8(int16x8_t out[], int ofs,
int stride) {
const int ofs0 = ROUND_POWER_OF_TWO(ofs + stride * 0, WARPEDDIFF_PREC_BITS);
const int ofs1 = ROUND_POWER_OF_TWO(ofs + stride * 1, WARPEDDIFF_PREC_BITS);
const int ofs2 = ROUND_POWER_OF_TWO(ofs + stride * 2, WARPEDDIFF_PREC_BITS);
const int ofs3 = ROUND_POWER_OF_TWO(ofs + stride * 3, WARPEDDIFF_PREC_BITS);
const int ofs4 = ROUND_POWER_OF_TWO(ofs + stride * 4, WARPEDDIFF_PREC_BITS);
const int ofs5 = ROUND_POWER_OF_TWO(ofs + stride * 5, WARPEDDIFF_PREC_BITS);
const int ofs6 = ROUND_POWER_OF_TWO(ofs + stride * 6, WARPEDDIFF_PREC_BITS);
const int ofs7 = ROUND_POWER_OF_TWO(ofs + stride * 7, WARPEDDIFF_PREC_BITS);
const int16_t *base =
(int16_t *)av1_warped_filter + WARPEDPIXEL_PREC_SHIFTS * 8;
out[0] = vld1q_s16(base + ofs0 * 8);
out[1] = vld1q_s16(base + ofs1 * 8);
out[2] = vld1q_s16(base + ofs2 * 8);
out[3] = vld1q_s16(base + ofs3 * 8);
out[4] = vld1q_s16(base + ofs4 * 8);
out[5] = vld1q_s16(base + ofs5 * 8);
out[6] = vld1q_s16(base + ofs6 * 8);
out[7] = vld1q_s16(base + ofs7 * 8);
}
static AOM_FORCE_INLINE uint16x4_t clip_pixel_highbd_vec(int32x4_t val,
int bd) {
const int limit = (1 << bd) - 1;
return vqmovun_s32(vminq_s32(val, vdupq_n_s32(limit)));
}
static AOM_FORCE_INLINE uint16x8x2_t clamp_horizontal(
uint16x8x2_t src_1, int out_of_boundary_left, int out_of_boundary_right,
const uint16_t *ref, int iy, int stride, int width, const uint16x8_t indx0,
const uint16x8_t indx1) {
if (out_of_boundary_left >= 0) {
uint16x8_t cmp_vec = vdupq_n_u16(out_of_boundary_left);
uint16x8_t vec_dup = vdupq_n_u16(ref[iy * stride]);
uint16x8_t mask0 = vcleq_u16(indx0, cmp_vec);
uint16x8_t mask1 = vcleq_u16(indx1, cmp_vec);
src_1.val[0] = vbslq_u16(mask0, vec_dup, src_1.val[0]);
src_1.val[1] = vbslq_u16(mask1, vec_dup, src_1.val[1]);
}
if (out_of_boundary_right >= 0) {
uint16x8_t cmp_vec = vdupq_n_u16(15 - out_of_boundary_right);
uint16x8_t vec_dup = vdupq_n_u16(ref[iy * stride + width - 1]);
uint16x8_t mask0 = vcgeq_u16(indx0, cmp_vec);
uint16x8_t mask1 = vcgeq_u16(indx1, cmp_vec);
src_1.val[0] = vbslq_u16(mask0, vec_dup, src_1.val[0]);
src_1.val[1] = vbslq_u16(mask1, vec_dup, src_1.val[1]);
}
return src_1;
}
static AOM_FORCE_INLINE void warp_affine_horizontal(const uint16_t *ref,
int width, int height,
int stride, int p_width,
int16_t alpha, int16_t beta,
int iy4, int sx4, int ix4,
int16x8_t tmp[], int bd) {
const int round0 = (bd == 12) ? ROUND0_BITS + 2 : ROUND0_BITS;
if (ix4 <= -7) {
for (int k = 0; k < 15; ++k) {
int iy = clamp(iy4 + k - 7, 0, height - 1);
int32_t dup_val = (1 << (bd + FILTER_BITS - round0 - 1)) +
ref[iy * stride] * (1 << (FILTER_BITS - round0));
tmp[k] = vdupq_n_s16(dup_val);
}
return;
} else if (ix4 >= width + 6) {
for (int k = 0; k < 15; ++k) {
int iy = clamp(iy4 + k - 7, 0, height - 1);
int32_t dup_val =
(1 << (bd + FILTER_BITS - round0 - 1)) +
ref[iy * stride + (width - 1)] * (1 << (FILTER_BITS - round0));
tmp[k] = vdupq_n_s16(dup_val);
}
return;
}
static const uint16_t kIotaArr[] = { 0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15 };
const uint16x8_t indx0 = vld1q_u16(kIotaArr);
const uint16x8_t indx1 = vld1q_u16(kIotaArr + 8);
const int out_of_boundary_left = -(ix4 - 6);
const int out_of_boundary_right = (ix4 + 8) - width;
#define APPLY_HORIZONTAL_SHIFT_4X1(fn, ...) \
do { \
if (out_of_boundary_left >= 0 || out_of_boundary_right >= 0) { \
for (int k = 0; k < 15; ++k) { \
const int iy = clamp(iy4 + k - 7, 0, height - 1); \
const uint16_t *idx = ref + iy * stride + ix4 - 7; \
/* We don't use vld1q_u16_x2 here as LLVM generates an incorrect \
* alignment hint for this intrinsic that causes a SIGBUS on Armv7 \
* targets when alignment checks are enabled. \
* (See bug: b/349455146) */ \
uint16x8x2_t src_1 = { { vld1q_u16(idx), vld1q_u16(idx + 8) } }; \
src_1 = clamp_horizontal(src_1, out_of_boundary_left, \
out_of_boundary_right, ref, iy, stride, \
width, indx0, indx1); \
int16x8_t rv0 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 0); \
int16x8_t rv1 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 1); \
int16x8_t rv2 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 2); \
int16x8_t rv3 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 3); \
tmp[k] = (fn)(rv0, rv1, rv2, rv3, __VA_ARGS__); \
} \
} else { \
for (int k = 0; k < 15; ++k) { \
const int iy = clamp(iy4 + k - 7, 0, height - 1); \
const uint16_t *src = ref + iy * stride + ix4; \
int16x8_t rv0 = vreinterpretq_s16_u16(vld1q_u16(src - 7)); \
int16x8_t rv1 = vreinterpretq_s16_u16(vld1q_u16(src - 6)); \
int16x8_t rv2 = vreinterpretq_s16_u16(vld1q_u16(src - 5)); \
int16x8_t rv3 = vreinterpretq_s16_u16(vld1q_u16(src - 4)); \
tmp[k] = (fn)(rv0, rv1, rv2, rv3, __VA_ARGS__); \
} \
} \
} while (0)
#define APPLY_HORIZONTAL_SHIFT_8X1(fn, ...) \
do { \
if (out_of_boundary_left >= 0 || out_of_boundary_right >= 0) { \
for (int k = 0; k < 15; ++k) { \
const int iy = clamp(iy4 + k - 7, 0, height - 1); \
const uint16_t *idx = ref + iy * stride + ix4 - 7; \
/* We don't use vld1q_u16_x2 here as LLVM generates an incorrect \
* alignment hint for this intrinsic that causes a SIGBUS on Armv7 \
* targets when alignment checks are enabled. \
* (See bug: b/349455146) */ \
uint16x8x2_t src_1 = { { vld1q_u16(idx), vld1q_u16(idx + 8) } }; \
src_1 = clamp_horizontal(src_1, out_of_boundary_left, \
out_of_boundary_right, ref, iy, stride, \
width, indx0, indx1); \
int16x8_t rv0 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 0); \
int16x8_t rv1 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 1); \
int16x8_t rv2 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 2); \
int16x8_t rv3 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 3); \
int16x8_t rv4 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 4); \
int16x8_t rv5 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 5); \
int16x8_t rv6 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 6); \
int16x8_t rv7 = vextq_s16(vreinterpretq_s16_u16(src_1.val[0]), \
vreinterpretq_s16_u16(src_1.val[1]), 7); \
tmp[k] = (fn)(rv0, rv1, rv2, rv3, rv4, rv5, rv6, rv7, __VA_ARGS__); \
} \
} else { \
for (int k = 0; k < 15; ++k) { \
const int iy = clamp(iy4 + k - 7, 0, height - 1); \
const uint16_t *src = ref + iy * stride + ix4; \
int16x8_t rv0 = vreinterpretq_s16_u16(vld1q_u16(src - 7)); \
int16x8_t rv1 = vreinterpretq_s16_u16(vld1q_u16(src - 6)); \
int16x8_t rv2 = vreinterpretq_s16_u16(vld1q_u16(src - 5)); \
int16x8_t rv3 = vreinterpretq_s16_u16(vld1q_u16(src - 4)); \
int16x8_t rv4 = vreinterpretq_s16_u16(vld1q_u16(src - 3)); \
int16x8_t rv5 = vreinterpretq_s16_u16(vld1q_u16(src - 2)); \
int16x8_t rv6 = vreinterpretq_s16_u16(vld1q_u16(src - 1)); \
int16x8_t rv7 = vreinterpretq_s16_u16(vld1q_u16(src - 0)); \
tmp[k] = (fn)(rv0, rv1, rv2, rv3, rv4, rv5, rv6, rv7, __VA_ARGS__); \
} \
} \
} while (0)
if (p_width == 4) {
if (beta == 0) {
if (alpha == 0) {
APPLY_HORIZONTAL_SHIFT_4X1(highbd_horizontal_filter_4x1_f1, bd, sx4);
} else {
APPLY_HORIZONTAL_SHIFT_4X1(highbd_horizontal_filter_4x1_f4, bd, sx4,
alpha);
}
} else {
if (alpha == 0) {
APPLY_HORIZONTAL_SHIFT_4X1(highbd_horizontal_filter_4x1_f1, bd,
(sx4 + beta * (k - 3)));
} else {
APPLY_HORIZONTAL_SHIFT_4X1(highbd_horizontal_filter_4x1_f4, bd,
(sx4 + beta * (k - 3)), alpha);
}
}
} else {
if (beta == 0) {
if (alpha == 0) {
APPLY_HORIZONTAL_SHIFT_8X1(highbd_horizontal_filter_8x1_f1, bd, sx4);
} else {
APPLY_HORIZONTAL_SHIFT_8X1(highbd_horizontal_filter_8x1_f8, bd, sx4,
alpha);
}
} else {
if (alpha == 0) {
APPLY_HORIZONTAL_SHIFT_8X1(highbd_horizontal_filter_8x1_f1, bd,
(sx4 + beta * (k - 3)));
} else {
APPLY_HORIZONTAL_SHIFT_8X1(highbd_horizontal_filter_8x1_f8, bd,
(sx4 + beta * (k - 3)), alpha);
}
}
}
#undef APPLY_HORIZONTAL_SHIFT_4X1
#undef APPLY_HORIZONTAL_SHIFT_8X1
}
static AOM_FORCE_INLINE void highbd_vertical_filter_4x1_f4(
uint16_t *pred, int p_stride, int bd, uint16_t *dst, int dst_stride,
bool is_compound, bool do_average, bool use_dist_wtd_comp_avg, int fwd,
int bwd, int16_t gamma, const int16x8_t *tmp, int i, int sy, int j) {
int32x4_t sum0 = gamma == 0 ? vertical_filter_4x1_f1(tmp, sy)
: vertical_filter_4x1_f4(tmp, sy, gamma);
const int round0 = (bd == 12) ? ROUND0_BITS + 2 : ROUND0_BITS;
const int offset_bits_vert = bd + 2 * FILTER_BITS - round0;
sum0 = vaddq_s32(sum0, vdupq_n_s32(1 << offset_bits_vert));
uint16_t *dst16 = &pred[i * p_stride + j];
if (!is_compound) {
const int reduce_bits_vert = 2 * FILTER_BITS - round0;
sum0 = vrshlq_s32(sum0, vdupq_n_s32(-reduce_bits_vert));
const int res_sub_const = (1 << (bd - 1)) + (1 << bd);
sum0 = vsubq_s32(sum0, vdupq_n_s32(res_sub_const));
uint16x4_t res0 = clip_pixel_highbd_vec(sum0, bd);
vst1_u16(dst16, res0);
return;
}
sum0 = vrshrq_n_s32(sum0, COMPOUND_ROUND1_BITS);
uint16_t *p = &dst[i * dst_stride + j];
if (!do_average) {
vst1_u16(p, vqmovun_s32(sum0));
return;
}
uint16x4_t p0 = vld1_u16(p);
int32x4_t p_vec0 = vreinterpretq_s32_u32(vmovl_u16(p0));
if (use_dist_wtd_comp_avg) {
p_vec0 = vmulq_n_s32(p_vec0, fwd);
p_vec0 = vmlaq_n_s32(p_vec0, sum0, bwd);
p_vec0 = vshrq_n_s32(p_vec0, DIST_PRECISION_BITS);
} else {
p_vec0 = vhaddq_s32(p_vec0, sum0);
}
const int offset_bits = bd + 2 * FILTER_BITS - round0;
const int round1 = COMPOUND_ROUND1_BITS;
const int res_sub_const =
(1 << (offset_bits - round1)) + (1 << (offset_bits - round1 - 1));
const int round_bits = 2 * FILTER_BITS - round0 - round1;
p_vec0 = vsubq_s32(p_vec0, vdupq_n_s32(res_sub_const));
p_vec0 = vrshlq_s32(p_vec0, vdupq_n_s32(-round_bits));
uint16x4_t res0 = clip_pixel_highbd_vec(p_vec0, bd);
vst1_u16(dst16, res0);
}
static AOM_FORCE_INLINE void highbd_vertical_filter_8x1_f8(
uint16_t *pred, int p_stride, int bd, uint16_t *dst, int dst_stride,
bool is_compound, bool do_average, bool use_dist_wtd_comp_avg, int fwd,
int bwd, int16_t gamma, const int16x8_t *tmp, int i, int sy, int j) {
int32x4x2_t sums = gamma == 0 ? vertical_filter_8x1_f1(tmp, sy)
: vertical_filter_8x1_f8(tmp, sy, gamma);
int32x4_t sum0 = sums.val[0];
int32x4_t sum1 = sums.val[1];
const int round0 = (bd == 12) ? ROUND0_BITS + 2 : ROUND0_BITS;
const int offset_bits_vert = bd + 2 * FILTER_BITS - round0;
sum0 = vaddq_s32(sum0, vdupq_n_s32(1 << offset_bits_vert));
sum1 = vaddq_s32(sum1, vdupq_n_s32(1 << offset_bits_vert));
uint16_t *dst16 = &pred[i * p_stride + j];
if (!is_compound) {
const int reduce_bits_vert = 2 * FILTER_BITS - round0;
sum0 = vrshlq_s32(sum0, vdupq_n_s32(-reduce_bits_vert));
sum1 = vrshlq_s32(sum1, vdupq_n_s32(-reduce_bits_vert));
const int res_sub_const = (1 << (bd - 1)) + (1 << bd);
sum0 = vsubq_s32(sum0, vdupq_n_s32(res_sub_const));
sum1 = vsubq_s32(sum1, vdupq_n_s32(res_sub_const));
uint16x4_t res0 = clip_pixel_highbd_vec(sum0, bd);
uint16x4_t res1 = clip_pixel_highbd_vec(sum1, bd);
vst1_u16(dst16, res0);
vst1_u16(dst16 + 4, res1);
return;
}
sum0 = vrshrq_n_s32(sum0, COMPOUND_ROUND1_BITS);
sum1 = vrshrq_n_s32(sum1, COMPOUND_ROUND1_BITS);
uint16_t *p = &dst[i * dst_stride + j];
if (!do_average) {
vst1_u16(p, vqmovun_s32(sum0));
vst1_u16(p + 4, vqmovun_s32(sum1));
return;
}
uint16x8_t p0 = vld1q_u16(p);
int32x4_t p_vec0 = vreinterpretq_s32_u32(vmovl_u16(vget_low_u16(p0)));
int32x4_t p_vec1 = vreinterpretq_s32_u32(vmovl_u16(vget_high_u16(p0)));
if (use_dist_wtd_comp_avg) {
p_vec0 = vmulq_n_s32(p_vec0, fwd);
p_vec1 = vmulq_n_s32(p_vec1, fwd);
p_vec0 = vmlaq_n_s32(p_vec0, sum0, bwd);
p_vec1 = vmlaq_n_s32(p_vec1, sum1, bwd);
p_vec0 = vshrq_n_s32(p_vec0, DIST_PRECISION_BITS);
p_vec1 = vshrq_n_s32(p_vec1, DIST_PRECISION_BITS);
} else {
p_vec0 = vhaddq_s32(p_vec0, sum0);
p_vec1 = vhaddq_s32(p_vec1, sum1);
}
const int offset_bits = bd + 2 * FILTER_BITS - round0;
const int round1 = COMPOUND_ROUND1_BITS;
const int res_sub_const =
(1 << (offset_bits - round1)) + (1 << (offset_bits - round1 - 1));
const int round_bits = 2 * FILTER_BITS - round0 - round1;
p_vec0 = vsubq_s32(p_vec0, vdupq_n_s32(res_sub_const));
p_vec1 = vsubq_s32(p_vec1, vdupq_n_s32(res_sub_const));
p_vec0 = vrshlq_s32(p_vec0, vdupq_n_s32(-round_bits));
p_vec1 = vrshlq_s32(p_vec1, vdupq_n_s32(-round_bits));
uint16x4_t res0 = clip_pixel_highbd_vec(p_vec0, bd);
uint16x4_t res1 = clip_pixel_highbd_vec(p_vec1, bd);
vst1_u16(dst16, res0);
vst1_u16(dst16 + 4, res1);
}
static AOM_FORCE_INLINE void warp_affine_vertical(
uint16_t *pred, int p_width, int p_height, int p_stride, int bd,
uint16_t *dst, int dst_stride, bool is_compound, bool do_average,
bool use_dist_wtd_comp_avg, int fwd, int bwd, int16_t gamma, int16_t delta,
const int16x8_t *tmp, int i, int sy4, int j) {
int limit_height = p_height > 4 ? 8 : 4;
if (p_width > 4) {
// p_width == 8
for (int k = 0; k < limit_height; ++k) {
int sy = sy4 + delta * k;
highbd_vertical_filter_8x1_f8(
pred, p_stride, bd, dst, dst_stride, is_compound, do_average,
use_dist_wtd_comp_avg, fwd, bwd, gamma, tmp + k, i + k, sy, j);
}
} else {
// p_width == 4
for (int k = 0; k < limit_height; ++k) {
int sy = sy4 + delta * k;
highbd_vertical_filter_4x1_f4(
pred, p_stride, bd, dst, dst_stride, is_compound, do_average,
use_dist_wtd_comp_avg, fwd, bwd, gamma, tmp + k, i + k, sy, j);
}
}
}
static AOM_FORCE_INLINE void highbd_warp_affine_common(
const int32_t *mat, const uint16_t *ref, int width, int height, int stride,
uint16_t *pred, int p_col, int p_row, int p_width, int p_height,
int p_stride, int subsampling_x, int subsampling_y, int bd,
ConvolveParams *conv_params, int16_t alpha, int16_t beta, int16_t gamma,
int16_t delta) {
uint16_t *const dst = conv_params->dst;
const int dst_stride = conv_params->dst_stride;
const bool is_compound = conv_params->is_compound;
const bool do_average = conv_params->do_average;
const bool use_dist_wtd_comp_avg = conv_params->use_dist_wtd_comp_avg;
const int fwd = conv_params->fwd_offset;
const int bwd = conv_params->bck_offset;
assert(IMPLIES(is_compound, dst != NULL));
for (int i = 0; i < p_height; i += 8) {
for (int j = 0; j < p_width; j += 8) {
// Calculate the center of this 8x8 block,
// project to luma coordinates (if in a subsampled chroma plane),
// apply the affine transformation,
// then convert back to the original coordinates (if necessary)
const int32_t src_x = (j + 4 + p_col) << subsampling_x;
const int32_t src_y = (i + 4 + p_row) << subsampling_y;
const int64_t dst_x =
(int64_t)mat[2] * src_x + (int64_t)mat[3] * src_y + (int64_t)mat[0];
const int64_t dst_y =
(int64_t)mat[4] * src_x + (int64_t)mat[5] * src_y + (int64_t)mat[1];
const int64_t x4 = dst_x >> subsampling_x;
const int64_t y4 = dst_y >> subsampling_y;
const int32_t ix4 = (int32_t)(x4 >> WARPEDMODEL_PREC_BITS);
int32_t sx4 = x4 & ((1 << WARPEDMODEL_PREC_BITS) - 1);
const int32_t iy4 = (int32_t)(y4 >> WARPEDMODEL_PREC_BITS);
int32_t sy4 = y4 & ((1 << WARPEDMODEL_PREC_BITS) - 1);
sx4 += alpha * (-4) + beta * (-4);
sy4 += gamma * (-4) + delta * (-4);
sx4 &= ~((1 << WARP_PARAM_REDUCE_BITS) - 1);
sy4 &= ~((1 << WARP_PARAM_REDUCE_BITS) - 1);
// Each horizontal filter result is formed by the sum of up to eight
// multiplications by filter values and then a shift. Although both the
// inputs and filters are loaded as int16, the input data is at most bd
// bits and the filters are at most 8 bits each. Additionally since we
// know all possible filter values we know that the sum of absolute
// filter values will fit in at most 9 bits. With this in mind we can
// conclude that the sum of each filter application will fit in bd + 9
// bits. The shift following the summation is ROUND0_BITS (which is 3),
// +2 for 12-bit, which gives us a final storage of:
// bd == 8: ( 8 + 9) - 3 => 14 bits
// bd == 10: (10 + 9) - 3 => 16 bits
// bd == 12: (12 + 9) - 5 => 16 bits
// So it is safe to use int16x8_t as the intermediate storage type here.
int16x8_t tmp[15];
warp_affine_horizontal(ref, width, height, stride, p_width, alpha, beta,
iy4, sx4, ix4, tmp, bd);
warp_affine_vertical(pred, p_width, p_height, p_stride, bd, dst,
dst_stride, is_compound, do_average,
use_dist_wtd_comp_avg, fwd, bwd, gamma, delta, tmp,
i, sy4, j);
}
}
}
#endif // AOM_AV1_COMMON_ARM_HIGHBD_WARP_PLANE_NEON_H_