From 221b7744a7ccbe4a0a11ee6c58596f8f3ef11764 Mon Sep 17 00:00:00 2001 From: Chun-Min Chang Date: Thu, 2 May 2024 22:25:29 +0000 Subject: [PATCH] Bug 1894408 - Import and build macOS/arm64 code in libvpx r=media-playback-reviewers,padenot This patch includes the macOS/arm64 code within libvpx through a two-step process: 1. Integration of macOS/arm64 code in the same revision (1) Change `revision` in moz.yaml to a random string. (2) Add code in the same revision via ``` ./mach vendor media/libvpx/moz.yaml -r \ 85dafa9c61f99330f484e77297684b42af6ff37d \ --patch-mode=none --ignore-modified ``` (3) Undo the changes for `revision` in moz.yaml 2. Local patch application: Run the command below ``` ./mach vendor media/libvpx/moz.yaml --patch-mode=only ``` Consequently, this enables the inclusion and successful building of macOS/arm64 code within the libvpx library, and facilitating its integration with Gecko. Differential Revision: https://phabricator.services.mozilla.com/D209086 --- media/libvpx/config/mac/arm64/vp8_rtcd.h | 204 +++ media/libvpx/config/mac/arm64/vp9_rtcd.h | 116 ++ media/libvpx/config/mac/arm64/vpx_config.asm | 98 ++ media/libvpx/config/mac/arm64/vpx_config.c | 10 + media/libvpx/config/mac/arm64/vpx_config.h | 109 ++ media/libvpx/config/mac/arm64/vpx_dsp_rtcd.h | 1187 +++++++++++++++++ .../libvpx/config/mac/arm64/vpx_scale_rtcd.h | 75 ++ media/libvpx/sources.mozbuild | 256 ++++ 8 files changed, 2055 insertions(+) create mode 100644 media/libvpx/config/mac/arm64/vp8_rtcd.h create mode 100644 media/libvpx/config/mac/arm64/vp9_rtcd.h create mode 100644 media/libvpx/config/mac/arm64/vpx_config.asm create mode 100644 media/libvpx/config/mac/arm64/vpx_config.c create mode 100644 media/libvpx/config/mac/arm64/vpx_config.h create mode 100644 media/libvpx/config/mac/arm64/vpx_dsp_rtcd.h create mode 100644 media/libvpx/config/mac/arm64/vpx_scale_rtcd.h diff --git a/media/libvpx/config/mac/arm64/vp8_rtcd.h b/media/libvpx/config/mac/arm64/vp8_rtcd.h new file mode 100644 index 000000000000..c8826511ff44 --- /dev/null +++ b/media/libvpx/config/mac/arm64/vp8_rtcd.h @@ -0,0 +1,204 @@ +// This file is generated. Do not edit. +#ifndef VP8_RTCD_H_ +#define VP8_RTCD_H_ + +#ifdef RTCD_C +#define RTCD_EXTERN +#else +#define RTCD_EXTERN extern +#endif + +/* + * VP8 + */ + +struct blockd; +struct macroblockd; +struct loop_filter_info; + +/* Encoder forward decls */ +struct block; +struct macroblock; +struct variance_vtable; +union int_mv; +struct yv12_buffer_config; + +#ifdef __cplusplus +extern "C" { +#endif + +void vp8_bilinear_predict16x16_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_bilinear_predict16x16_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_bilinear_predict16x16 vp8_bilinear_predict16x16_neon + +void vp8_bilinear_predict4x4_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_bilinear_predict4x4_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_bilinear_predict4x4 vp8_bilinear_predict4x4_neon + +void vp8_bilinear_predict8x4_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_bilinear_predict8x4_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_bilinear_predict8x4 vp8_bilinear_predict8x4_neon + +void vp8_bilinear_predict8x8_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_bilinear_predict8x8_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_bilinear_predict8x8 vp8_bilinear_predict8x8_neon + +int vp8_block_error_c(short *coeff, short *dqcoeff); +#define vp8_block_error vp8_block_error_c + +void vp8_copy32xn_c(const unsigned char *src_ptr, int src_stride, unsigned char *dst_ptr, int dst_stride, int height); +#define vp8_copy32xn vp8_copy32xn_c + +void vp8_copy_mem16x16_c(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +void vp8_copy_mem16x16_neon(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +#define vp8_copy_mem16x16 vp8_copy_mem16x16_neon + +void vp8_copy_mem8x4_c(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +void vp8_copy_mem8x4_neon(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +#define vp8_copy_mem8x4 vp8_copy_mem8x4_neon + +void vp8_copy_mem8x8_c(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +void vp8_copy_mem8x8_neon(unsigned char *src, int src_stride, unsigned char *dst, int dst_stride); +#define vp8_copy_mem8x8 vp8_copy_mem8x8_neon + +void vp8_dc_only_idct_add_c(short input_dc, unsigned char *pred_ptr, int pred_stride, unsigned char *dst_ptr, int dst_stride); +void vp8_dc_only_idct_add_neon(short input_dc, unsigned char *pred_ptr, int pred_stride, unsigned char *dst_ptr, int dst_stride); +#define vp8_dc_only_idct_add vp8_dc_only_idct_add_neon + +int vp8_denoiser_filter_c(unsigned char *mc_running_avg_y, int mc_avg_y_stride, unsigned char *running_avg_y, int avg_y_stride, unsigned char *sig, int sig_stride, unsigned int motion_magnitude, int increase_denoising); +int vp8_denoiser_filter_neon(unsigned char *mc_running_avg_y, int mc_avg_y_stride, unsigned char *running_avg_y, int avg_y_stride, unsigned char *sig, int sig_stride, unsigned int motion_magnitude, int increase_denoising); +#define vp8_denoiser_filter vp8_denoiser_filter_neon + +int vp8_denoiser_filter_uv_c(unsigned char *mc_running_avg, int mc_avg_stride, unsigned char *running_avg, int avg_stride, unsigned char *sig, int sig_stride, unsigned int motion_magnitude, int increase_denoising); +int vp8_denoiser_filter_uv_neon(unsigned char *mc_running_avg, int mc_avg_stride, unsigned char *running_avg, int avg_stride, unsigned char *sig, int sig_stride, unsigned int motion_magnitude, int increase_denoising); +#define vp8_denoiser_filter_uv vp8_denoiser_filter_uv_neon + +void vp8_dequant_idct_add_c(short *input, short *dq, unsigned char *dest, int stride); +void vp8_dequant_idct_add_neon(short *input, short *dq, unsigned char *dest, int stride); +#define vp8_dequant_idct_add vp8_dequant_idct_add_neon + +void vp8_dequant_idct_add_uv_block_c(short *q, short *dq, unsigned char *dst_u, unsigned char *dst_v, int stride, char *eobs); +void vp8_dequant_idct_add_uv_block_neon(short *q, short *dq, unsigned char *dst_u, unsigned char *dst_v, int stride, char *eobs); +#define vp8_dequant_idct_add_uv_block vp8_dequant_idct_add_uv_block_neon + +void vp8_dequant_idct_add_y_block_c(short *q, short *dq, unsigned char *dst, int stride, char *eobs); +void vp8_dequant_idct_add_y_block_neon(short *q, short *dq, unsigned char *dst, int stride, char *eobs); +#define vp8_dequant_idct_add_y_block vp8_dequant_idct_add_y_block_neon + +void vp8_dequantize_b_c(struct blockd*, short *DQC); +void vp8_dequantize_b_neon(struct blockd*, short *DQC); +#define vp8_dequantize_b vp8_dequantize_b_neon + +int vp8_diamond_search_sad_c(struct macroblock *x, struct block *b, struct blockd *d, union int_mv *ref_mv, union int_mv *best_mv, int search_param, int sad_per_bit, int *num00, struct variance_vtable *fn_ptr, int *mvcost[2], union int_mv *center_mv); +#define vp8_diamond_search_sad vp8_diamond_search_sad_c + +void vp8_fast_quantize_b_c(struct block *, struct blockd *); +void vp8_fast_quantize_b_neon(struct block *, struct blockd *); +#define vp8_fast_quantize_b vp8_fast_quantize_b_neon + +void vp8_loop_filter_bh_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +void vp8_loop_filter_bh_neon(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +#define vp8_loop_filter_bh vp8_loop_filter_bh_neon + +void vp8_loop_filter_bv_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +void vp8_loop_filter_bv_neon(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +#define vp8_loop_filter_bv vp8_loop_filter_bv_neon + +void vp8_loop_filter_mbh_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +void vp8_loop_filter_mbh_neon(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +#define vp8_loop_filter_mbh vp8_loop_filter_mbh_neon + +void vp8_loop_filter_mbv_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +void vp8_loop_filter_mbv_neon(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi); +#define vp8_loop_filter_mbv vp8_loop_filter_mbv_neon + +void vp8_loop_filter_bhs_c(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +void vp8_loop_filter_bhs_neon(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +#define vp8_loop_filter_simple_bh vp8_loop_filter_bhs_neon + +void vp8_loop_filter_bvs_c(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +void vp8_loop_filter_bvs_neon(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +#define vp8_loop_filter_simple_bv vp8_loop_filter_bvs_neon + +void vp8_loop_filter_simple_horizontal_edge_c(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +void vp8_loop_filter_mbhs_neon(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +#define vp8_loop_filter_simple_mbh vp8_loop_filter_mbhs_neon + +void vp8_loop_filter_simple_vertical_edge_c(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +void vp8_loop_filter_mbvs_neon(unsigned char *y_ptr, int y_stride, const unsigned char *blimit); +#define vp8_loop_filter_simple_mbv vp8_loop_filter_mbvs_neon + +int vp8_mbblock_error_c(struct macroblock *mb, int dc); +#define vp8_mbblock_error vp8_mbblock_error_c + +int vp8_mbuverror_c(struct macroblock *mb); +#define vp8_mbuverror vp8_mbuverror_c + +int vp8_refining_search_sad_c(struct macroblock *x, struct block *b, struct blockd *d, union int_mv *ref_mv, int error_per_bit, int search_range, struct variance_vtable *fn_ptr, int *mvcost[2], union int_mv *center_mv); +#define vp8_refining_search_sad vp8_refining_search_sad_c + +void vp8_regular_quantize_b_c(struct block *, struct blockd *); +#define vp8_regular_quantize_b vp8_regular_quantize_b_c + +void vp8_short_fdct4x4_c(short *input, short *output, int pitch); +void vp8_short_fdct4x4_neon(short *input, short *output, int pitch); +#define vp8_short_fdct4x4 vp8_short_fdct4x4_neon + +void vp8_short_fdct8x4_c(short *input, short *output, int pitch); +void vp8_short_fdct8x4_neon(short *input, short *output, int pitch); +#define vp8_short_fdct8x4 vp8_short_fdct8x4_neon + +void vp8_short_idct4x4llm_c(short *input, unsigned char *pred_ptr, int pred_stride, unsigned char *dst_ptr, int dst_stride); +void vp8_short_idct4x4llm_neon(short *input, unsigned char *pred_ptr, int pred_stride, unsigned char *dst_ptr, int dst_stride); +#define vp8_short_idct4x4llm vp8_short_idct4x4llm_neon + +void vp8_short_inv_walsh4x4_c(short *input, short *mb_dqcoeff); +void vp8_short_inv_walsh4x4_neon(short *input, short *mb_dqcoeff); +#define vp8_short_inv_walsh4x4 vp8_short_inv_walsh4x4_neon + +void vp8_short_inv_walsh4x4_1_c(short *input, short *mb_dqcoeff); +#define vp8_short_inv_walsh4x4_1 vp8_short_inv_walsh4x4_1_c + +void vp8_short_walsh4x4_c(short *input, short *output, int pitch); +void vp8_short_walsh4x4_neon(short *input, short *output, int pitch); +#define vp8_short_walsh4x4 vp8_short_walsh4x4_neon + +void vp8_sixtap_predict16x16_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_sixtap_predict16x16_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_sixtap_predict16x16 vp8_sixtap_predict16x16_neon + +void vp8_sixtap_predict4x4_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_sixtap_predict4x4_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_sixtap_predict4x4 vp8_sixtap_predict4x4_neon + +void vp8_sixtap_predict8x4_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_sixtap_predict8x4_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_sixtap_predict8x4 vp8_sixtap_predict8x4_neon + +void vp8_sixtap_predict8x8_c(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +void vp8_sixtap_predict8x8_neon(unsigned char *src_ptr, int src_pixels_per_line, int xoffset, int yoffset, unsigned char *dst_ptr, int dst_pitch); +#define vp8_sixtap_predict8x8 vp8_sixtap_predict8x8_neon + +void vp8_temporal_filter_apply_c(unsigned char *frame1, unsigned int stride, unsigned char *frame2, unsigned int block_size, int strength, int filter_weight, unsigned int *accumulator, unsigned short *count); +#define vp8_temporal_filter_apply vp8_temporal_filter_apply_c + +void vp8_rtcd(void); + +#include "vpx_config.h" + +#ifdef RTCD_C +#include "vpx_ports/arm.h" +static void setup_rtcd_internal(void) +{ + int flags = arm_cpu_caps(); + + (void)flags; + +} +#endif + +#ifdef __cplusplus +} // extern "C" +#endif + +#endif diff --git a/media/libvpx/config/mac/arm64/vp9_rtcd.h b/media/libvpx/config/mac/arm64/vp9_rtcd.h new file mode 100644 index 000000000000..b8fc782dfd9c --- /dev/null +++ b/media/libvpx/config/mac/arm64/vp9_rtcd.h @@ -0,0 +1,116 @@ +// This file is generated. Do not edit. +#ifndef VP9_RTCD_H_ +#define VP9_RTCD_H_ + +#ifdef RTCD_C +#define RTCD_EXTERN +#else +#define RTCD_EXTERN extern +#endif + +/* + * VP9 + */ + +#include "vpx/vpx_integer.h" +#include "vp9/common/vp9_common.h" +#include "vp9/common/vp9_enums.h" +#include "vp9/common/vp9_filter.h" + +struct macroblockd; + +/* Encoder forward decls */ +struct macroblock; +struct macroblock_plane; +struct vp9_sad_table; +struct ScanOrder; +struct search_site_config; +struct mv; +union int_mv; +struct yv12_buffer_config; + +#ifdef __cplusplus +extern "C" { +#endif + +void vp9_apply_temporal_filter_c(const uint8_t *y_src, int y_src_stride, const uint8_t *y_pre, int y_pre_stride, const uint8_t *u_src, const uint8_t *v_src, int uv_src_stride, const uint8_t *u_pre, const uint8_t *v_pre, int uv_pre_stride, unsigned int block_width, unsigned int block_height, int ss_x, int ss_y, int strength, const int *const blk_fw, int use_32x32, uint32_t *y_accumulator, uint16_t *y_count, uint32_t *u_accumulator, uint16_t *u_count, uint32_t *v_accumulator, uint16_t *v_count); +void vp9_apply_temporal_filter_neon(const uint8_t *y_src, int y_src_stride, const uint8_t *y_pre, int y_pre_stride, const uint8_t *u_src, const uint8_t *v_src, int uv_src_stride, const uint8_t *u_pre, const uint8_t *v_pre, int uv_pre_stride, unsigned int block_width, unsigned int block_height, int ss_x, int ss_y, int strength, const int *const blk_fw, int use_32x32, uint32_t *y_accumulator, uint16_t *y_count, uint32_t *u_accumulator, uint16_t *u_count, uint32_t *v_accumulator, uint16_t *v_count); +#define vp9_apply_temporal_filter vp9_apply_temporal_filter_neon + +int64_t vp9_block_error_c(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz); +int64_t vp9_block_error_neon(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz); +int64_t vp9_block_error_sve(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz); +RTCD_EXTERN int64_t (*vp9_block_error)(const tran_low_t *coeff, const tran_low_t *dqcoeff, intptr_t block_size, int64_t *ssz); + +int64_t vp9_block_error_fp_c(const tran_low_t *coeff, const tran_low_t *dqcoeff, int block_size); +int64_t vp9_block_error_fp_neon(const tran_low_t *coeff, const tran_low_t *dqcoeff, int block_size); +int64_t vp9_block_error_fp_sve(const tran_low_t *coeff, const tran_low_t *dqcoeff, int block_size); +RTCD_EXTERN int64_t (*vp9_block_error_fp)(const tran_low_t *coeff, const tran_low_t *dqcoeff, int block_size); + +int vp9_diamond_search_sad_c(const struct macroblock *x, const struct search_site_config *cfg, struct mv *ref_mv, uint32_t start_mv_sad, struct mv *best_mv, int search_param, int sad_per_bit, int *num00, const struct vp9_sad_table *sad_fn_ptr, const struct mv *center_mv); +int vp9_diamond_search_sad_neon(const struct macroblock *x, const struct search_site_config *cfg, struct mv *ref_mv, uint32_t start_mv_sad, struct mv *best_mv, int search_param, int sad_per_bit, int *num00, const struct vp9_sad_table *sad_fn_ptr, const struct mv *center_mv); +#define vp9_diamond_search_sad vp9_diamond_search_sad_neon + +void vp9_fht16x16_c(const int16_t *input, tran_low_t *output, int stride, int tx_type); +void vp9_fht16x16_neon(const int16_t *input, tran_low_t *output, int stride, int tx_type); +#define vp9_fht16x16 vp9_fht16x16_neon + +void vp9_fht4x4_c(const int16_t *input, tran_low_t *output, int stride, int tx_type); +void vp9_fht4x4_neon(const int16_t *input, tran_low_t *output, int stride, int tx_type); +#define vp9_fht4x4 vp9_fht4x4_neon + +void vp9_fht8x8_c(const int16_t *input, tran_low_t *output, int stride, int tx_type); +void vp9_fht8x8_neon(const int16_t *input, tran_low_t *output, int stride, int tx_type); +#define vp9_fht8x8 vp9_fht8x8_neon + +void vp9_fwht4x4_c(const int16_t *input, tran_low_t *output, int stride); +#define vp9_fwht4x4 vp9_fwht4x4_c + +void vp9_iht16x16_256_add_c(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +void vp9_iht16x16_256_add_neon(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +#define vp9_iht16x16_256_add vp9_iht16x16_256_add_neon + +void vp9_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +void vp9_iht4x4_16_add_neon(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +#define vp9_iht4x4_16_add vp9_iht4x4_16_add_neon + +void vp9_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +void vp9_iht8x8_64_add_neon(const tran_low_t *input, uint8_t *dest, int stride, int tx_type); +#define vp9_iht8x8_64_add vp9_iht8x8_64_add_neon + +void vp9_quantize_fp_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +void vp9_quantize_fp_neon(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +#define vp9_quantize_fp vp9_quantize_fp_neon + +void vp9_quantize_fp_32x32_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +void vp9_quantize_fp_32x32_neon(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +#define vp9_quantize_fp_32x32 vp9_quantize_fp_32x32_neon + +void vp9_scale_and_extend_frame_c(const struct yv12_buffer_config *src, struct yv12_buffer_config *dst, INTERP_FILTER filter_type, int phase_scaler); +void vp9_scale_and_extend_frame_neon(const struct yv12_buffer_config *src, struct yv12_buffer_config *dst, INTERP_FILTER filter_type, int phase_scaler); +#define vp9_scale_and_extend_frame vp9_scale_and_extend_frame_neon + +void vp9_rtcd(void); + +#include "vpx_config.h" + +#ifdef RTCD_C +#include "vpx_ports/arm.h" +static void setup_rtcd_internal(void) +{ + int flags = arm_cpu_caps(); + + (void)flags; + + vp9_block_error = vp9_block_error_neon; + if (flags & HAS_SVE) vp9_block_error = vp9_block_error_sve; + vp9_block_error_fp = vp9_block_error_fp_neon; + if (flags & HAS_SVE) vp9_block_error_fp = vp9_block_error_fp_sve; +} +#endif + +#ifdef __cplusplus +} // extern "C" +#endif + +#endif diff --git a/media/libvpx/config/mac/arm64/vpx_config.asm b/media/libvpx/config/mac/arm64/vpx_config.asm new file mode 100644 index 000000000000..82a96985ecad --- /dev/null +++ b/media/libvpx/config/mac/arm64/vpx_config.asm @@ -0,0 +1,98 @@ +@ This file was created from a .asm file +@ using the ads2gas.pl script. +.syntax unified +.equ VPX_ARCH_ARM , 1 +.equ VPX_ARCH_AARCH64 , 1 +.equ VPX_ARCH_MIPS , 0 +.equ VPX_ARCH_X86 , 0 +.equ VPX_ARCH_X86_64 , 0 +.equ VPX_ARCH_PPC , 0 +.equ VPX_ARCH_LOONGARCH , 0 +.equ HAVE_NEON_ASM , 0 +.equ HAVE_NEON , 1 +.equ HAVE_NEON_DOTPROD , 1 +.equ HAVE_NEON_I8MM , 1 +.equ HAVE_SVE , 1 +.equ HAVE_SVE2 , 1 +.equ HAVE_MIPS32 , 0 +.equ HAVE_DSPR2 , 0 +.equ HAVE_MSA , 0 +.equ HAVE_MIPS64 , 0 +.equ HAVE_MMX , 0 +.equ HAVE_SSE , 0 +.equ HAVE_SSE2 , 0 +.equ HAVE_SSE3 , 0 +.equ HAVE_SSSE3 , 0 +.equ HAVE_SSE4_1 , 0 +.equ HAVE_AVX , 0 +.equ HAVE_AVX2 , 0 +.equ HAVE_AVX512 , 0 +.equ HAVE_VSX , 0 +.equ HAVE_MMI , 0 +.equ HAVE_LSX , 0 +.equ HAVE_LASX , 0 +.equ HAVE_VPX_PORTS , 1 +.equ HAVE_PTHREAD_H , 1 +.equ CONFIG_DEPENDENCY_TRACKING , 1 +.equ CONFIG_EXTERNAL_BUILD , 1 +.equ CONFIG_INSTALL_DOCS , 0 +.equ CONFIG_INSTALL_BINS , 1 +.equ CONFIG_INSTALL_LIBS , 1 +.equ CONFIG_INSTALL_SRCS , 0 +.equ CONFIG_DEBUG , 0 +.equ CONFIG_GPROF , 0 +.equ CONFIG_GCOV , 0 +.equ CONFIG_RVCT , 0 +.equ CONFIG_GCC , 1 +.equ CONFIG_MSVS , 0 +.equ CONFIG_PIC , 1 +.equ CONFIG_BIG_ENDIAN , 0 +.equ CONFIG_CODEC_SRCS , 0 +.equ CONFIG_DEBUG_LIBS , 0 +.equ CONFIG_DEQUANT_TOKENS , 0 +.equ CONFIG_DC_RECON , 0 +.equ CONFIG_RUNTIME_CPU_DETECT , 1 +.equ CONFIG_POSTPROC , 0 +.equ CONFIG_VP9_POSTPROC , 0 +.equ CONFIG_MULTITHREAD , 1 +.equ CONFIG_INTERNAL_STATS , 0 +.equ CONFIG_VP8_ENCODER , 1 +.equ CONFIG_VP8_DECODER , 1 +.equ CONFIG_VP9_ENCODER , 1 +.equ CONFIG_VP9_DECODER , 1 +.equ CONFIG_VP8 , 1 +.equ CONFIG_VP9 , 1 +.equ CONFIG_ENCODERS , 1 +.equ CONFIG_DECODERS , 1 +.equ CONFIG_STATIC_MSVCRT , 0 +.equ CONFIG_SPATIAL_RESAMPLING , 1 +.equ CONFIG_REALTIME_ONLY , 0 +.equ CONFIG_ONTHEFLY_BITPACKING , 0 +.equ CONFIG_ERROR_CONCEALMENT , 0 +.equ CONFIG_SHARED , 0 +.equ CONFIG_STATIC , 1 +.equ CONFIG_SMALL , 0 +.equ CONFIG_POSTPROC_VISUALIZER , 0 +.equ CONFIG_OS_SUPPORT , 1 +.equ CONFIG_UNIT_TESTS , 0 +.equ CONFIG_WEBM_IO , 0 +.equ CONFIG_LIBYUV , 0 +.equ CONFIG_DECODE_PERF_TESTS , 0 +.equ CONFIG_ENCODE_PERF_TESTS , 0 +.equ CONFIG_MULTI_RES_ENCODING , 1 +.equ CONFIG_TEMPORAL_DENOISING , 1 +.equ CONFIG_VP9_TEMPORAL_DENOISING , 0 +.equ CONFIG_COEFFICIENT_RANGE_CHECKING , 0 +.equ CONFIG_VP9_HIGHBITDEPTH , 0 +.equ CONFIG_BETTER_HW_COMPATIBILITY , 0 +.equ CONFIG_EXPERIMENTAL , 0 +.equ CONFIG_SIZE_LIMIT , 1 +.equ CONFIG_ALWAYS_ADJUST_BPM , 0 +.equ CONFIG_BITSTREAM_DEBUG , 0 +.equ CONFIG_MISMATCH_DEBUG , 0 +.equ CONFIG_FP_MB_STATS , 0 +.equ CONFIG_EMULATE_HARDWARE , 0 +.equ CONFIG_NON_GREEDY_MV , 0 +.equ CONFIG_RATE_CTRL , 0 +.equ CONFIG_COLLECT_COMPONENT_TIMING , 0 + .section .note.GNU-stack,"",%progbits diff --git a/media/libvpx/config/mac/arm64/vpx_config.c b/media/libvpx/config/mac/arm64/vpx_config.c new file mode 100644 index 000000000000..3cf744dc1621 --- /dev/null +++ b/media/libvpx/config/mac/arm64/vpx_config.c @@ -0,0 +1,10 @@ +/* Copyright (c) 2011 The WebM project authors. All Rights Reserved. */ +/* */ +/* Use of this source code is governed by a BSD-style license */ +/* that can be found in the LICENSE file in the root of the source */ +/* tree. An additional intellectual property rights grant can be found */ +/* in the file PATENTS. All contributing project authors may */ +/* be found in the AUTHORS file in the root of the source tree. */ +#include "vpx/vpx_codec.h" +static const char* const cfg = "--target=arm64-darwin-gcc --enable-external-build --disable-examples --disable-install-docs --disable-unit-tests --enable-multi-res-encoding --size-limit=8192x4608 --enable-pic --disable-avx512 --log=/home/cm/Work/gecko-dev/media/libvpx/config/mac/arm64/config.log"; +const char *vpx_codec_build_config(void) {return cfg;} diff --git a/media/libvpx/config/mac/arm64/vpx_config.h b/media/libvpx/config/mac/arm64/vpx_config.h new file mode 100644 index 000000000000..098ae3eff4f6 --- /dev/null +++ b/media/libvpx/config/mac/arm64/vpx_config.h @@ -0,0 +1,109 @@ +/* Copyright (c) 2011 The WebM project authors. All Rights Reserved. */ +/* */ +/* Use of this source code is governed by a BSD-style license */ +/* that can be found in the LICENSE file in the root of the source */ +/* tree. An additional intellectual property rights grant can be found */ +/* in the file PATENTS. All contributing project authors may */ +/* be found in the AUTHORS file in the root of the source tree. */ +/* This file automatically generated by configure. Do not edit! */ +#ifndef VPX_CONFIG_H +#define VPX_CONFIG_H +#define RESTRICT +#define INLINE inline +#define VPX_ARCH_ARM 1 +#define VPX_ARCH_AARCH64 1 +#define VPX_ARCH_MIPS 0 +#define VPX_ARCH_X86 0 +#define VPX_ARCH_X86_64 0 +#define VPX_ARCH_PPC 0 +#define VPX_ARCH_LOONGARCH 0 +#define HAVE_NEON_ASM 0 +#define HAVE_NEON 1 +#define HAVE_NEON_DOTPROD 1 +#define HAVE_NEON_I8MM 1 +#define HAVE_SVE 1 +#define HAVE_SVE2 1 +#define HAVE_MIPS32 0 +#define HAVE_DSPR2 0 +#define HAVE_MSA 0 +#define HAVE_MIPS64 0 +#define HAVE_MMX 0 +#define HAVE_SSE 0 +#define HAVE_SSE2 0 +#define HAVE_SSE3 0 +#define HAVE_SSSE3 0 +#define HAVE_SSE4_1 0 +#define HAVE_AVX 0 +#define HAVE_AVX2 0 +#define HAVE_AVX512 0 +#define HAVE_VSX 0 +#define HAVE_MMI 0 +#define HAVE_LSX 0 +#define HAVE_LASX 0 +#define HAVE_VPX_PORTS 1 +#define HAVE_PTHREAD_H 1 +#define CONFIG_DEPENDENCY_TRACKING 1 +#define CONFIG_EXTERNAL_BUILD 1 +#define CONFIG_INSTALL_DOCS 0 +#define CONFIG_INSTALL_BINS 1 +#define CONFIG_INSTALL_LIBS 1 +#define CONFIG_INSTALL_SRCS 0 +#define CONFIG_DEBUG 0 +#define CONFIG_GPROF 0 +#define CONFIG_GCOV 0 +#define CONFIG_RVCT 0 +#define CONFIG_GCC 1 +#define CONFIG_MSVS 0 +#define CONFIG_PIC 1 +#define CONFIG_BIG_ENDIAN 0 +#define CONFIG_CODEC_SRCS 0 +#define CONFIG_DEBUG_LIBS 0 +#define CONFIG_DEQUANT_TOKENS 0 +#define CONFIG_DC_RECON 0 +#define CONFIG_RUNTIME_CPU_DETECT 1 +#define CONFIG_POSTPROC 0 +#define CONFIG_VP9_POSTPROC 0 +#define CONFIG_MULTITHREAD 1 +#define CONFIG_INTERNAL_STATS 0 +#define CONFIG_VP8_ENCODER 1 +#define CONFIG_VP8_DECODER 1 +#define CONFIG_VP9_ENCODER 1 +#define CONFIG_VP9_DECODER 1 +#define CONFIG_VP8 1 +#define CONFIG_VP9 1 +#define CONFIG_ENCODERS 1 +#define CONFIG_DECODERS 1 +#define CONFIG_STATIC_MSVCRT 0 +#define CONFIG_SPATIAL_RESAMPLING 1 +#define CONFIG_REALTIME_ONLY 0 +#define CONFIG_ONTHEFLY_BITPACKING 0 +#define CONFIG_ERROR_CONCEALMENT 0 +#define CONFIG_SHARED 0 +#define CONFIG_STATIC 1 +#define CONFIG_SMALL 0 +#define CONFIG_POSTPROC_VISUALIZER 0 +#define CONFIG_OS_SUPPORT 1 +#define CONFIG_UNIT_TESTS 0 +#define CONFIG_WEBM_IO 0 +#define CONFIG_LIBYUV 0 +#define CONFIG_DECODE_PERF_TESTS 0 +#define CONFIG_ENCODE_PERF_TESTS 0 +#define CONFIG_MULTI_RES_ENCODING 1 +#define CONFIG_TEMPORAL_DENOISING 1 +#define CONFIG_VP9_TEMPORAL_DENOISING 0 +#define CONFIG_COEFFICIENT_RANGE_CHECKING 0 +#define CONFIG_VP9_HIGHBITDEPTH 0 +#define CONFIG_BETTER_HW_COMPATIBILITY 0 +#define CONFIG_EXPERIMENTAL 0 +#define CONFIG_SIZE_LIMIT 1 +#define CONFIG_ALWAYS_ADJUST_BPM 0 +#define CONFIG_BITSTREAM_DEBUG 0 +#define CONFIG_MISMATCH_DEBUG 0 +#define CONFIG_FP_MB_STATS 0 +#define CONFIG_EMULATE_HARDWARE 0 +#define CONFIG_NON_GREEDY_MV 0 +#define CONFIG_RATE_CTRL 0 +#define CONFIG_COLLECT_COMPONENT_TIMING 0 +#define DECODE_WIDTH_LIMIT 8192 +#define DECODE_HEIGHT_LIMIT 4608 +#endif /* VPX_CONFIG_H */ diff --git a/media/libvpx/config/mac/arm64/vpx_dsp_rtcd.h b/media/libvpx/config/mac/arm64/vpx_dsp_rtcd.h new file mode 100644 index 000000000000..2c31ee4ef96f --- /dev/null +++ b/media/libvpx/config/mac/arm64/vpx_dsp_rtcd.h @@ -0,0 +1,1187 @@ +// This file is generated. Do not edit. +#ifndef VPX_DSP_RTCD_H_ +#define VPX_DSP_RTCD_H_ + +#ifdef RTCD_C +#define RTCD_EXTERN +#else +#define RTCD_EXTERN extern +#endif + +/* + * DSP + */ + +#include "vpx/vpx_integer.h" +#include "vpx_dsp/vpx_dsp_common.h" +#include "vpx_dsp/vpx_filter.h" +#if CONFIG_VP9_ENCODER + struct macroblock_plane; + struct ScanOrder; +#endif + + +#ifdef __cplusplus +extern "C" { +#endif + +unsigned int vpx_avg_4x4_c(const uint8_t *, int p); +unsigned int vpx_avg_4x4_neon(const uint8_t *, int p); +#define vpx_avg_4x4 vpx_avg_4x4_neon + +unsigned int vpx_avg_8x8_c(const uint8_t *, int p); +unsigned int vpx_avg_8x8_neon(const uint8_t *, int p); +#define vpx_avg_8x8 vpx_avg_8x8_neon + +void vpx_comp_avg_pred_c(uint8_t *comp_pred, const uint8_t *pred, int width, int height, const uint8_t *ref, int ref_stride); +void vpx_comp_avg_pred_neon(uint8_t *comp_pred, const uint8_t *pred, int width, int height, const uint8_t *ref, int ref_stride); +#define vpx_comp_avg_pred vpx_comp_avg_pred_neon + +void vpx_convolve8_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve8_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8_avg)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve8_avg_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_horiz_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_horiz_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_horiz_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8_avg_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve8_avg_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_vert_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_vert_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_avg_vert_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8_avg_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_horiz_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_horiz_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_horiz_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_vert_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_vert_neon_dotprod(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve8_vert_neon_i8mm(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +RTCD_EXTERN void (*vpx_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); + +void vpx_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve_avg_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_convolve_avg vpx_convolve_avg_neon + +void vpx_convolve_copy_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_convolve_copy_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_convolve_copy vpx_convolve_copy_neon + +void vpx_d117_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d117_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d117_predictor_16x16 vpx_d117_predictor_16x16_neon + +void vpx_d117_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d117_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d117_predictor_32x32 vpx_d117_predictor_32x32_neon + +void vpx_d117_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d117_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d117_predictor_4x4 vpx_d117_predictor_4x4_neon + +void vpx_d117_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d117_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d117_predictor_8x8 vpx_d117_predictor_8x8_neon + +void vpx_d135_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d135_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d135_predictor_16x16 vpx_d135_predictor_16x16_neon + +void vpx_d135_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d135_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d135_predictor_32x32 vpx_d135_predictor_32x32_neon + +void vpx_d135_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d135_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d135_predictor_4x4 vpx_d135_predictor_4x4_neon + +void vpx_d135_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d135_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d135_predictor_8x8 vpx_d135_predictor_8x8_neon + +void vpx_d153_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d153_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d153_predictor_16x16 vpx_d153_predictor_16x16_neon + +void vpx_d153_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d153_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d153_predictor_32x32 vpx_d153_predictor_32x32_neon + +void vpx_d153_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d153_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d153_predictor_4x4 vpx_d153_predictor_4x4_neon + +void vpx_d153_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d153_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d153_predictor_8x8 vpx_d153_predictor_8x8_neon + +void vpx_d207_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d207_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d207_predictor_16x16 vpx_d207_predictor_16x16_neon + +void vpx_d207_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d207_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d207_predictor_32x32 vpx_d207_predictor_32x32_neon + +void vpx_d207_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d207_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d207_predictor_4x4 vpx_d207_predictor_4x4_neon + +void vpx_d207_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d207_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d207_predictor_8x8 vpx_d207_predictor_8x8_neon + +void vpx_d45_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d45_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d45_predictor_16x16 vpx_d45_predictor_16x16_neon + +void vpx_d45_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d45_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d45_predictor_32x32 vpx_d45_predictor_32x32_neon + +void vpx_d45_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d45_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d45_predictor_4x4 vpx_d45_predictor_4x4_neon + +void vpx_d45_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d45_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d45_predictor_8x8 vpx_d45_predictor_8x8_neon + +void vpx_d45e_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d45e_predictor_4x4 vpx_d45e_predictor_4x4_c + +void vpx_d63_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d63_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d63_predictor_16x16 vpx_d63_predictor_16x16_neon + +void vpx_d63_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d63_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d63_predictor_32x32 vpx_d63_predictor_32x32_neon + +void vpx_d63_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d63_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d63_predictor_4x4 vpx_d63_predictor_4x4_neon + +void vpx_d63_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_d63_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d63_predictor_8x8 vpx_d63_predictor_8x8_neon + +void vpx_d63e_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_d63e_predictor_4x4 vpx_d63e_predictor_4x4_c + +void vpx_dc_128_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_128_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_128_predictor_16x16 vpx_dc_128_predictor_16x16_neon + +void vpx_dc_128_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_128_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_128_predictor_32x32 vpx_dc_128_predictor_32x32_neon + +void vpx_dc_128_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_128_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_128_predictor_4x4 vpx_dc_128_predictor_4x4_neon + +void vpx_dc_128_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_128_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_128_predictor_8x8 vpx_dc_128_predictor_8x8_neon + +void vpx_dc_left_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_left_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_left_predictor_16x16 vpx_dc_left_predictor_16x16_neon + +void vpx_dc_left_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_left_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_left_predictor_32x32 vpx_dc_left_predictor_32x32_neon + +void vpx_dc_left_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_left_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_left_predictor_4x4 vpx_dc_left_predictor_4x4_neon + +void vpx_dc_left_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_left_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_left_predictor_8x8 vpx_dc_left_predictor_8x8_neon + +void vpx_dc_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_predictor_16x16 vpx_dc_predictor_16x16_neon + +void vpx_dc_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_predictor_32x32 vpx_dc_predictor_32x32_neon + +void vpx_dc_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_predictor_4x4 vpx_dc_predictor_4x4_neon + +void vpx_dc_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_predictor_8x8 vpx_dc_predictor_8x8_neon + +void vpx_dc_top_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_top_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_top_predictor_16x16 vpx_dc_top_predictor_16x16_neon + +void vpx_dc_top_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_top_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_top_predictor_32x32 vpx_dc_top_predictor_32x32_neon + +void vpx_dc_top_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_top_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_top_predictor_4x4 vpx_dc_top_predictor_4x4_neon + +void vpx_dc_top_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_dc_top_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_dc_top_predictor_8x8 vpx_dc_top_predictor_8x8_neon + +void vpx_fdct16x16_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct16x16_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct16x16 vpx_fdct16x16_neon + +void vpx_fdct16x16_1_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct16x16_1_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct16x16_1 vpx_fdct16x16_1_neon + +void vpx_fdct32x32_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct32x32_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct32x32 vpx_fdct32x32_neon + +void vpx_fdct32x32_1_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct32x32_1_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct32x32_1 vpx_fdct32x32_1_neon + +void vpx_fdct32x32_rd_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct32x32_rd_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct32x32_rd vpx_fdct32x32_rd_neon + +void vpx_fdct4x4_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct4x4_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct4x4 vpx_fdct4x4_neon + +void vpx_fdct4x4_1_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct4x4_1_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct4x4_1 vpx_fdct4x4_1_neon + +void vpx_fdct8x8_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct8x8_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct8x8 vpx_fdct8x8_neon + +void vpx_fdct8x8_1_c(const int16_t *input, tran_low_t *output, int stride); +void vpx_fdct8x8_1_neon(const int16_t *input, tran_low_t *output, int stride); +#define vpx_fdct8x8_1 vpx_fdct8x8_1_neon + +void vpx_get16x16var_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +void vpx_get16x16var_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +void vpx_get16x16var_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +RTCD_EXTERN void (*vpx_get16x16var)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); + +unsigned int vpx_get4x4sse_cs_c(const unsigned char *src_ptr, int src_stride, const unsigned char *ref_ptr, int ref_stride); +unsigned int vpx_get4x4sse_cs_neon(const unsigned char *src_ptr, int src_stride, const unsigned char *ref_ptr, int ref_stride); +unsigned int vpx_get4x4sse_cs_neon_dotprod(const unsigned char *src_ptr, int src_stride, const unsigned char *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_get4x4sse_cs)(const unsigned char *src_ptr, int src_stride, const unsigned char *ref_ptr, int ref_stride); + +void vpx_get8x8var_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +void vpx_get8x8var_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +void vpx_get8x8var_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); +RTCD_EXTERN void (*vpx_get8x8var)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, int *sum); + +unsigned int vpx_get_mb_ss_c(const int16_t *); +#define vpx_get_mb_ss vpx_get_mb_ss_c + +void vpx_h_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_h_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_h_predictor_16x16 vpx_h_predictor_16x16_neon + +void vpx_h_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_h_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_h_predictor_32x32 vpx_h_predictor_32x32_neon + +void vpx_h_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_h_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_h_predictor_4x4 vpx_h_predictor_4x4_neon + +void vpx_h_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_h_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_h_predictor_8x8 vpx_h_predictor_8x8_neon + +void vpx_hadamard_16x16_c(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +void vpx_hadamard_16x16_neon(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +#define vpx_hadamard_16x16 vpx_hadamard_16x16_neon + +void vpx_hadamard_32x32_c(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +void vpx_hadamard_32x32_neon(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +#define vpx_hadamard_32x32 vpx_hadamard_32x32_neon + +void vpx_hadamard_8x8_c(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +void vpx_hadamard_8x8_neon(const int16_t *src_diff, ptrdiff_t src_stride, int16_t *coeff); +#define vpx_hadamard_8x8 vpx_hadamard_8x8_neon + +void vpx_he_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_he_predictor_4x4 vpx_he_predictor_4x4_c + +void vpx_idct16x16_10_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct16x16_10_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct16x16_10_add vpx_idct16x16_10_add_neon + +void vpx_idct16x16_1_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct16x16_1_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct16x16_1_add vpx_idct16x16_1_add_neon + +void vpx_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct16x16_256_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct16x16_256_add vpx_idct16x16_256_add_neon + +void vpx_idct16x16_38_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct16x16_38_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct16x16_38_add vpx_idct16x16_38_add_neon + +void vpx_idct32x32_1024_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct32x32_1024_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct32x32_1024_add vpx_idct32x32_1024_add_neon + +void vpx_idct32x32_135_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct32x32_135_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct32x32_135_add vpx_idct32x32_135_add_neon + +void vpx_idct32x32_1_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct32x32_1_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct32x32_1_add vpx_idct32x32_1_add_neon + +void vpx_idct32x32_34_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct32x32_34_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct32x32_34_add vpx_idct32x32_34_add_neon + +void vpx_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct4x4_16_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct4x4_16_add vpx_idct4x4_16_add_neon + +void vpx_idct4x4_1_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct4x4_1_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct4x4_1_add vpx_idct4x4_1_add_neon + +void vpx_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct8x8_12_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct8x8_12_add vpx_idct8x8_12_add_neon + +void vpx_idct8x8_1_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct8x8_1_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct8x8_1_add vpx_idct8x8_1_add_neon + +void vpx_idct8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride); +void vpx_idct8x8_64_add_neon(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_idct8x8_64_add vpx_idct8x8_64_add_neon + +int16_t vpx_int_pro_col_c(const uint8_t *ref, const int width); +int16_t vpx_int_pro_col_neon(const uint8_t *ref, const int width); +#define vpx_int_pro_col vpx_int_pro_col_neon + +void vpx_int_pro_row_c(int16_t hbuf[16], const uint8_t *ref, const int ref_stride, const int height); +void vpx_int_pro_row_neon(int16_t hbuf[16], const uint8_t *ref, const int ref_stride, const int height); +#define vpx_int_pro_row vpx_int_pro_row_neon + +void vpx_iwht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_iwht4x4_16_add vpx_iwht4x4_16_add_c + +void vpx_iwht4x4_1_add_c(const tran_low_t *input, uint8_t *dest, int stride); +#define vpx_iwht4x4_1_add vpx_iwht4x4_1_add_c + +void vpx_lpf_horizontal_16_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_horizontal_16_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_horizontal_16 vpx_lpf_horizontal_16_neon + +void vpx_lpf_horizontal_16_dual_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_horizontal_16_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_horizontal_16_dual vpx_lpf_horizontal_16_dual_neon + +void vpx_lpf_horizontal_4_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_horizontal_4_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_horizontal_4 vpx_lpf_horizontal_4_neon + +void vpx_lpf_horizontal_4_dual_c(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +void vpx_lpf_horizontal_4_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +#define vpx_lpf_horizontal_4_dual vpx_lpf_horizontal_4_dual_neon + +void vpx_lpf_horizontal_8_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_horizontal_8_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_horizontal_8 vpx_lpf_horizontal_8_neon + +void vpx_lpf_horizontal_8_dual_c(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +void vpx_lpf_horizontal_8_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +#define vpx_lpf_horizontal_8_dual vpx_lpf_horizontal_8_dual_neon + +void vpx_lpf_vertical_16_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_vertical_16_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_vertical_16 vpx_lpf_vertical_16_neon + +void vpx_lpf_vertical_16_dual_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_vertical_16_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_vertical_16_dual vpx_lpf_vertical_16_dual_neon + +void vpx_lpf_vertical_4_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_vertical_4_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_vertical_4 vpx_lpf_vertical_4_neon + +void vpx_lpf_vertical_4_dual_c(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +void vpx_lpf_vertical_4_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +#define vpx_lpf_vertical_4_dual vpx_lpf_vertical_4_dual_neon + +void vpx_lpf_vertical_8_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +void vpx_lpf_vertical_8_neon(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh); +#define vpx_lpf_vertical_8 vpx_lpf_vertical_8_neon + +void vpx_lpf_vertical_8_dual_c(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +void vpx_lpf_vertical_8_dual_neon(uint8_t *s, int pitch, const uint8_t *blimit0, const uint8_t *limit0, const uint8_t *thresh0, const uint8_t *blimit1, const uint8_t *limit1, const uint8_t *thresh1); +#define vpx_lpf_vertical_8_dual vpx_lpf_vertical_8_dual_neon + +void vpx_minmax_8x8_c(const uint8_t *s, int p, const uint8_t *d, int dp, int *min, int *max); +void vpx_minmax_8x8_neon(const uint8_t *s, int p, const uint8_t *d, int dp, int *min, int *max); +#define vpx_minmax_8x8 vpx_minmax_8x8_neon + +unsigned int vpx_mse16x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse16x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse16x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_mse16x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_mse16x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse16x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse16x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_mse16x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_mse8x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse8x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse8x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_mse8x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_mse8x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse8x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_mse8x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_mse8x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +void vpx_quantize_b_c(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +void vpx_quantize_b_neon(const tran_low_t *coeff_ptr, intptr_t n_coeffs, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +#define vpx_quantize_b vpx_quantize_b_neon + +void vpx_quantize_b_32x32_c(const tran_low_t *coeff_ptr, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +void vpx_quantize_b_32x32_neon(const tran_low_t *coeff_ptr, const struct macroblock_plane *const mb_plane, tran_low_t *qcoeff_ptr, tran_low_t *dqcoeff_ptr, const int16_t *dequant_ptr, uint16_t *eob_ptr, const struct ScanOrder *const scan_order); +#define vpx_quantize_b_32x32 vpx_quantize_b_32x32_neon + +unsigned int vpx_sad16x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad16x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad16x16_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x16_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x16_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad16x16_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad16x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x16x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad16x16x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad16x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad16x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad16x32_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x32_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x32_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad16x32_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad16x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad16x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad16x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad16x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad16x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad16x8_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x8_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad16x8_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad16x8_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad16x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad16x8x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad16x8x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad32x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad32x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad32x16_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x16_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x16_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad32x16_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad32x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x16x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad32x16x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad32x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad32x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad32x32_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x32_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x32_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad32x32_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad32x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad32x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad32x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad32x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad32x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad32x64_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x64_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad32x64_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad32x64_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad32x64x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x64x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad32x64x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad32x64x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad4x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad4x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad4x4 vpx_sad4x4_neon + +unsigned int vpx_sad4x4_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad4x4_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +#define vpx_sad4x4_avg vpx_sad4x4_avg_neon + +void vpx_sad4x4x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad4x4x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad4x4x4d vpx_sad4x4x4d_neon + +unsigned int vpx_sad4x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad4x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad4x8 vpx_sad4x8_neon + +unsigned int vpx_sad4x8_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad4x8_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +#define vpx_sad4x8_avg vpx_sad4x8_avg_neon + +void vpx_sad4x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad4x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad4x8x4d vpx_sad4x8x4d_neon + +unsigned int vpx_sad64x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad64x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad64x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad64x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad64x32_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad64x32_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad64x32_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad64x32_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad64x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad64x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad64x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad64x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad64x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad64x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad64x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad64x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +unsigned int vpx_sad64x64_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad64x64_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad64x64_avg_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +RTCD_EXTERN unsigned int (*vpx_sad64x64_avg)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); + +void vpx_sad64x64x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad64x64x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad64x64x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad64x64x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad8x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad8x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad8x16 vpx_sad8x16_neon + +unsigned int vpx_sad8x16_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad8x16_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +#define vpx_sad8x16_avg vpx_sad8x16_avg_neon + +void vpx_sad8x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad8x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad8x16x4d vpx_sad8x16x4d_neon + +unsigned int vpx_sad8x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad8x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad8x4 vpx_sad8x4_neon + +unsigned int vpx_sad8x4_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad8x4_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +#define vpx_sad8x4_avg vpx_sad8x4_avg_neon + +void vpx_sad8x4x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad8x4x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad8x4x4d vpx_sad8x4x4d_neon + +unsigned int vpx_sad8x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad8x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad8x8 vpx_sad8x8_neon + +unsigned int vpx_sad8x8_avg_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +unsigned int vpx_sad8x8_avg_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, const uint8_t *second_pred); +#define vpx_sad8x8_avg vpx_sad8x8_avg_neon + +void vpx_sad8x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad8x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad8x8x4d vpx_sad8x8x4d_neon + +unsigned int vpx_sad_skip_16x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_16x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_16x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x16x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_16x16x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_16x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_16x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_16x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_16x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_16x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_16x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_16x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_16x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_16x8x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_16x8x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_32x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_32x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_32x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x16x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_32x16x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_32x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_32x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_32x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_32x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_32x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_32x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_32x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_32x64x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x64x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_32x64x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_32x64x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_4x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_4x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad_skip_4x4 vpx_sad_skip_4x4_neon + +void vpx_sad_skip_4x4x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_4x4x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad_skip_4x4x4d vpx_sad_skip_4x4x4d_neon + +unsigned int vpx_sad_skip_4x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_4x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad_skip_4x8 vpx_sad_skip_4x8_neon + +void vpx_sad_skip_4x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_4x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad_skip_4x8x4d vpx_sad_skip_4x8x4d_neon + +unsigned int vpx_sad_skip_64x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_64x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_64x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_64x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_64x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_64x32x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_64x32x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_64x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_64x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_64x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_64x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +RTCD_EXTERN unsigned int (*vpx_sad_skip_64x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); + +void vpx_sad_skip_64x64x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_64x64x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_64x64x4d_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +RTCD_EXTERN void (*vpx_sad_skip_64x64x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); + +unsigned int vpx_sad_skip_8x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_8x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad_skip_8x16 vpx_sad_skip_8x16_neon + +void vpx_sad_skip_8x16x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_8x16x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad_skip_8x16x4d vpx_sad_skip_8x16x4d_neon + +unsigned int vpx_sad_skip_8x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_8x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad_skip_8x4 vpx_sad_skip_8x4_neon + +void vpx_sad_skip_8x4x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_8x4x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad_skip_8x4x4d vpx_sad_skip_8x4x4d_neon + +unsigned int vpx_sad_skip_8x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +unsigned int vpx_sad_skip_8x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride); +#define vpx_sad_skip_8x8 vpx_sad_skip_8x8_neon + +void vpx_sad_skip_8x8x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +void vpx_sad_skip_8x8x4d_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *const ref_array[4], int ref_stride, uint32_t sad_array[4]); +#define vpx_sad_skip_8x8x4d vpx_sad_skip_8x8x4d_neon + +int vpx_satd_c(const int16_t *coeff, int length); +int vpx_satd_neon(const int16_t *coeff, int length); +#define vpx_satd vpx_satd_neon + +void vpx_scaled_2d_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +void vpx_scaled_2d_neon(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_2d vpx_scaled_2d_neon + +void vpx_scaled_avg_2d_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_avg_2d vpx_scaled_avg_2d_c + +void vpx_scaled_avg_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_avg_horiz vpx_scaled_avg_horiz_c + +void vpx_scaled_avg_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_avg_vert vpx_scaled_avg_vert_c + +void vpx_scaled_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_horiz vpx_scaled_horiz_c + +void vpx_scaled_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const InterpKernel *filter, int x0_q4, int x_step_q4, int y0_q4, int y_step_q4, int w, int h); +#define vpx_scaled_vert vpx_scaled_vert_c + +int64_t vpx_sse_c(const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, int width, int height); +int64_t vpx_sse_neon(const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, int width, int height); +int64_t vpx_sse_neon_dotprod(const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, int width, int height); +RTCD_EXTERN int64_t (*vpx_sse)(const uint8_t *src, int src_stride, const uint8_t *ref, int ref_stride, int width, int height); + +uint32_t vpx_sub_pixel_avg_variance16x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance16x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance16x16 vpx_sub_pixel_avg_variance16x16_neon + +uint32_t vpx_sub_pixel_avg_variance16x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance16x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance16x32 vpx_sub_pixel_avg_variance16x32_neon + +uint32_t vpx_sub_pixel_avg_variance16x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance16x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance16x8 vpx_sub_pixel_avg_variance16x8_neon + +uint32_t vpx_sub_pixel_avg_variance32x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance32x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance32x16 vpx_sub_pixel_avg_variance32x16_neon + +uint32_t vpx_sub_pixel_avg_variance32x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance32x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance32x32 vpx_sub_pixel_avg_variance32x32_neon + +uint32_t vpx_sub_pixel_avg_variance32x64_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance32x64_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance32x64 vpx_sub_pixel_avg_variance32x64_neon + +uint32_t vpx_sub_pixel_avg_variance4x4_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance4x4_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance4x4 vpx_sub_pixel_avg_variance4x4_neon + +uint32_t vpx_sub_pixel_avg_variance4x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance4x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance4x8 vpx_sub_pixel_avg_variance4x8_neon + +uint32_t vpx_sub_pixel_avg_variance64x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance64x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance64x32 vpx_sub_pixel_avg_variance64x32_neon + +uint32_t vpx_sub_pixel_avg_variance64x64_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance64x64_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance64x64 vpx_sub_pixel_avg_variance64x64_neon + +uint32_t vpx_sub_pixel_avg_variance8x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance8x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance8x16 vpx_sub_pixel_avg_variance8x16_neon + +uint32_t vpx_sub_pixel_avg_variance8x4_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance8x4_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance8x4 vpx_sub_pixel_avg_variance8x4_neon + +uint32_t vpx_sub_pixel_avg_variance8x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +uint32_t vpx_sub_pixel_avg_variance8x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse, const uint8_t *second_pred); +#define vpx_sub_pixel_avg_variance8x8 vpx_sub_pixel_avg_variance8x8_neon + +uint32_t vpx_sub_pixel_variance16x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance16x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance16x16 vpx_sub_pixel_variance16x16_neon + +uint32_t vpx_sub_pixel_variance16x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance16x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance16x32 vpx_sub_pixel_variance16x32_neon + +uint32_t vpx_sub_pixel_variance16x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance16x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance16x8 vpx_sub_pixel_variance16x8_neon + +uint32_t vpx_sub_pixel_variance32x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance32x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance32x16 vpx_sub_pixel_variance32x16_neon + +uint32_t vpx_sub_pixel_variance32x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance32x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance32x32 vpx_sub_pixel_variance32x32_neon + +uint32_t vpx_sub_pixel_variance32x64_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance32x64_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance32x64 vpx_sub_pixel_variance32x64_neon + +uint32_t vpx_sub_pixel_variance4x4_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance4x4_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance4x4 vpx_sub_pixel_variance4x4_neon + +uint32_t vpx_sub_pixel_variance4x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance4x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance4x8 vpx_sub_pixel_variance4x8_neon + +uint32_t vpx_sub_pixel_variance64x32_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance64x32_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance64x32 vpx_sub_pixel_variance64x32_neon + +uint32_t vpx_sub_pixel_variance64x64_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance64x64_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance64x64 vpx_sub_pixel_variance64x64_neon + +uint32_t vpx_sub_pixel_variance8x16_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance8x16_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance8x16 vpx_sub_pixel_variance8x16_neon + +uint32_t vpx_sub_pixel_variance8x4_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance8x4_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance8x4 vpx_sub_pixel_variance8x4_neon + +uint32_t vpx_sub_pixel_variance8x8_c(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +uint32_t vpx_sub_pixel_variance8x8_neon(const uint8_t *src_ptr, int src_stride, int x_offset, int y_offset, const uint8_t *ref_ptr, int ref_stride, uint32_t *sse); +#define vpx_sub_pixel_variance8x8 vpx_sub_pixel_variance8x8_neon + +void vpx_subtract_block_c(int rows, int cols, int16_t *diff_ptr, ptrdiff_t diff_stride, const uint8_t *src_ptr, ptrdiff_t src_stride, const uint8_t *pred_ptr, ptrdiff_t pred_stride); +void vpx_subtract_block_neon(int rows, int cols, int16_t *diff_ptr, ptrdiff_t diff_stride, const uint8_t *src_ptr, ptrdiff_t src_stride, const uint8_t *pred_ptr, ptrdiff_t pred_stride); +#define vpx_subtract_block vpx_subtract_block_neon + +uint64_t vpx_sum_squares_2d_i16_c(const int16_t *src, int stride, int size); +uint64_t vpx_sum_squares_2d_i16_neon(const int16_t *src, int stride, int size); +uint64_t vpx_sum_squares_2d_i16_sve(const int16_t *src, int stride, int size); +RTCD_EXTERN uint64_t (*vpx_sum_squares_2d_i16)(const int16_t *src, int stride, int size); + +void vpx_tm_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_tm_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_tm_predictor_16x16 vpx_tm_predictor_16x16_neon + +void vpx_tm_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_tm_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_tm_predictor_32x32 vpx_tm_predictor_32x32_neon + +void vpx_tm_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_tm_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_tm_predictor_4x4 vpx_tm_predictor_4x4_neon + +void vpx_tm_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_tm_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_tm_predictor_8x8 vpx_tm_predictor_8x8_neon + +void vpx_v_predictor_16x16_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_v_predictor_16x16_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_v_predictor_16x16 vpx_v_predictor_16x16_neon + +void vpx_v_predictor_32x32_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_v_predictor_32x32_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_v_predictor_32x32 vpx_v_predictor_32x32_neon + +void vpx_v_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_v_predictor_4x4_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_v_predictor_4x4 vpx_v_predictor_4x4_neon + +void vpx_v_predictor_8x8_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +void vpx_v_predictor_8x8_neon(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_v_predictor_8x8 vpx_v_predictor_8x8_neon + +unsigned int vpx_variance16x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance16x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance16x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance16x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance16x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance16x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance16x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance32x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance32x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance32x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance32x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance32x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance32x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance32x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance4x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance4x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance4x4_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance4x4)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance4x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance4x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance4x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance4x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance64x32_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance64x32_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance64x32_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance64x32)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance64x64_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance64x64_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance64x64_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance64x64)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance8x16_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x16_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x16_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance8x16)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance8x4_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x4_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x4_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance8x4)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +unsigned int vpx_variance8x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x8_neon(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +unsigned int vpx_variance8x8_neon_dotprod(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); +RTCD_EXTERN unsigned int (*vpx_variance8x8)(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); + +void vpx_ve_predictor_4x4_c(uint8_t *dst, ptrdiff_t stride, const uint8_t *above, const uint8_t *left); +#define vpx_ve_predictor_4x4 vpx_ve_predictor_4x4_c + +int vpx_vector_var_c(const int16_t *ref, const int16_t *src, const int bwl); +int vpx_vector_var_neon(const int16_t *ref, const int16_t *src, const int bwl); +#define vpx_vector_var vpx_vector_var_neon + +void vpx_dsp_rtcd(void); + +#include "vpx_config.h" + +#ifdef RTCD_C +#include "vpx_ports/arm.h" +static void setup_rtcd_internal(void) +{ + int flags = arm_cpu_caps(); + + (void)flags; + + vpx_convolve8 = vpx_convolve8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8 = vpx_convolve8_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8 = vpx_convolve8_neon_i8mm; + vpx_convolve8_avg = vpx_convolve8_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8_avg = vpx_convolve8_avg_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8_avg = vpx_convolve8_avg_neon_i8mm; + vpx_convolve8_avg_horiz = vpx_convolve8_avg_horiz_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8_avg_horiz = vpx_convolve8_avg_horiz_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8_avg_horiz = vpx_convolve8_avg_horiz_neon_i8mm; + vpx_convolve8_avg_vert = vpx_convolve8_avg_vert_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8_avg_vert = vpx_convolve8_avg_vert_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8_avg_vert = vpx_convolve8_avg_vert_neon_i8mm; + vpx_convolve8_horiz = vpx_convolve8_horiz_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8_horiz = vpx_convolve8_horiz_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8_horiz = vpx_convolve8_horiz_neon_i8mm; + vpx_convolve8_vert = vpx_convolve8_vert_neon; + if (flags & HAS_NEON_DOTPROD) vpx_convolve8_vert = vpx_convolve8_vert_neon_dotprod; + if (flags & HAS_NEON_I8MM) vpx_convolve8_vert = vpx_convolve8_vert_neon_i8mm; + vpx_get16x16var = vpx_get16x16var_neon; + if (flags & HAS_NEON_DOTPROD) vpx_get16x16var = vpx_get16x16var_neon_dotprod; + vpx_get4x4sse_cs = vpx_get4x4sse_cs_neon; + if (flags & HAS_NEON_DOTPROD) vpx_get4x4sse_cs = vpx_get4x4sse_cs_neon_dotprod; + vpx_get8x8var = vpx_get8x8var_neon; + if (flags & HAS_NEON_DOTPROD) vpx_get8x8var = vpx_get8x8var_neon_dotprod; + vpx_mse16x16 = vpx_mse16x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_mse16x16 = vpx_mse16x16_neon_dotprod; + vpx_mse16x8 = vpx_mse16x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_mse16x8 = vpx_mse16x8_neon_dotprod; + vpx_mse8x16 = vpx_mse8x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_mse8x16 = vpx_mse8x16_neon_dotprod; + vpx_mse8x8 = vpx_mse8x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_mse8x8 = vpx_mse8x8_neon_dotprod; + vpx_sad16x16 = vpx_sad16x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x16 = vpx_sad16x16_neon_dotprod; + vpx_sad16x16_avg = vpx_sad16x16_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x16_avg = vpx_sad16x16_avg_neon_dotprod; + vpx_sad16x16x4d = vpx_sad16x16x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x16x4d = vpx_sad16x16x4d_neon_dotprod; + vpx_sad16x32 = vpx_sad16x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x32 = vpx_sad16x32_neon_dotprod; + vpx_sad16x32_avg = vpx_sad16x32_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x32_avg = vpx_sad16x32_avg_neon_dotprod; + vpx_sad16x32x4d = vpx_sad16x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x32x4d = vpx_sad16x32x4d_neon_dotprod; + vpx_sad16x8 = vpx_sad16x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x8 = vpx_sad16x8_neon_dotprod; + vpx_sad16x8_avg = vpx_sad16x8_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x8_avg = vpx_sad16x8_avg_neon_dotprod; + vpx_sad16x8x4d = vpx_sad16x8x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad16x8x4d = vpx_sad16x8x4d_neon_dotprod; + vpx_sad32x16 = vpx_sad32x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x16 = vpx_sad32x16_neon_dotprod; + vpx_sad32x16_avg = vpx_sad32x16_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x16_avg = vpx_sad32x16_avg_neon_dotprod; + vpx_sad32x16x4d = vpx_sad32x16x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x16x4d = vpx_sad32x16x4d_neon_dotprod; + vpx_sad32x32 = vpx_sad32x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x32 = vpx_sad32x32_neon_dotprod; + vpx_sad32x32_avg = vpx_sad32x32_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x32_avg = vpx_sad32x32_avg_neon_dotprod; + vpx_sad32x32x4d = vpx_sad32x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x32x4d = vpx_sad32x32x4d_neon_dotprod; + vpx_sad32x64 = vpx_sad32x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x64 = vpx_sad32x64_neon_dotprod; + vpx_sad32x64_avg = vpx_sad32x64_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x64_avg = vpx_sad32x64_avg_neon_dotprod; + vpx_sad32x64x4d = vpx_sad32x64x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad32x64x4d = vpx_sad32x64x4d_neon_dotprod; + vpx_sad64x32 = vpx_sad64x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x32 = vpx_sad64x32_neon_dotprod; + vpx_sad64x32_avg = vpx_sad64x32_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x32_avg = vpx_sad64x32_avg_neon_dotprod; + vpx_sad64x32x4d = vpx_sad64x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x32x4d = vpx_sad64x32x4d_neon_dotprod; + vpx_sad64x64 = vpx_sad64x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x64 = vpx_sad64x64_neon_dotprod; + vpx_sad64x64_avg = vpx_sad64x64_avg_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x64_avg = vpx_sad64x64_avg_neon_dotprod; + vpx_sad64x64x4d = vpx_sad64x64x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad64x64x4d = vpx_sad64x64x4d_neon_dotprod; + vpx_sad_skip_16x16 = vpx_sad_skip_16x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x16 = vpx_sad_skip_16x16_neon_dotprod; + vpx_sad_skip_16x16x4d = vpx_sad_skip_16x16x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x16x4d = vpx_sad_skip_16x16x4d_neon_dotprod; + vpx_sad_skip_16x32 = vpx_sad_skip_16x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x32 = vpx_sad_skip_16x32_neon_dotprod; + vpx_sad_skip_16x32x4d = vpx_sad_skip_16x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x32x4d = vpx_sad_skip_16x32x4d_neon_dotprod; + vpx_sad_skip_16x8 = vpx_sad_skip_16x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x8 = vpx_sad_skip_16x8_neon_dotprod; + vpx_sad_skip_16x8x4d = vpx_sad_skip_16x8x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_16x8x4d = vpx_sad_skip_16x8x4d_neon_dotprod; + vpx_sad_skip_32x16 = vpx_sad_skip_32x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x16 = vpx_sad_skip_32x16_neon_dotprod; + vpx_sad_skip_32x16x4d = vpx_sad_skip_32x16x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x16x4d = vpx_sad_skip_32x16x4d_neon_dotprod; + vpx_sad_skip_32x32 = vpx_sad_skip_32x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x32 = vpx_sad_skip_32x32_neon_dotprod; + vpx_sad_skip_32x32x4d = vpx_sad_skip_32x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x32x4d = vpx_sad_skip_32x32x4d_neon_dotprod; + vpx_sad_skip_32x64 = vpx_sad_skip_32x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x64 = vpx_sad_skip_32x64_neon_dotprod; + vpx_sad_skip_32x64x4d = vpx_sad_skip_32x64x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_32x64x4d = vpx_sad_skip_32x64x4d_neon_dotprod; + vpx_sad_skip_64x32 = vpx_sad_skip_64x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_64x32 = vpx_sad_skip_64x32_neon_dotprod; + vpx_sad_skip_64x32x4d = vpx_sad_skip_64x32x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_64x32x4d = vpx_sad_skip_64x32x4d_neon_dotprod; + vpx_sad_skip_64x64 = vpx_sad_skip_64x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_64x64 = vpx_sad_skip_64x64_neon_dotprod; + vpx_sad_skip_64x64x4d = vpx_sad_skip_64x64x4d_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sad_skip_64x64x4d = vpx_sad_skip_64x64x4d_neon_dotprod; + vpx_sse = vpx_sse_neon; + if (flags & HAS_NEON_DOTPROD) vpx_sse = vpx_sse_neon_dotprod; + vpx_sum_squares_2d_i16 = vpx_sum_squares_2d_i16_neon; + if (flags & HAS_SVE) vpx_sum_squares_2d_i16 = vpx_sum_squares_2d_i16_sve; + vpx_variance16x16 = vpx_variance16x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance16x16 = vpx_variance16x16_neon_dotprod; + vpx_variance16x32 = vpx_variance16x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance16x32 = vpx_variance16x32_neon_dotprod; + vpx_variance16x8 = vpx_variance16x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance16x8 = vpx_variance16x8_neon_dotprod; + vpx_variance32x16 = vpx_variance32x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance32x16 = vpx_variance32x16_neon_dotprod; + vpx_variance32x32 = vpx_variance32x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance32x32 = vpx_variance32x32_neon_dotprod; + vpx_variance32x64 = vpx_variance32x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance32x64 = vpx_variance32x64_neon_dotprod; + vpx_variance4x4 = vpx_variance4x4_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance4x4 = vpx_variance4x4_neon_dotprod; + vpx_variance4x8 = vpx_variance4x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance4x8 = vpx_variance4x8_neon_dotprod; + vpx_variance64x32 = vpx_variance64x32_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance64x32 = vpx_variance64x32_neon_dotprod; + vpx_variance64x64 = vpx_variance64x64_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance64x64 = vpx_variance64x64_neon_dotprod; + vpx_variance8x16 = vpx_variance8x16_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance8x16 = vpx_variance8x16_neon_dotprod; + vpx_variance8x4 = vpx_variance8x4_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance8x4 = vpx_variance8x4_neon_dotprod; + vpx_variance8x8 = vpx_variance8x8_neon; + if (flags & HAS_NEON_DOTPROD) vpx_variance8x8 = vpx_variance8x8_neon_dotprod; +} +#endif + +#ifdef __cplusplus +} // extern "C" +#endif + +#endif diff --git a/media/libvpx/config/mac/arm64/vpx_scale_rtcd.h b/media/libvpx/config/mac/arm64/vpx_scale_rtcd.h new file mode 100644 index 000000000000..b37136827560 --- /dev/null +++ b/media/libvpx/config/mac/arm64/vpx_scale_rtcd.h @@ -0,0 +1,75 @@ +// This file is generated. Do not edit. +#ifndef VPX_SCALE_RTCD_H_ +#define VPX_SCALE_RTCD_H_ + +#ifdef RTCD_C +#define RTCD_EXTERN +#else +#define RTCD_EXTERN extern +#endif + +struct yv12_buffer_config; + +#ifdef __cplusplus +extern "C" { +#endif + +void vp8_horizontal_line_2_1_scale_c(const unsigned char *source, unsigned int source_width, unsigned char *dest, unsigned int dest_width); +#define vp8_horizontal_line_2_1_scale vp8_horizontal_line_2_1_scale_c + +void vp8_horizontal_line_5_3_scale_c(const unsigned char *source, unsigned int source_width, unsigned char *dest, unsigned int dest_width); +#define vp8_horizontal_line_5_3_scale vp8_horizontal_line_5_3_scale_c + +void vp8_horizontal_line_5_4_scale_c(const unsigned char *source, unsigned int source_width, unsigned char *dest, unsigned int dest_width); +#define vp8_horizontal_line_5_4_scale vp8_horizontal_line_5_4_scale_c + +void vp8_vertical_band_2_1_scale_c(unsigned char *source, unsigned int src_pitch, unsigned char *dest, unsigned int dest_pitch, unsigned int dest_width); +#define vp8_vertical_band_2_1_scale vp8_vertical_band_2_1_scale_c + +void vp8_vertical_band_2_1_scale_i_c(unsigned char *source, unsigned int src_pitch, unsigned char *dest, unsigned int dest_pitch, unsigned int dest_width); +#define vp8_vertical_band_2_1_scale_i vp8_vertical_band_2_1_scale_i_c + +void vp8_vertical_band_5_3_scale_c(unsigned char *source, unsigned int src_pitch, unsigned char *dest, unsigned int dest_pitch, unsigned int dest_width); +#define vp8_vertical_band_5_3_scale vp8_vertical_band_5_3_scale_c + +void vp8_vertical_band_5_4_scale_c(unsigned char *source, unsigned int src_pitch, unsigned char *dest, unsigned int dest_pitch, unsigned int dest_width); +#define vp8_vertical_band_5_4_scale vp8_vertical_band_5_4_scale_c + +void vp8_yv12_copy_frame_c(const struct yv12_buffer_config *src_ybc, struct yv12_buffer_config *dst_ybc); +#define vp8_yv12_copy_frame vp8_yv12_copy_frame_c + +void vp8_yv12_extend_frame_borders_c(struct yv12_buffer_config *ybf); +#define vp8_yv12_extend_frame_borders vp8_yv12_extend_frame_borders_c + +void vpx_extend_frame_borders_c(struct yv12_buffer_config *ybf); +#define vpx_extend_frame_borders vpx_extend_frame_borders_c + +void vpx_extend_frame_inner_borders_c(struct yv12_buffer_config *ybf); +#define vpx_extend_frame_inner_borders vpx_extend_frame_inner_borders_c + +void vpx_yv12_copy_frame_c(const struct yv12_buffer_config *src_ybc, struct yv12_buffer_config *dst_ybc); +#define vpx_yv12_copy_frame vpx_yv12_copy_frame_c + +void vpx_yv12_copy_y_c(const struct yv12_buffer_config *src_ybc, struct yv12_buffer_config *dst_ybc); +#define vpx_yv12_copy_y vpx_yv12_copy_y_c + +void vpx_scale_rtcd(void); + +#include "vpx_config.h" + +#ifdef RTCD_C +#include "vpx_ports/arm.h" +static void setup_rtcd_internal(void) +{ + int flags = arm_cpu_caps(); + + (void)flags; + +} +#endif + +#ifdef __cplusplus +} // extern "C" +#endif + +#endif diff --git a/media/libvpx/sources.mozbuild b/media/libvpx/sources.mozbuild index 2b12bfa2ab8c..12c20ede9f3a 100644 --- a/media/libvpx/sources.mozbuild +++ b/media/libvpx/sources.mozbuild @@ -2142,6 +2142,262 @@ files = { 'libvpx/vpx_scale/vpx_scale_rtcd.c', 'libvpx/vpx_util/vpx_thread.c', 'libvpx/vpx_util/vpx_write_yuv_frame.c', +], + 'MAC_ARM64_EXPORTS': [ + 'libvpx/vpx/vp8.h', + 'libvpx/vpx/vp8cx.h', + 'libvpx/vpx/vp8dx.h', + 'libvpx/vpx/vpx_codec.h', + 'libvpx/vpx/vpx_decoder.h', + 'libvpx/vpx/vpx_encoder.h', + 'libvpx/vpx/vpx_ext_ratectrl.h', + 'libvpx/vpx/vpx_frame_buffer.h', + 'libvpx/vpx/vpx_image.h', + 'libvpx/vpx/vpx_integer.h', + 'libvpx/vpx/vpx_tpl.h', + 'libvpx/vpx_mem/include/vpx_mem_intrnl.h', + 'libvpx/vpx_mem/vpx_mem.h', + 'libvpx/vpx_ports/arm.h', + 'libvpx/vpx_ports/arm_cpudetect.h', + 'libvpx/vpx_ports/bitops.h', + 'libvpx/vpx_ports/compiler_attributes.h', + 'libvpx/vpx_ports/mem.h', + 'libvpx/vpx_ports/static_assert.h', + 'libvpx/vpx_ports/system_state.h', + 'libvpx/vpx_ports/vpx_timer.h', + 'libvpx/vpx_scale/vpx_scale.h', + 'libvpx/vpx_scale/yv12config.h', +], + 'MAC_ARM64_SOURCES': [ + 'libvpx/../config/mac/arm64/vpx_config.c', + 'libvpx/vp8/common/alloccommon.c', + 'libvpx/vp8/common/arm/loopfilter_arm.c', + 'libvpx/vp8/common/arm/neon/bilinearpredict_neon.c', + 'libvpx/vp8/common/arm/neon/copymem_neon.c', + 'libvpx/vp8/common/arm/neon/dc_only_idct_add_neon.c', + 'libvpx/vp8/common/arm/neon/dequant_idct_neon.c', + 'libvpx/vp8/common/arm/neon/dequantizeb_neon.c', + 'libvpx/vp8/common/arm/neon/idct_blk_neon.c', + 'libvpx/vp8/common/arm/neon/iwalsh_neon.c', + 'libvpx/vp8/common/arm/neon/loopfiltersimplehorizontaledge_neon.c', + 'libvpx/vp8/common/arm/neon/loopfiltersimpleverticaledge_neon.c', + 'libvpx/vp8/common/arm/neon/mbloopfilter_neon.c', + 'libvpx/vp8/common/arm/neon/shortidct4x4llm_neon.c', + 'libvpx/vp8/common/arm/neon/sixtappredict_neon.c', + 'libvpx/vp8/common/arm/neon/vp8_loopfilter_neon.c', + 'libvpx/vp8/common/blockd.c', + 'libvpx/vp8/common/dequantize.c', + 'libvpx/vp8/common/entropy.c', + 'libvpx/vp8/common/entropymode.c', + 'libvpx/vp8/common/entropymv.c', + 'libvpx/vp8/common/extend.c', + 'libvpx/vp8/common/filter.c', + 'libvpx/vp8/common/findnearmv.c', + 'libvpx/vp8/common/generic/systemdependent.c', + 'libvpx/vp8/common/idct_blk.c', + 'libvpx/vp8/common/idctllm.c', + 'libvpx/vp8/common/loopfilter_filters.c', + 'libvpx/vp8/common/mbpitch.c', + 'libvpx/vp8/common/modecont.c', + 'libvpx/vp8/common/quant_common.c', + 'libvpx/vp8/common/reconinter.c', + 'libvpx/vp8/common/reconintra.c', + 'libvpx/vp8/common/reconintra4x4.c', + 'libvpx/vp8/common/rtcd.c', + 'libvpx/vp8/common/setupintrarecon.c', + 'libvpx/vp8/common/swapyv12buffer.c', + 'libvpx/vp8/common/treecoder.c', + 'libvpx/vp8/common/vp8_loopfilter.c', + 'libvpx/vp8/common/vp8_skin_detection.c', + 'libvpx/vp8/decoder/dboolhuff.c', + 'libvpx/vp8/decoder/decodeframe.c', + 'libvpx/vp8/decoder/decodemv.c', + 'libvpx/vp8/decoder/detokenize.c', + 'libvpx/vp8/decoder/onyxd_if.c', + 'libvpx/vp8/decoder/threading.c', + 'libvpx/vp8/encoder/arm/neon/denoising_neon.c', + 'libvpx/vp8/encoder/arm/neon/fastquantizeb_neon.c', + 'libvpx/vp8/encoder/arm/neon/shortfdct_neon.c', + 'libvpx/vp8/encoder/arm/neon/vp8_shortwalsh4x4_neon.c', + 'libvpx/vp8/encoder/bitstream.c', + 'libvpx/vp8/encoder/boolhuff.c', + 'libvpx/vp8/encoder/copy_c.c', + 'libvpx/vp8/encoder/dct.c', + 'libvpx/vp8/encoder/denoising.c', + 'libvpx/vp8/encoder/encodeframe.c', + 'libvpx/vp8/encoder/encodeintra.c', + 'libvpx/vp8/encoder/encodemb.c', + 'libvpx/vp8/encoder/encodemv.c', + 'libvpx/vp8/encoder/ethreading.c', + 'libvpx/vp8/encoder/firstpass.c', + 'libvpx/vp8/encoder/lookahead.c', + 'libvpx/vp8/encoder/mcomp.c', + 'libvpx/vp8/encoder/modecosts.c', + 'libvpx/vp8/encoder/mr_dissim.c', + 'libvpx/vp8/encoder/onyx_if.c', + 'libvpx/vp8/encoder/pickinter.c', + 'libvpx/vp8/encoder/picklpf.c', + 'libvpx/vp8/encoder/ratectrl.c', + 'libvpx/vp8/encoder/rdopt.c', + 'libvpx/vp8/encoder/segmentation.c', + 'libvpx/vp8/encoder/temporal_filter.c', + 'libvpx/vp8/encoder/tokenize.c', + 'libvpx/vp8/encoder/treewriter.c', + 'libvpx/vp8/encoder/vp8_quantize.c', + 'libvpx/vp8/vp8_cx_iface.c', + 'libvpx/vp8/vp8_dx_iface.c', + 'libvpx/vp9/common/arm/neon/vp9_iht16x16_add_neon.c', + 'libvpx/vp9/common/arm/neon/vp9_iht4x4_add_neon.c', + 'libvpx/vp9/common/arm/neon/vp9_iht8x8_add_neon.c', + 'libvpx/vp9/common/vp9_alloccommon.c', + 'libvpx/vp9/common/vp9_blockd.c', + 'libvpx/vp9/common/vp9_common_data.c', + 'libvpx/vp9/common/vp9_entropy.c', + 'libvpx/vp9/common/vp9_entropymode.c', + 'libvpx/vp9/common/vp9_entropymv.c', + 'libvpx/vp9/common/vp9_filter.c', + 'libvpx/vp9/common/vp9_frame_buffers.c', + 'libvpx/vp9/common/vp9_idct.c', + 'libvpx/vp9/common/vp9_loopfilter.c', + 'libvpx/vp9/common/vp9_mvref_common.c', + 'libvpx/vp9/common/vp9_pred_common.c', + 'libvpx/vp9/common/vp9_quant_common.c', + 'libvpx/vp9/common/vp9_reconinter.c', + 'libvpx/vp9/common/vp9_reconintra.c', + 'libvpx/vp9/common/vp9_rtcd.c', + 'libvpx/vp9/common/vp9_scale.c', + 'libvpx/vp9/common/vp9_scan.c', + 'libvpx/vp9/common/vp9_seg_common.c', + 'libvpx/vp9/common/vp9_thread_common.c', + 'libvpx/vp9/common/vp9_tile_common.c', + 'libvpx/vp9/decoder/vp9_decodeframe.c', + 'libvpx/vp9/decoder/vp9_decodemv.c', + 'libvpx/vp9/decoder/vp9_decoder.c', + 'libvpx/vp9/decoder/vp9_detokenize.c', + 'libvpx/vp9/decoder/vp9_dsubexp.c', + 'libvpx/vp9/decoder/vp9_job_queue.c', + 'libvpx/vp9/encoder/arm/neon/vp9_dct_neon.c', + 'libvpx/vp9/encoder/arm/neon/vp9_diamond_search_sad_neon.c', + 'libvpx/vp9/encoder/arm/neon/vp9_error_neon.c', + 'libvpx/vp9/encoder/arm/neon/vp9_error_sve.c', + 'libvpx/vp9/encoder/arm/neon/vp9_frame_scale_neon.c', + 'libvpx/vp9/encoder/arm/neon/vp9_quantize_neon.c', + 'libvpx/vp9/encoder/arm/neon/vp9_temporal_filter_neon.c', + 'libvpx/vp9/encoder/vp9_alt_ref_aq.c', + 'libvpx/vp9/encoder/vp9_aq_360.c', + 'libvpx/vp9/encoder/vp9_aq_complexity.c', + 'libvpx/vp9/encoder/vp9_aq_cyclicrefresh.c', + 'libvpx/vp9/encoder/vp9_aq_variance.c', + 'libvpx/vp9/encoder/vp9_bitstream.c', + 'libvpx/vp9/encoder/vp9_context_tree.c', + 'libvpx/vp9/encoder/vp9_cost.c', + 'libvpx/vp9/encoder/vp9_dct.c', + 'libvpx/vp9/encoder/vp9_encodeframe.c', + 'libvpx/vp9/encoder/vp9_encodemb.c', + 'libvpx/vp9/encoder/vp9_encodemv.c', + 'libvpx/vp9/encoder/vp9_encoder.c', + 'libvpx/vp9/encoder/vp9_ethread.c', + 'libvpx/vp9/encoder/vp9_ext_ratectrl.c', + 'libvpx/vp9/encoder/vp9_extend.c', + 'libvpx/vp9/encoder/vp9_firstpass.c', + 'libvpx/vp9/encoder/vp9_frame_scale.c', + 'libvpx/vp9/encoder/vp9_lookahead.c', + 'libvpx/vp9/encoder/vp9_mbgraph.c', + 'libvpx/vp9/encoder/vp9_mcomp.c', + 'libvpx/vp9/encoder/vp9_multi_thread.c', + 'libvpx/vp9/encoder/vp9_noise_estimate.c', + 'libvpx/vp9/encoder/vp9_picklpf.c', + 'libvpx/vp9/encoder/vp9_pickmode.c', + 'libvpx/vp9/encoder/vp9_quantize.c', + 'libvpx/vp9/encoder/vp9_ratectrl.c', + 'libvpx/vp9/encoder/vp9_rd.c', + 'libvpx/vp9/encoder/vp9_rdopt.c', + 'libvpx/vp9/encoder/vp9_resize.c', + 'libvpx/vp9/encoder/vp9_segmentation.c', + 'libvpx/vp9/encoder/vp9_skin_detection.c', + 'libvpx/vp9/encoder/vp9_speed_features.c', + 'libvpx/vp9/encoder/vp9_subexp.c', + 'libvpx/vp9/encoder/vp9_svc_layercontext.c', + 'libvpx/vp9/encoder/vp9_temporal_filter.c', + 'libvpx/vp9/encoder/vp9_tokenize.c', + 'libvpx/vp9/encoder/vp9_tpl_model.c', + 'libvpx/vp9/encoder/vp9_treewriter.c', + 'libvpx/vp9/vp9_cx_iface.c', + 'libvpx/vp9/vp9_dx_iface.c', + 'libvpx/vp9/vp9_iface_common.c', + 'libvpx/vpx/src/vpx_codec.c', + 'libvpx/vpx/src/vpx_decoder.c', + 'libvpx/vpx/src/vpx_encoder.c', + 'libvpx/vpx/src/vpx_image.c', + 'libvpx/vpx_dsp/arm/avg_neon.c', + 'libvpx/vpx_dsp/arm/avg_pred_neon.c', + 'libvpx/vpx_dsp/arm/fdct16x16_neon.c', + 'libvpx/vpx_dsp/arm/fdct32x32_neon.c', + 'libvpx/vpx_dsp/arm/fdct4x4_neon.c', + 'libvpx/vpx_dsp/arm/fdct8x8_neon.c', + 'libvpx/vpx_dsp/arm/fdct_partial_neon.c', + 'libvpx/vpx_dsp/arm/hadamard_neon.c', + 'libvpx/vpx_dsp/arm/idct16x16_1_add_neon.c', + 'libvpx/vpx_dsp/arm/idct16x16_add_neon.c', + 'libvpx/vpx_dsp/arm/idct32x32_135_add_neon.c', + 'libvpx/vpx_dsp/arm/idct32x32_1_add_neon.c', + 'libvpx/vpx_dsp/arm/idct32x32_34_add_neon.c', + 'libvpx/vpx_dsp/arm/idct32x32_add_neon.c', + 'libvpx/vpx_dsp/arm/idct4x4_1_add_neon.c', + 'libvpx/vpx_dsp/arm/idct4x4_add_neon.c', + 'libvpx/vpx_dsp/arm/idct8x8_1_add_neon.c', + 'libvpx/vpx_dsp/arm/idct8x8_add_neon.c', + 'libvpx/vpx_dsp/arm/intrapred_neon.c', + 'libvpx/vpx_dsp/arm/loopfilter_neon.c', + 'libvpx/vpx_dsp/arm/quantize_neon.c', + 'libvpx/vpx_dsp/arm/sad4d_neon.c', + 'libvpx/vpx_dsp/arm/sad4d_neon_dotprod.c', + 'libvpx/vpx_dsp/arm/sad_neon.c', + 'libvpx/vpx_dsp/arm/sad_neon_dotprod.c', + 'libvpx/vpx_dsp/arm/sse_neon.c', + 'libvpx/vpx_dsp/arm/sse_neon_dotprod.c', + 'libvpx/vpx_dsp/arm/subpel_variance_neon.c', + 'libvpx/vpx_dsp/arm/subtract_neon.c', + 'libvpx/vpx_dsp/arm/sum_squares_neon.c', + 'libvpx/vpx_dsp/arm/sum_squares_sve.c', + 'libvpx/vpx_dsp/arm/variance_neon.c', + 'libvpx/vpx_dsp/arm/variance_neon_dotprod.c', + 'libvpx/vpx_dsp/arm/vpx_convolve8_neon.c', + 'libvpx/vpx_dsp/arm/vpx_convolve8_neon_dotprod.c', + 'libvpx/vpx_dsp/arm/vpx_convolve8_neon_i8mm.c', + 'libvpx/vpx_dsp/arm/vpx_convolve_avg_neon.c', + 'libvpx/vpx_dsp/arm/vpx_convolve_copy_neon.c', + 'libvpx/vpx_dsp/arm/vpx_convolve_neon.c', + 'libvpx/vpx_dsp/arm/vpx_scaled_convolve8_neon.c', + 'libvpx/vpx_dsp/avg.c', + 'libvpx/vpx_dsp/bitreader.c', + 'libvpx/vpx_dsp/bitreader_buffer.c', + 'libvpx/vpx_dsp/bitwriter.c', + 'libvpx/vpx_dsp/bitwriter_buffer.c', + 'libvpx/vpx_dsp/fwd_txfm.c', + 'libvpx/vpx_dsp/intrapred.c', + 'libvpx/vpx_dsp/inv_txfm.c', + 'libvpx/vpx_dsp/loopfilter.c', + 'libvpx/vpx_dsp/prob.c', + 'libvpx/vpx_dsp/psnr.c', + 'libvpx/vpx_dsp/quantize.c', + 'libvpx/vpx_dsp/sad.c', + 'libvpx/vpx_dsp/skin_detection.c', + 'libvpx/vpx_dsp/sse.c', + 'libvpx/vpx_dsp/subtract.c', + 'libvpx/vpx_dsp/sum_squares.c', + 'libvpx/vpx_dsp/variance.c', + 'libvpx/vpx_dsp/vpx_convolve.c', + 'libvpx/vpx_dsp/vpx_dsp_rtcd.c', + 'libvpx/vpx_mem/vpx_mem.c', + 'libvpx/vpx_ports/aarch64_cpudetect.c', + 'libvpx/vpx_scale/generic/gen_scalers.c', + 'libvpx/vpx_scale/generic/vpx_scale.c', + 'libvpx/vpx_scale/generic/yv12config.c', + 'libvpx/vpx_scale/generic/yv12extend.c', + 'libvpx/vpx_scale/vpx_scale_rtcd.c', + 'libvpx/vpx_util/vpx_thread.c', + 'libvpx/vpx_util/vpx_write_yuv_frame.c', ], 'WIN_AARCH64_EXPORTS': [ 'libvpx/vpx/vp8.h',